If, then, oxygen atoms and hydrogen atoms combine at random to form three-atom molecules containing at least one of each, then the laws of probability state that in any number of such three-atom molecules, the most probable distribution of each variety is as follows:

  Having combined oxygen and hydrogen, we ought now to test theory by observation. Suppose we’re super-microscopically small and can take out, from the mass of final substance, ten individual molecules, at random, and inspect them closely.

  What are the chances that all ten happen to be H-O-H, without a single one of the other varieties present? The chances are 1 out of 6 x 6 x 6 x 6 x 6 x 6 x 6 x 6 x 6 x 6 or about 1 out of 60,000,000. (Work it out yourself, if you don’t believe me.)

  Suppose you picked out twenty molecules; what are the chances that all twenty are H-O-H? The answer is 1 out of 3,600,000,000,000,000.

  You are welcome to figure out the chances of picking out ten billion molecules at random and finding them all H-O-H. The chances are as infinitesimal as are those of manufacturing a nucleoprotein molecule by pure luck.

  And yet—If you pick out ten billion molecules of the product of hydrogen-oxygen combination, you will find that all of them are H-O-H. There are no O-O-H, O-H-O, or H-H-0 molecules included.

  What’s wrong then? Are the laws of probability in error? No. It’s the people who think they are using the laws of probability that are generally in error.

  I started off, you see, by assuming that any three-atom combination of hydrogen and oxygen atoms was equally probable. My entire argument was based on that. My exact words were: ‘If then, oxygen atoms and hydrogen atoms combine at random—’

  And that’s the point. We have no right to assume they combine at random, and, as a matter of fact, they don’t. The chemical properties of hydrogen and oxygen atoms are such that the combination H-O-H is the only one that has any reasonable probability at all, so it is the only combination formed.

  The same fallacy exists in the du Noüy type of argument. Sticky marbles can stick together any old way and form any old pattern but there is no guide to the behavior of atoms. Atoms, real atoms, can only form a limited number of combinations with one another, and of that limited number, some are more probable than others.

  So one does not and must not ask: what are the chances that a nucleoprotein molecule is built up through the blind workings of chance?

  One must ask: what are the chances that a nucleoprotein molecule is built up through the known laws of physics and chemistry—the very definitely unblind workings of chance?

  To consider the possibilities, let’s take the nucleoprotein molecule apart.

  It can be done easily enough. All the really complex molecules made by living tissue are polymeric in nature; that is, they are made up of simple units, or atom-combinations, that are repeated over and over in a chain. The units are called monomers. In some cases, as in starch or in cellulose, there is only one type of unit making up the molecule. In the case of nucleoproteins (or proteins in general) the units vary.

  In general, the large molecules of living tissue can be broken down to the smaller units that compose them by adding the atoms of a water molecule at the joints between the units. This is called hydrolysis. The units can recombine by splitting out the water molecules. This is called condensation.

  Under the proper conditions, large molecules can hydrolyze into smaller units, and smaller units can condense into large molecules, either way.

  For instance, the nucleoprotein of a living virus, can be hydrolyzed into two parts: one, the protein part, and the other a nucleic acid part. Neither part by itself is living nor has any of the infectious characteristics of the original virus. If the two parts are mixed together and allowed to remain so for a while, a certain amount of recombination takes place. Either the number of possible ways of recombining is not very great, or else the ‘correct’ way is more probable than others because by the ‘blind workings of chance’, fully one per cent of the recombinations proved to be the original virus once more with all its infectious characteristics. (This was an actual experiment and, in a way, it represents the manmade creation of life out of non-life.)

  Well, then, if it can be shown that the simple molecules, water, carbon dioxide, ammonia and so on, can form the units out of which nucleoproteins are built up by condensation, then a large step has been taken.

  What are the units which are involved? Without going into the chemistry, Table XXIII gives the names and some idea of the variety of these units.

