500 Mr. A. W. Warrington on 



the liquid ; the temperature of the latter is then allowed to 

 change very slowly until a point is reached at which no 

 difference can be detected between the specific gravities of the 

 liquid and of the hydrometer. 



Fig. 2 shows the form used for solids. Determinations are 

 made in which the instrument has the same specific gravity 

 as water (1) loaded with mercury alone, (2) loaded with the 

 solid and mercury, care being taken that the temperature in 

 both determinations is about the same. 



The advantages claimed are : — 



1st. That the method is exceedingly sensitive. 



2nd. That the method is applicable within a considerable 

 range of temperature without appreciable change in 

 accuracy. My experiments ranged from 3°"5 to 35°. 



3rd. All the weighings are made in air and under conditions 

 which permit of repetition. 



4th. The elimination of the use of a suspending thread 

 should be beneficial. 



For liquids the results are accurate to 1 in 1,000,000 ; for 

 solids to 1 in 100,000. 



I have used the proposed method to determine : — 



(1) The coefficient of expansion of a hydrometer made of 

 normal Jena glass. 



(2) The temperature at which the density of water is at its 

 maximum. 



(3) The specific gravity of a piece of quartz. 



The Apparatus. 



To obtain and maintain the required temperature the 

 experiments were performed in an incubator. The incubator 

 was made of steel plate, coated with lead, furnished with a 

 double-walled isolated chamber and ha vino; two doors made of 

 the best plate-glass. The outside of the incubator was 

 covered with felt, and the bottom, which was funnel-shaped, 

 was made of copper. 



The top of the incubator was perforated with holes, to admit 

 of the thermometer, the connecting rods of the stirrer, and a 

 contrivance to clamp the hydrometer while stirring. 



The stirrer was made of a flattened ring of silver, and was 

 connected by two vertical rods to a cord which passed round 

 a pulley screwed into the ceiling. 



Through the centre hole in the roof of the incubator passed 

 a tube open at both ends and joined at its lower end to an 

 inverted glass funnel. Before stirring the liquid the funnel 

 was lowered until the hydrometer was clamped between it 

 and the bottom of the beaker, When it was desired to 



