397 



sheath. These become and , if we take 1=2 (as it pro- 



'^rO JL'OO 



bably is for gutta percha, nearly enough), and R='5, 1^=-^ 



8' 



R'=-0325, as the information given me by the Astronomer Royal 

 indicates. Whatever the theory may show for the influence of the 

 other wires, the result as regards retardation must be intermediate 

 between what it would be if the other wires were removed, and if 

 the one used were separated from them by a sheathing of its own. 

 We may therefore apply the theoretical result by taking c something 



between and . Hence if "the retardation" agree with 



2*45 I'oo 



the time corresponding to a in the diagrams, k must be intermediate 

 between 



" >x fo " x ro 



and 



^ X (180 X 5280)Xlog (|) T ^X(180X5280)*Xlog 6 (l) ' 



or again, if " the retardation" correspond to 9 a, k must be interme- 

 diate between 



. 



and 



^X(180X5280)Xlog (|) ^ X ( 180X5 280) 



I think it quite certain that what was observed as the retardation 

 must be in reality intermediate between a and 9a of the diagrams. 

 Hence the true value of fc for 1 foot of the wire must be between the 

 greatest and least of the preceding estimates, that is, between 



1 1 



and 



108X10 176X10' 



But the value of K (the "resistance" in British absolute electro-mag- 

 netic measure of 1 foot of the wire) must, according to Weber's obser- 

 vations on copper, be about 99810, or nearly enough 100,000*. Hence 

 a (the number of electro-statical units in the electro-magnetic unit) 



being equal to V , must be between 104,000,000 and 419,000,000. 



K 



* See a paper on the application of the general principle of mechanical effect to 

 the theory of electromotive forces, &c., published in the Philosophical Magazine, 

 Dec. 1851. 



