New study finds oxygenic photosynthesis 3.5 billion years ago

A new study published in Nature Geoscience by Masamichi Hoashi at Kagoshima University, and coworkers, proposes that oxygenic photosynthesis may have been going on 3.5 billion years ago… by far the oldest implication of cyanobacterial O2 production ever posited. The new study bases its conclusions on hematite (Fe2O3) crystals obtained from marine sediments laid down during the early Archean Eon, about 3.46 billion years ago.

The new study examines layered sediments from Archean rocks found in Australia, and proposes that seafloor hot spring activity delivered iron to ancient sea water as dissolved ferrous ions… which also would have been the dominant form of dissolved iron in the oceans back then, before free oxygen was present in the atmosphere. According to the authors, ferrous iron oxidized on the spot as it was emitted from hot springs, by oxygen gas permeating the water. Where would such oxygen gas come from? According to Hoashi et al., it came from oxygenic photosynthesizers – microbes capable of using sunlight energy for growth by taking in water and converting it to waste oxygen. The only problem with this model is that oxygenic photosynthesis probably didn’t appear on Earth prior to about 2.7 billion years ago. Widespread cyanobacteria didn’t start adding measurable amounts of O2 to Earth’s air until about 2.2 billion years ago… more than a billion years after Hoashi’s rocks were laid down.

What else could have oxidized all that ferrous iron, 3.5 billion years ago? Sunlight itself. Cyanobacteria use sunlight to work some complicated chemistry that results in O2 production, and O2 can rust iron. But sunlight can do the job directly, too. Ultraviolet light from the Sun can kick electrons from dissolved ferrous ions in water, forcing them to oxidize to ferric iron. Ferric is highly insoluble, and in Earth’s ancient oceans any ferric iron that formed would have sedimented out almost instantly as ferric oxides… which rapidly recrystallize to hematite in bottom sediments.

On the ancient Earth, an atmosphere without O2 meant a stratosphere without ozone, so the UV flux to the Earth’s surface would have been harsher than today. Surface ocean water – containing 100s of parts per million dissolved ferrous iron – irradiated by solar UV would have produced a constant slow trickle of ferric oxide rust, which would have drifted down through the water column to collect on the sea floor. In spots on the Earth’s sea surface where extra ferrous was coming in, say from sea floor hydrothermal activity below, hematite production at the surface would have been concomittantly more substantive. So why did the authors skip photolysis and jump straight to cyanobacteria?

Depth. The authors noted that sedimentary layers in their cores lacked any evidence of wave action or ripple marks that might occur in shallow sand and mud. Based on that – and apparently on that alone – the authors concluded that ferrous iron was belching into ancient seas filled with cyanobacteria, whose waste O2 also filled ancient sea water and quickly oxidized the ferrous to hematite.

Not so fast. I have serious doubts about this article, primarily because the authors appear to have leapt over a few logical steps in their rush to arrive at a flashy headline. If one stipulates that the authors’ observations are accurate, and no shallow-water sedimentary structures are apparent in their cores, that still doesn’t mean the sediments were laid down in deep water. On Earth today there are many marine settings where wave action is minimal and where waters are calm… such as behind barrier islands and sand bars, or within sheltered estuaries. There are simply too many options for sediments to avoid ripple marks, to justify a claim that the absense of such marks necessarily connotes depth.

In addition, even if the sediments derived from deep water there’s no reason to assume hematite had to form right there. Ferrous-iron photolysis occurs only in shallow water where UV can penetrate – typically in the upper few meters of water – but have you ever noticed what happens to ink dropped into water? It mixes. Ferrous iron delivered into 200 m deep ocean water by hot springs will naturally circulate with the water, mixing through the water column by advection and convection… until it passes through the upper few meters and gets oxidized by UV rays, at which point it forms ferric oxide solids and begins to sink back to the bottom. Hematite can build up in bottom sediment even if it gets oxidized in the photic zone, hundreds of meters above, simply through the action of gravity… with no anomalous, time-traveling cyanobacteria needed as an explanation.

This new paper by Hoashi et al. is an interesting study that provides some highly intriguing data… but only a weak argument that cyanobacteria evolved 800 million years earlier than we thought. In fact, if I were one of the peer-reviewers on the manuscript, I’d probably have sent this one back for revision, based on the objections I describe above. I don’t think the authors have made their case.