  Of these, the phosphate group exists as such in the ocean. It is an inorganic grouping scarcely more complicated than ammonia or carbon dioxide, so we don’t have to worry about it at all. The pentoses, purines, pyrimidines and amino-acids are all moderately complicated, their molecules being made up of from 10 to, at most 30 atoms apiece. And they are good, stable compounds; nothing fancy.

  Let’s concentrate on the amino-acids. They are the most various of the groups and the most complicated, in some ways.

  Suppose we mix water, ammonia, carbon dioxide, methane, hydrogen sulfide, and hydrogen and sit down and wait for amino-acids to be formed.—Bring lunch with you because you’ll be waiting a long time. Amino-acids won’t be formed in a billion years or a trillion or a trillion trillion. Just mixing is not enough.

  You see, in general, complicated molecules have more energy content than simple molecules. For simple molecules to be built up into complicated ones, energy must be added.

  In other words, water will not run uphill unless it is pumped. A rock will fall upward only if thrown. A scattering of bricks will come together to form a house only if someone takes an interest.

  In turning water, ammonia, etc. into amino-acids, the chemicals are moving uphill and that won’t happen unless, somehow, they are made to do so. Or, to be more precise, energy is supplied.

  Does that mean we have to abandon the unblind workings of chance after all? Not if we can find a source of energy that just happens to be hanging around the primordial earth where all this is happening.

  And we can! In fact, we can find two sources.

  One source of energy sufficiently concentrated to force chemical reactions to take place that wouldn’t otherwise—is the lightning bolt.

  The lightning bolt is with us today and it works. Our modern atmosphere contains nitrogen and oxygen. Nitrogen and oxygen can combine to form nitrogen oxides, if a lot of energy is supplied. The energy of a burning match isn’t enough (luckily!) The energy of the lightning bolt is. During the instant of flash, a small amount of nitrogen and oxygen in the air immediately surrounding it are forced together to form nitrogen oxides. These dissolve in the rainwater to form nitric acid. When the nitric acid hits the soil, it combines with compounds existing there and forms nitrates.

  Now the amount of nitrogen oxides formed by an individual lightning flash is infinitesimal and the amount of nitric acid in rainwater wouldn’t hurt gossamer, but take it over the entire Earth and you have something. It has been estimated that about 250,000 tons of nitrates are formed by thunderstorms each day, and that this is a significant factor in maintaining soil fertility.

  All right, then, the primordial lightning had no nitrogen and oxygen gas to play with, but it did have molecules of ammonia, carbon dioxide, methane, hydrogen sulfide, hydrogen, and, of course, water vapor for playthings, and it slammed them together most energetically.

  In 1952, a chemist named Miller circulated a mixture of ammonia, methane, water, and hydrogen past an electric discharge for a week, trying to duplicate primordial conditions. At the end of the week, the mixture was analyzed by paper chromatography (see Chapter Four) and amino-acids were present in the mixture. They were not the product of life-forms; the system had been carefully sterilized. They were not there to begin with; that had been checked. They had been formed from simpler compounds and energy. To be sure, only two or three of the simplest amino-acids were detected, but then Miller had only waited a week and he had a good deal less than a whole atmosphere of gases to play with.

  You may wonder, though, if thunderstorms and lightning-bolts existed on the primordial Earth. It seems hard to believe they didn’t, but let’s suppose that they didn’t. Does that knock everything to pieces?

  It does not. There’s a second source of energy that no one can possibly deny existed—the ultra-violet radiation of the sun. Experiments have been conducted in which simple compounds have been subjected to ultra-violet radiation and more complicated compounds have been formed.

  To be sure, amino-acids have not yet been reported in the ultra-violet experiments, as far as I know. One of the reasons for that is that they haven’t yet included ammonia among the compounds being subjected to the energy, to my knowledge, and without the nitrogen of ammonia, amino-acids can’t be built. You can’t have cake without flour.

