480 MR JOHN ALLAN BROUN ON THE BIFILAR MAGNETOMETER, 
while it is pointed out that the series is not sufficiently good to give this result 
for the variation of g’ a great value, it should not be forgotten that the rocks 
forming St Helena are of a highly magnetic nature, and that their magnetic force 
varies with the temperature.* 
28. The coefficient g’= 1:02 has been employed by me throughout. When 
k=0:000189, we find 
[7]=0-000193. 
The value of g was found by water experiments between the ten of 
60° to 90° Fahr. to be 
q = 0:000276. 
I am not certain that this includes the coefficient for the expansion of the wires, 
but it is suggested} that, on account of the variation of g with temperature, the 
true value might be more nearly 
q=0-000260. 
The value employed in the corrections of the fortnightly means in the ‘‘St Helena 
Observations” t seems to have varied; in 1842 and 1843 it was 1-47 sc. div. 
(=0:000278); in 1844 it was about 1-12 sc. div. =0-000213; and in 1845. about 
1-29 sc. div. = 0.000245. Assuming the value by water experiments = 0:000260, 
we have 
[qg]=0°742 gq. 
29. Hosarton, 1842-48.—The continuous observations (without weekly 
adjustments) commenced April 1842, and the observations in my possession 
terminate with September 1848.§ The torsion circle was turned Ist April 1846, 
causing a change of the coefficient 4, and therefore of g’. The following are the 
results :— 
174. April 1842 to June 18438. k=0-000120. 
TABLE VIII.—Sums or DIFFERENCES OF BIFILAR DAILY MEANS AT HOBARTON. 
Month. 
" , _ Se. div. 
1842, April : : 1842, Dec. 58°33 | 103-14 
May : : ; 1843, Jan. 60°46 |. 97-44 
June p . 32 Feb. 65°69 95-58 
July : 4 : Mar. 65:59 | 108-00 
Aug. ° : ; April 72°44 | 111-49 
Sept. ; 6 May 45°84 | 58-90 
Oct. ° : : June 17-65 18:62 






Nov. : : Readjustment. 
* At least I have found this to be the case with magnetic rocks in South India. 
+ Adjustments, &c., ‘‘ St Helena Observations,” vol. i., p. 34. 4 
t Ibid. p. 35. ‘This Table has some inaccuracies. The coeflicient varies, and the temperature 
rao is not always 60°, as stated. 
§ Hobarton Observations, vols. i., il, and iii., printed under the superintendence of Lieut 
Colonel E. SaBine. 

