ATMOSPHERES OF THE PLANETS RUSSELL 165 



the water has gone into hydrated minerals ; but how this could hap- 

 pen, unless there was much less there originally than on earth is 

 hard to understand. 



For the major planets we have to consider the course of events in 

 a cooling mass containing an excess of the lighter elements and 

 especially of hydrogen. The condensation of the refractory con- 

 stituents should take j^lace much as for a smaller body. The princi- 

 pal constituents of the rocks, however — potassium, sodium, magne- 

 sium, aluminium, calcium, and silicon — are not reduced from their 

 oxides by hydrogen, and would form rocks not unlike those of the 

 earth. But at high temperatures the oxides of iron are reduced by 

 hydrogen. My colleague. Prof. H. S. Taylor, to whom I am greatly 

 indebted for counsel on these problems of physical chemistry, re- 

 marks that the drops of molten lava falling through a hydrogen 

 atmosphere reproduce pretty closely the conditions of a blast fur- 

 nace. We may conclude then that most of the iron would go into 

 the core and less into the rocky shell. 



After the shell solidifies, the remainder of the mass will remain 

 fluid over a wide range of temperature. Its principal elementary 

 constituents will be hydrogen, helium, oxygen, carbon, and nitrogen, 

 with smaller quantities of the other inert gases, sulphur and the 

 halogens. 



The principal reactions which occur in such a gaseous medium at 

 different temperatures and pressures have been carefully studied, for, 

 in addition to their theoretical interest, they are of great practical 

 importance in chemical industry. 



IVlien oxygen, carbon, and hydrogen are considered the main 

 reaction is 



C02 + 4rL±^CH4 + 2H20 



The formation of methane is accompanied by diminution of 

 volume; hence it will be favored by high pressure. High tempera- 

 ture works the other way ; from the free-energy data it appears that, 

 at 1,0'00° C. and atmospheric pressure, the equilibrium inclines to 

 the side of carbon dioxide, even in the presence of a large excess of 

 hydrogen. Below 300° C. practically all the carbon should go into 

 methane; at about 600° the amounts of the two gases should be 

 comparable. 



With hydrogen and higher hydrocarbons the tendency of the 

 reaction is always toward methane at low temperatures. With sat- 

 urated hydrocarbons, this involves no change of volume and should 

 not be affected by pressure. Formation of methane from unsatu- 

 rated hydrocarbons should be favored by high pressure. The exclu- 

 sive presence of methane in the planets' atmospheres might thus have 

 been predicted. 



