An Underground Traverse. 407 



(5) The actual readings are apparently not recorded ; angles 

 taken out mentally, called " positive " and " negative/' after 

 transiting instrument. Working from zero point. 



(6) No indication of any check being adopted. 



{b) Levelling. In method (i), conducted by use of vertical 

 angles and no record of height of instrument, benchmarks, etc. ; (2) 

 carries on levels with azimuth readings ; correct checks to angular 

 measurements ; (3), (4), and (6) use dumpy level ; (5) uses angular 

 method without sufficient check. 



{c) Measurements. (2) works without decimals, which is 

 clumsy and contrary to Law ; (3) works to 3 places of decimals ; (4) 

 measures both fore and back sight, which, since he does not carry 

 out his levels, must necessitate adjustment of all plumb bobs to 

 height of instrument in order to serve as check on previous survey of 

 line. This is cumbersome and unnecessary, except at start of survey, 

 to check starting point. (5) measures fore and back distance, but 

 also takes vertical angles, which he reduces to horizontal. Writer 

 recommends only for inclined distances, e.g., shafts, winzes. (6) 

 gives no indication of method of checking. 



{d) Other Data. (i) and (2) adopt an off-set method for 

 position of drive, etc. ; (3) measures in reef and sketches in drive ; 

 (4), (5), and (6) apparently record no such data. 



Conclusion. 



It is hard to conceive that there is such a sad lack of uniformity 

 in the method of procedure. The writer is of opinion that in record- 

 ing work, all observations should be noted, and not merely the 

 deductions made at the instrument and at the time. In (a), the use 

 of the instrument, full advantage should be taken of the wonderful 

 accuracy of a transit theodolite when used in such a manner as to 

 eliminate errors of eccentricity, collimation, and of the standards. 

 Opinions must vary as to (b), levelling, whether best conducted in 

 conjunction with azimuth readings or by an independent survey with 

 a dumpy level. The writer would not be pedantic and say one or 

 the other should be adopted ; he would point out that the former 

 method, inasmuch as he must reduce all distances to the horizontal 

 and so also can calculate the vertical distance before calculating the 

 co-ordinates of any point, will enable the surveyor to be always up 

 to date in the elevation of his benchmarks. 



With regard to (c), measurements, individual practice is scarcely 

 disclosed in a study of these six methods. It is highly probable 

 that errors in underground traversing arise from (i) error in centering 

 the instrument, introducing small angular errors, and (2) errors in 

 measurement, due to the use of native assistants and the surveyor's 



