May 15, 1914] 



SCIENCE 



703 



von Sterneck of Austria and first used by 

 him about the year 1887. 



The method employed is to determine the 

 period of oscillation at a base station and 

 then at the new station and from the rela- 

 tion between the two periods the difference 

 in gravity between the two stations is deter- 

 mined. This is called the differential or 

 relative method of determining gravity. 

 The difference in gravity may be obtained 

 with an accuracy represented by a possible 

 error of about d= 0.0018 dyne, on an 

 average. 



The base station for the United States is 

 at the office of the Coast and Geodetic 

 Survey in Washington, D. C. The value 

 for it was determined by the relative 

 method, from the station at Potsdam, Ger- 

 many, at which the absolute gravity had 

 been determined from a long series of ob- 

 servations made with reversible pendulums. 



It is difficult to determine the absolute 

 gravity with a great precision, as is shown 

 by the determination of the difference in 

 gravity at Vienna and Potsdam, by the 

 relative method. Each of these stations 

 had its absolute value determined by many 

 years of observations and each value was 

 considered to be correct by an amount 

 represented by a probable error of only 

 a few thousandths of a dyne. The relative 

 determination showed an error between 

 them of 0.016 dyne. 



Of course, the same constant error which 

 may be present in the base station of a 

 system will also be present at each station 

 of that system determined by the relative 

 method. Potsdam is now generally used as 

 the base station for the world, although the 

 values are also occasionally referred to the 

 Vienna station. The values in the two 

 systems will differ by 0.016 dyne. 



A comparatively small error in the value 

 of gravity at the base station of the system 

 does not appreciably affect the flattening 



of the earth, derived from the gravity data 

 of the system. 



The investigations made in the Coast and 

 Geodetic Survey show that, on an average, 

 the gravity anomalies and the unaccounted 

 for residuals of the deflections of the vertical 

 are small for the United States. They also 

 show that the area of the United States as 

 a whole is practically in a state of perfect 

 isostatic equilibrium. But there are many 

 places where the gravity anomalies and 

 the deflections indicate departures from 

 the state of equilibrium. It is very im- 

 portant that additional data may be gotten 

 which will permit of a determination of the 

 degree of deviation from the perfect state 

 and of the horizontal extent of the affected 

 regions. 



In one of the illustrations^^ of the 

 second report on the gravity investigations, 

 which shows by contours the Hayford 

 anomalies, the evidence would lead one to 

 believe that the causes of the largest anom- 

 alies are confined to small areas. For 

 instance, there is an anomaly at Washing- 

 ton of + 0.037 dyne, while at Baltimore, 

 only 40 miles distant, the anomaly is 



— 0.011 dyne. At Seattle, the anomaly is 



— 0.093 dyne, while at Olympia it is 

 -(- 0.033 dyne, and the distance between 

 the two places is only about 50 miles. 

 There are other cases of decided differ- 

 ences in the anomalies in short distances, 

 and no doubt more will develop when addi- 

 tional gravity stations have been estab- 

 lished. 



The data shown in the illustration re- 

 ferred to lead to the conclusion that there 

 is no extensive area in the United States 

 which is very much under or over compen- 

 sated. This is equivalent to saying that 



11 No. 2 of the ' ' Effect of Topography and Iso- 

 static Compensation upon the Intensity of Grav- 

 ity" (second paper). 



