268 Dr. C. R. A. Wright on the Determination of 



Part II. § 54) be affected by any error, the values assigned 

 to the current passing in the present experiments are also 

 erroneous in the same ratio. In the majority of the experi- 

 ments described below the difference in the calculated amount 

 of hydrogen, according as the B.A. unit is assumed correct, 

 or as being ±0*5 or even ±1 per cent, in error, is less than 

 the error of measurement of the gas collected, together with 

 the error due to the rate of " diffusion discharge " not being 

 absolutely constantly equal to the values given above, owing 

 to unavoidable fluctuations of temperature. 



82. By comparing the quantity of gas collected after the 

 passage of a known average current for a known time with 

 the amount calculated from the current and time by the 

 formula 



W= (G-C')t x 



(where w is the weight of the hydrogen, C the average cur- 

 rent passing, Q f the current equivalent to the " diffusion 

 discharge," and t the time in seconds), numbers were obtained 

 according closely together : thus, for example, the observed 

 quantities of hydrogen collected in the experiment cited in 

 § 81 were as follows (voltameter No. 1 used) : — 



Period. Milligramme of hydrogen. 



Commencement to 6 hours. 

 6 hours to 21 hours. 



ZX. mm L*J •■ 



0-004 



0-0105 



0-0055 



Total . . 0-0200 



wdiilst the amount calculated from the formula is 



(0-000002203-0-0000004) x 29 x 60 x 60 x 0-000105 

 gramme = 0'0198 milligramme; 



or sensibly the calculated amount was actually obtained, the 

 difference between the two quantities being only 0*0002 milli- 

 gramme of hydrogen, representing about 0*002 cubic centim. 

 at 0° and 760 millims., an amount well within the experi- 

 mental errors. 



In precisely the same way the following numbers were ob- 

 tained in a series of experiments with the same voltameter 

 (No. 1). The currents are given in microwebers (1 micro- 

 weber =0*0000001, or 1 x 10"? C.G.S. current-unit) in the 

 first column, and the time in hours in the second; the third 

 column gives the value of C^, and the fourth of (C — C')t% 9 

 the two sets of values being given in order that the amount 

 of suppression of hydrogen due to " diffusion discharge " may 

 be rendered manifest. 



