34 JOSEPH BARRELL 



any individual curve, even though there may be no place where 

 the culminating phases of like sign all coincide and become additive. 

 These reasons are found in the subsurface location of those loads 

 due to outstanding density and also to the added stresses due to 

 isostatic compensation. These causes result in throwing a greater 

 stress upon the outer parts of the lithosphere and also serve to 

 broaden downward the stress diagrams. Furthermore, the stresses 

 due to isostatic compensation of continents would appear to be in 

 reahty much greater under the margins than the small values com- 

 puted by Love,^ since he has taken the continents as having the 

 broad sweeping surfaces of a harmonic nature, whereas, as a matter 

 of fact, the continents slope off rather abruptly to the depths of the 

 oceans. Facing the Pacific in fact, the two Americas show high 

 mountain elevations. This would cause the stresses in the vicinity 

 of these continental margins to resemble those imposed by a great 

 mountain chain and its isostatic compensation rather than those 

 imposed by the breadth of a continent. If isostatic compensation 

 is complete under mountain slopes, Love shows for the cases com- 

 puted by him that the maximum stress is about equal to that given 

 by a column of rock one-fourth the total height from mountain 

 crest to valley bottom. If the abyssal slopes of the continental 

 platforms be taken as averaging 3-4 km. in elevation and 50-100 

 km. in width, it is seen that, even if fully compensated, they add 

 stresses to the crust which may approach in magnitude one-half 

 of the stresses shown by curve A of Fig. 18. The extreme depths 

 of slope are much greater and it is clear that isostatic compensation 

 cannot be exact under these great reliefs. Therefore we may con- 

 clude that curve D does not overestimate the maximum stresses 

 imposed by the irregularities of the crust, both compensated and 

 uncompensated, as indicated by geodetic evidence within the 

 United States and especially along its ocean borders. This investi- 

 gation, however, has been of a general nature and is designed 

 merely to estabhsh an order of magnitude. It remains for future 

 work to make more precise analyses for each locality from the data 

 which may be acquired, and especially to investigate quantitatively 

 the problems offered by critical areas. 



■ Some Problems of Geodynamics (191 1), chap. ii. 



