20 EwiNG, The Structure of Metais. 



face exactly jone way. How does that agree with the 

 views I am placing before you now? In the first place 

 the polarity we are now dealing with — the polarity that 

 has to do with crystal-building — is not a uni-axial polarity 

 as the other was. It is a polarity along three independent 

 axes. In the second place this polarity is not a magnetic 

 polarity at all, because the application of magnetising force 

 does not cause the brickbats composing the crystals to 

 revolve. Consequently we must suppose that the mole- 

 cule is really a much more complex thing than these 

 models would seem to show. We must suppose that in 

 addition to this polarity along three axes, which deter- 

 mines the position of the brickbat among its neighbours 

 in the crystal, there is in addition a uni-axial polarity 

 which is capable of being rotated with respect to the 

 other. In other words, we must think of the molecule 

 as having its three axes capable of being held fixed in 

 space by the neighbouring molecules, and at the same 

 time possessing some sort of internal structure that is- 

 capable of being turned round so as to get the magnetic 

 axes to point one way without interfering at all with the 

 orientation of the molecule as a whole. 



