THE IMAGINARY RANGE AND TROUGH. 43 



excess of those due to the Imaginary Range, but this is due to 

 the fact that Major Crosthwait's calculations include all topography 

 within 2,564 miles of the station, and therefore the whole of the 

 highlands of Central Asia, whereas those for the Imaginary Ranee 

 only include topography within 100 miles distance. If we turn to the 

 compensated deflections this great difference disappears and we find 

 that Major Crosthwait's calculations give rather smaller values for the 

 northerly deflections. At the stations north of the boundary fault, that 

 is to say within the Himalayan region proper, the difference varies 

 from 6" to 2", an irregularity which finds a natural explanation 

 in the irregularity of the contour of the actual Himalayas and in 

 the deep cut valleys which penetrate it. At stations outside the 

 Himalayas, where these irregularities have less effect, a greater uni- 

 formity is observable and a closer agreement ; the greater difference 

 at Jalpaiguri is doubtless due to the inclusion in Major Crosthwait's 

 calculation of the southerly pull of the highlands of the Assam 

 Range and the Peninsula. 



From this comparison two conclusions may bo drawn. Firstly 

 that the limitation, of the extent of topography considered, to that 

 lying within 100 miles of the station is justified by the smallness of 

 the effect of more distant topography, when the opposite effect of 

 its compensation is taken into consideration ; in none of the 

 stations does the effect of the topography beyond 100 miles, and 

 up to 2.564 miles, differ materially from about a couple of seconds 

 of arc, and in every one of them it is in the same, northerly direc- 

 tion, so that no change is introduced in the difference between the 

 calculated deflection for any pair of stations. Secondly it appears 

 that the Imaginary Range will serve the purpose for which it was 

 intended ; that the deflections calculated from it are, on the average 

 of the same order of magnitude as those which would be deduced 

 from the actual topography ; and that the departures from the 

 deflections calculated from Mr. Hayford's tables which would result 

 from a variation in the hypothesis of compensation will a^ree in 

 character and order of magnitude with those which would result 

 from the application of a similar hypothesis to the more complicated 

 topography of a station, similarly situated, in the Himalayas. 



No more than this is, at present, required, so far as the range 

 representing the Himalayas proper is concerned ; but for the greater 

 part of its length the main range is bordered by a tract of lower 



[191 ] 



