OP THE UNITED STATES. i273 



give only individual results, yet it may be proper to men- 

 tion, approximately, the ratio they give between the length 

 of the toise and of the metre, omitting the toises of Lalande. 

 The comiiinations which they give occasion to make, all 

 give the metre, in parts of the toise, between 0",513162 and 

 ",513137. The Committee of Weights and Measures 

 adopted, in the construction of the metres, the ratio, 0"51317. 

 M, Delambre gives, in the Base Metrique, 0",513111185 at 

 l6io centigrade. 



Description of the .Apparatus for measuring Base Lines. 



In all surveys of considerable extent, the exact determi- 

 nation of the line, whicii forms the base of the whole trian- 

 gulation, is of the greatest importance. 



This line forms the absolute unit on which all future units 

 depend. It is expressed in terms of the unit of Icnglh cm- 

 ployed in its admeasurement ; and the extreme distances of 

 the whole survey referred to it must correspond to the places 

 which astronomical observations assign to them on the earth. 



The measurement of a line may appear simple and easy 

 in common life, where no minute degree of accuracy is re- 

 rpiircd, and where commonly the line itself is of no conside- 

 rable Icnglh. 



In the application to large surveys this forms the most te- 

 dious part of the work; and presents, in its meciianical exe- 

 cution, difficulties, which have always called forth the inven- 

 tive genius of the oj)erators. 



This is not the place to expect a history of the different 

 means employed in determining tliis line, nor to comment 

 on their comparative advantages or dinicultics. 



It may be easily conceived tliat the most mitnite care is 

 required to determine the fundatncnial unit Icnglh of a Ijar 

 or chain to be used in the measurement of a !)ase, from the 



