PHYSICS: ROSA AND VINAL 
63 
of five nationalities using three different types of instruments, and yet 
they agree to about 1 part in 100,000. 
The national laboratories were in correspondence on the matter of 
specifications when the war in Europe began. Since then the Bureau 
of Standards has prepared and published its specifications for the volta- 
meter and it will recommend these for adoption when it is again possible 
to do so after the restoration of peace. 
The specifications which the Bureau of Standards has formulated 
have been made as broad as is consistent with work of the highest 
accuracy. They do not specify any particular form of voltameter, 
but rather the conditions which must be fulfilled by the voltameter. 
The concentration of the solution is given rather wide limits because 
it has been found that identical results may be obtained with pure 
solutions over a considerable range. 
An important question as to the purity of the silver obtained has only 
recently been settled. The impurities are very small, as indeed pre- 
vious observers, including Van Dijk in Holland, Jaeger and von Stein- 
wehr in Germany, and Boltzmann in Austria, have previously shown. 
The Bureau of Standards has found the average amount of foreign 
matter in its standard deposits to be about 0.004%. It is so uniform 
in the different deposits that the results given by the voltameter as a 
measuring standard are sufficiently accurate without the necessity of 
determining the foreign matter in each individual deposit. Without 
the greatest precautions the error introduced in attempting to deter- 
mine the inclusions will be greater than the total amount of the inclusions. 
In some of the experiments at the Bureau of Standards our absolute 
current balance has been used in connection with the silver voltameter. 
This has permitted the determination of the absolute electrochemical 
equivalent of silver. It was found to be 1.11804 mg. per coulomb, 
but when corrected for the inclusion of foreign material this becomes 
1.11800 mg. This figure is exactly the value assigned to this constant, 
although at the time it was adopted by the London Conference, eight 
years ago, the best information available indicated a higher value by 
about 30 parts in 100,000. In Holland, Prof. Haga, has obtained 1.11802 
using a tangent galvanometer. We beheve that the closest figure that 
can be assigned at the present time to this constant is the round num- 
ber 1.11800 mg. per absolute coulomb, and is this precisely the value 
fixed by international agreement in 'international coulombs.' 
Since one absolute coulomb of electricity deposits 0.00111800 gram 
of silver, the number required to deposit a gram-equivalent of silver, 
