268 The Temperature of Water 



how the existence of any invisible mechanical operation, 

 the progress of which does not immediately fall under 

 the cognizance of our senses, can ever be demonstrated. 

 As the experiment made with the ball heated in boil- 

 ing water appeared to me to be very interesting, I 

 repeated it twice, and its results were always nearly the 

 same. The mean results of these three experiments 

 were as follows : 



By comparing the mean results of these experiments 

 with the mean results of those in which the ball was at 

 the temperature of 60 or less, we may see how much 

 more rapid the communication of heat in the cold water 

 from above downwards was when the metallic ball was 

 relatively cold than when it. was much warmer ; but we 

 must not consider of too much importance the deter- 

 mination of the relative rapidity thus made, because it 

 is more than probable that it was not till after the 

 conical metallic point had been considerably cooled by 



