252 Journal of Agricultural Research voi. iii, N0.3 



The density determinations were made by what is commonly known 

 as the method of hydrostatic weighing. By this method a sinker or 

 plummet of known mass and volume is suspended in the liquid under 

 examination and weighed. The density of the liquid is then calculated 

 by means of the equation 



A y^ 



in which Z)( = density ' of the liquid at the temperature t; 

 W^= weight of sinker in vacuo; 

 w = apparent weight of sinker in liquid; 

 ,o=air density; 

 8.4= assumed density of brass weights; 

 1^2 = volume of sinker at temperature t. 



This method of determining densities, though very accurate when used 

 under suitable conditions, is open to criticism when applied to a non- 

 homogeneous liquid, such as milk or cream. There is, of course, a con- 

 stant tendency for the fat of the sample under investigation to separate 

 out and rise to the surface and for the heavier components to sink to the 

 bottom. The density of a nonhomogeneous liquid determined by this 

 method will therefore tend to be too low if the sinker is suspended near 

 the surface of the liquid and too high if suspended near the bottom. 

 The difficulty, however, may be largely overcome by the frequent stirring 

 of the sample and still more effectively by the use of a sinker of such a 

 length as to reach nearly from the top to the bottom of the liquid. The 

 average density of the displaced liquid will then be nearly the same as 

 the average density of the whole mass of liquid, and the density deter- 

 mined will be nearly the average density of the sample. 



DESCRIPTION OF APPARATUS 



The apparatus employed in making the density determinations is de- 

 scribed in publications of the Bureau of Standards.^ Its essential parts 

 are as follows : 



The sample to be tested is placed in a tube surrounded by a water 

 bath kept in constant circulation. This bath is in turn surrounded by 

 another, which is also kept in constant circulation. The temperature of 

 the outer bath may be kept constant or varied at will by the adjustment of 

 the energy through an electric heating coil and by the flow of refrigerat- 

 ing brine in a coil, around which the water in the bath may be made to 

 circulate. A sinker of known mass and volume is suspended from one 

 arm of a sensitive balance placed above the other apparatus. The tem- 

 perature is read from two mercury thermometers placed in water in a 

 second tube similar to that in which the sample is placed, the two tubes 

 being placed side by side within the inner circulating bath. The ther- 

 mometers are suspended from a movable cover, which may be rotated to 

 bring them successively into position for reading. 



' Throughout this paper the term "density" is used to denote the mass per unit of volume, and is ex- 

 pressed in crams per miUiliter. The densities are therefore numerically the same as specific gravities ia 

 terms of water at 4^ C. as unity. 



2 Bearce. H. W. Density and thermal expansion of linseed oil and turpentine. . . U.S. Dept. Com. 

 and Labor, Bur. Stand.. Technol. Paper 9. 27 p.. igiz. 



Osborne. N. S.. McKelvy. E. C. and Bearce. H. W. Density and thermal expansion of ethyl alcohol 

 imd of its mixtures with water. U. S. Dept. Com., Bur. Stand. Bui., v. 9, no. 3. p. 327-474. li fig . 1913. 



