640 MR JAMES RUSSELL ON 



subject of cross magnetisation, or, more correctly, of superposed magnetic inductions 

 in iron, was investigated, apart from the subject of shielding in hollow iron cylinders. 

 It has been thought advisable to introduce this now, although the experiments were 

 made at a later date than those following on magnetic shielding. Their bearing on 

 this latter subject, however, will become evident the moment more magnetic forces 

 (circular or longitudinal) than that due to the transverse field fall to be considered. 



In § 2 various references to previous workers in this subject will be found. 

 The general results appear to have been that, apart from vibration effects and under 

 steady forces, the permeability of iron to a given magnetising force is lowered, if, at 

 the same time, it be subjected to a magnetising force at right angles. So far as known 

 to the writer, the two components of the resultant magnetic induction have not 

 been co-ordinated the one with the other, when two magnetising forces at right 

 angles to each other are superposed the one upon the other. 



§ 15. Shield B was in the first place used for these experiments. The two 

 primary windings (§5) supplied the means whereby two magnetising forces, circular 

 and longitudinal, could act upon the iron. It may be repeated that the circular force 

 H c is, when acting alone, the true magnetising force in the iron. The longitudinal 

 force is its calculated value due to the ampere turns of the solenoid merely. 



Table VIII. gives the results of experiments, when upon increments from zero 

 of the circular force H c is superposed (CL * conditions) and repeatedly reversed 

 (CLL * conditions) four different values of the longitudinal force H ; . It will suffice 

 to describe the two sets of readings when Hj = 3 C.G.S. units, the lowest value used. 

 In the first set the ballistic galvanometer is connected with the exploring coil measuring 

 either circular induction or the circular component of the resultant induction. The 

 first increment of H c = - 96 C.G.S. units gives a throw of "9 scale divisions of 

 the galvanometer (B.G.). The superposition of Hj gives an additional throw of 

 '6 scale divisions in the same direction, making a total of 1*5 scale divisions. The 

 next column, B, converts the results into C.G.S. units of induction. H z is now 

 repeatedly reversed, the ballistic throw being noted at each reversal. The final total 

 is 2*2 scale divisions, or, as in the last column, B= 850. The iron is now demagnetised 

 by decreasing reversals of the circular field, and the second increment of H c =r6, 

 made from zero, resulting in a throw of 2 '4 scale divisions. The superpositions 

 of Hj are made as before, and so on until the maximum increment of H,. has 

 been reached. It w r ill be noted that the first few reversals of H ? causes the circular 

 component of the induction to increase, and this irrespective of whether H; be 

 positive or negative, but this effect soon ceases, and further reversals produce no 

 increase of the induction. It is, of course, necessary that the iron be demagnetised 

 before each increment of H c . In the third, fifth, and seventh sets the same operations 

 are followed for higher values of the longitudinal field, viz., H ? =ir9, 21 "5, and 

 39 "6 C.G.S. units respectively. 



* See footnote on following page. 