~ by Planetologist on March 22, 2009.

19 Responses to “New study finds oxygenic photosynthesis 3.5 billion years ago”

  1. I’ve been reading your paper (2002) Haas J. R. and DiChristina T.J., and it brought up some interesting thoughts.

    I already mentioned the possibility of a photosynthetic cycle based on iron (rather than sulfur, or oxygen as today), but reading the paper I realized that the energy yield (per electron) from ferrous oxidation should be about 1/10th of that from oxygen (if I got my numbers right). This, in turn, means that the energy costs of photosynthesis (using it as a hydrogen source) would be lower than sulfur.

    It seems reasonable to me that a complete cycle of photosynthesis and oxidative decomposition might have been present before the appearance of oxygen photosynthesis. An interesting point regarding use of ferric iron as an oxidant is that, because it is normally present as an insoluble solid, an amoeboid life form would have a great advantage in using it. I can envision an ocean floor with falling detritus (containing reduced carbon) and hematite, with an ecology of primitive Eukaryotes (pre-mitochondrial) living on it.

    Of course, that might date Eukaryotes as far back as 3.46 GYA (assuming a photosynthetic source for the hematite), but AFAIK there is no evidence that they aren’t that old, only absence of evidence.

    • Yep, all these other metabolisms were invented long ago by the Eubacteria and Archaea. There are bacteria that use sunlight to oxidize Fe(II) to Fe(III), and others that oxidize sulfide to elemental S, and they occupy very deep positions on the universal tree of life. Eukaryotes came along fairly late, and we’re limited to only a few metabolisms… fermentation and aerobic respiration, mainly, along with oxygenic photosynthesis (and that’s accomplished by chloroplasts). All the other metabolisms, including sulfate reduction, sulfur oxidation, iron reduction, iron oxidation, denitrification, nitrification, methanogenesis, blah blah blah… all those are denied to poor, one-trick-pony Eukaryotes. Our advantage is meta-structural, mostly… multicellularity gives us physical complexity, but the bacteria and Archaea simply own us when it comes to basic metabolic diversity.

      Methanogenesis, photosynthetic iron oxidation, and photosynthetic sulfide oxidation to S are probably in the first set of main actors on Earth in terms of metabolism. Also somewhere early on it’s likely sulfate reduction and Fe(III) reduction came along, using the waste products of the other metabolisms, or maybe they got going with abiotic photolysis products. The sequence isn’t clear.

      • Thanks, Planetologist, I hadn’t known iron photosynthesis was currently present on Earth.

        Eukaryotes came along fairly late, and we’re limited to only a few metabolisms… fermentation and aerobic respiration, mainly, along with oxygenic photosynthesis (and that’s accomplished by chloroplasts). All the other metabolisms, including sulfate reduction, sulfur oxidation, iron reduction, iron oxidation, denitrification, nitrification, methanogenesis, blah blah blah… all those are denied to poor, one-trick-pony Eukaryotes.

        We should probably specify mitochondrial Eukaryotes here. I know the current fad is that Eukaryotes rose in one “big bang” from some sort of fusion of a Eubacterium with one or more Archaea, but there doesn’t seem to be consensus on the subject, and IMO it’s much more plausible that modern (mitochondrial) Eukaryotes rose from endosymbiosis of a Eubacterium (related to modern purple bacteria) in an amitochondrial Eukaryote that was otherwise similar to mitochondrial Eukaryotes.

        The fact that endosymbiosis of one prokaryote within another has been observed doesn’t invalidate the theory that mitochondria arose from an amoeboid amitochondrial Eukaryote establishing a Eubacterial lineage as endosymbiote(s). It just means there may be other options. AFAIK the current consensus is that all modern Eukaryotes derive from this endosymbiotic lineage (however it arose), including those lineages that lack fully functional mitochondria. But this doesn’t mean that there wasn’t a full ecology of pre-mitochondrial Eukaryotes, potentially much older, of which all became extinct except for those derived from the mitochondrial lineage. Among these may have been lineages that reduced iron III, only distantly related to those that later incorporated Eubacteria into mitochondria.

