THE SUPPORT OF THE HIMALAYAS. 103 



station is the larger valley of the Tista, similarly penetrating the 

 range. I am unable to determine the exact amount of the defect 

 of attraction due to these valleys, but an approximate estimate, 

 made from the 32-mile contoured map of India, shows that the north- 

 erly deflection at Senchal is in defect by an amount which is of the 

 order of 10" to 12" of arc, as compared with the average of stations 

 at the same distance from the outer boundary of the hills, or with 

 what would have been found at a station situated twenty miles 

 or so west of its actual position. Applying this correction to the 

 observed deflection we find that there remains a small northerly 

 residual of unexplained deflection at this station, instead of the 

 considerable southerly difference shown in table 29. 



The effect of these deep- cut river valleys must be felt, though 

 to a lesser degree, at Kurseong, but is there neutralised by the 

 purely local topography, which gives an excess of attraction amount- 

 ing to about — 1" of arc, and at this station the difference between 

 the estimated and observed deflections amounts to — 3". 



In this group we have only a single check on the estimates, in 

 Major Crosthwait's calculation of the residual at Kurseong, where he 

 made it amount to — 23", of which — 15" would be accounted for by 

 the estimate of the effect of the trough adopted in table 28, leaving 

 an unexplained residue of — -8" to be accounted for in some other 

 way. In part this is doubtless due to the estimate of the effect of 

 the trough being too small, but the difference between the actual 

 and the assumed dimensions cannot possibly amount to 50 per cent., 

 as would be required if this was a complete explanation, and part 

 of the northerly residual must remain unexplained after full allow- 

 ance has been made for any possible effect of the trough. 



In both the eastern and the western series of stations we have 

 the same feature of only small differences between the actual and the 

 calculated deflections at stations near the outer edge of the hills, if 

 we allow for the effect of the lesser density of the material filling the 

 Gangetic trough, and a high northerly residual of unexplained 

 deflection at stations situated 30 to 40 miles in. This difference 

 amounting to 10" to 12" is repeated in the three stations situated 

 between the two groups and may be accepted as not only real, but 

 directly connected with the structure and compensation of the 

 range, rather than with four independent, fortuitous, variations in 

 the density of the rocks which in every case act in the same direction 

 and to the same amount. 



[ 251 J 



