Febeuaby 10, 1911j 



SCIENCE 



201 



There is an excess of mass in some columns 

 and a defect in others, but the evidence in- 

 dicates that the average excess in the areas 

 of undercompensation is properly repre- 

 sented by a stratum only 250 feet thick 

 having the same density as the surface por- 

 tion of the earth (2.67). Similarly, the 

 average defect of mass in the areas of over- 

 compensation corresponds to a stratum 

 only 250 feet thick. 



Contrast these small excesses and defects 

 of 250 feet with the actual elevations in 

 the United States, 2,500 feet on an average 

 for the whole country. If there were no 

 isostatie compensation these actual eleva- 

 tions would represent excesses of mass. 

 The compensation may propetly be char- 

 acterized as departing from completeness 

 only one tenth on an average. 



These are the facts, established by abun- 

 dant geodetic evidence. These facts may 

 not be removed or altered by showing that 

 difficulties are encountered when one at- 

 tempts to make them fit existing theories 

 geological or otherwise. The theories must 

 be tested by the facts and modified if 

 necessary. 



A close approach to the condition called 

 isostasy certainly exists. It is uncertain 

 how this condition has been produced; 

 upon that point the geodetic observations 

 furnish no direct evidence. 



The recognition of isostasy in a definite 

 and reasonable manner in the computa- 

 tions of the figure and size of the earth 

 from astronomical observations and trian- 

 gulation has nearly doubled the accuracy 

 of the computed results. This recognition, 

 combined with other improvements in 

 methods of computation, has enabled the 

 Coast and Geodetic Survey to compute the 

 equatorial radius and the flattening of the 

 earth from observations in the United 

 States alone with greater accuracy than it 

 yras formerly possible to compute it from 



all the observations of the world combined 

 — by such computations, for example, as 

 those made by Bessel and Clark. 



The evidence is clear that the present 

 isostatie compensation is not an initial 

 condition which has persisted since early 

 geologic times. There is abundant geo- 

 logical evidence" that within the interval 

 covered by the geologic record many thou- 

 sands of feet of thickness of material have 

 been eroded from some parts of the United 

 States and adjacent regions and deposited 

 in other parts, that changes of elevation of 

 the surface amounting to thousands of feet 

 have been produced in this and other ways, 

 and that these changes have continued to 

 take place in recent time. Hence it is evi- 

 dent that if there had been complete iso- 

 statie compensation in early geologic time, 

 and no readjustment toward the isostatie 

 condition had taken place since, the depar- 

 ture from complete compensation would 

 now be measured by strata thousands of 

 feet thick upon an average. In fact, the 

 present departures from complete com- 

 pensation are measured by strata only a 

 few hundred feet thick — 250 feet on an 

 average. It is certain that a readjustment 

 toward isostasy has been in progress dur- 

 ing the period covered by geologic record. 



Let us consider the tendency of gravi- 

 tation to produce readjustment toward 

 isostasy. Eecur to the case indicated in 

 Fig. 1. Columns A and B have been as- 

 sumed to contain equal masses. There is 

 complete isostatie compensation. The 

 pressures at the bases of the two columns 



° The paper entitled " Paleogeography of North 

 America," by Charles Sehuchert, pp. 427-606 of 

 volume 20 of the Bulletin of the Geological So- 

 ciety of America may be cited as an example of 

 such evidence marshaled in systematic form. 

 Consult the fifty maps at the close of this publi- 

 cation for a graphic indication of the changes 

 which have probably taken place on this con- 

 tinent. 



