io SCIENCE PROGRESS 



largely modified for the invisible rays, and the ratio ot the 

 absorbing coefficients comes out : 



Earth _ 60 

 Mars 99 



These figures give a value for the mean temperature ot Mars 

 equal to 22 C. ; this, however, has to be reduced owing to 

 difference in the rate of loss of heat, and the final result 

 is 8° C. 



The question now arises : if Prol. Poynting's value of the 

 absorption coefficient is incorrect, how is it that his theory 

 gives an accurate value lor the earth's mean temperature ? 

 Prof. Poynting's theory gives the result : 



6 = o"93# E = 0*93 x (eS/n-o-)' 



where 6 and $e are respectively the average and equatorial 

 temperatures of the earth (in absolute Centigrade degrees), 

 e is the coefficient ol absorption, S is the solar constant (in 

 ergs per sq. cm. per sec), and o- is the radiation constant 

 (i.e. the constant of Stefan's Law). Now Prof. Poynting gets 

 three different values for 6, owing to uncertainty in the value 

 of S. These are 52 C, 29 C, and 17 C, corresponding to the 

 values of S obtained by Angstrom, Langley, and Rosetti 

 respectively. Expressed in ergs per square centimetre per 

 second, these values of S are respectively 0*28 x io 7 , 021 x io 7 , 

 and 0*175 x io 7 . Prof. Poynting takes the third value of the 

 temperature as being that most in accordance with the fact, 

 under the assumption that e is 09. But if we put e = o*6, we 

 get the values of 6 as 22 C, i° C, and — n° C, and from this 

 it is clear that it is merely a question of choosing the solar 

 constant suitably. But now a serious difficulty arises. 



Since Prof. Poynting's paper appeared, a long series of 

 determinations of the solar constant has been carried out by 

 Prof. Abbot. Of a number of values obtained in 1902-4, and 

 given in the Encyclopaedia Britaunica, 1 the mean is 2' 12 cal. per 

 sq. cm. per min., or ©"148 x io 7 ergs per sq. cm. per sec. A 

 later value given in a paper read before the American Philo- 

 sophical Society 2 in 191 1 is 1*93 cal. per sq. cm. per min. The 

 latest of all 3 is i'933 cal. per sq. cm. per min., or 0*135 ergs per 



1 Art. " Meteorology." 



* See Nature, Ixxxvi. p. 534 (June 15, 191 1). 



3 See Nature, xciii. p. 198 (April 23, 1914). 



