624 



Once the magnet was adjusted in its place, it was marked by means 

 of two plummets suspended to llie crjostat and maikiiifj two pointers 

 on tiie yoke of the magnet, for the magnet had tempoiarily to be 

 removed to afford opportunity of bringing the Dewar vessels Vhc 

 and F^ (fig. 3) into place. The large magnet is very heavy and there 

 was no device for moving the magnet siigthiy in horizontal direction, 

 so the horizontal adjustment was accompanied by great difficulties 

 and possibilities for iiuiccuracy. 



During the operations with liquid iielium and liquid hydrogen the 

 cryostat, forming one whole with the liquefactor, moved slightly in 

 an irregular way as a consequence of the changing temperature 

 circumstances in the different parts. By means of pulling rods the 

 initial position with respect to the magnet was restored. 



As far as the adjustment in vertical direction is concerned, it 

 must be pointed out that the distance (at atmospheric temperature) 

 from the centre of the mass to the centre of the field is considered 

 as "place" of the sample in the magnetic field. This place determines 

 the values of the constants in formulae (2) — (5}. In the measure- 

 ments in liquid hydrogen and in liquid helium tiiis place has changed 

 really by the shortening of the carrier in consequence of its cooling. 



dH . dH . 



The influence on -r— will be very small as t;— is maximum, but for 

 az ■ 02 



the same reason the influence on H has to be taken into consider- 

 ation. In itself there is reason for a correction. In the (rather 

 unfavorable) case that the carrier up to 20 cm above the sample 

 has the temperature of the boiling point of liquid hydrogen and the 

 othei' part is at atmospheric temperalui-e, a shortening of 0,3 mm 

 would follow from the data of Ch. Lindemann.') H would be 0,006 

 ƒ/„ smaller than corresponds to formula (2) i. e. about 0,7 " '„. Yet 

 no coi-reclion has been applied, because it would have required a 

 accurate determination of the place of the substance during the 

 measurements as the sinking of the liquid level changed the tempe- 

 rature distribution along the carrier and thus the place of the sample. 

 Moreover in the measurements in liquid hydrogen and in liquid 

 helium (and the experiments only refer to these temperatures) the 

 correction is nearly equal when the liquid level is on the same 

 height, as the expansion coefficient at these low temperatures rapidly 

 decreases to zero. 



g. Finally it must be mentioned that no trace has been observed 

 of the powder particles getting directed or remaining directed by 

 the magnetic forces. 



1) Physik. Zs. 13, (1912), p. 737. 



I 



