some Mercury Resistance Standards. 73 



any two is '00012 B.A. unit, or rather more than 1 in 10,000 ; 

 and no appreciable change was produced by passing the mer- 

 cury through the tubes. The observations on No. 39 are not 

 quite so close. On January 6 a distinct change was caused 

 by passing the mercury through ; after that date the results 

 are fairly consistent, and no further alteration was observed ; 

 the mean of the values found after this date may be taken as 

 the resistance of the tube to an accuracy of about 1 in 10,000. 

 The two results therefore are: — 



No. 37 1-01106 B.A. units. 



No. 39 1-01032 B.A. units. 



The observations in 1885 were reduced to Legal Ohms, 

 using the value for the resistance of mercury in terms of the 

 B.A. unit adopted by the B.A. Committee in that year, and 

 based on Lord Rayleigh's experiments. According to this 

 1 Legal Ohm = l-0112 B.A. units. 



If we take these values we have as the resistances in Legal 

 Ohms the following : — 





Benoit. 



K. T. G. 1885. 



E. T. G. 1891. 



37 .... 



.. 1-00045 



•99990 



•99986 



39 .... 



... -99954 



•99917 



•99913 



Comparing the last two columns we see that the tubes have 

 apparently fallen in value relative to the standard Flat in the 

 5 J years by '00004, but this quantity is too small to be really 

 certain of. The results of the experiments therefore prove 

 within this limit that the platinum-silver coil Flat has not 

 altered relative to the mercury-tubes in this interval. The 

 difference between the values found by Benoit and myself 

 depends on the value used for the resistance of mercury in 

 B.A. units. According to Lord Bayleigh *, the resistance of 

 a column of mercury 100 cm. long, 1 sq. mm. in section, at 

 0° O.j is '95412 B.A. unit, and this is the value which has 

 been used in the above. The value found by Mr. Fitzpatrick f 

 and myself for this resistance was '95352. Practically the 

 same number has been obtained by Wuilleumeier J, Hutchin- 

 son §, and Salvioni []. If this number be adopted, then 



1 Legal Ohm=106 x '95352 B.A. unit. 

 = 1-01073 B.A. units. 



* Phil. Trans. 1883. 



t Phil. Trans. A. 1888. 



t Wuilleumeier, Comptes Rendus, cvi., 1888. 



§ Hutchinson, Johns Hopkins University Circulars, 1890. 



|| Salvioni, Rend, della R. Acad, del Lincei, vol. v. fasc. 7. 



