Chap, xix.] SALJMETER. 203 



lower surface for carrying bodies is needed here. 

 The stem, rising from the blown-out part, carries a 

 plate for weights, as in Nicholson's hydrometer. 



These hydrometers are of constant volume, but of 

 variable it-eight, because they are always immersed to 

 the same depth, and displace always the same volume 

 of liquid, the weights being altered to accomplish this. 

 Another type of hydrometer is the reverse of these, of 

 constant weight, but variable volume, where the instru- 

 ment is always loaded to the same extent, and the 

 specific gravity of different fluids is indicated by the 

 depth to which the instrument sinks. If a hydrometer 

 of this kind is put into water it sinks to a mark 

 on the stem. It must sink to the indicated extent 

 before it displaces sufficient water to give an upward 

 pressure equal to the weight of the instrument. 

 If it is now put into a fluid of less specific gravity 

 it will sink farther, because the same volume of this 

 fluid does not create sufficient upward pressure, and a 

 greater volume is required. If put, on the other 

 hand, into a fluid of greater density, the same volume 

 of this fluid gives rise to a greater upward pressure 

 than the weight of the hydrometer ; consequently 

 the instrument rises for some distance 

 higher out of the water than the mark, 

 because a diminished volume gives the 

 required upward force. Such an instru- 

 ment is shown in Fig. 98. It is made 

 of a glass tube, one part being blown 

 out, and terminated by a small bulb 

 containing mercury. On immersing it 

 in a liquid it floats upright, having sunk 

 to a distance that can be read off by Fig iJf- t 7 r Sali " 

 means of the marks on the stem, the 

 distance varying -with the density of the liquid. 

 The graduation of the instrument must be performed 

 empirically, however. Thus, let such an instrument 



