2i8 American Quarterly Microscopical Journal. 



series of comparisons made at the Coast Survey Office during 

 the Summer of 1878 gave 67.7°, 



The bar which I have for the present adopted as a standard 

 is of steel, and has the yard and its subdivisions on one side of 

 the edge, and the meter and its subdivisions on the other. The 

 bar is 40^ inches long, i^ inches deep, and ^ inch wide. 



COMPARISON OF THE STEEL YARD WITH THE STANDARD. 



Both the steel bar and the brass standard were placed in a 

 wooden trough, with their graduated surfaces nearly in the 

 same horizontal plane. The stops were set to correspond 

 nearly to the end lines of the standard, after the trough had 

 been filled with water having nearly the temperature of 62° F. 

 The temperature of the room was then brought up to such a 

 point that after several hours the thermometer immersed in 

 water indicated 62", The following comparisons were then 

 made : 



146=^4- .\b ■ 

 \2,2=a-\- .lb 

 6$=a-\-2.ib 

 4-j^a + 4.8b 

 25=a + 6.sb 

 ^i=a + 6.sb 

 Solving by least squares, we get : 



^=+131 

 b= — 17 

 For 62' we have for the error (E): 

 E=<z4-i.o6 



=131 — 17=4-114 

 At 62", therefore, the steel yard is : 



.00114 inch too long. 



