OF THE MAGNETIC PERMEABILITY OF MAGNETITE. 



575 



22. Experiment. — In the practice of the foregoing method a composition of levigated 

 iron, beeswax, and resin was first tried. This, however, gave trouble in two ways. 

 First, it was almost impossible to secure homogeneity, the iron settling immediately on 

 the cessation of stirring and before the wax and resin had solidified. Secondly, on 

 attempting to make a long cylinder of the mixture by pouring it into a glass tube, it 

 was found that the contraction on solidification was such that the composition shrank 

 from the sides of the tube, thus giving rise to an irregularly shaped body quite useless 

 for the end in view. 



A composition of levigated iron and plaster of Paris, well mixed with pestle and 

 mortar in the dry state and so used, was then tried, and it seemed quite satisfactory. 



To imitate correctly the size and shape of the piece of magnetite, the mixture was 

 rammed into a boat made by wrapping copper foil round the magnetite specimen, taking 

 it off and soldering up the edges. The long cylinder was obtained by ramming the 

 same mixture into a glass tube about 100 diameters long. Before taking a pair of 

 readings for the kick with this composition, the primary current was rapidly reversed a 

 number of times ; as, without this precaution, the induction was found to diminish 

 materially if kicks were taken at the ordinary intervals of reading. A few reversals, 

 however, at once restored the mixture to its standard magnetic condition. 



23. The secondaries S and S' being balanced, the kicks obtained were as shown in 

 the table. 



Magnetic Substance. 



Galvanometer Kick. 



Specimen of mag 



[netite, 1*3 cm. square, 4'9 cm. long. 



202 scale divisions. 



182 



1 Mean of these 

 199 . 5 f two = 202-25. 



Mixture. 



Iron (gm.). 



Plaster of Paris (gm.). 



No. 1 

 2 

 3 



25 

 30 



28-5 



5 

 5 



5 



The constants of the apparatus, using symbols as explained in Art. 21, were as 



follows : — 



p = 7 '9 no. per cm. 



a = 5'62 sq. cm. a' = 5 '41 sq. cm. 



s = 159. s' = 136. 

 A = 0-36 sq. cm. 



C= -0485 c.g.s. units. Thus H' = y(4irpC) = y X 4'82 cg.s. units. 

 N= -005* 

 B=z 3 x 10 9 c.g.s. units. 



* Ewing's Magnetic Induction in Iron and other Metals, p. 32 ; Maxwell's Electricity and Magnetism, 3rd edition, 

 vol. ii. pp. 69, 70. 



VOL. XXXVIII. PART III. (NO. 15). 4 K 



