168 On fhe Conttadion of Water lij Heat. 



Though these experiments, and some others of a similar 

 nature, carried conviction to my mind, and perfectly satis- 

 fied me respecting the reality of the anomaly of water, I 

 determined to vary somewhat the mode of making the ex- 

 periment, so as to obtain still more striking results. 



For the fifth experiment, I used an apparatus which 

 consisted of a still taller jar. It was 2] inches high, and 4 

 in diameter. I adjusted at the middle of its height a per- 

 forated bason of tinned iron, 2 inches in depth, and 10 in 

 diameter. As this bason embraced the middle of the jar, 

 1 could, by filling it with hot water, or a frigorific mixture, 

 apply heat or cold to the middle portion of the fluid in the 

 jar, and thence, by the thermometer, learn what course 

 the heated or cooled fluid should take. 



Experiment V. 



I filled the jar with water at 32". I placed it upon seve- 

 ral folds of thick carpet, previously cooled to the same de- 

 gree. The air of the room going from 33" to 35", I intro- 

 duced two thermometers, as in experiments first and se- 

 cond. I then poured water of temperature 6S° into the 

 bason, and by means of a spout arising from the side of it, 

 and a syphon connected with a reservoir of water at the 

 temperature now mentioned, I renewed the contents of the 

 bason frequently, but without causing any agitation. 



Bottom. Top. Air. 



At coimmencement 32" 32° 33—33 



In 10 minutes 35 32 



— 15 r 36— 32 



— 20 • 36+ 32 



— 25 37 33 



— 30 •• 38 33 *.iff* From this time I 



— 38 38 -f- 33 charged the bason with 



— 45 ■ — - 39-T 33 water of temperature 



— 50 39+ 44 88", and renewed it fre- 



— 55 39+ 43 quentjy. 



— 60— • 39+ 48 



Nothing cm be more decisive with regard to the question 

 in dispute, than the particulars of this experiment. Heat 

 is applied to the middle of a colunm of ice-cold water. 

 The heated portion has an equal share of the column of 

 cold fluid above it and beneath it. There is nothing to de- 

 termine its course in one direclion or another, excepting its 

 actual change of density. 



The thermometer evinces that the waVm current sets 

 downwards, and carries the increased temperature to the 

 bottom. There, this instrument indicates the successive 



rise 



