National Geographic

Scientists finish a 53-year-old classic experiment on the origins of life

In 1958, a young scientist called Stanley Miller electrified a mixture of simple gases, designed to mimic the atmosphere of our primordial lifeless planet. It was a sequel to one of the most evocative experiments in history, one that Miller himself had carried five years earlier. But for some reason, he never finished his follow-up. He dutifully collected his samples and stored them in vials but, whether for ill health or dissatisfaction, he never analysed them.

The vials languished in obscurity, sitting unopened in a cardboard box in Miller’s office. But possessed by the meticulousness of a scientist, he never threw them away. In 1999, the vials changed owners. Miller had suffered a stroke and bequeathed his old equipment, archives and notebooks to Jeffrey Bada, one of his former students. Bada only twigged to the historical treasures that he had inherited in 2007. “Inside, were all these tiny glass vials carefully labeled, with page numbers referring Stanley’s laboratory notes. I was dumbstruck. We were looking at history,” he said in a New York Times interview.

By then, Miller was completely incapacitated. He died of heart failure shortly after, but his legacy continues. Bada’s own student Eric Parker has finally analysed Miller’s samples using modern technology and published the results, completing an experiment that began 53 years earlier.

Miller conducted his original 1953 experiment as a graduate student, working with his mentor Harold Urey. It was one of the first to tackle the seemingly insurmountable question of how life began. In their laboratory, the pair tried to recreate the conditions on early lifeless Earth, with an atmosphere full of simple gases and laced with lightning storms. They filled a flask with water, methane, ammonia and hydrogen and sent sparks of electricity through them.

The result, both literally and figuratively, was lightning in a bottle. When Miller looked at the samples from the flask, he found five different amino acids – the building blocks of proteins and essential components of life.

The relevance of these results to the origins of life is debatable, but there’s no denying their influence. They kicked off an entire field of research, graced the cover of Time magazine and made a celebrity of Miller. Nick Lane beautifully describes the reaction to the experiment in his book, Life Ascending: “Miller electrified a simple mixture of gases, and the basic building blocks of life all congealed out of the mix. It was as if they were waiting to be bidden into existence. Suddenly the origin of life looked easy.”

Over the next decade, Miller repeated his original experiment with several twists. He injected hot steam into the electrified chamber to simulate an erupting volcano, another mainstay of our primordial planet. The samples from this experiment were among the unexamined vials that Bada inherited. In 2008, Bada’s student Adam Johnson showed that the vials contained a wider range of amino acids than Miller had originally reported in 1953.

Miller also tweaked the gases in his electrified flasks. He tried the experiment again with two newcomers – hydrogen sulphide and carbon dioxide – joining ammonia and methane. It would be all too easy to repeat the same experiment now. But Parker and Bada wanted to look at the original samples that Miller had himself collected, if only for their “considerable historical interest”.

Using modern techniques, around a billion times more sensitive than those Miller would have used, Parker identified 23 different amino acids in the vials, far more than the five that Miller had originally described. Seven of these contained sulphur, which is either a first for science or old news, depending on how you look at it. Other scientists have since produced sulphurous amino acids in similar experiments, including Carl Sagan. But unbeknownst to all of them, Miller had beaten them to it by several years. He had even scooped himself – it took him till 1972 to publish results where he produced sulphur amino acids!

The amino acids in Miller’s vials all come in an equal mix of two forms, each the mirror image of the other. You only see that in laboratory reactions – in nature, amino acids come almost entirely in one version. As such, Parker, like Miller before him, was sure that the amino acids hadn’t come from a contaminating source, like a stray bacterium that had crept into the vials.

Imagine then, a young and violent planet, wracked with exploding volcanoes, noxious gases and lightning strikes. These ingredients combined to brew a “primordial soup”, fashioning the precursors of life in pools of water. On top of that, meteorites raining down from space could have added to the accumulating molecules. After all, Parker found that the amino acid cocktail in Miller’s samples is very similar to that found on the Murchison meteorite, which landed in Australia in 1969.

These are powerful images, so why aren’t people more excited? Echoing many sources I spoke to, Jim Kasting, who studies the evolution of Earth’s atmosphere, said, “I am underwhelmed by it.” The main problem with the study is that Miller was probably wrong about the conditions on early Earth.

By analysing ancient rocks, scientists have since found that Earth was never particularly teeming in hydrogen-rich gases like methane, hydrogen sulphide or hydrogen itself. If you repeat Miller’s experiment with a more realistic mixture – heavy in carbon dioxide and nitrogen, with just trace amounts of other gases – you’d have a hard time finding amino acids in the resulting brew.

