Joly — On a Hydrostatic Balance. 349' 



increase ; the sensibility increasing approximately as the square 

 root of the power of the balance. Thus, if we double the diameter 

 of the wire, the balance will now indeed indicate nothing smaller 

 than double the least weight formerly causing displacement ; but, 

 on the other hand, we may assume a quadrupled carrying capacity. 

 This leaves out of consideration the effect of viscosity of the 

 liquid. 



The effect of viscosity will hardly be to reduce the sensibility, 

 but rather to render more tedious the use of floats having large 

 displacements. As, however, the tangential resistance to the 

 motion of a solid surface, in the act of communicating a shearing 

 strain to a liquid, is proportional to the extent of surface, and as this 

 area increases at a slower rate than the volume inclosed by it, it 

 appears that the tediousness attending operations is, again, not 

 fairly assumed to be an attendant disadvantage which increases 

 proportionally with increase of power of the balance The effect is 

 indeed, probably, complicated by the presence of currents or eddies 

 in the liquid. 



As regards the effects of solid friction, contact between the 

 movable and immovable parts might, indeed, be altogether 

 avoided. Thus we might attach the wire externally to a flat 

 cantilever, or flat spiral spring, so that it is retained in the centre of 

 the tubulure by the horizontal rigidity of the spring, while the 

 spring may possess such small vertical rigidity as not to interfere 

 with the sensibility of the balance. It will be seen, however, from 

 the figures, that this plan is not resorted to. It appears indeed 

 unnecessary to do more than guard against contact down the wall 

 of the tubulure ; and this is provided for in the little projecting 

 collar placed at the point where the tubulure meets the wider 

 nozzle. The diameter of the passage here provided for the wire is 

 about 1*5 mm. ; the tubulure is about 3 mms. in diameter. The 

 edge of the collar is sharpened to a knife edge all round, but just 

 burnished smooth. With this arrangement, if the precaution be 

 taken of using a smooth piece of wire, there appears but little 

 retardation due to friction : this, doubtless, is partly due to the 

 position of the collar within the liquid, the liquid acting as a 

 lubricant. The effect of substituting a collar of burnished agate 

 for the brass collar has been tried as in the balance, fig. 2, but 

 with hardly appreciable gain in freedom. This little balance (fig. 2) 



