18 Land Magnetic Observations, 1905-10 



that it swings freely, after which, when the ampHtude has been sufificiently reduced, 

 a setting of north end of needle is made on the zero of the graduated arc of the 

 compass box and the horizontal circle read; a similar setting and reading of hori- 

 zontal circle for south end of needle; after lifting needle and letting it down gently, 

 settings and readings in reversed order, viz., south end, then north end, thus 

 comprising four circle readings; (c) four settings and readings in the order as before, 

 making the north end stand at the 5 east mark of the graduated arc and the south 

 end at 5 west; (d) four settings and readings, north end 5 west and south end 

 5 east; (e) four settings and readings same as (b); (/) finally, mark readings in 

 reversed order to (c). Throughout the needle settings, -the needle is preferably 

 kept in slight vibration and every care is taken to detect any sticking or irregularity 

 of swing, thus revealing injury to pivot or to jewel bearing. Additional sets are 

 obtained as conditions permit. See specimen form of observations and of com- 

 putations, p. 40. 



Earth inductor observations. A single set after the rotation axis of the coil has 

 been oriented in the magnetic meridian by means of a compass needle consists of: 

 (a) vertical circle reading when axis of coil is vertical as shown by attached level 

 direct and reversed ; (b) vertical circle reading for circle east and righthand rotation 

 of coil, of line of inclination, viz., position of axis of coil when no current is produced 

 and galvanometer shows, hence, no deflection; (c) repetition of (b); (d) repeating 

 (6) for lefthand rotation of coil; (e) repetition of (d); (/) four settings with read- 

 ings made as before except that vertical circle is west; (g) same as (a). For speci- 

 men set of observations and of computation, see p. 41. 



Diurnal Variation and Magnetic Perturbations. 



In order to eliminate the error due to uncertainty of the diurnal variation 

 correction, the declination observations are made at various times of the day, 

 whenever conditions will permit. In ordinary latitudes, declination observations 

 made about 10 to 11 a. m. or between 4 and 6 p. m. (local time) will correspond 

 fairly closely with the average for the day (24 hours) ; so will likewise the mean of 

 observations made between 7 to 9 a. m. and 12 to 2 p. m. Local disturbances 

 frequently exceed the diurnal variation correction; hence, in disturbed regions and 

 if the time be limited, multiplicity of stations rather than great accuracy at one 

 station must be the endeavor. Occasionally, once every two or three months, 

 diurnal variation observations are made in the field by taking 5-minute declination 

 readings throughout the day; these will be used, together with observatory data, 

 in the reduction to mean of day or month. For diurnal variation correction in 

 inclination or in intensity, reliance must be placed upon the data being obtained 

 by the existing observatories. In general, except during magnetic storms, the 

 diurnal variation corrections for the latter elements are on the order of the error 

 of observation. 



When the observations give indication of pronounced magnetic perturbations, 

 correction on account of which would be difficult to determine with certainty from 

 the available observatory data, the best course is to reject them entirely. Accord- 



