552 BEPORT— 1886. 



are we likely to be. Weber has made the assumption that a coBstant force tends 

 to restore the molecule to its original position. The simpler assumption seems to 

 be that in ideally soft iron the molecules take up positions depending upon the 

 influence of the others. I suppose that when there is no magnetisation the mole- 

 cules naturally group themselves in pairs with poles of opposite name in juxta- 

 position, and that the value of the resolved part of the magnetic force required to 

 separate them must reach a certain value K before any deflection takes place, and 

 then the deflected molecule sets itself in the direction of the magnetic force. This 

 seems to be a very likely hypothesis for iron of ideal softness. If H be the magnetic 

 force and 6 the inclination of the axis of a pair of molecules to the magnetic force, 

 the molecules acted upon and set with their axis parallel to the direction of the 



force are those for which sin 6 is greater than —. . The number of these varies 



11 



as the magnetisation. Calling the magnetisation I, we have 



7r 



1= a sin ed6 = ^^W^^EF 



1 sin - 1 K 

 -^ H 



for magnetising forces less than H = K, 1 = 0, for other values we have 



H=K 1=0 

 = 1-2K =-55 



= 1-5K =-74 



= 2-OK =-86 



= 6-OK =-98 



= IQ-OK = -995 



The curve deduced from this has the characteristics of extremely soft annealed 

 ii'on. The only want of resemblance is the suddenness in the rise of the curve 

 after H has reached the value K. This would doubtless be smoothed by any want 

 of perfect mobility of the molecules such as is involved in Maxwell's addition to 

 Weber's hypothesis, which must be taken into account in explaining the behaviour 

 of steel which has no connection with the cause of hysteresis. 



After a certain magnetisation is attained the magnetisation of the iron (if there 

 be no demagnetising influence of ends) must retain each molecule in its axial 

 direction, and the demagnetisation cannot commence until a reverse magnetic force 

 is applied. 



The effects deduced from this hypothesis are the same in character as those 

 deduced from Swing's friction hypothesis, but it seems to me that perfect mobility 

 rather than friction is more likely to be the explanation of a property specially 

 possessed by the softest kinds of iron, and I have thought that the hypothesis is at 

 least worthy of consideration. 



16. On a new System of Electrical Control for Uniform-motion Clocks. 

 By Howard Grubb, F.B.S. 



The two systems of electrically controlled clocks in use for di-iving astronomical 

 telescopes possess in common some disadvantages. 



1. At best they can only correct the rate of the clock itself, and have no power 

 to correct any error in the train of wheels between the clock and the endless screw 

 which drives the instrument. 



2. The checking apparatus acts on the clock governor itself, altering its rate ; 

 and as that portion of the instrument has a considerable vis inertice there is a 

 liability to a slight oscillation in rate after a correction is made. 



In the new system the author has endeavoured to avoid these disadvantage.s. 



The ' detector ' portion of the apparatus is attached directly, or almost so, to the 

 screw spindle itself, and the acceleration or retardation is effected by the temporary 

 introduction of a differential gearing into the train of wheels between the clock and 

 the screw spindle. 



