95 



Thus we are compelled to attribute to the homogeneous magnetic 

 Id tin symmetry of the group C* previously mentioned. 



It is worth while remarking here, that this result is essentially 



IH iult nt on the symmetry attributed above to the electric field, 

 to the electric current (C). Indeed, the connection between 

 le different physical phenomena, as proved by experience, makes 



mressary that definite relations must also exist between their 



vial symmetries. If for some reason or other we had primarily 

 ittributed the symmetry C* to the electrostatic field, we should 

 ive to give to the magnetic field the symmetry previously attributed 



the electric field, i.e. C. The electro-magnetic phenomena them- 

 ;lves determine this reciprocal relation: and the whole question 



as closer examination shows, evidently settled, as soon as it has 

 clear what one wishes properly to consider as the "mirror- 

 re of an electro-magnetic field" 1 ). 



If it be postulated that also in "the mirror-image of the elec- 

 ro-magnetic field", the general relations between electric and mag- 

 ?tic quantities shall preserve their validity, and that therefore 

 le said mirror-image also shall have the function of a possible electro- 

 lagnetic system, then we have to decide which of the two following 

 tandpoints we wish to adopt: 



a. Either in the mirror-image we can take as electric vectors 

 electric force, current, dielectric polarisation) the mirror-images 



of the original electric vectors, and as magnetic vectors (magnetic 

 >rce, magnetic induction, etc.) the inversed mirror-images of the 

 riginal magnetic vectors; 



b. Or in the mirror-image we can take as magnetic vectors the 

 lirror-images of the original magnetic vectors, and as electric 

 actors the inversed mirror-images of the electric vectors in the 

 riginal electro-magnetic field. 



In fixing our choice in the way first mentioned, we have in a 

 lomogeneous electric field symmetry-planes passing through the 

 lines of force, in the magnetic field however a single symmetry-plane 

 perpendicular to the lines of force. But in fixing our choice in the 

 second way, the functions of the electric and magnetic fields are 

 exactly interchanged. 



Now there are "mechanical" theories of the electro-magnetic 

 field, which are founded on the first conception; but there are also 



i) On this side of the problem my attention was kindly drawn by prof. 

 H. A. Lorentz, to whom I am indebted for some valuable remarks here. 



