The agonic lines on the surface of the earth are in reality the 

 positions where the influences of terrestrial magnetism and local 

 attraction are equalized. The earth is a great magnetic field having 

 its north magnetic pole on the western shore of Boothia Felix in 

 Canada and its southern magnetic pole in that great south polar 

 region recently explored by Amundsen, Shark leton and Scott. 



This directive force of the earth's magnetic currents is divided 

 into horizontal and vertical components, the first being measured 

 by the ''Magnetometer" and the second by the "Dip Needle". 

 The first manifestation is what produces declination commonly 

 known by surveyors as the variation of the needle and the second 

 produces the unequal balance that must be compensated in the 

 northern hemisphere by slipping a little movable weight over the 

 S. end of the needle. This identification will always distinguish 

 the South from the North end of the needle. At the equator the 

 magnetic forces play horizontally over the surface of the earth and 

 the needle requires no artificial balance, but in the southern hemi- 

 sphere the counterweight must be transferred to the north end. 



A needle balanced for the latitude of Rochester will not 

 necessarily float freely elsewhere. To readjust this condition simply 

 unscrew the compass glass and, removing the needle, keep sliding 

 the counterweight along until, by repeated trial, a perfect balance is 

 secured. Wipe the needle free from finger stains on the final 

 replacement. Our method of mounting the compass glass makes 

 the needle easily accessible but keeps it free from moisture. 



Our needles are made of tubular tungsten steel of the lightest 

 possible weight, averaging about 0.165 grms. per inch (seep. 127). 

 The unusual superficial area permits the highest limits of magnetic 

 saturation and the lightness reduces friction on the pivot, making 

 the needle very responsive. An accurate needle is scarcely ever in 

 absolute repose, for the pen points at the ends and the point of 

 support should be as nearly as possible in the same straight line. 



Magnetic bearings are read on an extended traverse as a check 

 against such errors as arise from using the wrong tangent screw or 

 in making an incorrect reading on the vernier; but to guard against 

 the effects of local attraction, magnetic records of both forward and 

 back-sights should be taken and corrections applied. In this event, 

 however, nothing is gained over doubling every observation on the 

 vernier plates to prove the first reading by seeing that the second is 

 double its value. 



The sum of the interior angles of a polygon, as deduced from 

 magnetic observations, will not necessuri it the theoretical 



sum; for whether right or wrong, the aggregate will apparently 

 prove the work. Such angles, if read to the n . will often 



check up by compensation, or chuncr, \%hile the same angles care- 

 fully determined on the graduated plate \\ill tail to check. 