        For that matter, I’m not convinced by the notion that the earliest protein-using (ie modern) life looked like modern prokaryotes. While the development of the nucleus may have come later, I suspect that the endoplasmic reticulum and much of the system of vesicles was present prior to the invention of proteins. My favorite guess is that the common ancestor had a double membrane (like most prokaryotes) with a lumen contiguous with that of the ER. A cellular skeleton made from polyglycans was contained within this lumen, which is topologically identical to the typical structure of Gram-negative Eubacteria today, assuming the ER component to have degenerated to nothing. (Of course, in the modern case it’s mostly glycoproteins; indeed one possible origin for the entire protein system is that it began as a system for controlling the manufacture of polyglycans.)

        From that ancestral condition, I would speculate two developmental paths: that leading to Eukaryotes brought the ER entirely into the cytoplasm, sacrificing the outer membrane in the interest of endocytosis, and replacing the old endoplasmic skeleton with a modern reusable system based on proteins (actin). Indeed, the membranous actin-binding protein Ponticulin may be derived from an older structure that still used part(s) of the original polyglycan skeleton.

        The other developmental path (IMO) was essentially degenerating (in a structural sense), and was taken independently by Eubacteria and Archaea (possibly many times). The strong structural similarity among prokaryotes would thus be a matter of convergent evolution. However, many or most of the early enzyme developments may have taken place prior to this degeneration. Or, perhaps, they took place in more structurally complex relatives and moved by lateral gene transfer to the more simplified relatives.

        AFAIK there is no evidence against this theory, rather a prejudice among most observers that evolution must involve an increase in structural complexity. IMO recent work in complexity theory renders this notion invalid. Not to mention that, even if such complexity had to evolve (rather than being present prior to the appearance of life), it could easily have evolved prior to the invention of proteins.

      • A couple of corrections, before I get into the later points. Eukaryotes are defined by a number of attributes, among which are the presence of a nucleus enclosing their DNA, and the optional presence of mitochondria and chloroplasts. There are a few Eukaryotes that lack mitochondria, but in such cases those strains are limited to fermentation-based metabolism. Mitochondria provide the capacity to undertake aerobic respiration, of which their host cells are incapable without mitochondria. Chloroplasts, of course, undertake oxygenic photosynthesis and are not found in animals (except for a few weird examples, like some nudibranchs that steal chloroplasts from algae).

        The case for mitochondria and chloroplasts originating as endosymbiotized Eubacteria that originally were free-living is pretty much closed, at this point. There’s really no controversy about this, mainly because once genetic sequencing became available it was discovered (by Carl Woese, at the University of Illinois at Champagne-Urbana) that the ribosomal RNA of both “organelles” correspond almost exactly with those of free-living aerobic bacteria (mitochondria) and cyanobacteria (chloroplasts). The DNA inside both has been pared down over the ages, so that neither organelle is really capable of sustained life outside their hosts, but their RNA is essentially intact and recognizably similar to their ancestral free-living cell lines. A lot of this work was pioneered by Lynn Margulis, who originated the endosymbiosis hypothesis in the 1960s, but the confirming genetic evidence had to wait for the invention of sequencing technology. There’s a nice online summary of this – including a phylogenetic tree of life based on rRNA sequencing, written by Norman Pace, at Berkeley. I use a similar phylogenetic map when I teach sections on the evolution of microbes through Earth’s history.

        But you’re probably right about Eukaryotes having needed a few pre-adaptations before they could endosymbiotize other bacteria. For one thing, they had to embiggen ($0.25 goes to Phil Plait, there) themselves. Proto-Eukaryotes had to develop structural polymers that enabled larger cell sizes, such that other bacteria could physically fit inside them. On Earth, that invention apparently went first to some fermenting microbes, who eventually managed to corral mitochondria into breathing for them, and later some of those corralled cyanobacteria into photosynthesizing for them, too. Or maybe the endosymbiotes invaded like parasites, instead of being engulfed by their hosts. The order there isn’t clear.

      • I’ve tried three times to leave a response here (this thread, I’ve posted at other threads). This is a test, without any links or much size.

      • For one thing, they had to embiggen ($0.25 goes to Phil Plait, there) themselves. Proto-Eukaryotes had to develop structural polymers that enabled larger cell sizes, such that other bacteria could physically fit inside them.

        While researching one of my next blog posts, I just discovered Why the Phosphotransferase System of Escherichia coli Escapes Diffusion Limitation by Christof Francke, Pieter W. Postma, Hans V. Westerhoff, Joke G. Blom, and Mark A. Peletier.

