322 



Profs. J. A. Fleming and J. Dewar. 



The balls were accordingly all weighed in liquid oxygen contained 

 in a vacuum vessel placed over the pole piece of the magnet. This 

 vacuum vessel contained mercury in its vacuum space and was of an 

 unusually excellent kind. In it liquid oxygen could be preserved 

 for periods of many hours without a trace of ebullition, and no 

 difficulty was experienced in making the weighings with great 

 accuracy. These weighings of course served also to determine the 

 density of the liquid oxygen used. The results are embodied in the 

 following tables. 



In each case the weighings give the apparent susceptibility of the 

 liquid oxygen, and these figures have to be corrected by adding or 

 subtracting a number representing the absolute susceptibility of the 

 ball at the liquid oxygen temperature. Thus in the case of the 

 silver ball the figure subtracted is two, as the nearest integer repre- 

 senting the susceptibility of silver at 182 C. In the case of the 

 bismuth ball the figure subtracted is sixteen, in the case of the glass 

 balls it is zero or at most unity, and in the case of the copper ball 

 the correction is additive, depending on the value of the field. 



Table Y. Determinations of the Magnetic Susceptibility of Liquid 



Oxygen. 



[I. With the Silver Ball. 



Volume of ball = 12'684 c.c. at temperature of liquid oxygen. 

 Density of liquid oxygen = 1*1376. 



Distance of centre of ball from pole of magnet = 5*37 cm. 

 Magnetic susceptibility of silver ball = 1'73 x 10~ 6 . 



Mean value of susceptibility of liquid oxj'gen as determined 

 the silver ball = 326 x 10~ 6 . 



