NO. 9 DETERMINATION OF OZONE WULF 11 



ficiently close to the area defined in the above work to yield a satis- 

 factory determination of the ozone. 



For these eight days, then, the logarithms of the logarithms of the 

 transmission coefficients were plotted against the logarithm of the 

 wave length and a straight line passed through the values lying to 

 both sides of the ozone absorption. The differences of the anti- 

 logarithms of those points lying in the ozone region from the anti- 

 logarithms of the corresponding points on the straight line were read. 

 These dift'erences were then plotted against wave length to the same 

 scale as that described above and the areas planimetered and divided 

 by the area corresponding to i mm. of ozone path, determined as the 

 result of the work described in this paper. Thus an apin-o.ximate 

 determination of the ozone on these days was afforded, utilizing an 

 area practically the same as that described above, defined by the 



Table I. — Atmospheric Ozone for Eight Days 



Path mm. o, 



, * _, 



Date Present work Dobson 



Aug. I, 1928 2.1 2.24 



Oct. 3, 1928 2.9 2.28 



Nov. 4, 1928 2.5 1.97 



Dec. 8, 1928 1.9 1.98 



Dec. 18, 1928 1.8 2.53 



Dec. 19, 1928 2.4 2.43 



Mar. 24, 1929 2.8 3.42 



Apr. 23, 1929 2.4 3.08 



Mean 2.35 2.49 



positions in which these points lay. The results are shown in Talkie i. 

 For these eight days the values of the ozone over Table Mountain as 

 determined by the Dobson method were also available and they are 

 given in this table for comparison. From these results it appears that 

 the average amount of ozone given by the two methods is essentially 

 the same. The independent values, however, are not in good accord, 

 which may be entirely due to the uncertainties in the present determi- 

 nations. It must be emphasized that this is due to the insufficiency of 

 the data used to make such a determination, which had been collected 

 from solar-constant work, and not to the method employed. It is to 

 be noted especially that differences in the relative values for the eight 

 days obtained by the two methods of Dobson and of Wulf cannot be 

 due to the method of the ozone absorption measurements described in 

 this paper, but must be contained in the roughness of the data. For 

 quite independent of the evaluation of the areas in terms of ozone, 

 these areas should be closely proportional to the ozone values on these 

 days. That they are not is actually due largely to the uncertainties in 



