426 



THE BRITISH ASSOCIATION 



probable variations in the values of the coils throughout their whole existence . 

 The values are plotted in Fig. i. It is apparent from this diagram that the 

 values for 1876 are all comparatively high. At this time the temperature 

 was only observed to o- 1° C., and an error of this amount for the platinum 

 coils would lower all the values to those shown by the dotted lines, which 

 values are regarded as rather more probable than those tabulated. 



The B,A. Standard Resistance Coils. 



Values in terms of the Original B.A. Unit (1867). 



A B — Platinum- Iridium 



C — Gold-Silver 



D E — Platinum 



FGH & 



lOOSOO 



I OOlSO 



lOOlOO 



1 000 so 



I 00000 



Flat 



Platinum-Silver 



999 so 



1865 



The curves show clearly the remarkable constancy of the coils in the last 

 twenty- four years, even in the case of coils F and G, which had changed 

 very considerably before that time. This constancy is probably the result 

 of the storage of the coils at constant temperature throughout the whole 

 period. On one or two occasions between 1879 and 1888 the coils had 

 been tested in melting ice, and it was considered that this was probably the 

 cause of some of the observed changes. Throughout the whole of the recent 

 observations the temperature has not differed from 16 • 0° by more than i° C. 



