( 810 ) 



of electrons, unless it he niodilied by some new livpotliesis, wonld 

 nii(lout)te(Ily i-ecpiire the (wisleiice ot' sncli a couple. Jn oi-der to 

 see this, it will sut'tice h> consider a condenser with aetJier as 

 dielectricinn. It may Ik^ shown that in e\ cry eh'ctrostatic system, 

 movijiii' witii a velocity iv» ^), there is a certain amonnt of "electi'O- 

 magjietic momentnm". If we re|)resent this, in dii-ection and ma.uni- 

 tndo, hv a \'(M*tor Q\ the conj»le in ([nestion will he detei-mined hy 

 the vector j trod net ') 



K^>-iv| (1) 



Now, if the axis of : is chosen perpendicnUir to the condenser 

 j)lates, the velocity >r iia\ing' any direction we like, and if [■ is 

 the energy of the condenser, calcniated in the ordinary way, the 

 components of (?*> are given ^) hy the foHo\\ ing foi-mnlae, which are 

 exact n|» to the tii'st order: 



2t' ,, 2U ^. 



^r = — \i\ ^\/ — — \v,f, C^>. = 0. 



Snbstitnting these valnes in (J), we get for the components of 

 the conjde, njt to terms of the second order, 



2ir 2U 



— \\\, IV- — - — IÜ, iv- . 



These expressions show that the axis of the couple lies iji the 

 plane of the plates, perpendicnlar to the Ir-anslation. If « is the angle 

 between the velocity and the noi-mal to the plates, the moment of the 



conple will be — ir' .■^ii/ '2(( : it tends to tni'ii the condenser into snch 

 (■" 



a position that the plates are parallel to the Earth's motion. 



In the apparatns of Troitox and Xohi-k the condenser was 



tixed to the beam of a toi-sion-balance, snl'liciejitly delicate to be 



detlecte<] by a conple of the above order of magnit nde. No effect 



could howex'er be obserx'ed. 



^ 2. The expei-iments of which I have spoken are not the oidy 

 reason for which a n(M\ examination of the problems conne«'ted 

 with the motion of the Karth is desirable. Poincark ') has c>bjected 



') A vci'lor will be donolf'd l)y a (ierman letter, its niagnitiide by the con-e- 

 spondini;- lialin letter. 



~) See my article: Weilerbildung der MAXWELL'sclien Theorie, Eleetronentlieorie 

 ill tlie Mathem. Encyclopiidie V li, § 21, a. (This aitiele will be quoted asM.E.) 



■') M. E. § 5Ü, c. " 



*) PoiNCARÉ, Rapports du Congres de physique de 1900, Paris, 1, p. 22, 23. 



