It’s long been known that the atmosphere of the early Earth probably didn’t include the gasses used by Stanley Miller and Harold Urey in their famous Miller-Urey experiments. Conventional wisdom among origin-of-life theorists holds that when you try to conduct Miller-Urey type experiments with the actual gasses present on the early Earth, you don’t get amino acids.

Theorist David Deamer explains: “This optimistic picture began to change in the late 1970s, when it became increasingly clear that the early atmosphere was probably volcanic in origin and composition, composed largely of carbon dioxide and nitrogen rather than the mixture of reducing gases assumed by the Miller-Urey model. Carbon dioxide does not support the rich array of synthetic pathways leading to possible monomers…” (David W. Deamer, “The First Living Systems: a Bioenergetic Perspective,” Microbiology & Molecular Biology Reviews, Vol. 61:239 (1997).)

Likewise, the journal Science states:

The so-called Miller-Urey experiment simulated the prebiotic atmosphere by mixing molecules they presumed were present on the early Earth: methane, ammonia, hydrogen, and water. They then zapped this soup with an electrical charge to mimic lightning, which in turn produced small amounts of amino acids–the building blocks of proteins, which are critical to all living things. “[That study] had a tremendously important role in making chemists aware that the whole question of origin of life could be approached by lab experiments,” says NSCORT’s Arrhenius. “It became an acceptable field.”

Yet today, Arrhenius and many other researchers dismiss the experiment itself because they contend that the early atmosphere looked nothing like the Miller-Urey simulation. Basically, Miller and Urey relied on a “reducing” atmosphere, a condition in which molecules are fat with hydrogen atoms. As Miller showed later, he could not make organics in an “oxidizing” atmosphere.

(Jon Cohen, “Novel Center Seeks to Add Spark to Origins of Life,” Science, Vol. 270:1925-1926 (December 22, 1995) (emphasis added).)

Every once in a while, however, we hear resurrected claim that amino acids can be produced under realistic early-Earthlike conditions. These claims rarely pan out. But that hasn’t stopped Wikipedia from recounting them. On its page dedicated to the Miller-Urey experiment, Wikipedia states:

[S]ome evidence suggests that Earth’s original atmosphere might have had a different composition from the gas used in the Miller-Urey experiment. There is abundant evidence of major volcanic eruptions 4 billion years ago, which would have released carbon dioxide (CO₂), nitrogen (N₂), hydrogen sulfide (H₂S), and sulfur dioxide (SO₂) into the atmosphere. Experiments using these gases in addition to the ones in the original Miller-Urey experiment have produced more diverse molecule.

Wikipedia acknowledges that the atmosphere of the early Earth might have been very different from what Stanley Miller used in his experiments. This is good. But then the online encyclopedia makes a false claim that Miller-Urey type experiments using realistic atmospheres “have produced more diverse molecules.”I’m willing to believe this claim, if the evidence backs it up. And this is one of those times where Wikipedia actually provides a citation, so we can fact-check its claims.

The citation is to a 2006 article in New Scientist titled “Right-handed amino acids were left behind.” Does the article show that amino acids (or other molecules necessary to form life) were produced under conditions mimicking the early Earth? Not in the least.

The brief article in New Scientist is about an experiment, published in Nature in 2006, that was nothing like Miller-Urey. In fact, it wasn’t trying to explain how amino acids could have been produced on the early Earth. Instead, it was trying to solve another problem that origin of life theorists face — the chirality problem.

Virtually all known non-biological chemical processes produce equal amounts of right- and left-handed amino acids. But all life uses left-handed amino acids. A challenge for origin of life theorists is thus to explain how mixtures biased towards left-handed amino acids could have been produced without a biological organism to produce them. Hence the chirality problem.

The research cited by Wikipedia doesn’t even attempt to explain how amino acids could have been created on the early Earth by natural chemical processes. New Scientist explains what the experiment tried to do:

A clue to the mystery of how nature selected left-handed amino acids rather right-handed ones may lie in the way the substances behave as they dissolve in water. … Donna Blackmond at Imperial College London and colleagues dissolved a mixture of solid L and D versions of the amino acid serine in water. They found that a small difference in the initial proportion of one version gets amplified in the resulting solution.

So the experiment simply dissolved one particular amino acid (serine) in water. It found that if you start with a solution of serine molecules that’s already biased towards left-handed serine molecules, then that bias will be amplified and you’ll end up with an even greater bias. It’s interesting chemistry, which perhaps takes origin-of-life chemistry one step further towards solving the chirality problem for one particular amino acid. (Note: chirality is hardly the largest problem facing origin-of-life theorists.) But, as the New Scientist article states, this experiment assumes that you must start with a mixture that is already slightly biased towards left-handed amino acids: “This effect could have amplified a slight excess of L-amino acids in nature. Why there was a slight excess to start with is another question.” Why there were even amino acids to begin with is another question.