        Based on what I read, Proto-Eukaryotes would have had to evolve a whole new signalling system, as well.

  2. This post, and the comments that followed, were fascinating.

    I would so go back to school and become a planetologist if I weren’t so bad at the maths.

    I guess I have to stick to writing sci fi instead.

    • I’m terrible at maths. I scored adequately at best in college, in all the math courses I took. At that time, that weakness pissed me off, so I declared and suffered through a math minor just so the damn subject couldn’t be said to have defeated me. Silly thinking, but what else can one do at that age?

  3. Two comments still awaiting moderation? OK, here’s one more. No early Cyanobacteria? You wrote…

    “…a weak argument that cyanobacteria evolved 800 million years earlier than we thought.” “The only problem with this model is that oxygenic photosynthesis probably didn’t appear on Earth prior to about 2.7 billion years ago. Widespread cyanobacteria didn’t start adding measurable amounts of O2 to Earth’s air until about 2.2 billion years ago… more than a billion years after Hoashi’s rocks were laid down.”

    Not so fast. Without oxygenic photosynthesis there is no PLAUSIBLE alternative to explain the organic carbon that is present across a wide variety of facies and in a variety of environments in older rocks. The rocks studied in this paper contain organic carbon. Where did it come from? The 3.75 Byr Isua banded iron formations contain several percent isotopically-light organic carbon, some of which has “fossils” that look like it was produced by planktonic organisms. Like these rocks, the iron facies of Isua also display negative or neutral Ce-anomalies, positive Eu-anomalies, but no positive Ce-anomlies…expected if oxidation was UV-induced. Before you say that it was all done by non-oxygenic photosynthetic bacteria please remember that an ocean filled with dissolved ferrous iron cannot contain the dissolved sulfide that those bugs use.. sulfide that would have to be present world-wide…and for a billion years?

    “On the ancient Earth, an atmosphere without O2 meant a stratosphere without ozone, so the UV flux to the Earth’s surface would have been harsher than today.”

    Harsher indeed! And this is an argument for an oxygen-free world? The early Sun was putting out more UV than it does today. Without ozone the entire globe was a giant “ozone hole”. Light-requiring photosynthesis is supposed to evolve and proliferate under those conditions? Talk about a weak argument…

    “Surface ocean water – containing 100s of parts per million dissolved ferrous iron – irradiated by solar UV would have produced a constant slow trickle of ferric oxide rust…”

    Sounds simple, doesn’t it? But those who have speculated quantitatively about this theory come up with GLOBAL amounts of iron that are enormous. 10^16 gm Fe per year? That is absurdly large to be acceptable. It is equal to the entire global sedimentation rate! The world oceans would be overrun with this “slow trickle”. And, as I said earlier, not a single mol of this could be used as a “sink” for free oxygen.

    • Wow… quite a bit of scene-chewing we have here.

      First, it is not a disputed issue that photosynthesis is accomplished today by bacteria that use older enzymatic processes than oxygenic photosynthesis… such as photosynthetic iron and sulfur oxidizers. Those clades predate cyanobacteria, and appear to have contributed genes to the later evolution of cyanobacteria. Are you claiming those bacteria don’t exist? Because I can show you where to go see them in the field.

      Second, an ocean filled with Fe(II) can, actually, co-exist in a world where H2S emissions from volcanic activity go on. If you bothered to do the calculations, you’d see that in aqueous solution you can have a significant amount of dissolved Fe(II) and H2S without deposition of pyrrhotite (or precursers, e.g. greigite, makinawite, etc..). In fact, in aqueous solution at 25 C and 1 bar, at a pH of 4 representing a mildly acidic Archean ocean, you can dissolve up to a millimolar Fe(II) and H2S and FeS(sub 1-x) is still undersaturated. This remains true at a pH of 8. It’s easy to jump to conclusions regarding solubility, but it’s more appropriate to actually do the calculations first.

      Third, enhanced UVB does not penetrate very far into the photic zone. Early photosynthesizers, such as associated with Archean stromatolites and the bacterial mats thought to have been responsible for the Isua graphite layers, only needed to reside at a depth of a few meters below the sea surface to escape the UVB flux. In addition, modern research on the evolution of oxygenic photosynthesis shows that the genes that encode for handling peroxide in photosystem II probably evolved to cope with peroxide present in the water column from atmospheric photolysis of H2O vapor. Bacteria coped, and we have the evidence.

