THE NATURE AND INTERPRETATION OF GEODETIC EVIDENCE. 21 



method of determining the departure of the plumb-line from the 

 vertical at any one station, all that can be measured is the difference 

 of the deflections at one station as compared with another. A 

 station is therefore selected as the station of origin, in India it is 

 Kalianpur, and the figures published represent the difference in 

 deflection between the direction of the plumb-line at that station 

 and the other station of observation. Further it must be noted 

 that the calculation of the deflections necessitates the use of certain 

 assumed figures as representing the mean dimensions of the earth, 

 dimensions which are known with approximate, but not absolute, 

 accuracy. In the publications of the Great Trigonometrical Survey 

 the published deflections of the plumb-line are based on an assumed 

 zero deflection of the plumb-line at Kalianpur, and the dimensions 

 of the Everest spheroid, which has an equatorial diameter of 

 20,922,932 ft. and a flattening of 1/300-8. It seems certain that 

 this is not the closest approximation possible to the true dimen- 

 sions of the earth, and in the more recent publications of the 

 Survey of India the Bessel-Clarke spheroid J has been adopted, 

 which has an equatorial diameter 3,270 ft. greater than the Everest 

 and a polar flattening of 1/299-15 ; but the deflections are still 

 calculated and published in terms of the Everest spheroid, and will 

 be used without any attempt to adjust them to more modern 

 dimensions of the earth. Any such adjustment would only give an 

 illusory appearance of accuracy, for the difference in the absolute 

 deflection at Lambatach, the station most distant from the refer- 

 ence station of Kalianpur, does not amount to more than 3" of 

 arc, and the differences, with which we are concerned, would not 

 be altered by more than a single second in any of the groups of 

 stations considered, an amount which is trivial in comparison 

 with the differences actually observed. 



A more important correction, which will be applied, depends 

 on the probable existence of a southerly deflection at the reference 

 station of Kalianpur, where no deflection is assumed in the pub- 

 lished figures. Of the reality of this southerly deflection there 

 seems no possibility of doubt, but the amount is open to uncer- 

 tainty. In 1905 Sir S. G. Burrard adopted a value of +6", in 1912 

 a value of +4" is used in Major Crosth wait's investigation and, 

 being the latest authoritative estimate, it has been used and a 

 correction of -f4" has been applied to the published figures. It 



l See Phil. Tran*., Series A, CCV, 1905, p. 29P 



[ 169 ] 