  In any case, the principle that ultra-violet will drive compounds uphill is definitely established.

  Picture, then, the primordial ocean, as simpler compounds are converted into more complicated compounds under the lash of ultra-violet and of lightning. Amino-acids, purines, pyrimidines, pentoses and many other types of compounds can be formed and as time passed they would thicken the ocean into a soup. As more and more of them were formed, they would collide with one another more and more frequently, and with energy spurring them on, they would frequently stick together.

  But mind you, they would not stick together in a random manner. There would always be a limited number of ways in which they could stick together, sometimes not more than two or three ways.

  For instance, a purine or pyrimidine could combine with a pentose and a phosphate in not more than six or eight likely ways to form what are called nucleotides.

  Two nucleotides could combine with one another in not more than three likely ways, or two of the simpler amino-acids could combine with one another in not more than two likely ways, to form double molecules.

  A double molecule may collide and combine with another nucleotide or amino-acid to form a triple molecule and so on. When enough of these units combine, the multiple amino-acids have become protein and the multiple nucleotides have become nucleic acid. And then, finally, the day will come when a nucleic acid molecule and a protein molecule will collide and stick together in such a way as to form a nucleoprotein—a nucleoprotein sufficiently complicated and properly constructed to be able to autoreproduce.

  And when that happens, we have life.

  The mark of those chance encounters exist in the proteins and nucleic acids of today. We have learned how to determine the order of amino-acids in the proteins and the order of nucleotides in nucleic acids. Where we have actually done so, the order appears quite random.

  Of course, you may wonder how amino-acids and nucleotides, put together at random, can turn out to serve the needs of life so neatly. It seems too much to ask of chance. There is an intellectual trap here; we tend to put the cart before the horse.

  There were all the oceans and up to a billion years as the space and time in which nucleoproteins (and other molecules) might form at random (within the limits, always, of the laws of physics and chemistry). All that space and all that time, multiplied a millionfold, would not suffice to make the formation of a particular nucleoprotein more than infinitesimally probable; that is, one with particular amino-acids and nucleotides arranged in a particular order.

  But if we are counting on the production of any old nucleoprotein with any old arrangement of parts, the time and space is more than sufficient. To be sure, every different order of parts makes for a final molecule with a different set of properties, but, so what? Whatever the final properties, those will be the raw materials of life. Some nucleoproteins might have properties that make for better survival? Those will survive.

  To suppose that the properties of the chemicals within living tissue are adapted to the needs of living tissue, rather than vice versa, is what 1 meant by putting the cart before the horse.

  It is as though we congratulated Nature on placing ears where she did on the human head, since that was just the right distance for the ear-pieces of spectacles to fit round. Or to be grateful that the rotation of Earth has been so designed as to last exactly 24 hours to the second, thus making a convenient whole number to work with. Or to wonder why the sun is wasted by having it shine in the daytime when it is light anyway, rather than in the night when it is dark and we could use a little light.

  But let’s move on. There are two final points to consider. Can life still be created out of non-life by natural processes on Earth today? Can we suppose that life may be created out of non-life on planets other than Earth?

  To answer the first question, there seem to be very good reasons indeed for doubting that the process can be repeated today.

  First, as life advanced to the stage where photosynthesis became possible and oxygen and nitrogen replaced the ammonia and carbon dioxide of the atmosphere, some of the oxygen was converted by the impinging ultra-violet into the more energetic ozone. (Ordinary oxygen molecules are made up of two oxygen atoms apiece; ozone molecules of three. Again ultra-violet light is converting the simple into the complex.)

  The ozone thus formed absorbs ultra-violet strongly, with interesting consequences. In today’s atmosphere, for instance, there is a layer of ozone fifteen miles up, formed by the ultra-violet impinging on the upper atmosphere. That layer absorbs ultra-violet and prevents it from reaching the surface of the Earth. A good thing, too, because modern life, not adapted to ultra-violet light, probably could not survive if the U-V came crashing through. Nevertheless, the rays of the sun that hit our modern oceans are comparatively weak and tame and much less efficient at producing complicated molecules out of simple ones.

