426 WILLIAM BOWIE 



The average gravity anomaly with regard to sign for stations 

 in the United States by the adopted method of reducing for isostasy 

 is —0.003 dyne, while the average without regard to sign for these 

 stations is o . 020 dyne. An anomaly is the difference between the 

 observed and computed values of gravity. The computed value 

 has corrections applied for the elevation of station above sea-level 

 and for the effect of the attraction of the topography of the whole 

 world and of the opposite effect of the isostatic compensation of 

 the topography. The topography is considered to be that material 

 on the continents and on islands which is above sea-level and the 

 deficiency of material in the oceans. 



An anomaly of 0.020 dyne in terms of mass is here given in 

 order that the reader may have a clear conception of the magnitude 

 of the deviation of the gravity from normal. If we should have a 

 disk of material directly under a gravity station and if the disk 

 should be of normal density, 20 km. in diameter and about 600 feet 

 thick, the attraction on a gram mass at the station would be 0.020 

 dyne. An anomaly of 0.001 dyne represents the attraction of a 

 disk of material of indefinite horizontal extent and about 30 

 feet in thickness on a gram mass located near the center of the 

 surface. 



The anomalies by the isostatic method varied from +0.059 

 at Minneapolis, Minnesota, to —0.093 dyne at Seattle, Washington. 

 There were only ten anomalies which were greater than o . 050 dyne. 



When a correction was applied to the computed value of gravity 

 for the effect of the topography, but none for the isostatic com- 

 pensation, the mean anomaly with regard to sign for the stations 

 in the United States was —0.037 dyne. The corresponding mean 

 without regard to sign for the stations in the United States was 

 0.050 dyne. 



The fact that the gravity anomalies were more nearly eliminated 

 by the isostatic method of reduction than by the methods where 

 isostatic compensation was not considered is strong evidence in 

 favor of the former method. 



We must conclude, however, that no method can be near the 

 truth unless it has a rather general application to different sections 

 of the country and to different classes of topography. In order 



