THE STRENGTH OF THE EARTH'S CRUST 541 



number and sufficiently close to a straight line to permit the applica- 

 tion of the principles previously discussed. 



The residuals are given in the north-south and east-west direc- 

 tions. The broken lines give the resultants. Their convergence 

 indicates that there are two large and controlling positive masses 

 marked on the map as 2 and 5. To account for the local variations 

 shown by the resultants from station to station it is necessary, 

 however, to locate smaller masses of positive or negative nature 

 approximately as shown at i, 3, 4, and 6. There must be of course 

 many other centers of moderate disturbance within the area of 

 400,000 km. which is shown, but such as exist are far enough from 

 the line of section not to exert an appreciable influence. It is 

 noteworthy that gravity stations only 200 km. from the Hne of the 

 traverse can show anomalies as large as —0.029, +0.032, and 

 —0.027 dyne without the masses which give these anomalies 

 showing appreciable control over the deflections on the line of 

 traverse. Their areas of influence are therefore restricted. The 

 limited influence of these masses giving anomalies somewhat 

 above the average and at a moderate distance, and the small 

 masses locally modifying the deflections both serve to show the 

 importance of nearness of location. This limitation of control 

 over the deflections, restricted to distances of less than 100 to 200 

 km., is itself an indication that these outstanding masses lie within 

 the zone of compensation, otherwise their effects would be more 

 far-reaching. 



The residuals permit, however, a much more detailed solution 

 to be made. As a first approximation assume the outstanding 

 masses to be spheres. Figs. 12B and 12C. show the results. This 

 is not merely an arbitrary adjustment of curves and one of a num- 

 ber which might be devised. On the contrary, it has been shown 

 in the discussion of Section A and especially in Fig. 1 1 that the ratio 

 of the two maxima of the deflection components, Fy and Fx, and 

 the ratio of EE' to E'M hold a definite relation to the distance and 

 depth of the center of the sphere. Therefore if curve 2 be drawn in 

 proper proportion and as shown in B in order to satisfy the demands 

 of the y component, then the maximum value of Fx must not be 

 over 40 per cent of the maximum value for Fy, even if the center 



