On the Polarized State of the Gas in a Crookess Layer. 119 



which is red hot, and the other below the temperature of boiling 

 water, and with an interval between them of a quarter of a 

 millimetre, then one-iiftieth of the light will be restored in an 

 analyzer which was so turned as to extinguish the beam before 

 transmission between the plates. Such an effect could be easily 

 observed, but as it is very unlikely that the whole of the strain 

 in a Crookes's Layer is due to a difference of density of the gas in 

 different directions, it is very unlikely that so great an effect would 

 be produced. One object, however, in trying the experiment 

 would be to determine to what extent the strain is due to a 

 difference of densities. 



K 2 



