MERCURY STANDARDS OF RESISTANT!. F.TC. 



113 



Summary. 



It is not difficult to summarise the resulta The difficulty of defining and 

 constructing a practical invariable mercury standard of resistance has been shown to 

 be rather under- than over-estimated. The possibility that the axis of the tube may 

 be undulating seems difficult to remove, though the final results indicate that its 

 effect is small. The presence of films of moisture and of air is disconcerting, though 

 again the results indicate the constancy of these under definite conditions. Of the 

 methods chosen to measure the resistance of mercury standards, that of the Kelvin 

 double bridge is certainly best, the tube being erected as indicated in Method II. 

 As already mentioned for permanent work, all rubl)er may be removed, the tul)es 

 themselves l>eing ground as stoppers. The temperature of observations is advisably 

 that of melting ice, for while there is little uncertainty concerning the equality of 

 temperature of the mercury and a well-stirred liquid, the thermometry then plays a 

 considerable part in the determinations. The change in resistance is determinable to 

 O'OOOl per cent., equivalent to one-thousandth of a degree on the centigrade scale. 



The primary objects in view have, it would appear, been achieved with a fair 

 measure of success. The eleven mercury standards constructed from theoretical 

 considerations enable a resistance to be evaluated in international ohms with an 

 accuracy of at least O'OOl per cent. This is evident from Table IX. The international 

 ohms, constructed at the Phya-Tech. Reichsanstalt, have indirectly l>een compared 

 with those constructed at the National Physical Laboratory. In the Report of the 

 British Association (Section A) for 1903, the ratio of the unit of resistance employed 

 at the Phya-Tech. Reichsanstalt to that of the National Physical Laboratory is 

 discussed. A table of measurements of certain manganin coils, ranging from 0*1 ohm 

 to 10,000 ohms, is given. These measurements were made partly at the Reichsanstalt 

 (the corresponding values l>eing international ohms), and in part at Teddington (the 

 values, presumably, l>eing expressed in terms of the absolute ohm [10 9 C.G.S. units]). 

 For the sake of completeness, the table is given here : 



TABLE XVI. Results of Measurements of Various Coils at the Reichsaustalt 

 and at the National Physical Laboratory, March, l'J03. 



VOL. tX^IV. A. 



