■366 Report concerning the new Metrical Syjiem. 



weight. The delicate and laborious tafk of determining the unity of weight was en- 

 trufted by the National Inftitute to Lefevre Gineau, together with Fabroni of Florence; 

 and a fpecial commiflion was afterwards employed in examining all the jegifters of ob- 

 fervation and experiment, and verifying all the computations. The unity of weight muft 

 necefTarily confift of fome folid, the magnitude of which (hall be determined from the 

 lineal, meafures of the fyftem, and the material of which (hall be fuch as can at all times be 

 procured of one and the fame uniform denfity. The magnitude of the folid adopted for this 

 purpofe was the cube of the decimeter, and the fubftance itfelf diftilled water. ,. Hence 

 the experiment was reduced to determining the relation between the new unity of weight 

 and thofe of the ancient fyfliem, by an aftual experiment with fuch a folid of water. Ex- 

 perimental philofophers are aware that there are two methods of doing this; that is to 

 fay, by a£tually weighing the water contained in a veflel of known dimenfions, or other- 

 wife by finding the weight of the quantity of water difplaced by a folid, of which the di- 

 menfions muft alfo be known. The latter method was preferred, and with juftice, on the 

 prefent occafion. 



A hollow cylinder of brafs, internally fupported by ^dge-bar-work, was conftrufled by 

 C. Fortin, together with a guage capable of afcertaining longitudinal meafures, to the pre- 

 cifion of the four thoufandth part of a line of the ancient meafure. On the bafes of this^ 

 cylinder were drawn twelve diameters, interfering each other at equal angles, and each, 

 diameter lying in the fame plane with a correfpondent diameter upon the oppofite bafe. 

 Three concentric circles were drawn on each bafe, alfo correfponding by pairs with each other. 

 The points of interfetStion, including the centre itfelf, were confequently 37 ; and by mea- 

 furing all the feveral diftances between each of thefe points, and its correfpondent point on, 

 the other bafe, the true figure of thefe boundaries, and mean length of the cylinder, were 

 then deducible. Eight circles were alfo defcribed on the convex furface, and twelve' riglit 

 lines were drawn joining the extreme points of the diameter that had been drawn on the 

 bafes. Thefe lines and circles afForded ninety-fix interferons, to every pair of which the 

 guage being duly applied, gave the meafure of forty-eight diameters. From all thefe mea- 

 iures, fuitably reduced, it was found that the volume of the cylinder at the temperature oF 

 17,6 of the centigrade thermometer was 0,0112900054 of the cubic metre, or more thaiv. 

 eleven times the magnitude of the intended unity. 



The balances ufed for weighing were extremely accurate. One of thefe, charged with ra- 

 ther more than two pounds poids de marc, in each bafon, fhews the millionth part of the 

 weight ; that is to fay, one fiftieth of a grain, and it turns with one tenth of a grain when, 

 each arm is loaded with 23 pounds. 



The weights were arbitrary, though nearly the intended ftandard, it being of more confe- 

 quence that they fhould be adjufted to the moft precife equality, than that they fhould agree 

 with arry definite weight. The fub-divifions were in the decimal order. 



The cylinder was made hollow, in order that it might load the balance as little as poffible^ 

 under a given magnitude, but it was neceflary that it fliould be heavy enough to fuik in water. 



I There 



