Propagation of Heat in Liquids. 283 



tained in the annular vessel in consequence of their 

 acquired heat and levity, the progress of the heat from 

 the wooden tube to the sides of the large vessel K L, 

 was interrupted, and all the heat that passed through the 

 sides of the wooden tube was by these means turned aside 

 in such a manner that it could no longer disturb the 

 progress of the experiment, nor affect the certainty of its 

 results. 



Before I proceed to give an account of the result of 

 this inquiry, I shall take the liberty to recall the atten- 

 tion of the Assembly to the most important circum- 

 stances of the experiment. 



On pouring boiling water in a small uninterrupted 

 stream into the funnel, the hollow conical point which 

 terminates the vertical tube belonging to the vessel N O 

 was heated, and kept at a constant temperature little 

 under that of boiling water. 



This point was surrounded by a small quantity of 

 water contained in the cavity of the lower part of the 

 wooden tube, and as this water could not change its 

 place nor be displaced by the surrounding cold water, 

 being enclosed and protected by the sides of the wooden 

 tube, it would necessarily become very hot in a short 

 time. 



But this small quantity of hot water lay immedi- 

 ately upon a stratum of 'cold water, which separated it 

 from the bulb of the thermometer, placed directly under 

 it at the distance of only half an inch. 



If heat could pass in the water from above downwards, 

 it would no doubt pass from the lower stratum of hot 

 water contained in the open end of the wooden tube to 

 the bulb of the thermometer, which lay immediately 

 below it and at so small a distance. 



