THE UNDERGROUND FORM OE FLOOR OF GANGETIC TROUGH. 69 



local departure from the average, in this case represented by the 

 defect of density in the Gangetic trough, and it will be seen that 

 either of the suppositions considered in table No. 19, if combined 

 with the average conditions assumed in calculating the observed 

 deflections, would largely reduce the differences between the cal- 

 culated and observed values. They may therefore be regarded as 

 approximations to the actual form of the trough, but it is not pos- 

 sible to obtain a closer approximation, with any degree of certainty, 

 owing to the uncertainty in which we are as to the density of the 

 alluvium in the lower layers of the deposit, as to the nature and 

 extent of the separate compensation of the trough, and as to the 

 presence and character of any independent cause which would 

 affect the direction of the plumb-line. The effect of the last two 

 elements of uncertainty has been dealt with in the last chapter, 

 and need not be enlarged on here. With regard to the possible 

 increase in density of the lower layers of the alluvium, the depth, 

 indicated by the observations, of 20,000 to 25,000 feet, even if 

 allowance is made for the possible increase due to a separate com- 

 pensation, is not so great as to necessitate or suggest a condensation 

 of the sands and clays of which the alluvium is composed to a 

 much greater density than the 2-16 which was assumed as the 

 mean density of the deposit, and against this must be placed the 

 fact that the upper layers have certainly a considerably lower 

 density than that assumed as the mean of the whole deposit. 



Allowing for all these possible modifications of the conclusions 

 come to if the whole of the deflections, so far as they are not 

 accounted for by the visible topography, are due to the alluvial trough, 

 the fact remains, that the published deflections agree so well with 

 those which should result from a cross section and dimensions of 

 the trough which are in accord with those suggested by a wholly 

 independent line of research, as to render it probable that this 

 is the preponderating, if not the sole, influence at work ; and we 

 reach the conclusion that the maximum depth of the trough lies 

 at, or a little south of, the edge of the hills and need not exceed 

 about 25,000 feet; it can hardly be less than 20,000 and is not 

 likely to exceed 30,000 feet, so far as the indications of this group 

 of latitude stations are concerned. 



The northern part of the section, which was not represented 

 in the group just considered, is covered, further west, by a very 



[ 217 J 



