ON THE COMPARISON OF LINE AND END 
STANDARDS. 
BY 
Louis Albert Fischer. 
[Read before‘the Philosophical Society of Washington, May 28, 1898.] 
The most formidable errors in the comparison of line 
standards are due to the unknown temperatures of the stand¬ 
ards, but with proper facilities, such as a room of practically 
constant temperature and specially designed comparators, 
results may readily be obtained the probable errors of which 
will not exceed the - go - ooo ' oo - part of the standards, or 0.2 of a 
micron in the case of a metre. In addition to the temper¬ 
ature, other difficulties are, however, presented when effort 
is made to determine the relation of an end to a line stand¬ 
ard—difficulties due to the indirect methods that must be 
employed and which unfortunately not only cause greater 
accidental, but introduce constant errors as well. 
The method that has been employed in the more recent 
and important comparisons of this kind is that suggested by 
Fizeau. It was used at the International Bureau of Weights 
and Measures in 1881* to determine the relation of the Metre 
of the Archives to the Provisional metre, with which the 
present international and the various national prototypes 
were afterwards compared. 
It was also used in this country in 1889 by Mr. 0. H. 
Tittmann t to compare the Committee metre with the Rep- 
sold metre belonging to the United States Army engineers. 
* Travaux et Memoires du Bureau International des Poids et Mesures, 
Tome X. 
f Bulletin No. 17, U. S. Coast and Geodetic Survey. 
35—Bull. Phil. Soc., Wash., Vol. 13 
(241) 
