EAKTH'S CEUST — WASHINGTON. 313 



Austria-Hungary, and then to the east the low-lying plains of south 

 Russia. East of these are the Ural Mountains, and then (bending 

 somewhat southerly) we pass through Turkestan and Persia, and 

 reach the very high Pamirs, the " roof of the world." Thence the 

 surface slopes down across China, rises again in Japan, and again 

 drops to the depths of the Pacific Ocean. 



Let us now see how the rock densities, or rather the rock specific 

 volumes, correspond with the elevations. This inner graph, it is to 

 be remembered, represents the specific volumes, that is the reciprocals 

 of the densities, so that it is inverse to what the graph of the den- 

 sities would be — that is, the heavier the average rock the nearer to 

 the center it is on this graph, and the lighter the farther away. 



Starting with California, we find its specific volume arc above 

 the average, and that of Nevada-Utah to the east still higher. The 

 average of Colorado is a trifle lower, though the elevation is higher, 

 and this is one of the very few notable exceptions to the general rule. 

 We are ignorant of the igneous rocks beneath Kansas; they are in- 

 dicated as but a little above the average, which is probably not very 

 far wrong. There is a decided rise below the Ozark Ridge (with 

 its greater elevation), while the arc below the Mississippi Valley 

 is represented by the small arc for the Lake Superior rocks, which 

 are high in iron oxides and with high average density. Of the rocks 

 below "Kentucky" (east of the Mississippi) we know little, but 

 the Lake Superior arc is continued here because of the sporadic oc- 

 currences of some heavy peridotites in Kentucky mentioned above. 

 The graph rises sharply in the arc beneath the Appalachians, falling 

 again beneath New England, which is distinctly below the average. 



With the Atlantic floor we descend to an arc beneath it that is 

 well toward the center, as its rocks are of very high density. The 

 arc for Great Britain is scarcely above that of the Atlantic floor, that 

 of France distinctly higher, though still below the average, while the 

 arc below Germany is just above the average. 38 With the Alpine 

 and Tyrol arc we rise well above the average and here, just as in 

 the elevation graph, we reach the culminating point of Europe. The 

 arc beneath southern Russia (dotted) is placed at a level but slightly 

 different from that of Great Britain, because, though igneous rocks 

 are rare in this district, there are occurrences in Volhynia of very 

 heavy iron-bearing basalts. The arc beneath the Urals is but slightly 

 above this, corresponding with the heavy rocks of these mountains 

 which, it is to be remembered, are low and little more than large 



88 The specific volume arc for Germany should be but little above that of France to corre- 

 spond with the relative elevations ; it appears to be much higher because very many of the 

 German analyses of the heavier rocks (diabases, basalts, etc.), the analysis of which is 

 most liable to error, are of very poor quality, and are therefore omitted. 



