STANDARDS OF RESISTANCE, 1865-1932 



429 



which can be measured most accurately), assuming they were accurate in 

 1888 and estimating from the records the changes which had occurred since 

 that time. The value obtained was — 



I international ohm ^= i •013 67 B.A. unit (1903), 



and this value has been used at the Laboratory for some years. It is to be 

 noted that, in the light of our present knowledge, neither of the above values 

 represents the original B.A. unit. It follows from Table VI that this unit 

 is determined as nearly as can be ascertained by the relation — 



I international ohm ^ i -014 39 B.A. unit (1867). 



Table IX. 

 Relative Values of the Mean B.A. Unit. 



Of course since 1903 mercury tubes have formed the fundamental 

 standards, and the more recent values have not influenced any actual 

 measurements. 



Appendix. 

 The Manganin Resistance Coils. 



In 1895 the British Association obtained from Otto Wolff of Berlin 

 resistance standards of the pattern used at the Physikalisch-Technische 

 Reichsanstalt, and described by Dr. St. Lindeck in the British Association 

 Report for 1892. The wire is of manganin insulated with silk, and after 

 winding the coil was heavily coated with shellac varnish and then baked. In 

 use the coil is completely immersed in oil, and the outer case was perforated 

 so as to allow the oil to come into contact with the wire and thereby control 

 its temperature. Four such standards each of nominal resistance 1 ohm 

 were obtained, and in due course deposited at the National Physical 

 Laboratory. The temperature coefficients of these coils were found to be 

 only I to 2 parts in 100,000 per 1° C, and it became a comparatively 

 simple matter to compare them with an accuracy of i part in a million. 

 They therefore largely displaced the old B.A. coils for general standardising 

 work. 



The history of these coils is shown in the following table and chart. Two 

 of them, Nos. 1690 and 780, have shown remarkable constancy, and even 

 at the present time they are probably the most satisfactory standard coils 

 which the Laboratory possesses. In 191 1 each of these two coils was pro- 

 vided with an oil-tight case, which was filled with paraffin oil and her- 

 metically sealed. The improvement in performance which followed this 

 treatment is very striking, and leaves little doubt that most of the changes in 

 resistance previously observed were due to the variations in atmospheric 

 humidity. The coils numbered 381 and 147 were not hermetically sealed. 

 The change in resistance of 381 from 1908 to 1932 is of the same order as 

 that shown by most of the older B.A. coils ; the change in 147 has been very 



