614 



of tlie method, more accurate calihratioiis and sludy of the coirec- 

 tioris. E. g. concerning the direct resnlts of the observations tiie 

 considerable deviations ') from tlie proportionality between the 

 force {F) and the square of the magnelic force between the poles 

 of the electromagnet occurring al hydrogen temperatures (cf. fig. 1, 

 taken from the paper mentioned) are particularly striking and, if 

 the results are given in terms of IjANGEVin's theory of paramagnetic 

 gases (cf. fig. 2, taken from Leiden Comm. Suppl. N°. 44a) it may 

 be asked whether no systematic errors occurred. 



1.0 



0.6 



0.6 



04 



0.2 



(J. 



-^ 



2 

 Fig. 2. 



The method for measuring the magnetisation, it sources of errors 



1) In the paper these deviations have been mentioned as probably due to 

 inaccuracy in the topography. 