Parker accepts the problem, but he suggests that a few specific places on the planet may have had the right conditions. Exploding volcanoes, for example, throw up masses of sulphurous compounds, as well as methane and ammonia. These gases, belched forth into lightning storms, could have produced amino acids that rained out and gathered in tidal pools. But Kasting still isn’t convinced. “Even then the reduced gases would not be as concentrated as they are in this experiment.”

Even if our young planet had the right conditions to produce amino acids, that’s a less impressive feat than it appeared in the 1950s. “Amino acids are old hat and are a million miles from life,” says Nick Lane. Indeed, as Miller’s experiments showed, it’s not difficult to create amino acids. The far bigger challenge is to create nucleic acids – the building blocks of molecules like RNA and DNA. The origin of life lies in the origin of these “replicators”, molecules that can make copies of themselves. Lane says, “Even if you can make amino acids (and nucleic acids) under soup conditions, it has little if any bearing on the origin of life.”

The problem is that replicators don’t spontaneously emerge from a mixture of their building blocks, just as you wouldn’t hope to build a car by throwing some parts into a swimming pool. Nucleic acids are innately “shy”. They need to be strong-armed into forming more complex molecules, and it’s unlikely that the odd bolt of lightning would have been enough. The molecules must have been concentrated in the same place, with a constant supply of energy and catalysts to speed things up. “Without that lot, life will never get started, and a soup can’t provide much if any of that,” says Lane.

Deep-sea vents are a better location for the origins of life. Deep under the ocean’s surface, these rocky chimneys spew out superheated water and hydrogen-rich gases. Their rocky structures contain a labyrinth of small compartments that could have concentrated life’s building blocks into dense crowds, and minerals that would have catalysed their get-togethers. Far away from visions of languid soups, these churning environments are the current best guess for the site of life’s hatchery.

So Miller’s iconic experiment, and its now-completed follow-ups, probably won’t lay out the first steps of life. As Adam Rutherford, who is writing a book on the origin of life, says, “It’s really a historical piece, like finding that Darwin had described a Velociraptor in one of his notebooks.”

If anything, the analysis of Miller’s vials is a testament to the value of meticulous scientific work. Here was a man who prepared his samples so cleanly, who recorded his notes so thoroughly, and who stored everything so carefully, that his contemporaries could pick up where he left off five decades later.

Reference: Parker, Cleaves, Dworkin, Glavin, Callahan, Aubrey, Lazcano & Bada. 2011. Primordial synthesis of amines and amino acids in a 1958 Miller H2S-rich spark discharge experiment. PNAS http://dx.doi.org/10.1073/pnas.1019191108

Photos by Carlos Gutierrez and Marco Fulle

More on origins:

There are 28 Comments. Add Yours.

  1. steve
    March 21, 2011

    Thank you for fleshing out this story in a completely readable and understandable manner. Great work, as always.

  2. Adam Rutherford
    March 21, 2011

    Hola, fab stuff this, as it really helped redefne the study of the origin of life, and fleshed out an experiment that Oparin and Haldane (yes, that Haldane) has speculated on. I was lucky enough to film with the senior author on this paper Jeffrey Bada, who was one of Miller’s students, and is the curator of the kit.

    http://bit.ly/fabnhU about 20 mins in I get to do some analysis on the original samples too, though the kit itself is off limits as it has ungrounded 10kV running through it.

  3. tideliar
    March 21, 2011

    Excellent post Ed

  4. Chris Lindsay
    March 21, 2011

    Instead of focusing on replicating what we believe to be the conditions of an early Earth, wouldn’t just making self-replicators from amino acids in ANY condition be a major discovery?

  5. Adam Rutherford
    March 21, 2011

    Chris, this is the top down approach you describe, and lo, it has been done. Jerry Joyce’s group at Scripps in San Diego produced a ribozyme (an RNA enzyme) which catalyses it’s own replication.

  6. stonemason89
    March 21, 2011

    Indeed Chris; and who knows what they’d find if they added other elements to the mix, like say boron? They’d discover a whole new family of potentially useful compounds, for sure!

  7. drcharles
    March 22, 2011

    Enjoyed reading this post. The more we know it seems the less we can truly understand, which fills me with a reverence for existence.

  8. Philosophacles
    March 22, 2011

    This experiment is outdated. The conditions for an early earth were discovered to contain entirely different gases. Amino acids are still light years away from life anyways.

  9. amphiox
    March 22, 2011

    re #8 Philosophacles;

    Err, that’s exactly what the last half of the post says.