      Fourth, I wasn’t talking about BIFs when I mentioned photolysis of Fe(II) to explain the kind of small-scale iron deposits found at 3.5 Ga. Global deposition of BIFs came much later… and was probably microbially produced, anyway.

      Also, you might want to try the decaf.

      • A few combined replies to several of your reactions:

        The detrital uraninite/pyrite situation has always been a source of debate because it is not just the oxidation, it is a combination of several factors, not the least of which is time of exposure during transport, particle size and, yes, even pO2. Kinetics. Stream-borne uraninites exist today at 2i% O2 and pyrites wash around in between ripples in shallow waters. Both minerals seem to do fine in museum cases. Could easily have existed earlier at 1-2% of PAL O2. BTW, some of the so-called detrital uraninites are actually debris flows. Not “convincing” evidence.

        Once again, ANY explanation for the oxidation of iron in BIFs that does NOT use molecular O2 simply makes it harder, if not impossible, for the computer modelers to keep their model atmospheres anoxic. They need all the Fe2+ they can find to act as their “sink” for oxygen, 4 mol Fe per mol O2. This, after all, was one of the original reasons why geologists thought the early atmosphere was anoxic… the BIFs took up all the O2… they rusted. And then, voila… oxygen began to rise.

        Microbial explanations also suffer from the fact that the oxide facies of the BIFs contain precious little Corg. No microfossils. It is the siderite (FeCO3) facies that contain the Corg…and with it the isotopic signal characteristic of photosynthesis. Gallionella is a microaerophile. Indeed, many of the microorganisms that are primitive (e.g. Aquificales) are also microaerophiles, many using elemental sulfur. Elemental sulfur is unstable under reducing conditions and is reduced abiologically in seawater at thermophile temperatures. These Aquifex-Hydrogenobacter so-called “knallgas microbes” use trace oxygen (0.5%) but are all more primitive than Cyanobacteria. During their evolution this oxygen came from stratospheric photodissociation of water vapor (as you mentioned too).

        Alternate explanation such as photolysis? I thought I dealt with that nutty photolysis idea. It won’t work for a whole host of reasons that I mentioned. Pardon the pun, but that explanation is a red herring.

        Wow, sulfide in excess of Fe2+ and for billions of years? I’ll have to re-read Walker and Brimblecome (1985) and Kump and Seyfried (2005). Not sure they agree with all that but maybe they didn’t do the appropriate calculations? How then did PS-II ever get started with all that sulfide around? Cyanobacteria placed into micromolar amounts of sulfide switch off the PS-II and switch back to PS-I. PS-II is repressed! With all your sulfide everywhere why bother to evolve PS-II?

        It is true that UVB doesn’t penetrate that deeply into the photic zone but that’s because it is absorbed by the dissolved organic carbon in many waters. Clear ocean water does not absorb UV well. But, you suggest those bugs at Isua were happily photosynthesizing with sulfide and PS-I while the UV was oxidizing the Fe2+ around them? Not exactly a compelling explanation.

        Bacteria coping with photodissociated oxygen? Now you’re making sense. But, that’s not support for an early anoxygenic world…rather the reverse. Look at the evolution of the thermophiles. Look at RuBisCo. Look at the evolution of the superoxide dismutases… the Fe, Mn enzymes occur in obligate anaerobes and photosynthetic bacteria which, like the Aquificales, are more primitive than the Cyanobacteria. Makes sense in an early atmosphere with a low but important level of free O2. Doesn’t make much sense otherwise.

        I’ll go try some decaf if you’ll stop pretending your a skeptic. On this topic you sound more like a man protecting the forces of ignorance by following the paradigm. Get skeptical!

      • I am skeptical. I’m so skeptical, in fact, that I’ve done a bit of literature searching. It appears that this argument has been going on for a while… at least to you. Not really to anyone else, so far as I can find. It appears that you’ve been publishing quite a bit on what appears to be your pet theory that the early Archean was oxidizing… mostly in recent years as letters to the editor, comment papers, and so forth, but without a lot of new ideas. I also checked on the papers that did cite your earlier work crusading for early O2 – back into the late 1970s, it appears – and they provide an interesting pattern. Citations of your papers, such as in Frimmel (2004), Sreenivas and Murakami (2005), Vieira-Silva and Rocha (2008), do so to point out your position on the topic as a minority view that isn’t supported by the bulk of evidence. I didn’t find a lot of citations, of any kind, on your more recent op-ed pieces, recent being from the late 1990s on. Perhaps people just moved on with better ideas.

