As a matter of fact, though, hydrogen and helium are nearly impossible to detect spectroscopically at planetary temperatures. (At solar temperatures, they’re very easy to detect, but that’s another matter.) It was only quite recently and by rather indirect means that the hydrogen-helium nature of Jupiter’s atmosphere was deduced. Before that, astronomers were much more aware of certain other components of Jupiter’s atmosphere which, while present only in comparatively small quantity, happen to have strong absorption bands that are easily observed spectroscopically. What are these impurities?

  Checking Tables XVI and XVIII, you might suppose that the chief impurities would be oxygen, nitrogen and neon in that order. You’d be right as far as neon is concerned. The chances are that it is present in Jupiter’s atmosphere in a concentration of something below one per cent. You’d be wrong about oxygen and nitrogen, though.

  Oxygen and nitrogen in the presence of a vast surplus of hydrogen would form compounds with the hydrogen, particularly under the pressure conditions in a large atmosphere. One atom of oxygen combines with two of hydrogen to form water (H2O). One atom of nitrogen combines with three of hydrogen to form ammonia (NH3). Water and ammonia would be more stable than oxygen and nitrogen themselves under hydrogen-helium atmosphere conditions.

  Similar statements can be made for most of the other common elements listed in Table XVI. Helium and neon are out of it. They combine with no other element under any condition. They exist in splendid isolation. The others form hydrogen compounds if they can. If they can’t, they form oxygen compounds, oxygen being the next most common compound-forming element.

  Thus one atom of carbon combines with four atoms of hydrogen to form methane (CH4). One atom of sulfur combines with two atoms of hydrogen to form hydrogen sulfide (H2S). Silicon, magnesium, and iron won’t combine with hydrogen. They combine with oxygen instead, forming silicon dioxide (SiO2), magnesium oxide (MgO) and ferric oxide (Fe2O3) respectively.

  Sulfur and carbon will combine with oxygen as well as with hydrogen. Oxygen is a lot less available than hydrogen, but both sulfur and carbon prefer oxygen to hydrogen by quite a bit, so a certain amount of sulfur dioxide (SO2) and carbon dioxide (CO2) would form (especially after much of the hydrogen has leaked away on smaller planets).

  In Table XIX are listed these common compounds and their boiling points in degrees absolute. (Incidentally, I should mention that the boiling points given in Tables XVIII and XIX are the values at Earth’s atmospheric pressure. The values vary with pressure, going up as the pressure does as far as what is called the critical point but no further. Let us use the ordinary values given in the tables to avoid complications. They will serve to compare one element or compound with another, and the line of argument would not be much affected by boiling point values that would take pressure into consideration.)

  Looking at Table XIX, we see that magnesium oxide, silicon dioxide and ferric oxide could never form part of an atmosphere under any planetary conditions. On Earth, in fact, these three compounds, plus aluminum oxide (which boils at 2320 degrees absolute) form at least 80 per cent of the solid crust of the Earth.

  Water would be a gas on Mercury, a volatile liquid on Venus and Earth (and on Mars at its warmest) but frozen solid and not volatile on the outer planets. Sulfur dioxide is in the same situation plus the fact that at Earth temperatures and below it tends to react with water to form an even less volatile compound.

  Ammonia is a gas as far out as Mars and remains fairly volatile as far out as Uranus. The same for hydrogen sulfide and carbon dioxide. Methane remains a gas on Jupiter (always neglecting the pressure effect) and would be volatile even on Pluto.

  As far as Jupiter is concerned then, the impurities in its atmosphere consist of ammonia, methane, carbon dioxide, neon and hydrogen sulfide; possibly in that order. Neon, like hydrogen and helium, is almost impossible to spot spectroscopically in the cold. Carbon dioxide and hydrogen sulfide are present in minor traces. That leaves ammonia and methane, and those are both easily detectable in Jupiter’s atmosphere.

