170 ANNUAL REPORT SMITHSONIAN INSTITUTION, 194 



rived from the mean of the values of the resistances of this primary 

 group. 



The present evidence is that the mean value of this group has 

 drifted at the rate of about 1 part per million per year since the 

 group was established in 1908. With the hope of eliminating this 

 drift in resistance standards, J. L. Thomas undertook the develop- 

 ment of a new precision 1-ohm coil at the National Bureau of 

 Standards in 1928 (1).- The manganin resistance after winding 

 was annealed in vacuum at a red heat (550° C.) to remove all 

 locked-up stresses, and special precautions were taken to avoid strain- 

 ing the coil afterward while assembling it in its double- walled annu- 

 lar container. These coils have shown remarkable stability, at least 

 relative to one another. The relative changes within the group have 

 been of the order of 1 part in a million in the past 5 years, compared 

 with similar changes of the order of 10 parts per million in the coils 

 of the older group. In 1932 the 10 coils of the standard gTOup 

 were replaced by coils of the new type. 



THE STANDARD OF ELECTROMOTIVE FORCE 



The national standard of electromotive force was maintained up 

 to 1937 by a group of 20 saturated Weston cells, with many other.s 

 in reserve. At that time 9 acid cells which had shown gratifying 

 stability were added, and 3 cells were discarded. The present pri- 

 mai-y group of 26 cells contains 17 cells that have remained honored 

 members of the group since it was established in 1906. These cells 

 are kept at a constant temperature. The NBS "international unit" 

 of electromotive force is derived from the mean value of the 26 

 cells of the primary group. 



DECISION TO DEFINE THE ELECTRICAL STANDARDS IN ABSOLUTE UNITS 



At the conclusion of the work of the Washington conference of 

 1910 the standards of resistance and electromotive force of the United 

 States, England, France, Germany, Japan, and Russia had been 

 brought into good agreement. But subsequent comparisons showed 

 (hat the standards were slowly drifting apart, and by 1930 differences 

 as large as 1 part in 10,000 were found in the standards of electro- 

 motive force of the different nations, while similar but smaller changes 

 had taken place in the standards of electrical resistance. 



A discrepancy also exists between the "international" electrical 

 units now in force and the mechanical units. For example, if the 

 same amount of power were measured first in terms of the electrical 

 units now in use and then in terms of the mechanical units, the differ- 



* Numbers iu iiaientlieses refer to bibliography at end of article. 



