18 HAROLD C. UREY 



corresponds to the decomposition of 20 g of water per cm^ during 4-5 ;•: 10^ 

 years. This would not account for the elementary oxygen of the Earth's atmos- 

 phere, but the calculated loss may be in error sufficiently to make agreement 

 possible, as has often been done in the past. However, hydrogen, methane, 

 ammonia, carbon monoxide and hydrogen sulphide escape from the interior of 

 the Earth in unknown amounts. These are oxidized and this can occur without 

 decrease in atmospheric oxygen only by the escape of hydrogen from the Earth. 

 Also, unknown amounts of ferrous iron of the Earth's crust have been oxidized 

 to ferric iron. In order to have oxidized the carbon, sulphur, nitrogen and 

 ferrous iron observed in the crust of the Earth, water to the extent of about 

 10% of the oceans must have been decomposed and the hydrogen lost from the 

 Earth. We do not know how extensive the loss of hydrogen has been, but it 

 seems most likely that the rate of loss has been much greater for much of 

 geologic time than the calculated rate given above. 



For the purposes of the calculation of the rate of loss of hydrogen mentioned 

 above, the concentration of hydrogen as water or molecular hydrogen above the 

 tropopause was taken as 2 ppm. Light of the Schumann region cannot penetrate 

 to the tropopause and below because molecular oxygen absorbs in this region. 

 In the absence of molecular oxygen only carbon dioxide will absorb in this 

 region and this gas is removed in a high degree from the Earth's atmosphere by 

 reaction with silicate rocks ([9], p. 148 flf.). Hence, at the present time, water 

 below the tropopause is not decomposed photochemically and hence excess 

 amounts of hydrogen do not escape above the tropopause and thus to the escape 

 layer and away from the Earth. However, before oxygen appeared in the atmos- 

 phere this must have occurred and hence far more hydrogen escaped to the high 

 atmosphere and the rate of escape from the Earth was far greater than it is now. 

 Thus there are good reasons to suppose that the rate of escape during the early 

 history of the Earth was very large. 



We outline the history of the atmosphere approximately as follows. Molecular 

 hydrogen escaped so rapidly that this occurred essentially during the period of 

 accumulation of the Earth. Methane and ammonia were decomposed rapidly, 

 but carbon compounds soluble in water were formed which dissolved rapidly 

 in the oceans. Some volatile hydrogen compounds, e.g. CH4, C2H2 etc., remained 

 in the atmosphere and were decomposed by light and the hydrogen escaped. If 

 these were absent, water below the tropopause was decomposed and the hydrogen 

 escaped. When carbon dioxide appeared, it reacted with the silicate rocks to 

 form limestone. Only when oxygen appeared did the rapid escape of hydrogen 

 cease due to the absorption of the Schumann region by molecular oxygen and 

 the present low rate of escape begin. There is no difficulty in understanding some 

 very large rate of escape in the past and a very small rate at the present time. 



When did free oxygen first appear in the Earth's atmosphere? An attempt 

 was made to answer this question some years ago [9]. Unfortunately, no addi- 

 tional evidence has come to light during the past years, (i) Thode, MacNamara 

 & Fleming [11] have observed that •'^'S is concentrated in sulphates relative to 

 the sulphides and they present evidence that this began some lo^ years ago. 

 This may mark the time at which free oxygen first appeared in the atmosphere. 