        I’ll dispense with this, and move on. It’s clear that you hold this position on the rise of oxygen occurring in the early Archean rather… well, let us say firmly. Despite a lot of work being done by a lot of other people in a lot of journals that appear to contradict your summary opinion. Based on my own survey of the primary literature over the years, and my involvement with geochemical research topics that touch on questions of exobiology, the origins of life, and Earth’s history, I simply disagree with your assessment. There is a lot of data out there supporting an anoxic early Archean, but in my survey of your papers I don’t find a lot of data contradicting that work. In fact I don’t find any data at all, only opinions based on anomaly-hunting, which even after you’ve been contradicted in the literature you appear to maintain steadfastly. That is interesting. It’s a pattern that raises a red flag. Still, that’s not my reason for disagreeing, even as the alarm bells ring.

        You’ve thrown out a lot of points to argue over, in quick succession, jumping from topic to topic in as few words as possible. That’s another red flag… sometimes it’s called a Gish Gallup. You may or may not be familiar with the term. Anyway, I’ll direct the reader to the articles I’ve mentioned above for a more thorough discussion of the facts. I don’t really have time to deal with a full-out crusade on this one point. Yes, I know you do… but please do it elsewhere. This is my blog, and I’m only interested in science, not dogma. Well… except to pour derision on dogma, because that’s always fun.

        So, where to begin? The detrital uraninite/pyrite debate actually seems pretty settled, in favor of the minerals having been deposited not hydrothermally but at the surface, in the weather, which at the time was anoxic. Sulfur isotope data back this up… something I noticed you failed to mention. It’s really hard to fool sulfur isotopes. Anyway, yes, today sometimes pyrite washes up on the beach… and you get to see it for a brief time, until it oxidizes away. Pyrite in mine tailings oxidizes right away, too… mostly through the action of Thiobacillus ferooxidans. The result is acid mine drainage. Pyrite doesn’t hang out today in riverbeds for very long, and certainly not long enough to form babbling brooks of shiny fool’s gold cobbles under every foot. And then there’s the uraninite… which lasts not even that long at the surface as pyrite. I know this, because I work with UO2 in my lab, and I have to go to great lengths to keep it from oxidizing once water hits it.

        And, once again right back at you, BIFs probably did form in the presence of O2 during the late Archean… that’s the prevailing model, in fact. There are also iron oxide deposits in the early Archean, but they’re small and unusual, and probably represent isolated settings where the conditions were right for enough Fe(II) photolysis and sediment to make it happen… as oasis environments surrounded by a world of rocks that directly contradict the idea of a lot of O2 in the bulk atmosphere. Perhaps cyanobacteria were already limping along on peroxide at the time, here and there. And perhaps not… they could have a deep genetic lineage as niche critters in highly restricted habitats, but that will need to be determined with more evidence than the lack of ripple marks in a few rock samples.

        BIFs don’t have a lot of organic C or microfossils, but neither do many other rocks from that time. Neither do most stromatolites, which were directly precipitated as sand on mats of organic slime. Most stromatolites don’t have microfossils, either. That’s a red herring… sort of like that other point you like to bring up, about there couldn’t have been enough Fe(II) from hydrothermal activity to fill the oceans, when no one argues that in the first place. Continental weathering at the time would have been sending…. wait for it…. dissolved Fe(II) from weathering of primary igneous olivine, pyroxenes, amphiboles and micas from exposed Fe-bearing rocks in a reducing atmosphere.

        You didn’t deal with the idea of photolysis, you simply dismissed it. Of course the bulk ocean surface wasn’t cranking.. no pun intended… out Fe(III) left and right due to photolysis alone back then. But in unusual geographic environments, perhaps shallow restricted estuaries, as at Shark Bay today, things could become ideal for Fe(III) photo-oxidation…. a process that’s a lot easier at high chloride activities, as it happens, and those are conditions you can get in a restricted basin environment quite easily. I think it’s interesting, though, that you try to support your minority opinion by calling opposing opinions “nutty”. Red flag.

