Suppose some scientists built a large vat, sterilised it and filled it with distilled water. Then they start dissolving compounds into the water such as amino acids or what ever they deem conducive to life. They can also control temperature, radiation and other environmental factors.

They periodically capture a sample of the water and examine the results.

Would they be able to inadvertently create self replicating molecules and the precursors to life?

Would this kind of experiment be worth attempting?

  • 1
    $\begingroup$ There are groups attempting this. [Check out this (paywalled) paper from 2010]( (liebertpub.com/doi/abs/10.1089/ast.2009.0402)) that tries to see if life could form from such a primordial soup on a simulation of Titan. Plenty of others too! $\endgroup$
    – James
    Commented May 12, 2018 at 11:30

2 Answers 2


That question was asked in 1953 by a classic experiment by Miller and his mentor Urey. You can find it on Wikipedia under "Miller_Urey_experiment" Basically they zapped basic molecular building blocks to life with electricity and more complex molecules (amino acids) formed. The experiment, with updates is nicely discussed here. The experiment showed that the molecules required for life could be formed spontaneously on Earth from lightning. You would need amino acids, but also bipolar lipids, which can spontaneously form compartments. However, attempting to create life this way would not be worth attempting for two reasons, relating to time and scale. First (time), life has evolved only one time on Earth over 4 billion years. Second the scale of the entire Earth over billions of years is a pretty big experimental "test tube". It happened once in all this time, over all that space. A small side-issue; the fact that the codon table is optimized (and pretty universal, with only small differences), such that common amino acids have more ways to make them (more codons), and codons close to stop codons are for rare amino acids. This indicates that some early life experimented with alternative suboptimal codes that didn't make it.

  • $\begingroup$ Earlier description in Goethe’s Faust: “Es wird ein Mensch gemacht”. Publication lacks experimental details though. $\endgroup$
    – David
    Commented May 6, 2018 at 21:06
  • $\begingroup$ A much more modern and updated version of the experiment can be found here, wired.com/2009/05/ribonucleotides got all the way to RNA. After all codons are meaningless without nucleic acids. $\endgroup$
    – John
    Commented Jul 19, 2018 at 14:43

Suppose some scientists built a large vat, sterilised it and filled it with distilled water. Then they start dissolving compounds into the water such as amino acids or what ever they deem conducive to life. They can also control temperature, radiation and other environmental factors.

Your direct question is whether this experiment would be worth attempting, and as other answers have said, it has been attempted many times.

I'll address the question implied in the title "Creating ideal conditions for life".

You talk of dissolving compounds into water and controlling temperature, radiation "and other environmental factors"; I assume you mean this to imply all the possible variables are being manipulated, but that is not the case. "Being in a large vat" is also an environmental factor, and there is good reason to think that large vats don't provide ideal conditions to develop life at all.

If that sounds silly, consider all the chemical reactions that could be happening in your vat. Chemical reactions are typically reversible: molecules A and B can meet and interact and form molecules C and D, and molecules C and D can in turn meet and interact and form molecules A and B. In a large vat, nothing is preventing all those molecules from meeting and interacting with each other; this leads to an equilibrium, depending on the rates of both reactions, and you'll end up with a stable quantity of A, B, C and D.

Now imagine instead you have two vats, with a system between them that transfers molecule C from vat 1 to vat 2. Now molecules A and B continue meeting and forming molecules C and D, but as that happens C gets transferred from vat 1 to 2, so you get a buildup of molecule D in vat 1 while A and B gradually run out because they keep forming C and D but not being reformed in turn, while in vat 2 you have a surplus of molecule C, so any molecule D that is formed from the interaction of A and B is likely to meet a C to re-form A and B, meaning you'll end up with more molecules A, B and C and very little D.

In other words, you got very different results from the case where everything was in a single vat. The macroscopic and microscopic environment your molecules are reacting in is a variable you cannot ignore.

As for large vats not being conducive to developing life, this 2010 article by Nick Lane, John Allen and William Martin summarizes those issues with the "large vat" experiments, and the "primordial soup" hypothesis in general:


(full article downloadable here: http://nick-lane.net/publications/luca-make-living-chemiosmosis-origin-life/)

Setting aside the absence of geochemical evidence that a primordial soup ever existed, there are grave difficulties with the soup theory. To give a single example, polymerisation into RNA requires both energy and high concentrations of ribonucleotides. There is no obvious source of energy in a primordial soup. Ionizing UV radiation inherently destroys as much as it creates. If UV was the primordial source of energy, why does no life today synthesise ATP from UV radiation? Worse, every time an RNA molecule replicates itself, the nucleotide concentration falls, unless nucleotides are replenished at an equal rate. UV radiation is an unlikely energy source for rapid polymerisation and replication, and an unpromising initiator of natural selection.

The reason that nucleotides and other organic molecules are reluctant to react further in a soup is that they are at thermodynamic equilibrium. They have already reacted, and the homogeneous soup has no internal free energy that would allow them to react further. Life is not just about replication; it is also a coupling of chemical reactions – exergonic ones that release energy and endergonic ones that utilise it, preventing the dissipation of energy as heat.

That article was in 2010, and those people have been investigating the hypothesis that life began in hydrothermal vents under specific thermodynamic conditions since; Nick Lane recently published the book The Vital Question on the subject, that I highly recommend to anyone interested in how life might have begun.

Their hypothesis for how life formed also implies very different kinds of experiments to try and reproduce it. Putting chemicals in a large vat is one thing, trying to recreate the conditions of a hydrothermal vent is another.

Here is a video of Nick Lane presenting their research; at 30:15 he briefly talks about where they're at experimentally, with a picture of their setup.


ETA: Stealing John's comment here: Creating ideal conditions for life

The experiments described in the article that yielded RNA are also more complex than just dumping molecules in a vat, and the cycles of heating and evaporation described are the kind of energy input and mechanisms to force disequilibrium that can make it thermodynamically possible for those molecules to form. This highlights another variable that can matter, which is variations in the environment over time as well as over space.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .