INTRODUCTION. XXXiX 
divisions, are given, with the periods to which they apply, in the following table, 
and at the foot of each page of the Daily Observations. 
The adopted value of 
6= 71° 12’. 
See section 7. 
TABLE 18.—Values of & and g for the Balance Magnetometer, and the periods to which 
they apply. 
Periods to which the Quantities 
apply. Value of k, Value of g. 
1841. Mic. Div. 
4d. 
ad. oh. . he 
July 11 20—July 0:0000127 5-74 
Aug. 
4 20—Aug. 0-0000143 5-10 
Sept. 5 20—Sept. 0-0000127 5-74 
Oct. 6 20—Nov. 0-0000118 6-20 
1842. 
Jan, 12 20—Jan. 0-0000127 5-74 
Jan. 23 20—March 21 0-0000140 5:20 
March 21 20—April 16 0-0000118 6-20 
April 17 20—Oct. 8 0-0000130 5-60 
Oct. 9 20—Dec. 31 0-0000143 5-10 
§ 6. OBSERVATIONS FOR THE TEMPERATURE CORRECTIONS OF THE BIFILAR AND 
BALANCE MAGNETS. 
46. June 16. 1841.—Experiments were made to determine the effect of tempera- 
ture on the bifilar and balance magnets. For this purpose, the magnet was placed 
in a copper trough, and surrounded with ice, or water of the desired temperature. 
The trough was placed on a stand upwards of 5 feet east of the declination instru- 
ment, so that the axis of the magnet should be at right angles to that of the decli- 
nometer. A thermometer was placed in the trough beside the magnet. Seven 
readings of the declinometer were made in each of the following positions, N. pole E., 
N. pole W., N. pole E., in order to eliminate the changes of declination. The ob- 
servations are not worth recording, owing to the imperfections of the method ; the 
following is the final result for the bifilar magnetometer, g = 0-000353. 
January 1, 2, &c., 1842. Experiments were made on the balance magnet. The 
same method was adopted as on June 16, excepting that the balance needle was placed 
nearer the suspended magnet (within 4 feet); that the needle was repeatedly re- 
versed, and the bifilar magnetometer observed during the experiments. 
The discordances of partial results were so great, that only the final result is 
given here, 
q = 00001283. 
