Mr. 0. Heaviside on Wheatstone's Bridge. 93 



or require special precautions, such as the use of extra-flat 

 glass and greater pressure to ensure close contact in the print- 

 ing. For preliminary experiments I should be inclined to try 

 mica as a support, whose flexibility would facilitate a close 

 contact. I may mention that I have done copies of the 3000- 

 line grating oq sheets of mica, which may be obtained very 

 thin and smooth from the photographic dealers. For more 

 convenient manipulation in the preliminary stages of prepara- 

 tion, the mica should be mounted on a sheet of glass of the 

 same size as itself. A small drop of water interposed will en- 

 sure a sufficiently close adhesion. 



I have tried to take copies of copies, but with indifferent 

 success, even when the performance of the first was not per- 

 ceptibly inferior to that of the original. 



Gratings may be copied without the aid of photography by 

 simply taking a cast. Following Brewster, I have obtained a 

 fair result by allowing filtered gelatine to dry after being 

 poured on the 3000 Nobert. This method, however, is at- 

 tended with much more risk to the original, and is besides 

 open to other objections, sufficient, I think, to prevent its com- 

 peting with photography. 



[To be continued.] 



XIII. On Wheatstone's Bridge. 

 By Oliver Heaviside, Esq. 



To the Editors of the Philosophical Magazine and Journal. 



Newcastle-on-Tyne, 

 Gentlemen, January 11, 1874. 



MR. BROUGH, in the last Number of this Journal, has 

 called attention to the fact that, in a paper " On the best 

 Arrangement of Wheatstone's Bridge/' in the Philosophical Ma- 

 gazine for February 1873, I did not start from the general equa- 

 tion for the strength of the current through the galvanometer 

 irrespective of a balance, but took that form which the expression 

 assumes when at a balance. In fact I assumed (ad —be) to be 

 indefinitely small. My reason for doing so was, that the ques- 

 tion was to find the best arrangement at a balance, the great 

 object of resistance-measuring being to get a balance; besides 

 which, the course I adopted led to much greater simplicity and 

 no less accuracy. 



To take the special case upon which Mr. Brough dwells, viz. 

 to find the best resistance for the galvanometer. This amounts 

 to finding the resistance external to the galvanometer ; and when 

 a balance is obtained, as the galvanometer and battery are conju- 



