MAGNETIC SHIELDING IN HOLLOW IRON CYLINDERS. 669 



the case with the transverse and circular fields. There was in no case, however, the 

 same numerical correspondence with the theoretical formulae. 



II. — Superposed Magnetic Inductions in Iron. 



Of the two magnetising forces at right angles to each other, let Hj be the force first 

 acting, H 2 the force superposed. Each force acting alone produces the normal B-H 

 induction curve. Also let B x and B 2 be the two components of the resultant induction 

 in the direction of H x and H 2 respectively. 



First. — When upon a pre-existing induction due to H t increments of H 2 ascending 

 from zero to a maximum are superposed, the Bi component of the resultant induction 

 always lies above the B 2 component. For low fields the Bj curve is above the normal 

 induction curve ; but as the fields are increased a point is reached where the curves 

 cross, the B x component then falling below the normal curve. The first increase is a 

 relatively large effect, and appears to be due to molecular disturbance caused by the 

 superposition of H 2 increasing the permeability of the iron to the constant force Hj ; the 

 final decrease to lowered permeability. 



Second. — When repeated reversals of H 2 are superposed upon a pre-existing induction 

 due to H 1? the B x component lies still further above the B 2 component. For low fields 

 the B t curve is also further increased above the normal curve ; but as the fields are 

 increased, a point is reached where the curves cross, the Bi component then falling 

 below the normal induction curve. The increase is the well recognised effect due to 

 molecular vibration ; the final decrease to lowered permeability. Within due limits if the 

 effect of the first superposition of H 2 is large, repeated reversals have relatively less effect. 



Third. — When upon a pre-existing induction due to Hj increments of H 2 ascending 

 from zero to a maximum are superposed, the B 2 component of the resultant induction 

 always lies below the B x component — an effect of hysteresis. For low fields the B2 

 curve lies above the normal induction curve, but as the fields are increased a point is 

 reached where the curves cross, the B 2 component then falling below the normal curve. 

 The first increase is relatively a small effect, and appears to be caused by the two fields 

 acting together hastening the second or steep portion of the normal curve for both 

 components. For any two values of H4 (including H 1 = 0) the B 2 curves cross, the 

 curve for the lower value of H^ being at first below and afterwards above that for the 

 higher value of H x . The decrease of B 2 is a large effect and has been repeatedly 

 observed. 



Fourth. — When repeated reversals of H 2 are superposed upon a pre-existing induc- 

 tion due to H l5 the B 2 component lies still further below the B x component. The B 2 

 curve throughout appears to lie below the normal induction curve. If there exists an 

 initial rise, it must occur at lower minimum values than those used in these experiments 

 and close to the vertical axis. 



Fifth. — When H 2 carries the B 2 induction component in steps round a series of 



