wenner: wire resistance standards 417 



less than that indicated by the table since the error in comparing 

 a 10,000-ohm standard with a 1-ohm standard, as formerly made, 

 may have amounted to several parts per million. 



To what extent the mean value of the reference set has remained 

 constant and may be expected to remain constant, or the mean 

 value of any other reference set may be expected to remain con- 

 stant, is a matter of prime importance. The history of a standard 

 or set of standards giving the resistances on different dates should 

 in most cases indicate the further changes to be expected. 



The standards constituting the reference set were made at 

 different times, by different persons using wire made under differ- 

 ent conditions. Therefore, when we find their relative values 

 and their mean values as compared with a large number of stand- 

 ards of different denominations remaining remarkably constant, 

 it is reasonable to suppose that their mean value is also remaining 

 very constant. This supposition, however, is not entirely justifi- 

 able since the resistance of most of the standards may conceiv- 

 ably be changing at an approximately uniform rate, and if so the 

 intercomparisons made between the various standards furnish no 

 means of detecting such a change. We are therefore desirous 

 of knowing more about the constancy of the mean value of the 

 reference set than can be obtained by comparing standards, in all 

 of which the resistance material is manganin. It is therefore 

 desirable to construct additional standards using other resistance 

 materials. If two or three sets of standards each of a different 

 resistance material are found to remain relatively constant, then 

 there is much more justification for assuming the mean value of a 

 selected set of standards as constant. 



In this a beginning has been made in that a few standards in 

 which the resistance material is the new alloy "therlo" have been 

 under observation for more than a year. 



The alloy is one of a number developed by Wilbur B. Driver 

 and has many advantages for use as a resistance material in stand- 

 ards and various kinds of resistance apparatus. The indications 

 are that standards in which it is used will be found fully as reliable 

 as those in which manganin is used. In such standards the 

 resistance increases for a few months following the annealing but 



