ESTIMATES OF TIME 839 



yearly by the sun and earth was the basis of the soundest methods for 

 determining the limits of duration of those bodies. 



It was calculated that if the energ-y of the sun were derived from 

 gravitational infall of its own mass and had been constant since the 

 beginning, the sun could not be over 18,000,000 years old. Thomson 

 showed, however, that the age could be extended beyond this narrow limit, 

 but his guarded conclusion was that the sun most probably has not 

 illuminated the earth for 100,000,000 years and almost certainly not for 

 500,000,000 years. 



On the assumption that the temperature gradient of the earth was 

 the result of simple cooling through geologic time from a molten begin- 

 ning, Thomson calculated also the age of the earth. His first estimates 

 were between 20,000,000 and 400,000,000 years; probably within 100,- 

 000,000 years. Later, in 1897, he reduced the time limits to between 

 20,000,000 and 40,000,000 years. 



Clarence King, in 1893, argued from the melting curve of diabase and 

 the temperature gradient of the earth that the earth could not possess its 

 known tidal rigidity and be more than 24,000,000 years old. This argu- 

 ment rests, however, on a very nice use of very un'certain quantities. 

 These estimates based on the thermal gradient ignore the great quanti- 

 ties of heat brought up by igneous activity. It is clear that the rise of 

 Archean and later granites must have disturbed profoundly the previous 

 temperature gradients. It has been found, furthermore, that radio- 

 activity gives such an embarrassingly large quantity of heat that it has 

 been necessary to assume the restriction of uranium and thorium with 

 their observed percentage to the outer 40 miles of the earth's crust, since 

 otherwise the earth would be heating up with geological rapidity, instead 

 of being a body slowly cooling or in thermal equilibrium. -. 



The discovery of radioactivity cuts out all solid basis for calculating 

 age from the flow of solar energy or the temperature gradient of the 

 earth. The subject of age as based on temperature gradient must still be 

 discussed, however, because of its inherited influence on geologic thinking. 



A recent attempt has been made by Becker to revive the argument 

 from temperature gradient by including the new data of isostasy and 

 radioactivity.^^ Becker takes the curve of fusion of diabase with respect 

 to pressure as given by Carl Barus, a .curve which is quite certainly wrong 

 for high pressures, since Barus assumed that the fusion point increased 

 in a rectilinear ratio with all pressures. Furthermore, Barus's measure- 



**- Age of a cooling globe in which the initial temperature increases directly as the 

 distance from the surface. Science, vol. xxvii, 1908, pp. 227-283 ; also in later publica- 

 tions. 



