Comparisons of Magnetic Standards, 1915-21 



449 



magnetometer 19 showed that the decrease had taken place practically as a linear 

 function of the time during which the instrument was in field service. 



The absolute declination-observations on September 25, 1919, were erratic for 

 some reason, and the values for Rio de Janeiro in the declination-series of 1919 are, 

 therefore, based upon magnetograph scalings, using the observations of September 24 

 only. The base-line value finally adopted and communicated by Dr. Morize in his 

 letter of February 23, 1921, viz, 11 58'. 8, as determined by absolute observations with 

 magnetometer No. 25 at station A, was used. 



The horizontal-intensity results for the Observatory depend upon values of the 

 distribution coefficient, P', derived from the comparison observations only. The 

 Observatory values for horizontal intensity given in Table 14B may, therefore, be con- 

 sidered as provisional. Had the value of P' used for the reductions of the 1919 work 

 at Vassouras with magnetometer 25 been used in computing the results for 1915, the 

 resulting value of (I. M. S. Rio de Janeiro) would have been 0.00292// instead of 

 0.0032//. Dr. Morize states that, as a result of intercomparisons made on November 

 22, 1916, at Vassouras between Observatory magnetometer Cooke No. 25 and Cooke 

 No. 18, the value obtained for (Rio de Janeiro 25 Rio de Janeiro 18) = +2O7 or 

 +0.00081//. Accepting from Table 14B the mean value (I. M. S. Rio de Janeiro 25) 

 = -0.0034/7, we have (I. M. S Rio de Janeiro 18) = -0.0026//. 



Because of the erratic behavior of needles in dip circle 221 during the comparisons 

 of 1919, it was necessary to deduce from the horizontal-intensity and vertical-intensity 

 magnetograms a base-line value for inclination depending upon the mean of the six 

 observations made with the three needles at piers C and A with circle 221 on September 

 26 and 27. The mean temperatures in the magnetograph room for the intervals 

 covered during the inclination observations were the same for the first three and the 

 last three sets; the variation in inclination, A/ expressed in minutes of arc, was determined 

 by the formula (AZ and AH being in gammas): 



A/ = 0.131 AZ +0.035 AH 



Table 14A. Results of Declination Comparisons at the Rio de Janeiro (Vassouras) Observatory, 1915 and 1919. 



' All values are referred to station A; in 1915, A = B + 0'A. 