  As one moves out from Jupiter, away from the Sun, to Saturn, Uranus and Neptune, the ammonia absorption bands get steadily weaker and the methane absorption bands steadily stronger. This is probably not due to any change in overall composition but only to the fact that as the temperature drops, ammonia becomes less and less volatile; there is less and less ammonia vapor in the atmosphere; and methane, which remains volatile all the way out, has less competition.

  We can summarize then by saying that large, moderately cold planets have hydrogen-helium atmospheres with ammonia as the chief impurity, while large, excessively cold planets have hydrogen-helium atmospheres with methane as the chief impurity.

  But so far we have talked only of large planets. What about small planets? What about the Earth?

  To begin with, the Earth is closer to the sun than are any of the large, outer planets and is therefore at a higher temperature. The molecules in its original atmosphere moved faster than those on Jupiter and its colder brethren. Either Earth could not collect the particularly nimble hydrogen and helium in the first place, or, having collected them, she could not hold them. In either case, Earth (and all the inner planets, for that matter) was built up out of the ‘impurities’ of the Universe—the elements other than hydrogen and helium.

  This accounts for the great differences between the inner and outer planets and explains why the inner planets are so much smaller and denser than the outer ones.

  Now one frequently thinks of the Earth, at its beginning, as a molten globe that slowly cooled down and solidified. If this were so, one would have to use most ingenious arguments to explain the persistence of any atmosphere at all.

  If, however, Earth were formed by gradual accretion of matter in a turbulent maelstrom of interstellar material rather than by way of a solar catastrophe, the original temperatures (of Earth’s outer crust, at least) might never have been startlingly higher at the beginning than now—say not above the boiling point of water.

  Let’s suppose that and see where it takes us.

  To begin with, let’s consider the atomic or molecular weights of the gases that are likely to occur in the Earth’s atmosphere originally. These are listed in Table XX. Remember—the smaller the atomic or molecular weight, the more likely Earth is to lose that particular gas.

  The gases listed in Table XX fall into three groups. The light gases, hydrogen and helium, leak away or are never collected. In either case they are not in Earth’s atmosphere except in minute traces. A second group, consisting of the heavy gases, hydrogen sulfide, carbon dioxide, and sulfur dioxide, would remain in the atmosphere even if Earth’s surface temperature were rather higher than it is today.

  The third group, methane, ammonia, water, and neon, requires more attention. At today’s temperature. Earth could hold them. If the temperature were higher by 50 degrees they would slip away slowly. Judging from the molecular weights: 16, 17, 18, and 20, they ought all to slip away at the same rate, just about. This is not so; other factors intervene.

  At a temperature of, say, 340 degrees absolute, water is still liquid and only a small portion of the substance is in the atmosphere as vapor and only that small portion is available for leakage. Methane and neon, on the other hand, are gases and are all available for leakage. Ammonia is in an intermediate position. It, too, is a gas, but it is a gas which is extremely soluble in water (while methane and neon are only very-slightly soluble in water). Much of the ammonia is safely tucked away in the oceans where it is safe from leakage.

  We can reason then that most of the methane and neon is lost; most of the water stays; and that ammonia is betwixt and between.

  We end up with a planet which has an atmosphere composed mainly of ammonia and carbon dioxide, with hydrogen sulfide and water vapor as minor impurities and with sulfur dioxide, methane, and neon present in traces.

  We can summarize, then, the only types of atmosphere-structures that may be expected in the Universe on the basis of atom abundances alone:

  (1) Large, excessively cold planets—Hydrogen/helium plus methane impurity (example, Neptune).

  (2) Large, moderately cold planets—Hydrogen/helium plus ammonia impurity (example, Jupiter).

  (3) Small, cool planets—Ammonia/carbon dioxide (example, early Earth).

  (4) Small, hot planets—No atmosphere (example, Mercury).

  (Note that I am omitting large, hot planets from consideration. No such thing is possible. Any planet close enough to a sun to be hot loses its hydrogen and helium and the elements that are left can only make a small planet.)

  But if the atmospheres listed above are the only ones to be expected, that leaves out precisely the one type of atmosphere most important to us—the nitrogen/oxygen atmosphere on Earth today. How did that come about?

