260 Proceedings of the Royal Physical Society. 



same order as r v and since p\ 7^ 1 2 is two dimensions in r x 

 lower than •fyr 1) we must retain terms of the form 2{<pr x h x m }, 

 &c, to the fifth order in h x Jc x \ . 



Also, since the particles of ether are symmetrically 

 arranged, if fr be any function of r, m 2{/r h p 1& l r ] vanishes, 

 unless p q r are all even, and then its value is the same in 

 whatever manner qr be interchanged. Hence from (7) 

 and (10) we find 



(1 r A 1 7 0 \ d 2 u 



Put E = s 2 I pr, V 



6 



Then it can be easily shown that 



t V V } = g ^ { 9 r x h, * 1* } = I 2 { 9 r x k> I*} = E . 



And let ^ 2 7 'i 2 = &c = S * 



Then equation (11) becomes 



d® r d(t r 

 dx x 1 dx 



= (3 R + S)£ + (E + S)5 + ( R + S )S( 12 ) 

 Similarly it may be shown that 



Also by (8) and (9) 



dy dz 



0 (15) 



