OP ARTS AND SCIENCES. 293 



« 

 ates the inconvenience of placing the defining lines at the bottom of 

 wells in order to reach the plane passing through the neutral axis, and 

 it enables one to place the terminal lines and the lines marking the 

 subdivisions in the same plane. 



The steel bar prepared in this way is designated i?j. Three sets of 

 graduations are traced upon the polished surfaces of the steel plugs. 

 They are designated as follows : — 



Lines 7?^"* define the meter and the yard provisionally at 62°.0 Fahr. 

 The corrections required for these defining lines having been ascer- 

 tained from a sufficient number of observations, they were applied in 

 tracing the remaining graduations. The lines of the group i?^ vary 

 in width between 0.6 jx and 5 /x. They are therefore well adapted to 

 furnish the data required in order to determine the effect of the size 

 of the lines upon the error of focus under the objective. 



III. TTie Bronze Yard and Meter, designated R^. 



The material of this standard is an alloy known as Bailey's metal. 

 It consists of 16 parts of copper, 2|^ parts of tin, and 1 part of zinc. 

 It is identical in composition with the Imperial Yard, and differs 

 from it only in having the graduations upon platinum-iridium plugs, 

 which have their polished surfaces in the plane of one surface of the 

 bar when it is supported at its neutral points, instead of having them 

 upon plugs of gold situated at the bottom of wells sunk to the plane of 

 the neutral axis. 



This standard forms the basis of the units which the writer under- 

 took to obtain for Messrs. Pratt and Whitney, of Hartford, Connecti- 

 cut, and upon which their system of gauges depends. 



Both the yard and the meter are nearly standard at 62° Fahr. 

 The defining lines of the meter were obtained by transfer from the 

 Tresca meter. The defining lines of the yard were obtained indirectly 

 from the Imperial Yard, in the following manner. 



Early in 1880, Mr. Chaney, Warden of the Standards, did me the 

 great service to compare a yard traced upon the surface of a steel bar 

 with the Imperial Yard. According to the observations of Mr. Chaney, 

 the yard traced upon this bar was .000316 of an inch too short at 62° 

 Fahr. Unfortunately, this bar had been nickel-plated, and the gradu- 

 ations were traced upon the nickel surface. Subsequent investigations 

 revealed the fact that the expansion and contraction of this bar were 