  Well, the four cases listed above are those that may be expected on the basis of atom abundances alone. On Earth, a new factor enters in—the presence of life.

  Life, in general, exists by making use of the energy that can be evolved from chemical reactions among the substances in its vicinity. Several possible schemes for doing this exist among the life-forms of Earth. There are life-forms that take advantage of energy-forming reactions among sulfur compounds, iron compounds and nitrogen compounds. Such life-forms never evolved past the bacterial stage. The raw materials they use for energy are too specialized.

  The real success lay with those organisms that learned to extract energy from the most common substance on Earth—which happens to be water. (The lucky fellow who learns how to make delicious and nourishing soup out of sawdust is going to make a lot more money than one who learns how to make it out of peacock tongues.)

  One type of organism (ancestral to the green plant) learned how to make use of solar energy to break up the water molecule into hydrogen and oxygen. The hydrogen was used to convert carbon dioxide (the second most available substance on early Earth) into starch and in this way solar energy was stored as chemical energy to be tapped as needed. The oxygen from the water was a by-product, not needed, and so was released into the air.

  Observe that the net result, to the atmosphere, of this process (photosynthesis) is to consume carbon dioxide and release oxygen. As the green plants multiplied and spread through the oceans and invaded the land, carbon dioxide was used up and oxygen produced at an ever greater rate.

  There was the reverse tendency, too. When plant life died, the bacterial action involved in decay consumed oxygen and produced carbon dioxide. The development of animal life was also a factor in consuming oxygen and producing carbon dioxide. However, by the time equilibrium was established almost all the carbon dioxide was gone from the atmosphere (0.03 per cent of our modern atmosphere is carbon dioxide, no more). In its place was oxygen.

  In the presence of this vast surplus of the active element, oxygen, any methane present was slowly converted to carbon dioxide and water. The water joined the oceans and the carbon dioxide was replaced by more oxygen through plant action. Hydrogen sulfide was converted to water and sulfur dioxide.

  Finally, oxygen combined with the hydrogen atoms of the ammonia molecule to form water. The nitrogen atom of the ammonia molecule does not combine with oxygen except under drastic conditions and it went free to tie up in pairs as nitrogen molecules.

  The result was that by the time equilibrium was reached and photosynthesis had completed its work by changing the atmosphere, both the carbon dioxide and the ammonia were gone. In its place was nitrogen (from the ammonia) and what was left of the triumphant oxygen. And so a new type of atmosphere must be added to the others:

  (5) Small, cool planets, with life—Nitrogen/oxygen (example, modern Earth).

  There remains, of course, the possibility of intermediate situations. For instance, a large planet with the proper temperature might have methane and ammonia in approximately equal concentrations in its atmosphere, and have a hybrid atmosphere intermediate between cases (1) and (2). Saturn and Uranus might be examples of such.

  A planet of intermediate size and intermediate temperature, say one lying where the asteroid belt is now and somewhat smaller than Uranus in size, might lose most but not all of its hydrogen and helium and end up with an atmosphere in which hydrogen, helium, ammonia, methane, and carbon dioxide are ail present in respectable proportions. This would be a hybrid of atmospheres (1) and (3), of which there are no known examples.

  A planet considerably smaller than Earth or considerably warmer might lose most of its atmosphere but not quite all, retaining a wispy kind of air rich in carbon dioxide. This is a hybrid of atmospheres (3) and (4) and an example of that is Mars (complicated by the possible presence of plant life).

  Finally, a planet might be in the process of developing life, with some of the carbon dioxide and ammonia consumed and free oxygen and nitrogen appearing in the air. This is a hybrid of atmospheres (3) and (5) and there are no known examples.

  I have now covered, as far as I can tell, every type of atmosphere that there is any likelihood of encountering anywhere in the Universe.

  Any reasonable likelihood.

  Let us, however, throw off the shackles of probability and devote some attention to atmospheres that are, in the main, wildly improbable.

