848 MILLER [chap. 32 



organic compounds formed during the reducing conditions would have been 

 oxidized rather rapidly by the molecular oxygen (Miller, 1957a). Whether the 

 surface carbon of the present Earth was all part of the initial atmosphere or 

 whether it has been escaping from the Earth's interior in a somewhat reduced 

 condition is not important to the overall picture to be presented. 



A . The Escape of Hydrogen 



We have learned in recent years that the temperature of the high atmosphere 

 is 2000°K or more, and there is no reason to suppose that the same temperature 

 was not present in the past. One might expect that a reducing atmosphere 

 would be cooler than an oxidizing atmosphere because methane and ammonia 

 can emit infrared radiation while nitrogen and oxygen cannot. However, 

 Curtis and Goody (1956) have shown that carbon dioxide is ineffective in emit- 

 ting infrared radiation in the high atmosphere. This is due to the low efficiency 

 of energy transfer from the translational and rotational to the vibrational 

 degrees of freedom, and it seems likely that this would apply to methane as 

 well. 



The loss of hydrogen from the Earth is now believed to be limited by the 

 diffusion of H2 to the high atmosphere, since almost all the water is frozen out 

 before it reaches the high atmosphere. Urey (1959) has discussed this problem 

 and finds that the loss is entirely due to these effects and not to the Jean's 

 escape formula. 



The present rate of escape is 10'^ atoms of hydrogen cm~- sec~i, and it is 

 proportional to the concentration of molecular hydrogen in the atmosphere, 

 which is now 10~^ atm at the Earth's surface. This rate would result in an 

 escape of hydrogen equivalent to 20 g of water/cm'^ in the last 4.5 x 10^ years. 

 This rate is not sufficient to account for the oxygen in the atmosphere (230 

 g/cm2). 



In addition, we must account for the oxidation of the carbon, ammonia and 

 ferrous iron to their present states of oxidation. The oxidation of the 3000 g/cm"^ 

 of surface carbon on the Earth from the to +4 valence states (i.e. from C or 

 H2CO to CO2) would require the loss of 1000 g/cm^ of hydrogen. At the present 

 rate of escape this would require 2.5 x lO^' years. If this escape was to be 

 accomphshed in 2.5 x 10^ years (i.e. from 4.5 to 2.0 x 10^ years ago) the pressure 

 of hydrogen at the surface of the Earth would be 0.7 x 10""^ atm. In order also 

 to oxidize the nitrogen, sulfur and iron, even larger losses and a higher pressure 

 of hydrogen are needed. We shall use a figure of 1.5 x 10"^ atm for the hydrogen 

 pressure in the primitive atmosphere. 



These calculations are much oversimplified since methane and other volatile 

 hydrogen compounds would be decomposed in the high atmosphere, and, there- 

 fore, a higher concentration of hydrogen might exist in the high atmosphere than 

 indicated by surface partial pressures. However, the results of the calculation 

 would be qualitatively the same for hydrogen pressures different from the 

 chosen value by an order of magnitude. 



