On the Measurement of Electric Currents. 179 



of this by the help of an imposing array of multiple integrals. 

 But this would be the sort of thing which I have called 

 " playing with symbols/' i. e. using them instead of thought, 

 while their proper function is to assist thought. A mathema- 

 tical demonstration does not necessarily imply the use of 

 symbols, any more than that of diagrams : — and, when we 

 find an author continually using symbols to establish what is 

 obvious without them, we very naturally question the validity 

 of his symbolical processes when they are employed for their 

 legitimate purpose. I still think the assumption above a 

 legitimate and indeed almost an obvious one ; but it is strange 

 that an objection of this kind should come from a writer like 

 Prof. Boltzmann, who (see head Second above) has made, and 

 still defends, a fundamental assumption (of the class to which 

 he applies the term " unbewiesene Yoraussetzung ") which 

 most clamantly demands proof. 



Finally, as Prof. Boltzmann objects alike to Greek, and to 

 English, quotations, although they have Plato and De Morgan 

 for their authors, what does he say to the Latin one 



" Quis tulerit Gracchos de seditione querentes " ? 



XXI. On the Application of the Electrolysis of Copper to the 

 Measurement of Electric Currents. By Thomas Gray, 

 B.Sc, F.R.S.E.* 



IN the Philosophical Magazine for November 1886 I pub- 

 lished results of experiments on this subject which had been 

 conducted in the physical laboratory of Glasgow University. 

 Since that time the electrolysis of copper has been extensively 

 employed in that laboratory for standardizing Sir William 

 Thomson's new Standard Electric Balances, and considerable 

 additional experience has thus been obtained as to the accuracy 

 of the method. As the result of that experience, it may be con- 

 fidently asserted that the constant of an electric-current instru- 

 ment can be obtained with certainty within a twentieth per 

 cent, of absolute accuracy by copper electrolysis. Although 

 the mode of manipulation that seemed most likely to lead to 

 satisfactory results, and the precautions to be attended to, 

 were pretty fully given in the paper above referred to, Sir 

 William Thomson has called my attention to one or two 

 defects in it which seem to render a supplementary paper 

 desirable. 



When discussing the proper value to be taken as the electro- 



* Communicated by the Author. 



