458 Prof. D. Mendeleeff. Experimental 



(6) For other balances considerably slower, as for the balance of 

 Nemetz : 



P 1147 grms. 1359 grms. 2213 grms. 3442 grms. 



k 0-0192 ings. 0-0234 ings. 0'0481 mgs. 0'0670 mgs. 



D 15 1/0321 1-0392 1-0337 1-0313 



t L5 30-5 sees. 32-8 sees. 33'3 sees. 33'7 sees. 



The decrement D 15 in some balances with increasing load (a) 

 increases as with the above-mentioned balance of Ruprecht, or (6) 

 decreases as with a balance of Collot designed to carry a load of 

 40 kilograms on each pan : 



P grm. _ 



-. _ . v -L. ^J. 4. O. 



409'o grm. 



P . . 29,879 grm. 38,069 grm. 46,259 grm. 62,640 grm. 95'402 grm. 

 D 15 .. 1-0459 1-0316 1-0259 1-0172 1-0103 



t l5 . . 55-9 sees. 53'0 sees. 49'3 sees. 43'5 sees. 35'9 sees. 



In this case the variation of D in relation to p corresponds in first 

 approximation to a hyperbola with an asymptote 1 or (c), a combi- 

 nation of the preceding (a) and (6). 



III. Variation of D 15 and / 15 with the Variation of the Sensibility of a 



Balance I/A;. 



If k increases, all other conditions being the same, then Di 5 and 15 

 decrease as with the balance of Nemetz. (P = 2213 grams.) 



k 0'034 mgrms. 0'055 mgrms. 



t l5 37-3 sees. 30'6 sees. 



D 15 1-0363 1-0286 



TV. Variation of the Time of one Oscillation depending upon the 

 Friction of the Knife-edges (Prisms) against the three Planes. 



When the pans and the movable parts used for suspending them 

 are removed and the balance beam alone is oscillating, there is 

 friction only at the middle knife-edge, and the time of one oscillation 

 is shortened, as with the above mentioned balances of Collot and 

 Ruprecht ; for the latter we have found 



With pans but without 



any load. Beam only. 



P 1231 grms. 678 grms. 



k 0-0310 mgrms. 0'0290 mgrms. 



ti 5 27-3 sees. 15'0 sees. 



D 15 1-013 1-007 



If in the balance of Nemetz the pans are suspended from the beam, 



