MAGNETIC SHIELDING IN HOLLOW IRON CYLINDERS. 661 



Conclusions under TC conditions. 



S 48. We may now proceed to discuss more fully the shielding ratio curves given 

 on Plate IV. 



It was found that under the TC conditions the shielding ratio curves coincide 

 (except near their origin) with the theoretical curves ((g)) (§ 39), and that when 

 various values of the transverse field are used, the differences which exist in the 

 initial values of the shielding ratio diminish finalty as the circular field increases 



(§ 42). 



Circles have been added to fig. xxn. (shield A) and fig. xxiii. (shield B) at various 



points corresponding to the theoretical shielding ratio ((g)) curves of fig. iv. in order to 



show to what extent the theoretical and actual shielding ratios approximate the one to 



the other for various values of the transverse field. When H c = 0, and the transverse 



field is the only magnetising force in the iron, the initial shielding ratios, minus unity, 



are proportional to the values of B/H (see § 11), but as the circular field is increased the 



quantities rapidly become proportional to the values of clB'dH. where H is the circular 



magnetising force ; always provided that the values of H e are not pushed too high (see 



fig. xxii. (shield A), when H, = 64 - 25 and 80). As the values of the transverse field 



are reduced, this approximation takes place at lower values of H c , and reference to the 



curves shows that when H, = 4\37, the lowest value experimented with, the theoretical 



and experimentally determined shielding ratios are, within the limits of error, identical 



for all values of H c greater than - 5 C.G.S. units. 



But we have just seen that the permeability of the iron to the transverse field, under 

 the same conditions, likewise responds to the superposed circular force H c . Hence the 

 following two effects are observed under the TC conditions of field superposition. First, 

 the number of magnetic lines of the transverse field converging upon iron follows, although 

 necessarily in a limited way as shown in § 46, the permeability of the iron impressed 

 upon the shields by the superposed circular force H c . Second, the experimentally 

 determined ratios between the undisturbed transverse field and the weakened fields 

 within the shields, diminished by unity, have been found to be proportional to the 

 permeability, where n is defined as the rate at which B c is changing with respect to the 

 superposed force H c . 



These two effects are co-related the one with the other under the given conditions. 

 But the theoretical formula assumes (§ 3) the permeability of the iron to be uniform all 

 round the shields, and the absence of retentivity and coercive force in the iron. Hence 

 we may conclude that when values of the circular force increased from zero are 

 superposed upon a pre-existing induction due to the transverse field, the theoretical 

 conditions hold ; and that the experimentally determined shielding ratios are not effected 

 by the phenomena usually associated with hysteresis in iron. 



The TC conditions may therefore fairly be regarded as the normal conditions of field 



