216 GROVE KARL GILBERT— DAVIS ltoMOI "vo A ™xx£ 



sea-level value, to make certain hypotheses concerning the density of the plateau or mountain 

 mass, as well as concerning the density of that part of the crustal shell which lies beneath them. 

 Naturally the most acceptable hypothesis as to distribution of densities will be the one under 

 which observed values of gravity reduced to sea level agree most closely with calculated sea- 

 level values. 



THEEE ESSAYS OF 1895 



It was particularly in framing and criticizing various hypotheses as to crustal density that 

 Gilbert's later contacts with the problem of isostasy were made. Three essays on this aspect 

 of the subject were published in 1895. The first 3 was notable in establishing a closer relation 

 between geology and geodesy than had usually been reached. In preparation for this study 

 visits were made, on the way eastward from field work in Colorado in the autumn of 1894, to 

 a number of Coast Survey gravity stations from Denver to Cleveland, with the object of deter- 

 mining the average density of the underlying rocks. The fundamental crystallines were taken 

 to have a density of 2.70; beneath Chicago 2,800 feet of stratified rocks were assigned an aver- 

 age density of 2.52, and beneath Denver 15,800 feet of such rocks, 2.42. After the completion 

 of this study it was stated that corrections for the densities of the local formations may advisedly 

 be applied to gravity determinations under the theory of terrestrial rigidity but not under 

 that of terrestrial isostasy; yet even if such corrections are applied the departures of gravity 

 values at individual stations, after reduction to sea level, from the mean of these reduced values 

 average six times greater than if the calculations are made under the theory of isostasy. 



The second essay 4 is a further discussion of the same problem, but with the inclusion 

 of 16 stations east and west of those before considered; and here Gilbert's competence to deal 

 with mathematical concepts is well illustrated. Few geologists will care to give the time or 

 thought necessary to follow his "reduction to the mean plain," and fewer still would of them- 

 selves perceive that, under the theory of rigidity, the apparent altitude and the altitude cor- 

 rection as usually applied is too small, because "under that theory the sealevel, or geoid of 

 reference, rises in continental regions above the spheroid of reference," bat the error thus 

 introduced is thought not to affect the qualitative results of the inquiry. However, a number 

 of the more important steps taken and the conclusions reached from them may be quoted. 

 First to be noted is the explicit statement of the underlying isostatic principle: 



The ideal result of the correction of pendulum observations [of gravity] is uniformity. 



That is, if all corrections, including those for the variations of density in the crust and nucleus 

 of an isostatic earth as well as those for topography, altitude, and latitude could be accurately 

 applied, the resulting values of gravity would be everywhere alike; hence, if the resulting values 

 differ, some corrections, presumably those calculated from certain assumptions as to crustal or 

 nucleal density or from the fundamental assumption of isostasy, must be wrong; and it is pre- 

 cisely in the indications that these differing values give as to the distribution of rock density 

 within the earth that they attain a geological value and interest. Then comes a reenforcement 

 of the conclusions reached in the previous paper: 



The measurements of gravity [by the Coast Survey] appear far more harmonious when the method of reduc- 

 tion postulates isostasy than when it postulates rigidity. Nearly all the local peculiarities of gravity admit of 

 simple and rational explanation on the theory that the continent as a whole is approximately isostatic, and 

 that the interior plateau [between the Alleghenies and the Rocky Mountains] is almost perfectly isostatic. 

 Most of the deviations from the normal arise from an excess of matter and are associated with uplift. 



Unhappily, later observations do not wholly support this simple generalization. 



It is next explained more fully than in the previous essay that, while gravity measures at 

 high-level stations must, under the theory of rigidity, be corrected for the attraction of the 

 rock mass between the station and sea level, no such correction should be made under the 

 theory of isostasy which is now by reason of its above-noted success to be further pursued; 



! A report on a geological examination of some Coast and Geodetic Survey gravity stations. U. S. Coast Survey rept. for 1894, 1895, pt. 2, 

 51-55. 



' Notes on gravity determinations reported by G. R. Putnam. Bull. Phil. Soc. Wash., liii, 1895, 61-75. 



