58 



DRS. J. A. BARKER AND P. CHAPPUIS ON A 



It will be observed that the largest residual is only 0'0158 mean bridge-\\hv 

 unit, or about '00003 ohm, showing that Avhen all the contacts are kept clean the 

 uncertainties due to variability of the contact-resistance are exceedingly small, and 

 that the form of contact-maker employed is extremely constant in its action. 



The largest residual observed in any standardization was O'OOOOS ohm, and the 

 average was about a quarter of this amount. 



XIX. Changes in the Resistance- Coils. 



In order to give an idea of the magnitude of the changes which took place during 

 the work, we give in the following table the values obtained for the wire coils in the 

 first and last standardizations expressed in mean bridge-wire units. In the fifth 

 column is shown the change which took place in each coil, not taking into account 

 the variation of unit. The figures were obtained by dividing the values in the fourth 

 column for the several coils by the corresponding values in the third. The absolute 

 magnitude of the changes cannot be deduced with certainty, but from other experi- 

 ments made by one of us with manganine wires it seemed probable that the total 

 change in any coil is in reality a combination of two distinct effects, the one IKMIIIJ a 

 change in the specific resistance of the wire throughout its entire length, and the 

 other an effect confined to a small length at each end, which was very strongly 

 heated during the operation of hard-soldering it to the copper tags. 



Examination of the appended results shows that the change in the value of the 

 lower coils is relatively much greater than in the case of the higher ones. This is in 

 accordance with what we should expect, if the statement above were true, and both 

 changes tended in the same direction. 



