272 LIGHT SCIENCE FOR LEISURE HOURS. 



twenty-six inches per mile. These discrepancies must 

 not only be regarded as insignificant in themselves, but 

 must appear yet more trifling when it is remembered 

 that they are not cumulative, inasmuch as the prelim- 

 inary triangulation is itself dependent on the great 

 trigonometrical survey. 



Let us understand clearly what are the various forms 

 of survey which are or have been in progress in India. 

 There are three forms to be considered : (1) The Great 

 Trigonometrical Surveys ; (2) The Revenue Surveys ; 

 and (3) the Topographical Surveys. 



Great trigonometrical operations are extended in a 

 straight course from one measured base to another. 

 Every precaution which modern skill and science can 

 suggest is taken in the measurement of each base-line, 

 and in the various processes by which the survey is 

 extended from one base-line to the other. The accu- 

 racy with which work of this sort is conducted may be 

 estimated from the following instance: During the 

 progress of the Ordnance Survey of Great Britain and 

 Ireland, a base-line nearly eight miles long was meas- 

 ured near Loch Foyle in Ireland, and another nearly 

 seven miles long on Salisbury Plain. Trigonometrical 

 operations were then extended from Loch Foyle to 

 Salisbury Plain, a distance of about 340 miles ; and 

 the Salisbury base-line was calculated from the obser- 

 vations made over this long arc. The difference 'between 

 the measured and calculated values of the 'base-line was 



