306 De/cr'iptlon of a Support appUcahh to Balanets, 



Thefe bodies are ufually fufpended by a very thin metallic wire; but, by placing two 

 veflels as I have done, and fufpending to each of the bafons a wire of metal of the fame 

 diameter, one to fupport the body, and the other to be fimply immerfed in the water, the 

 magnitude of the diameter will have no influence on the accuracy of the experiment. In 

 fa£l,'if we fuppofe the machine to be fo adjufted that at the commencement of the operation 

 the difhes (hail be in equilibrio (which is eafily efFetted by varying the height of the water 

 in the veflels), thefe two vvires will continue to form an equilibrium, when the beam F F" 

 being raifed ftill preferves its horizontal pofition. Whence it follows that if to one of 

 the wires there be fufpended an immerfed body, and in the oppofite bafon there be placed 

 out of the water a weight capable of maintaining the equilibrium, that equilibrium will 

 continue, whatever may be the elevation or depreflion of the whole balance, provided it 

 continues horizontal. For the lengths of the wires raifed out of the water being equal, the 

 ditFerence between the fpecific gravities of the water and the metal will have the fame in- 

 fluence on the two extremities of the beam. It is evident that the fame advantage could 

 not be had by employing only one wire; namely, that which fupports the fufpended body, 

 and that the equilibrium correfponding with a certain elevation, and the horizontal pofition 

 of the beam, will not agree with other elevations and the fame horizontal pofition. 



It muft be obferved that my procefs not only compenfates for the excefs of fpecific 

 gravity of the wires beyond that of the water, but alfo for the adhefion of the fluid and 

 that portion which hangs round the wires, and is raifed with them. 



The whole of the preceding account is applicable to balances fufpended in the ufual clip, 

 or forked piece, terminating in rings for fupporting the fulcrum. But in order to render 

 my fupport abfolutely univerfal, it muft be adapted to a beam alone, the knife edges of 

 which fliall find a proper place to zQ: upon : and in this I have fuccecded, by contriving a 

 forked piece, which hangs from the fork « q, and is capable of receiving the knife edges 

 of any beam whatever. In the drawing this piece is properly placed in the machine, and 

 its conftrudlion is as follows: 



A piece S / (fig. 2.) is pierced at / by a tapped hole, into which the fcrew d d enters to 

 half its length. Another hole made in S, perpendicular to the former, receives the pin^^, 

 (fig. I.) which fufpends the whole lower apparatus, and anfwers to the hole in the upper 

 part of the fork of common' balances. 



Two other vertical pieces r r, Fig. 4, have at their upper part holes not tapped, in 

 which tlie fcrew d ^ may freely move. Thefe upper ends are to be placed at an arbitrary 

 diftance, where they are retained, and fixed by four fcrew-nuts u, each piece being thus 

 confined between two fcrews. A cylindrical rod /; h pafles through the lower extremities 

 of thefe pieces r r, and is fixed by prefling fcrews ; fo that the upper and lower extremities 

 of thofe pieces arc invariably at the fame diftance. 



Each of thefe pieces r r has on its face, which is perpendicular to the direftion of d d 

 and k A,. a long opening ff, and a circular perforation X, (fig. i. and 5.) the lower part 

 of which, is armed with a fmall face of polilhed fteel, upon which the edge of the fulcrum 



reds; 



