of Lightning- Conductors. 225 



inches without any alternative path, it remains able to juinp, 

 Bay, 1| inches when offered as an alternative a copper rod a 

 quarter of an inch thick and six or seven yards long. This 

 gives a rough notion of the kind of results obtained, and it 

 shows that the extremities of the rod experience almost the 

 whole of the original E.M.F. of the condenser. 



Some experiments on much the same lines had been previ- 

 ously made by Prof. Hughes and M. Guillemin (see Comptes 

 renduSj 188(3, Annates Telegrapluques, 1865, Address to So- 

 ciety of Telegraph Engineers, 1886) ; but they used a fine 

 wire instead of an air-space, and tried what conductor would 

 protect the fine wire from being deflagrated. 



Under these conditions the experiment is practically a com- 

 parison between the impedances of two conductors — one of 

 which has its inertia term the bigger, while the other has its 

 resistance term the bio-aer. 



The general theory of divided circuits has been given by 

 Lord Eavleigh (Phil. Mag. 1886, pp. 377 et seqq.), and inas- 

 much as in the present case there is practically no mutual 

 induction between the two conductors, and the frequency of 

 alternation is very great, the resultant resistance and induction 

 take the following forms : — 



-p_ RjBg (LxEg — I^R^ 2 . xqx 



*~R 1 + V (Ri + R^ + L^* ■ • • W 



L =^t 2 < 10 > 



The resultant impedance is, as usual, 



P =X /( |R2+ L} (11) 



Perhaps, however, it is hardly fair to assume that the dis- 

 charge will remain oscillatory when one of the branches of the 

 divided circuit is permitted to have a high resistance. Cer- 

 tainly one cannot apply to such degenerate formula) for crite- 

 rion conditions. 



The general expressions are : — 



B.R, n* (L^-L-.R,) 2 



K 1 + R 2 " , "R, + E 2 • (R 1 + R 2 ) 2 + » 2 (L l + L 2 ) 2 ' ' ' yli > 

 _ Ljy + LJi, 2 _ (LiRa-L^,) 2 

 (B.+R,)' (R 1 + R 2 ) 2 (L 1 + L 2 ) 



(L.B , - U I m ) 2 ,, ,. 



+ (L, + L 2 ) { (E, + K 2 / + n *(L, + L 2 )H ; ^ 6 > 

 1 R 2 



» 4 =LS-S? (") 



Phil. Mng. 8. 5. Vol. 26. Xo. 150. Aug. 1888. Q 



