490 



PROFESSOR KNOTT ON SOME RELATIONS BETWEEN 



Table I. — For Iron Wire, 0*8 



mm. in diameter — continued. 



Tension Kg.-Wt. 

 pei Sq. Cm. 



Line Current 

 in Amperes. 



Longitudinal 



Field Eloctrom, 



Units, C.G.S. 



Twist ± H in 



Thousandths of 



Radians. 



Twist ± C in 



Thousandths of 

 Radians. 



Curve Number. 



T C 



II 



ht 



ct 



No. 



907 



304 









(11) 







48-3 



5-42 



315 









76-2 



3-80 



2-10 









97-7 



3 00 



1-64 









180-5 



1-43 



0-63 





907 



1-06 



2-7 



0-38 



0-50 



(12) 







8-5 



4-1G 



0-74 









15-1 



4-28 



0-76 









363 



3-23 



0-53 









57-8 



2-21 



0-36 









86-1 



1-68 



0-19 









106-0 



1-28 



015 









132-1 



1-11 



0-06 









243-2 



0-78 



-008 





The corresponding curves are shown in Plate I. ; and to facilitate reference each set 

 having the same tension and current is numbered in accordance with the numbers given 

 in the last column of the table. For each set there are two curves, since there are two 

 sets of twists to be plotted in terms of the longitudinal fields. 



5. Discussion of the Results for Iron. — The collocation of curves on Plate I. shows at 

 a glance many of the features which have been discussed in my former papers. For 

 example, the property of the maximum twist is well marked in all, and when the field 

 is high enough the wire twists in the opposite direction as first noticed by Bidwell. 

 In the present experiments, this phenomenon shows itself only when the twist is obtained 

 by reversal of the line current. With higher fields, however, there is little doubt that 

 the change of direction of twist would be produced by reversal of the field also. 



Again, if we group the curves according to the tensions, we shall find that each group 

 {e.g., 1, 2, 3, 4) brings out very distinctly the occurrence of the maximum twist in 

 higher fields as the line current is taken stronger and stronger. The effect also of 

 increased tension in diminishing the twist for the same combination of magnetising 

 forces can be clearly traced — a point which was specially discussed in Part I. 



Passing to the consideration of the new features, we see at a glance that the field- 

 reversal twist and the current-reversal twist are, except in very special cases, quite 

 different. When the field is high enough, the field-reversal twist is always the greater ; 

 in the curves, that one of each pair which is uppermost in fields higher than 1 5 is the one 

 obtained from the field-reversal twists. The current-reversal twist, however, is the 

 greater in low enough fields — excepting the case of No. 9. 



Leaving this single case out of account, we see that there is a particular field for each 

 current, for which the twist is the same whether the field or current is reversed. This 

 particular field corresponds, of course, to the point of intersection of the corresponding 



