ELECTRODYNAMTC QUALITIES OF METALS. 
71 
Diagram XXIY. 
August 2, ll h 5 m to 12 h 40 m . 
Stretching weight, 18 lbs. 
Zero of magnetometer, at end 95. 
Diagram XXY. 
August 2, 4 1 * 0™ to 5 h 20 m . 
Stretching weight, 20 lbs. 
Zero of magnetometer, at beginning 95, at end 85. 
Diagram XXVI. 
August 3, 10 h 25 m to ll h 45 m . 
Stretching weight, 22 lbs. 
Zero of magnetometer, at end 90. 
Diagram XXVII. 
August 3, 12 h 0 ra to l h 50 m . 
Stretching weight, 24 lbs. 
Zero of magnetometer, at beginning 90, at end 100. 
Diagram XXVIII.* 
August 3, 4 h 0 m to 5 h 30 m . 
Stretching weight, 26 lbs. 
Zero of magnetometer, at end 90. 
Diagram XXIX. 
August 4, 12 h 20 ra to l h 50 m . 
Stretching weight, 28 lbs. 
Zero of magnetometer, at beginning 90, at end 100. 
225. From the curves it will be seen that the amount of the effect of torsion in 
diminishing the magnetization is not greatly influenced by the differences of pull from 
10 lbs. to 20 lbs., but that it is greatly diminished by increase of the pull above 20 lbs. 
For simplicity and uniformity in the comparison, take the amount of the diminution of 
magnetization by the first application of torsion in the direction called negative, from 
the whole range of from +50° to —200° on the scale of torsion (abscissas of the curves). 
We find with pulls of from 10 lbs. to 20 lbs. various amounts of from 180 to 230 
magnetometer scale divisions. The seemingly irregular differences between these 
amounts showed no regular dependence on the amount of pull, but seemed rather to 
depend upon previous conditions of the wire. But when the weight exceeded 20 lbs. 
there seemed a somewhat regular diminution in the effect of torsion with increase of 
pull, as shown in the following table ; thus:— 
With pull of 20 lbs. the effect was 193 
22 „ 177 
24 „ 150 
„ 26 „ 130 
„ 28 „ 65 
* In this diagram the two dotted lines between +40 and —200 coincide. 
