617 



The tube (6, cf. tlie diagram of this detail in fig. 3) containing 

 the rnagneti(t substance, lias been made and placed to obtain a 

 symmeirical distribution of glass with respect to a horizontal plane 

 passing through the centres of the poles of the electromagnet. In 

 this way the attraction exerted by the magnet on the glass has 

 been minimised and may be neglected. The dimensions have been 



chosen sucii that (he sample is at the place of maximum — - — , if 



oz 



the tube has been placed symmetrically in the field. The lower part 



(6,) of the tube has been evacuated, in the upper part [b^) a small 



quantity of helium gas has been introduced in order to improve 



the temperature equilibrium of the powder and the surroundings 



and of the particles of the powder mutually. The substance is 



enclosed between two glass disks, one of which has been melted 



on the tube, the other is free but is kept in its place by a small 



plug of cotton wool. Two flattened spiral springs, V^, V, prevent 



a lateral displacement of the carrier. The lower one has been 



attached to the carrier and not, as in previous work, to the tube, 



so that the tubes may be replaced without changing the position 



of the carrier. 



The end faces of the large size Weiss magnet have a diameter 

 of 4 cm and are 26,5 mm apart. The semi-angle of the conesliaped 

 boundary faces is 60°. 



The compensating force as function of the intensity of the current 

 in the coil D has often been determined as carefully as possible by 

 suspending weights to the tube. Notwithstanding all precautions 

 unexplained differences subsisted between the different calibrations. 

 The extreme ones differ about 2 "/„. In calculating a series of 

 observations use was made of the mean of the calibrations "before" 

 and "aftei". 



The specific magnetisation, ö, is calculated from the force measui'ed 

 by means of the relation 



Fg = ma—- (1) 



^ az 



where F represents- the force (in grammes) exerted on the mass m. 

 The ^-coordinate is measured along a xertical from the middle of 

 the interferrum; H is the magnetic force at tiie point indicated by 

 z; g — 9S1.3. 



1) If the susceptibility does not depend on the field strength, the maximum of 

 3 is preferable. [Note added in the translation]. 