  10. Craig
    March 22, 2011

    Aww, this is a bit of a shame, I really liked the idea of the Miller-Urey experiments. Never mind – better informed now!

    I know that there are some polymers that assemble and disassemble (actin?) and so essentially ‘move’ if both processes occur at the same rate. Are these the kinds of processes that require a constant source of energy?

  11. Curious Wavefunction
    March 22, 2011

    If sulfur from volcanoes could contribute to sulfur-containing amino acids and OOL, why not metal sulfides from hydrothermal vents? This is especially interesting because for most of his life Miller was quite hostile to the “ventrists”. The present discovery could make that fact quite ironic.

  12. Robert S-R
    March 22, 2011

    Deep sea vents, huh? Funny how an ecosystem never thought to exist fifty years ago becomes the prime suspect for the origin of life. Any guesses how life migrated up to the surface? The deep ocean seems like a significant barrier to migration for unicellular life.

  13. ohwilleke
    March 22, 2011

    “These are powerful images, so why aren’t people more excited?”

    It is harder to get excited about things that you have long assumed to be true. Only when you have another scenario in mind that is ruled out is it profound.

    Also, given how far back we can trace the tree of life, I do think that saying “Amino acids are old hat and are a million miles from life,” as Nick Lane does, really does understate the importance of this step. We have the steps written pretty much from the simplest viruses to humans. We have the steps written from nucleosynthesis in stars (with a fair understanding of the rules of nature that make that so) to amino acids. We have even made considerable gains in understanding DNA and cells at a biochemical level.

    There is a gap from amino acids to simple viruses, but is it really all that vast?

  14. Stolen Dormouse
    March 22, 2011

    Somewhere on YouTube is a film that Julia Child made with Dr. Urey for the National Science Foundation, titled something like “Julia Child makes Primordial Soup.” It is hilarious, since she is her usual self.

  15. Wang-Lo
    March 22, 2011

    @Robert S-R: “The deep ocean seems like a significant barrier to migration for unicellular life.”

    I agree. But not, perhaps, to the convection of simple self-replicators and other building blocks.

    -Wang-Lo.

  16. Douglas Watts
    March 22, 2011

    Excellent post, Ed. Thank you. The original Miller-Urey experiments have great historical value because of their audacity and simplicity. People who say ‘amino acids are millions of miles from life’ are creating a strawman. We all know this; but you can’t have life without amino acids. That’s the value of basic research. You learn something, and hopefully, it inspires you to learn more.

    Any guesses how life migrated up to the surface?

    How does it now? Anyway, at some point chloroplasts evolved in phytoplankton. If Tom Gold’s idea of a deep biosphere is true then you would not be surprised to have chemo-bacterial colonies in vents close enough to the ocean surface to allow for chloroplasts to be a useful adaptation.

  17. amphiox
    March 22, 2011

    @Robert S-R, @Wang-Lo;

    We need not assume that the vents on early earth have to be “deep” ocean vents. It is certainly possible that some of these vents were in relatively shallow water. Neither should the deep sea necessarily pose that great a barrier, given sufficient time. There is a steady gradient of depth and pressures between deep seas and shallow seas, and currents mixing the waters between them. Unicellular life would have had plenty of time to spread and gradually adapt to the changing environment, from the seafloor on up, one pressure and depth threshold at a time.

  18. amphiox
    March 23, 2011

    To add to my #17, currents that would have swept unicellular organisms up from the depths to shallower waters, or laterally away from the heat and resources of the vents, would have been routine. And counter currents that would have had a tendency of carrying these same cells back to the haven of the vents would also have existed. So there would have been selective pressure promoting cells with increasing abilities to survive, for increasing periods of time, away from the vents. This situation is really all that would be needed to eventually produce cells that could survive indefinitely independent of the vents, in both deep and shallow waters (not necessarily both for all cells, but at least some cells that would be able to survive at both sorts of depths). And from there on the colonization of the rest of the world’s oceans would just be a matter of cell division, diffusion, and currents.

  19. Ed Yong
    March 23, 2011

    @Douglas Watts – Sure, and no one I spoke to actually slated the Miller experiments. Regardless of whether they actually have much bearing on the origins of life, they really captured the imagination and basically launched an entire field of study. That counts for a lot. But Lane’s point (which he elaborates more thoroughly in his book) is that the primordial soup discussions that resulted from Miller’s experiments ignore the thermodynamic aspects of the origin of life. It’s not simply enough to create building blocks – you need to give them an energetic nudge.

