306 Lieutenant Bartlett on the Expansion 



burgh Encyclopaedia, article Expansion, we find the decimal 

 .00000944' to express the same for hammered copper. From 

 these data, the actual length of the measuring rod was calculated 

 for each experiment, knowing its length at 60° FaTirenheit. 

 But to abridge the calculation, the difference in length between 

 the stone and measuring rod, as shown by the wedge W, was 

 substracted from the length of the rod, before making the re- 

 duction for the temperature of the latter. The length of the 

 copper part, and that of the wooden part, were calculated sepa- 

 rately, on account of their different expansibilities. The result 

 of this calculation is the following Table. 



It is probable that many of the discrepancies here noticed 

 were owing to the hygrometric state of the stone, and perhaps 



