16i 



SCIENCE. 



[N. S. Vol. XXI. No. 527. 



try using a foot of different length. The 

 confusion resulting from this lack of uni- 

 formity prompted the French in 1799 to 

 adopt a new unit of length, and remember- 

 ing how surely and elegantly the unit of 

 time is fixed by the rotation of the earth, 

 they sought to make the meter, the new 

 standard of length, permanent and inflex- 

 ible by basing it upon the dimensions of 

 the earth. The meter was chosen to be one 

 ten-millionth part of the distance from 

 the equator to the pole of the earth at a 

 particular meridian, and was fixed in con- 

 crete form as the length of a platinum bar, 

 which has been carefully preserved in 

 Paris. Subsequent and more accurate 

 measurements have given a slightly differ- 

 ent value for the circumference of the 

 earth, so that the meter is known not to be, 

 as originally intended, just one ten-mill- 

 ionth of a particular quadrant of the 

 earth. The meter has, however, not 

 been changed, its value being fixed by the 

 length of the platinum standard and not 

 by the earth. Thus the platinum bar has 

 become the primary standard of length, 

 instead of a secondary standard as was 

 originally intended. This is a happy re- 

 sult, for the difficiilties of comparing a 

 meter with the dimensions of the earth is 

 too great to make the dimensions of the 

 earth of any value as a standard of length. 

 The original standard meter has been 

 reproduced many times in platinum and 

 iridio-platinum, and many of the civilized 

 nations of the earth possess such duplicates. 

 We have two of them at the bureau of 

 standards in Washington, one of which was 

 recently taken to Paris by Mr. Fischer, and 

 rccompared with the standards of the inter- 

 national bureau. The results showed al- 

 most perfect agreement with the compari- 

 son made fifteen years previously, the dif- 

 ference, if any, being not greater than 

 about 0.5 of a micron, that is, 1/50,000 

 inch. This is one part in 2,000,000 of the 



length of the bar and represents about the 

 limit of accuracy obtainable in comparisons 

 of this nature, although the computed 

 probable error of the observations was only 

 .02 of one micron, or less than a millionth 

 of an inch. 



The third fundamental unit, that of mass, 

 has likewise varied in different countries 

 and in different ages. The most widely 

 used unit was the pound, and before the 

 metric system came into use there were him- 

 dreds of dift'erent pounds in use in Europe, 

 differing from country to country and from 

 province to province, and varying also ac- 

 cording to the commodity to be measured. 

 The ancient Roman pound was equivalent 

 to a little less than twelve of oi;r avoirdu- 

 pois ounces, and from it were derived the 

 various Italian pounds, varying in value 

 from the Venice light pound, equivalent 

 to about eleven of our avoirdupois ounces, 

 and the Naples silk-pound and the Milan 

 ligh1 pound of about twelve ounces to the 

 Piedmont pound of about thirteen ounces 

 and the Venice heavy pound of about sev- 

 enteen ounces. There were silk pounds, 

 and chocolate pounds, and table pounds, 

 and goldsmith pounds and medicina 

 pounds ; there were light poimds, an 

 heavy pounds, and half-heavy pounds an 

 extra-heavy pounds. There were pound 

 of 12, 14, 15, 16, 17, 18, 20, 21, 22, 24, 28, 

 30 and 36 ounces, and the ounces had vary 

 ing values in different countries and i 

 different provinces of the same country. 



To remedy this distressing confiisiou the 

 French, in 1799, at the same time the mete 

 was chosen, adopted the kilogram as th 

 unit of mass, fixing it concretely in a cylin 

 drical mass of platinum, which was intend- 

 ed to be equal to the mass of a cubic deci 

 meter of water at the temperature of it 

 maximum density. This, like the meter, 

 was designed to be a natural unit that 

 could be derived originally at any subse- 

 quent time and in any country. But, a 