        Actually, the calculations I did are fairly straightforward, and yes, as it turns out you can dissolve Fe(II) and sulfide in water up to the respective solubility limit of FeS(1-x). That’s what is actually meant by a solubility limit, by the way. And how did PS II get started in the presence of H2S? Why, maybe it didn’t… maybe it was like everyone else has been saying all along, that H2S was a limiting nutrient at the time, that it wasn’t erupting everywhere. Where it did, H2S oxidizers got a foothold, but at the fringes of those regions there could have been all sorts of cyanobacterial pre-adaptations getting sorted out. That’s sort of the point of them evolving the capacity to use H2O instead of H2S, freeing them from such restricted settings. You declare a false dichotomy: H2S atmosphere versus O2 atmosphere. H2S even today doesn’t mix evenly through the atmosphere, because it’s heavy. Try driving to a salt marsh, where it stinks of microbial H2S… and a kilometer away it doesn’t, because the gas doesn’t travel that far without a strong wind.
        False dichotomy to support argument: another red flag.

        Ocean water absorbs UV weaker than it does the red end, but it still attenuates below just a few meters. It attenuates to where bacteria can handle it, mostly using weird little adaptations called pigments. Pigments that evolved before photosynthesis co-opted them. Pigments that had no reason to be there in a world without eyes, unless they served some adaptive advantage… dare I suggest protection from UV? That would seem the most parsimonious explanation to me. Also, you seem to overlook the fact that light is a finite incoming resource… and if part of the incoming UV is absorbed in the atmosphere to make peroxide, and in the upper photic zone to partially oxidize metals or dissociate organics, there’s less left over to bombard bacterial mats living deeper. That’s more or less how O3 saves us today, by using up the UV before it hits us way down here.

        I’ll stop “pretending” to be a skeptic when you stop pretending to be unbiased. Sorry, but when someone publishes op-eds and opinion articles and commentaries on the same thing for 30 years without any new data, and gets cited mainly by people contradicting those ideas, but yet keeps repeating those bad ideas, I start to detect bias. That’s the final red flag.

        Look, if more data comes in that supports all the evidence, and points more clearly to an oxic early Archean, I’ll blog about it here with amazement. I have no stake in this race. But the points you’re somehow finding the time to repeat here (although I suppose that for you this is just like posting an op-ed somewhere else, really) have been asked and answered. And the answer, so far, is “no”.

        Which will be my answer to further posts from you, repeating the same dogmas uncritically. Try some skepticism, yourself. But please try it somewhere else. I only have time for reality, here.

      • And you didn’t make any friends by getting upset that your comments weren’t posted instantaneously. Some of us still have jobs, you know.

  4. Don’t know what happened to my first comment, perhaps the moderator didn’t like it? But here’s one on another issue: the water depth.
    You wrote: “If one stipulates that the authors’ observations are accurate, and no shallow-water sedimentary structures are apparent in their cores, that still doesn’t mean the sediments were laid down in deep water. On Earth today there are many marine settings where wave action is minimal and where waters are calm… such as behind barrier islands and sand bars, or within sheltered estuaries. There are simply too many options for sediments to avoid ripple marks, to justify a claim that the absense of such marks necessarily connotes depth.”

    This assessment overlooks the fact that all of the shallow water locales that you mention are subject to periodic storms. The Hamersley iron formation (Dales Gorge member; oxide facies) goes for miles without significant disturbance and with no clastic input. Most specialists agree that it was deposited well below storm wavebase…300-500 meters.

    • Fair enough, the authors’ samples might very well have been from deep water. But whether they were or not, I still would like to see them examine all other possibilities – including alternate explanations for Fe(II) oxidation, such as photolysis – before leaping to the conclusion that cyanobacteria had to be around back then. That’s a stretch, and – as I said – constitutes extraordinary evidence that the authors have not provided. If such data becomes available, I’d consider it fascinating…. despite those fat checks the Anti-Oxygen Dogma Society sends me every month. 😛

