THE ATMOSPHERES OF THE PLANETS 567 



in proving from theoretical investigations that certain of the bands agreed 

 in position with bands of ammonia and that others agreed in position 

 with bands of methane or marsh-gas. These theoretical conclusions 

 were confirmed by Dunham, who, with the 100-inch telescope using much 

 higher dispersion than had been available to Slipher, was able to obtain 

 a more complete resolution of the bands into their component lines and 

 found a complete coincidence. Dunham estimated that the quantity of 

 ammonia gas producing the absorptions in the spectrum of Jupiter is 

 equivalent to a layer 30 feet thick under standard conditions. The amount 

 is less in Saturn. The ammonia absorptions are not detected in the 

 spectra of Uranus and Neptune. 



Methane is present in much larger amount. Adel and Slipher, in 

 1935, found that a 45-metre path of methane, at a pressure of 40 atmo- 

 spheres, gave bands intermediate in intensity between those of Jupiter 

 and Saturn. The much greater strength of the methane absorptions in 

 Uranus and Neptune is probably accounted for by the lower temperatures 

 of these planets. The ammonia must be frozen out of their atmospheres, 

 making it possible to see through them to a greater depth. Adel and 

 Slipher estimated that 25 miles of methane at atmospheric pressure would 

 be required to give absorptions as strong as those of Neptune. 



The higher gaseous hydrocarbons, ethane, ethylene and acetylene, have 

 been looked for in vain in the spectra of the outer planets. All the 

 absorption bands appear to be accounted for by ammonia and methane. 

 It is a grand slam. 



The presence of ammonia and methane in the atmospheres of the 

 large planets is not surprising. It is to be expected as a consequence of 

 the large amount of hydrogen in the atmospheres. The picture, as 

 painted by Russell, of the successive developments is as follows. When 

 the major planets were hot, the hydrogen and helium was mixed with 

 water-vapour, nitrogen and carbon dioxide. When the temperature fell 

 below about 300° c, the carbon dioxide reacted with some of the hydro- 

 gen to produce methane and water-vapour, the partially reduced oxides 

 of iron on the rocky surface exposed to hot hydrogen acting as a catalytic 

 agent. With further cooling, at about the temperature at which the 

 moisture began to condense, the free nitrogen would react with hydrogen 

 to produce ammonia. There would then be an atmosphere of hydrogen, 

 helium and other inert gases, mixed with methane, ammonia and water- 

 vapour, but with little or no carbon dioxide or free nitrogen. Below this 

 there would be a deep ocean, strongly alkaline from the ammonia in 

 solution. As the temperature fell still further, the ocean would freeze. 

 It may be mentioned that an ocean consisting of one part of ammonia 

 to two parts of water would freeze at — 100° c. ; all the four major planets 

 are colder than this. The only constituents in the atmospheres that are 

 capable of detection are ammonia and methane. 



It used to be thought that the rapid changes shown by the markings 

 on Jupiter were indications that the planet was hot. It was believed that 

 it still retained a great amount of its original heat. The theoretical 

 considerations of Jeffreys and the direct measurement of the temperature 

 of Jupiter — which give a value of about — 135° C. — have shown that 

 Jupiter must be intensely cold. The presence of ammonia and methane 



