166 Mr. F. Jenkin on the Question of the 



tion Reports the argument in favour of this coherent system of 

 measurement is very convincingly put forward. The mercury 

 stands alone an arbitrary definition, while every one of the Com- 

 mittee's units will be an approximation to the natural or mathe- 

 matical unit. Now this relation between the several units is of 

 practical convenience in the daily testing of submarine cables, and 

 in all other electrical measurements. 



But Dr. Siemens says, first, your so-called permanent unit 

 does not represent your definition even with reasonable accuracy ; 

 and secondly, it will not be permanent. 



With reference to the first point, Dr. Siemens has fairly 

 pointed out the great difficulties of the determinations; but I 

 think he also admits that the names of those who actually con- 

 ducted the experiments are some guarantee that care would be 

 taken, and that the names of the Committee at large are a 

 further guarantee that the results would not be adopted without 

 a strong conviction that they were trustworthy. But even if 

 they were not, which I entirely disbelieve, the B.A. unit would 

 not be at all more arbitrary than Dr. Siemens' s unit, although I 

 allow there would then be no reason for its adoption except its 

 permanency. 



Will it then be permanent ? Time alone will show whether 

 any of the ten material standards will remain perfectly perma- 

 nent ; but we have at least as good a guarantee for their per- 

 manence as Dr. Siemens has for his. 



If his 1864 issue were correct, and I heartily wish we could 

 be sure of this, the B.A. unit would be equal to T0486 metre of 

 mercury at 0° C. of 1 square millimetre section; and this resist- 

 ance can be obtained just as easily as the resistance of one metre. 

 If, therefore, all our material standards vary, we are no worse off 

 than Dr. Siemens. But we have what, while the present dis- 

 crepancies in mercury-determinations exist, I venture to think is 

 an extremely important addition to our security, in the probabi- 

 lity of the permanence of the standards already made. We have 

 all Dr. Siemens has, and a great deal more. 



Every additional metal of which the specific resistance is ac- 

 curately determined gives a fresh guarantee; and it is possible 

 that some day this method of ensuring permanence may be more 

 important than the mere conservation of material standards. As 

 yet I think this is not so. 



To sum up the arguments hitherto used. 



The mercury unit as denned is arbitrary, varying, has no prac- 

 tical advantage, and forms part of no coherent system of general 

 electrical measurements. 



The British-Association unit is based on the natural mathema- 

 tical unit, is permanent, and has the practical advantage of form- 



