S EXPERIMENTS ON 



the shortest way was to mark with an asterisk (*) 

 those cases where it had failed ; and by inspecting 

 the tables, it will appear that the number of these 

 is comparatively small. 



The quantity of refraction, as entered in the ap- 

 propriate columns, has been computed, accoixling to 

 those respective cases of depression or elevation, 

 founded on reasonings too generally known to re- 

 quire any minute explanation. 



It will only be necessary to state, that, with regard 

 to the measured line, the formula r=^ A — D-^ a* 

 has been used in preference to that of y— --^-r^+pv ^ 



for tlie obvious reason, that a greater reliance was to 

 be placed on the perpendicular depression, obtained 

 during the process of measurement, than on corres- 

 ponding angular depressions, taken at the S. end of 

 the base line, even though an equal number of ob- 

 servations had been taken at each end. 



The refraction, affecting the elevation of the 

 Miaitapiimlf, was necessarily computed by means of 

 the elevations and depressions, taken at both places re- 

 spectively, i^—^-c^-12 being the appropriate formula, 



where E expresses the elevation, and is less than D. 



An example of each will suffice to render the sub- 

 ject perfectly clear. 



* Where r=. Refraction. 



\ A'=. One half the contained arc. 

 i)r= The greater depression. 

 D'zn The less depression. 

 And azr. The angle subtended by the perpendicular depres- 

 sion of S. end of measured line. 



t A small stone building, on a very conspicuous rising ground, 

 about four miles N. of Bangalore. 