  Again, the lightning bolt has only nitrogen, oxygen and water vapor to work on in our modern atmosphere and the nitric acid produced is not a stepping stone on the way to life. Missing are the large quantities of carbon atoms (in carbon dioxide) and hydrogen atoms (in ammonia) that were present in the primordial atmosphere. Without carbon and hydrogen, life as we know it cannot form though all Jove’s thunderbolts flashed at once.

  Does this sound unduly pessimistic? Are there no sources of life-yielding energy other than the sun and the storm? Is Nature so unresourceful as to yield no third possibility or am I so unimaginative as not to see one?

  Unfortunately, whether there are other sources of energy or not doesn’t matter. There is another difficulty of another type that puts the final quietus on present-day formation of life from non-life.

  The primordial ocean was a dead ocean. Large molecules could slowly be built up in peace and thicken in concentration till the oceans were practically nothing more than a nutrient broth. Nowadays, though, any organic molecule that happened to come into existence through some fortunate collision of simpler molecules would promptly be absorbed by some minute sea-creature and either broken down for energy or incorporated into living tissue.

  The modern ocean teems with life, and long before new life could possibly be formed, the raw materials out of which it might have been formed would be gobbled up voraciously by the life that already exists.

  Now what about other planets?

  Proposition 1: Given a planet at a distance from its sun such as to give it a temperature in the range where water is a liquid at least part of the time, then (barring exceedingly unusual characteristics of the interstellar stuff out of which the planet is formed—either in quantity or in the abundance of the elements) an ammonia-carbon dioxide atmosphere is inevitable.

  Proposition 2: Given an ammonia-carbon dioxide atmosphere and a source of energy such as the ultra-violet light from the sun, life is inevitable.

  It follows, then, if the line of deductions is reasonable, that life exists on any Earth-like planet. (Note, I say nothing about humanoid life, or even intelligent life. I say, simply—life. About anything beyond that, I make no predictions. Nor am I saying anything about anything resembling life which may exist on a completely different chemical basis from our own—non-nucleoprotein life, in other words—on such planets as Jupiter or Mercury.) Is there any way of checking this conclusion?

  There is one partial check we can make. We have a variety of worlds in the Solar System and among them is one world, other than Earth, which fulfils the conditions set above—just barely. That world is Mars. (Venus might be another, but we know practically nothing about it.)

  Mars is almost too small to suit, but it manages to retain just a bit of atmosphere and water. It is almost too cold to suit, but water just manages to be liquid part of the time. It is almost too far from the sun to suit, but it picks up some ultra-violet from the sun (less than half of what the primordial Earth did).

  So Mars is a severe test of our line of reason. A cold, nearly dry, nearly airless world——We could excuse our-selves if it failed.

  But let’s see, is there life on Mars?

  Despite all the odds against it, despite the poorness of the planet, the answer seems to be: possibly, yes. At least, the green areas on Mars seem to signify some kind of vegetation. The vegetation might be very primitive and undiversified, nothing like the teeming life of Earth, but it would be life.

  And if Mars can do it, then it is my belief that any Earth-like planet can do it.

  NOTE

  Since July, 1956, when this article was first written, chemists have continued the kind of work Miller initiated and have shown in great detail the manner in which even some quite complicated chemicals can be formed without the intervention of cellular life, although they have come nowhere near life itself even yet.

  Tiny traces of organic materials of the kind pointing toward life have been found in meteorites: and these show signs of having been formed without the intervention of living cells. In the 1970s, astronomers detected simple organic molecules in the dust-clouds of outer space. Apparently, the kind of molecules that seem to point toward life form whenever they are given the slightest chance to do so. (For more details see my essay entitled “The Inevitability of Life” which appears in my book Of Matters Great And Small.)