14 



NATURE 



[September 3, 1914 



of relief in the other; for slow subsidence .yives rise 

 to the conditions that favour deposition and the up- 

 lifting of a range results in the increased energy of 

 eroding streams. 



Thus there was a natural desire to see if Dutton's 

 theory agreed with the variations of gravity. If the 

 ups and downs are balanced, the apparently large 

 mass of a mountain-range ought to be compensated by 

 lightness of material in and below it. Dutton was 

 aware of the fact that this was approximately true 

 regarding the great continental plateaux and oceanic 

 depressions ; but he imagined that the balance was 

 delicate enough to show up in a small hill-range of 

 3000 to 5000 feet. 



The data required to test this theory, accumulated 

 during the triangulation of the United States, have 

 been made the subject of an elaborate analysis by 

 J. F. Hayford and VV. Bowie.'"' They find that, by 

 adopting the hypothesis of isostatic compensation, the 

 differences between the observed and computed deflec- 

 tions of the vertical caused by topographical inequali- 

 ties are reduced to less than one-tenth of the mean 

 values which they would have if no isostatic com- 

 pensation existed. According to the hypothesis 

 adopted, the inequalities of gravity are assumed to 

 die out at some uniform depth, called the depth of 

 compensation, below the mean sea-level. The columns 

 of crust material standing above this horizon vary in 

 length according to the topography, being relatively 

 long in highlands and relatively short under the ocean. 

 The shorter columns are supposed to be composed of 

 denser material, so that the product of the length of 

 each column by its mean density would be the same 

 for all places. It was found that, by adopting 122 

 kilometres as the depth of compensation, the deflec- 

 tion anomalies were most effectually eliminated, but 

 there still remained unexplained residuals or local 

 anomalies of gravity to be accounted for. 



Mr. G. K. Gilbert, ^^ who was one of the earliest 

 geologists to turn to account Dutton's theory of 

 isostasy, has recently offered a plausible theory to 

 account for these residual discrepancies between the 

 observed deflections and those computed on the 

 assumption of isostatic compensation to a depth of 

 122 kilometres. An attempt had already been made 

 by Hayford and Bowie to correlate the distribution 

 oiF anomalies with the main features of the geological 

 map and wnth local changes in load that have occurred 

 during comparatively recent geological times. For 

 example, they considered the possibility of an in- 

 creased load in the low Mississippi valley, where there 

 has been In recent times a steady deposition of sedi- 

 ment, and therefore possibly the accumlatlon of mass 

 slightly in advance of isostatic adjustment. One 

 would expect in such a case that there would be locally 

 shown a slight excess of gravity, but, on the contrary, 

 there Is a general prevalence of negative anomalies in 

 this region. In the Appalachian region, on the other 

 hand, where there has been during late geological 

 times continuous erosion, with consequent unloading, 

 one would expect that the gravity values would be 

 lower, as isostatic compensation would naturally lag 

 behind the loss of overljurden ; this, however, is also 

 not the case, for over a greater part of the Appalachian 

 region the anomalies are of the positive order. 

 Similarly, in the north central region, where there has 

 been since Pleistocene times a removal of a heavy 



20 J. F. Hayford, "The Figure of the Earth and Isostasy, ' U.S. Coast 

 and Geodetic Survey, WashinEtiorj. 1909. " Supplementary Investigation," 

 Washington, iqio. See also " Science," New Series, vol. xxxiii., p. 199, 

 1911. J. F. Hayford and W. Bowie, " The Effect of Topography and 

 Isostatic Con.pensation upon the Intensity of Gravity," " U.S. Coast and 

 Geodetic Survey Special Publication No. 10," Washington, 1912. 



21 "Interpretation of Anomaliei of Gravity," "U.S. Geol. Surv. Pro- 

 fessional Paper," 85-C, 1913, p. 29. 



ice-cap, there is still a general prevalence of positive 

 anomalies. 



These anomalies must, therefore, remain unex- 

 plained by any of the obvious phenomena at the com- 

 mand of the geologist. G. K. Gilbert now suggests 

 that, while it may be true that the product of the 

 length of the unit column by its mean density may be 

 the same, the density variations 'within the column 

 may be such as to give rise to different effects on 

 the pendulum. If, for instance, one considers two 

 columns of the same size and of exactly the same 

 weight, with, in one case, the heavy material at a 

 high level and In the other case with the heavv 

 material a low level, the centre of gravltv of the 

 former column, being nearer the surface, will mani- 

 fest itself with a greater pull on the pendulum; these 

 columns would be, however, In isostatic adjustment. ^^ 



Gilbert's hypothesis thus differs slightly from the 

 conception put forth by Hayford and Bowie; for 

 Gilbert assumes that there is still appreciable hetero- 

 geneity In the more deep-seated parts of the earth, 

 •vhile Hayford and Bowie's hypothesis assumes that 

 in the nuclear mass density anomalies have practic- 

 ally disappeared, and that there is below the depth 

 of compensation an adjustment such as would exist In 

 a mass composed of homogeneous concentric shells. 



In order to make the Indian observations compar- 

 able with those of the United States as a test of the 

 theory of isostasy, Major H. L. Crosthwait *^ has 

 adopted Hayford 's system of computation and has 

 applied it to 102 latitude stations and 18 longitude 

 stations in India. He finds that the unexplained 

 residuals In India are far more pronounced than they 

 are In the United States, or, in other words, it would 

 appear that isostatic conditions are much more nearly 

 realised In America than In India. 



The number of observations considered In India is 

 still too small for the formation of a detailed map of 

 anomalies, but the country can be divided into broad 

 areas which show that the mean anomalies are com- 

 parable with those of the United States only over the 

 Indian peninsula, which, being a mass of rock prac- 

 tically undisturbed since early geological times, may 

 be regarded safely as having approached isostatic 

 equilibrium. To the north of the peninsula three 

 districts form a wide band stretching west-north-west- 

 wards from Calcutta, with mean residual anomalies 

 of a positive kind, while to the north of this band lies 

 the Himalayan belt, in which there is always a large 

 negative residual. 



Colonel Burrard ** has considered the Himalayan 

 and sub-Himalayan anomalies In a special memoir, 

 and comes to the conclusion that the gravity deficiency 

 Is altogether too great to be due to a simple geosyn- 

 clinal depression filled with light alluvium such as 

 we generally regard the Gangetic trough to be. He 

 suggests that the rapid change In gravity values near 

 the southern margin of the Himalayan mass can be 

 explained only on the assumption of the existence of 

 a deep and narrow rift in the sub-crust parallel to 

 the general Himalayan axis of folding. A single 



22 It is interesting to note that the idea suggested by G. K. Gilbert in 

 1913 was partly anticipated bv Major H. L. Crosthwait in 1912 (" Survey of 

 India, Professional Paper," No. 13, p. 5). Major Crosthwait, in discussing 

 the similar gravity anomalies in India, remarks oarenthetically : " Assuming 

 the doctrine of isostasy to hold, is it not possible that in any two columns of 

 matter extending from the surface down to the depth of compensation there 

 may be the same mass, and yet that the densitv mav be very differentlv 

 distributed in the two columns? These two columns, though in isostatic 

 equilibrium, would act differently on the plumb-line owing to the unequal 

 disiribution of mass. 



"The drawback to treating this subject by hard and fast mathematical 

 formulae is that we are introducing into a discussion of the constitution of 

 the earth's crust a uniform method when, in reality, probablyno uniformity 

 exists." 



2:) " Survey of India, Professional Paper," No. 13, 1912. 



-^ " Survey of India, Professional Paper," No. 12, 1912. 



NO. 2340, VOL. 94] 