  Life depends, as I said, on the utilization of energy. The way this is handled on Earth, stripped to its bare essentials, is this:

  Plants, utilizing solar energy, split water to hydrogen and oxygen, storing the hydrogen (in the form of compounds) in their tissues. Animals (and plants, too, for that matter) make use of the chemical energy of the stored hydrogen. Animals eat food which consists of plant tissue or animal tissue derived from plant tissue and combine its hydrogen with the oxygen they breathe. In other words, we have a cyclic water/hydrogen-oxygen system. Plants push in one direction and animals in the other, the whole remaining in balance.

  Furthermore, one of the members of the system is a liquid present in sufficient quantities to form oceans and one of the others is a gas forming a major portion of the atmosphere. So let’s say that in order to have life-as-we-know-it, we need a cyclic system with one member a liquid and another a gas.

  What other systems are possible? Is there anything we can substitute for oxygen? Something which, like oxygen, will produce energy if combined with hydrogen and something which is a gas and which produces a liquid on combination with hydrogen.

  Well, to substitute for oxygen it has to be an active chemical and the only low-boiling elements that will bear comparison with oxygen as far as activity is concerned are sulfur, chlorine, fluorine and bromine. To give you an idea of the kind of pickle we’re in, Table XXI gives the atomic abundance of these substances in comparison with oxygen (on a silicon equal to 10,000 basis).

  From Table XXI, you can see at once how improbable it is that the atom distribution over sizable volumes of space should be so abnormal as to create planets in which sulfur, chlorine, fluorine or bromine are the major components of the atmosphere in the place of oxygen.

  But we’ll ignore that and just consider the cyclic systems that result. They are:

  (a) Hydrogen sulfide/hydrogen-sulfur

  (b) Hydrogen bromide/hydrogen-bromine

  (c) Hydrogen chloride/hydrogen-chlorine

  (d) Hydrogen fluoride/hydrogen-fluorine

  In Table XXII, some data are given on the components of these systems.

  If we take sulfur first, we can see from Table XXII (and Table XVII) that sulfur is a gas not even under extreme Mercurian conditions, and that at any temperature at which sulfur is gaseous, hydrogen sulfide is certainly gaseous. However, who says it is sulfur that has to be the gaseous component of the cycle? At any temperature between 393 and 718 (which covers the normal temperature range of Venus as well as Mercury) it is possible to have a hydrogen sulfide atmosphere and a liquid sulfur ocean.

  The same inversion holds true in the cases of bromine and chlorine. Neither a bromine nor chlorine atmosphere is admissible since in both cases there would be no liquid component of the cycle. Hydrogen bromide and hydrogen chloride would also be gaseous. But at a temperature range of 266 to 332 (Earth and Mars), one could have a hydrogen bromide atmosphere and oceans of liquid bromine; while at a temperature range of 188 to 239 (asteroid belt) one could have a hydrogen chloride atmosphere and oceans of liquid chlorine.

  In all three cases plants would have to breathe in hydrogen sulfide (or hydrogen bromide or hydrogen chloride), break it up to hydrogen and sulfur (or bromine or chlorine), store the hydrogen in their tissues and excrete liquid sulfur (or bromine or chlorine). Animals would have to eat the plants and drink the liquid sulfur (or bromine or chlorine), reform the hydrogen sulfide (or hydrogen bromide or hydrogen chloride) and belch it out periodically.

  This may sound complicated and unpalatable to you but the big drawback is that when hydrogen and chlorine combine they yield only one-third the energy that the combination of hydrogen and oxygen does. Hydrogen and bromine yield only one-eighth as much and hydrogen and sulfur only one-tenth as much. Life is such an energy-consuming thing that that alone should eliminate the bromine and sulfur system (at least for anything over the micro-organism stage) and make the chlorine system pretty shaky.

  Fluorine is another thing altogether. No inversion is necessary here. At temperatures between 190 and 293 (Mars), it is possible to have a fluorine atmosphere and a hydrogen fluoride ocean, and fluorine combines with hydrogen to yield 1½ times as much energy as the hydrogen-oxygen combination would produce. This seems the best bet (if we could only forget how rare fluorine is in the universe compared to oxygen).