The experiment cited by Wikipedia did not intend to explain the existence of amino acids on the early Earth. Rather, the experiment assumes it. The experiment certainly does not relate in any way to Wikipedia’s claim that atmospheres of “carbon dioxide (CO₂), nitrogen (N₂), hydrogen sulfide (H₂S), and sulfur dioxide (SO₂)” have “produced more diverse molecules.”

So what we have here is Wikipedia making a false claim, backed up by a citation bluff, pure and simple. No matter how many times they’ve been debunk, some icons of evolution are just really hard to let go.

UPDATE: I e-mailed an ID-friendly biochemist about the research cited by the New Scientist article, and my correspdonent sent back the following response:

Donna Blackmond is really good at studying reactions, especially from a theory standpoint, and has made a name for herself in the “where-did-homochirality-come-from” area of chemistry. She benefits, of course, from the credulity of most scientists when it comes to OOL stuff; after all, we know it happened, right?! This paper makes perhaps slightly more headway towards the chirality problem than most, in that the results were somewhat surprising (at least to me). Here is how I would summarize it:

(A) There is an “artificial” reaction (the aldol, a type I think that occurs in biological systems, but this is a specific one with non-natural reactants) that is known to be catalyzed by the amino acid proline. If the proline is not racemic (racemic = 50:50 mixture of mirror images) then the product will have more of one mirror image than the other. Whether the enantiomer amounts vary linearly with the optical purity of the proline or not has been a point of dispute (she refers to this early on). There are one or two scenarios where theory says it doesn’t have to be linear. This is important to her because in nature the furthest from racemic you can hope for is not much. As far as I know, non-racemic AA are found only in carbonaceous meteorites, and the largest ratio I have read of is 57:43. So she needs a very non-linear catalysis to hope to get other molecules that are further from racemic. Interestingly, the amino acid she observes the best results with (serine) is entirely absent from meteorites (the Murchison meteorite anyway).

(B) In studying amino acids in water, she found that IF the amino acid (AA) was not racemic AND the concentration was high enough the not all of the AA was dissolved (some present as solid), then what was dissolved could have quite a bit more of one enantiomer than the solid did. I think you could think of this as being caused by the racemic material being much less soluble than one enantiomer, especially for serine (99+%) but high or significant for six others. I gather that these seven are all that she found worth reporting, as I imagine she studied all twenty of the common biogenic AAs. So to answer your question, it “works” best for serine but to a lesser extent for six others. Even if you take this work as a solution to the homochirality problem, this leaves 2/3 of the AA entirely unaccounted for.

(C) She suggests (and I think shows in one or two cases) that this could be a way to catalyze aldol reactions to produce products of higher enantiomer ratios than the initial enantiomer imbalance in the amino acids. This is the only angle on the homochirality problem this work represents. She may be content to leave the reader with the impression that this is a way to get homochiral amino acids, but she doesn’t try to go there because it would require some intelligent intervention in the form of isolating solutions from solids at appropriate places, and even then only for serine.

(D) Note that she doesn’t try to suggest that this is a mechanism for getting homochiral proteins, for several reasons. (a) AA coupling reactions hardly occur in any naturalistic scenario; all the ones I have seen (concentrated salt solutions, heat or carbonyl sulfide) all have serious problems, not to mention that without appropriate protecting groups an unreactive AA dimer forms (and I’m told these dimers also racemize easily). (b) Except perhaps for serine, her numbers are not high enough; there is too much D AA present. AA coupling reactions have never been found to have any significant selectivity for L-L over D-L formation, so if both L and D are present, they both will be incorporated (assuming you can achieve coupling at all). No one suggests that you can hope to get a living system with mixtures of L and D. (c) Her scenario does not have enough amino acids showing the enrichment of solution-over-solid; you can’t make any important proteins from the few she found this behavior with.

(E) The bottom line is that these scenarios do NOT represent what would happen by unguided chemical reactions on the early earth or anywhere else. They need specialized intervention from a chemist — i.e. an intelligent agent — to carefully manipulate the conditions to get them to work. This paper is kind of like Sutherland’s: one of the better pieces of work in the area but only convincing (or even impressive) for the already-convinced or the uninformed. Unfortunately, these kinds of papers leave non-specialists with a false impression that significant progress has been made. As you know, this is made much worse by a media that loves to promote these impressions.

Yes, I do know.