424 Scientific Proceedings, Royal Dublin Society. 



magnetization at each pitch-angle which requires the susceptibility to be the 

 same in all directions : (2) that the demagnetizing effect can be neglected. 

 The deviations from this result give some information on the relative 

 importance of these factors. 



The results for anhysteretic spiral magnetization are shown in the two 

 graphs of fig. 3 for S' = 10 and S' = 20. In these observations a spiral field 

 is applied given by H' = S' sin a and F= S' cosa; and the tube is then subjected 

 to the diminishing alternating field, as in the case of the twist, while under 

 the influence of this spiral field. It will be seen that the graph follows very 

 closely the broken line y = I L sin a where I L is the intensity of magnetiza- 

 tion when the pitch-angle of the spiral field becomes equal to 90°, that is, for 

 a simple longitudinal field of the same strength as the spiral field. It has a 

 curious lack of symmetry lying above this line for values 0°-45°, and below 

 it for values 45°-90 G , of which it has not been possible to give an 

 explanation. 



The applied alternating field superposes a longitudinal oscillation on the 

 steady spiral field. This oscillation is asymmetric about the final rest-point, 

 either from the point of view of magnitude or direction, except for small or 

 large pitch-angles. Again, the problem is complicated by the demagnetizing 

 factor, whose influence it is difficult to follow out in the present instance. 

 In fact, the investigation of the anhysteretic spiral magnetization can only 

 be regarded as showing that the experimental conditions give approximately 

 a constant intensity of magnetization, as required by the theory of the twist 

 given in this paper, subject, however, to some small disturbing effects, which 

 cannot be separated in the complex method which is used to give anhys- 

 teretic magnetization. 



Initial spiral magnetization. — At this stage it became necessary to return 

 to the second method of getting spiral magnetization, viz., to arrange for the 

 circular and longitudinal fields to grow up from zero simultaneously, so that 

 the lines of magnetic force formed a spiral of constant pitch-angle, in which 

 the intensity increased from zero to the required value. This would magne- 

 tize the tube everywhere in lines inclined at an angle a to its axis. This 

 type of magnetization may be termed the "initial" spiral magnetization. 

 Here also we should have 



Iia = Is' sin a, 



using the same notation, provided the same set of assumptions hold. Of 

 these it may be remarked that (la) is satisfied by the method of obtaining 

 the spiral field. We can evaluate to a certain extent the influence of (2), 

 which will allow us to deduce the effect of (1 6). 



