Siemens Mercury Unit of Electrical Resistance. 301 



exactly on this account the neighbourhood of masses of metal 

 ought to excite suspicion. 



I have been unable to avoid a criticism of the method by 

 which the " British -Association Unit M was obtained, since the 

 question is how to explain a difference the cause of which I 

 cannot, to the best of my knowledge, find in my measurement. 

 For the rest, it is not impossible that many of the above objec- 

 tions might fall away if the results of the observations were 

 published more in detail ; and the want of such publication is 

 the more to be regretted, as it causes a deficiency in the other- 

 wise in some respects classical reports. 



The question as to which unit of resistance is best fitted for 

 general use hardly belongs to a scientific investigation. For the 

 sake of Physics itself we can but rejoice in the competition be- 

 tween the Siemens unit and that of the British Association ; for 

 thereby the best means is afforded of testing the unalterable- 

 ness of both, and this is all that comes into account for scientific 

 application. 



With regard to practice, the position occupied by Mr. Siemens 

 in relation to telegraphy must have secured a considerable start 

 for his unit ; while a no less important circumstance is that the 

 judiciously arranged and, so far as I know, well-adjusted 

 Sieinens's Resistance- scales have become widely distributed. 

 Moreover it can scarcely be denied that the definition of the 

 unit by means of mercury is intelligible to practical men, whereas 

 the definition of the other unit would be intelligible to but few 

 in the first instance. But further, a considerable additional 

 weight must be laid in the scale on the side of the Siemens unit, 

 if the other unit, as appears to be the case, does not approach in 



any important degree more closely to the absolute — -= 



than the mercury unit does by accident. The British-Associa- 

 tion unit is in reality (even in the view taken of it by the Com- 

 mittee ; comp. Report for 1864, page 346) not an absolute 

 measure but a concrete standard; and for use it is quite imma- 

 terial whether its degree of approximation to the absolute unit 

 amounts to 2 per cent, or to 3 per cent. Lastly, if the possibility 

 of reproducing the British-Association unit is to be made to 

 depend, not upon a repetition of the absolute measure (see 

 ' Report ' for 1864, page 348), but upon the conductivity of 

 metals, in which case it is self-evident that the preference would 

 be given to mercury, there does not seem to be any reason what- 

 ever why a column of mercury of simple and convenient dimen- 

 sions should not be chosen. 