  5. Several comments. Firstly, sedimentary hematite, especially euhedral or coarse-grained hematite, is never precipitated directly. It is a secondary dehydration product of diagenesis or metamorphism. The euhedral hematite crystals found by these authors are recrystallized from the initial colloidal hydrous ferric oxides that are the normal oxides precipitated in hydrothermal systems.
    Secondly, the hematite found in most of the banded iron formations is primary in the sense that it was not oxidized later in geologic history. These oxides were deposited at depth and far from the influence of clastic sediments…sands and clays. We know this because ferric oxyhydroxides precipitated in surface waters will adsorb the insoluble rare earth cerium-IV colloids and transport them to the bottom. This scavenging process creates a positive Ce-anomaly in the sediments. This is seen today in red clays in the deepest parts of the ocean. The ferric oxides in the BIFs, however, do not show this; they show negative (or neutral) Ce-anomalies (characteristic of oxidized seawater) and positive Eu-anomalies (characteristic of hydrothermal waters). This is what one sees today in marine hydrothermal systems.
    Thirdly, the idea of UV-induced oxidation of iron can be (and has been) dismissed for several reasons. (1) It would also generate positive Ce-anomalies, but even if it did not the oxides would not have positive Eu-anomalies. (2) It requires that the ferrous iron be present only in surface waters above the banded iron formation basins but not elsewhere. (3) It would work against the evolution of photosynthesis, especially in the shallow waters where the stromatolites flourished. (4) It would eliminate any sulfides that might be used by the non-oxygen producing photosynthetic bacteria, assuming any could survive the high UV itself. (5) It removes the necessary oxygen sink that is assumed to have sopped up the cyanobacterial oxygen to keep the atmosphere anoxic. Every mol of Fe2+ that is UV-oxidized is a mol that cannot also be oxidized by free oxygen.
    But this evidence is not the first to posit the presence of free oxygen very early in Earth history. There have been numerous papers published that presented evidence implying an early oxygenic atmosphere. They are routinely ignored by the dogmatists who insist that the early Earth was free of oxygen.

    • No, the moderator (me) was just busy this week… I just got reviews back on a paper, and am trying to get those done.

      Thanks for your comments, and I was with you more or less until the “dogmatists” thing. You still make some good points, though. I think the issue isn’t settled, regarding exactly when O2 began to built wholesale in the atmosphere. It does seem pretty obvious from the geochemical record that O2 was negligible prior to the late Archean, after which O2 levels probably built to maybe 1% or so… but that’s mostly an upper limit. But it becomes difficult to justify a view of O2 in the atmosphere as far back as 3.5 Ga, because at the same time there were apparently placer uraninite and pyrite deposits being laid down on the continents. With uraninite especially, you can’t have even a whiff of O2 in the air or you starting seeing uranyl minerals. There is a large literature on this, as I’m sure you are aware. And yes, I’ve done a lot of work with iron oxides / iron-reducing bacteria, so the news that hematite isn’t a primary mineral in most cases isn’t actually news to me. 😉

      Recent work does in fact show that BIFs probably didn’t form solely (or predominantly, at least) as a result of UVB photolysis, and were probably a product of microbial oxidation (Konhauser et al., 2007). That’s fine. I’m not arguing against BIFs… I’m arguing that the particular study mentioned in my original post does not provide convincing evidence of cyanobacteria at 3.5 Ga. Until more information is forthcoming from the researchers, I think my position is reasonable.

  6. Seem to me we’ve got a bit of chauvinism here, as well as with the 2.7GYear date: the assumption that the only major forms of organic photosynthesis possible are those we see today.

    AFAIK there’s no reason that photosynthesizing bacteria couldn’t have used ferrous ions has direct electron sources, creating ferric ions which would have precipitated sunk to the bottom. Even the banded iron formations could be the result of periodic growth cycles of this sort of life.

    Another likely electron source pre-oxygen is ammonia, yielding elemental nitrogen. I know there are recent arguments regarding widespread nitrogen on other plants (which I’m not qualified to evaluate), but arguments from elemental nitrogen from terrestrial vulcanism are completely invalid: the mantle has been circulating atmospheric nitrogen for at least 3 GYears, and oxygen for at least 400 MYears, it’s highly likely that reduced nitrogen was long ago oxidized to elemental.

    Something else you missed is that the prevailing opinion seems to be that the deeper ocean was anoxic until after the Cambrian, which sort of rules out deep oxidation of ferrous iron.

  7. I’m inclined to agree with your assessment. Where the possibilities you mentioned even speculated about in the study?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: