JANUARY 5, 1912] 
precision of measurement increase, the na- 
tional standardizing laboratory must main- 
tain so high a plane of excellence that those 
best qualified to judge have the fullest con- 
fidence in its fundamental standards and 
in the accuracy of its work. The endeavor 
to improve these standards is not merely a 
matter of doing patiently the same thing 
over and over again. It means a whole 
train of investigations, through which an 
intimate knowledge of the behavior of these 
standards is obtained, disturbing influences 
removed or taken account of, impurities in 
the materials eliminated, and measure- 
ments made with what a few years ago 
would have seemed almost impossible aceu- 
racy. The by-products of such researches 
are often of great value, and become useful 
in many other directions. 
To illustrate, let me refer to the silver 
voltameter, the concrete standard of elec- 
trie current. In 1908 it was believed by 
many that the chemistry of the Rayleigh 
voltameter was simple, and that the most 
recently published work upon it had 
cleared up a mystery of long standing, rel- 
ative to the disagreement of the Rayleigh 
and Richards forms of voltameters. Further 
study at the Bureau of Standards showed 
complexities due to the presence of filter 
paper which astonished chemists, and three 
years of continuous work have not ans- 
wered all the questions which have arisen 
as to the effect of traces of organic im- 
purity or of traces of acid or alkali, in the 
salt, or slight variations in the physical 
condition of the anode, or the volume and 
concentration of the electrolyte, or the 
density of the current, or the influence of 
dissolved gases. In addition to excellent 
facilities for purifying materials and ma- 
king chemical determinations, and an un- 
surpassed equipment for measuring the 
current and weighing the deposits, the 
microscope and ultramicroscope have been 
SCIENCE 11 
brought into service. And although the 
outstanding discrepancies are only a few 
thousandths of one per cent., we have not 
felt justified in closing the work until the 
voltameter as a _ scientific instrument 
should be as thoroughly understood as 
possible. 
The construction of standard cells has 
been beset with similar difficulties. The 
preparation of the materials has been 
studied with great pains, and hundreds of 
cells have been set up and carefully stud- 
ied. To test their portability, they have 
been carried from country to country and 
around the world, and even sent through 
the mails to Europe and elsewhere. To 
try to determine the source of small differ- 
ences between different lots of cells, and be- 
tween different cells of the same lot, one 
component at a time has been varied, and 
materials prepared at different times and 
in different countries have been used side by 
side. The result has been a considerable 
improvement in standard cells, so that for 
most purposes they are satisfactory, but as 
standards there is still room for improve- 
ment. 
One reason for desiring more perfect 
conerete electrical standards is to facilitate 
absolute measurements. We now know the 
value of the absolute ampere better than 
the value of the absolute ohm, but during 
the next few years the absolute ohm will 
probably be realized as well as the abso- 
lute ampere now is. Our international 
electrical units are now so well fixed that 
one can measure electrical power in inter- 
national watts with great accuracy. With 
a better knowledge of the absolute ohm, 
we shall be able also under proper condi- 
tions to measure electrical and mechanical 
power in watts or in absolute units by 
means of electrical instruments with high 
precision. 
The work at the Bureau of Standards 
