ABSOLUTE OBSERVATIONS AND ADJUSTMENTS. 



123 



Vertical Intensity (Y"). 



The absolute value of the vertical component of the Earth's magnetic force Y corresponding 

 to a given reading of the balance magnetometer, was found from each value of the inclina- 

 tion 5, in conjunction with the corresponding value of the horizontal intensity X by the 

 formula, 



Y = X tan. 9. 



These 61 values of Y were reduced to the same scale reading of the balance magnetometer ; 

 five of them were rejected by Pierce's criterion, and the mean of the remainder, • 6176, was 

 adopted as corresponding to the scale reading 1500, and with the scale value found below. 

 Table 11 was computed for the reduction of the variation observations. 



The probable error of this mean was found to be • 0004. 



VARIATION OBSERVATIONS. 



The observatory for the variation instruments was a log hut, 19 ft. 9 in. (6 m.) X 16 ft. 

 6 in. (5 m.), and from 7 ft. 6 in. (2-5 m.) to 15 ft. (4-5 m.) in height. 



The floor was fastened with wooden pegs, the windows with copper nails. The walls were 

 of wood and mud, the fireplace of mud and stone, which latter had no effect on the magnets. 



The projection of the fireplace on either side screened the balance magnetometer and 

 declinometer from the direct heat rays of the fire ; the bifilar was screened by a table, which 

 was nailed to the floor. The accompanying plan shows these details and the ^Josition and 

 distance apart of the diff'erent instruments, which were mounted on wooden pillars about 

 0'2 metre in diameter, sunk about 1 m. in the ground. 



WINDOW 



BAROMETER 



A. Declinometer. 



B. Balance Magnetometer 



C. Bifilar. 

 D- Spare, declinometer. 



Window door 



:^ — 



Bifilar Magnetometer. 

 The horizontal intensity was recorded l)y means of the bifilar. A transportable Weber magne- 

 tometer with 3-inch (7 cm.) needle, hung in a loop of unspun silk fibre, was adjusted by placing 

 the instrument with the telescope to the North, and in the magnetic meridian, the interval 

 between the suspending threads being so regulated that when the torsion circle was turned 

 through 138°, the reflection of the centre division of the scale coincided with the cross 

 wire of the telescope. 



Although it was found that the instrument thus adjusted was slightly too sensitive, it was 

 thought best to leave it untouched, rather than to break the continuity of the observations 

 by altering the adjustment. 



Q 2 



