of Liquids disposed to unite with exch other. 32 i 



After the little apparatus mentioned above had been 

 left to itself for twenty-four hours, I entered the cellar, 

 taking a light in order to note the progress of the ex- 

 periment, and I found that the little ball had risen 3 lines. 



The next day, at the same hour, I observed the 

 ball again, and I found that it had risen about 3 lines 

 more ; and thus it continued to ascend about 3 lines a 

 day for six days, when I put an end to the experiment. 



I afterwards made nearly similar experiments with 

 saturated aqueous solutions of nitrate of potash, car- 

 bonate of potash, and carbonate of soda. In each of 

 these experiments the surface of the saturated solution 

 was covered with a layer of distilled water 3 inches in 

 thickness, but the surface of this layer of water was 

 not covered by a layer of olive oil ; it was exposed to 

 the air, and this circumstance was, without doubt, the 

 reason that the daily results of a single experiment were 

 not always the same two days in succession. 



The little ball of oil of cloves, which served as an 

 index to mark the progress of the mixture of the satu- 

 rated solution with the distilled water resting upon it, 

 ascended usually 2 or 3 lines in twenty-four hours, but 

 sometimes I found that it had left its position and had 

 risen to the very surface of the water. 



In such cases it was, without doubt, borne upwards 

 by the air which it had attracted from the liquid ; for 

 when I allowed a fresh drop of the same oil to fall into 

 the water, I found that it never failed to descend im- 

 mediately in the liquid, and to take up its position 2 or 

 3 lines above the level at which, the day before, I had 

 found the ball which had now left its place. 



In the experiments made with solutions of carbonate 

 of soda and carbonate of potash, the balls of oil 