  20. David Kroll
    March 23, 2011

    Indeed, we need to be critical but before we become flippant or dismissive, do remember this: amino acids are the primary building blocks in de novo nucleic acid biosynthesis. Energetics aside, recall that all of the nitrogens in purines are derived from amino acids: aspartic acid, glycine, and glutamine (the remainder of the rings are from N10-formyl-tetrahydrofolate and CO2). For pyrimidines, aspartic acid is again used with the other nitrogen coming from the ATP-dependent fusion of bicarbonate and ammonia into carbamoyl phosphate.

    Yes, yes, many steps away from a non-enzymatic synthesis. But all too often my students (and colleagues) are misled that amino acids cannot be the precursors of nucleic acids.

  21. AmoebaMike
    March 23, 2011

    @ Ed Yong,

    Are there respectable theories that life may have independently come about more than once? That is, the Archaebacteria, Eubacteria, and Eukaryotes are each separated from one another in the tree of life, but shouldn’t there be a common ancestor? My understanding is that Archaebacteria are possibly the common ancestor, but also, possibly not.

    Then again, if all three groupings have things in common, they don’t necessarily have to be related. The whole analogous/homologous thing where bees and bats developed structures (wings) for flight independent from one another. Whereas the bat’s wings are way more similar to our (humans) arms, which aren’t used for flying–or really locomotion at all.

  22. Ed Yong
    March 24, 2011

    @AmoebaMike – I like the “ring of life” model where archaea and bacteria shared a common ancestor and eukaryotes are a fusion of the two – an archaea that swallowed a bacterium, which then became a mitochondrion. See here

    Regardless, it’s very unlikely that any of the three domains evolved independently. The universal genetic code very strongly suggests a common origin.

  23. jaron
    March 25, 2011

    “The amino acids in Miller’s vials all come in an equal mix of two forms, each the mirror image of the other. You only see that in laboratory reactions – in nature, amino acids come almost entirely in one version.”-par.10

    Within this matter of fact statement of the characteristics of laboratory reactions can be found a fundamental problem with the conclusions often drawn from the Miller experiments. It is as impossible to produce amino acids of all one type with random electrical charges as it would be to take a deck of well-shuffled cards and deal a hand of all one color off the top.

    Even before tackling the problem of self-replicating DNA, an experiment successfully demonstrating the origins of life would have to identify the environment and conditions necessary to allow random chance to produce the unilateral array of amino acids.

    This is not merely a technicality involving the nature of laboratory research, but one of the most important factors in an accurate analysis of the data procured through these experiments.

  24. Kristy
    March 26, 2011

    Amazing..bunch of bottles left to sit on a shelf since 1953 and we get the answers today in 2011.
    I wonder who else has similar old things given to them by their mentor. Could be more answers and stuff to write about too!

  25. Michael Meadon
    March 26, 2011

    Superb writing, and a fascinating story.

  26. AmoebaMike
    March 26, 2011

    @Ed Yong, Thanks for the reference post. Yes I agree that it’s very unlikely that more than one domain evolved independently. I’m just not sure it’s out of the realm of possibilities.

    Some think life may have been brought to Earth, so in that case it would be interesting if the same building blocks that lead to life could have come about independently under the same conditions, but in two different environments. And then when brought together, there was some intermingling of genes.

    Really, this is just thinking out loud. I do appreciate the perspective though.

  27. Dan
    March 26, 2011

    Nicely written story. I was wondering, isn’t this something very similar to the published research someone else conducted long time ago?

    http://www.springerlink.com/content/qx55749700j61652/

    With Melvin Calvin’s group:
    http://nobelprize.org/nobel_prizes/chemistry/laureates/1961/calvin-bio.html

  28. Bill Crofut
    March 19, 2012

    Re: “…Jim Kasting…said, “I am underwhelmed by it.” The main problem with the study is that Miller was probably wrong about the conditions on early Earth.”

    Prof. Kasting’s assessment of the claims for the early-Earth atmosphere is far less critical (and over two decades later) than that of a pair of geologists:

    “Geological evidence often presented in favor of an early anoxic atmosphere is both contentious and ambiguous. The features that should be present in the geological record had there been such an atmosphere seem to be missing….Ever since the work of Oparin…and the success of the experiments conducted by Miller…the dogma has arisen that Earth’s early atmosphere was anoxic, probably highly reducing…Conjecture and speculation, based on a knowledge of the chemistry of living matter, gave to them the composition of their starting materials, and it would have been surprising if they had not achieved the results they did.”
    [Harry Clemmey and Nick Badham. 1982. Oxygen in the Precambrian Atmosphere: An Evaluation of the geological Evidence. GEOLOGY, March, p. 141]

Add Your Comments

All fields required.

Related Posts