20 OLl)HAM: THE STRUCTURE OF THE HIMALAYAS, ETC. 



station, an increase in the depth of compensation is accompanied 

 by a decrease in its effect, or in other words an increase in the net 

 effect of the attraction of the visible mass and of its compensation. 



From the general considerations which have been set forth we 

 may conclude that the existence of a residual, or a divergence 

 between a computed and an observed deflection of the plumb-line 

 at any station may be explained in one or other of three different 

 ways, or by a combination of more than one of them. It may 

 indicate 



(1) that the compensation of the visible topography is not 



exact, but either in excess or defect ; 



(2) that the compensation is exact, but lies at a greater 



or less depth than that assumed in the calculation ; 



(3) that there is an excess of density, either in the surface 



rock or at some depth from the surface, which has not 

 been allowed for in the calculation. 



One or other of these conclusions is indicated, and it is only by 

 the comparison of a number of separate observations in the same 

 region that a decision can be reached as to which is the most pro- 

 bable explanation of the observed deflections. 



Before leaving this subject it will be necessary to devote a few 

 words to the nature of the evidence available. In practice the 

 deflection of the plumb-line from the vertical can only be deter- 

 mined in the two directions of the meridian and the prime ver- 

 tical ; the latter is more difficult than the former and more liable 

 to small errors, the observations moreover are too few in number, 

 in the region under consideration, to be made use of, but this is 

 a matter of small importance, seeing that the general trend of the 

 Himalayas, and of the Gangetic trough, approaches more nearly 

 to an east and west, than to a north and south, direction ; conse- 

 quently, the effect on the plumb-line is much greater in a north 

 and south than in an east and west direction. Only incidental 

 reference will, therefore, be made to the few available determina- 

 tions of the deflections in the prime vertical, and attention con- 

 centrated on the more numerous and important determinations 

 of the iefleetion in the meridian ; and in getting at the meaning 

 of them it k important to remember that the published figures 

 represent differences, not actual deflections. There is no known 



[ 168 J 



