534 Wlmher the ExpmtnenU of Count Rutnfird 



contaft with the Ice, and, of courfe, melt it. The Count's ingenuity, never without 

 refources, enabled him to prove completely tliat the ice employed in his experiment was 

 aftually melted in that manner. For when he covered the ice partially with flips of wood, 

 that part which was (haded by the wood was not melted; and when he covered the whole 

 of the ice with a thin plate of tin, having a circular hole in the middle, only the part exactly 

 under the hole was melted. From thefe fa£ts it certainly may be concluded that the Ice 

 was melted by defcending currents of water. 



Bnt the point to be proved is not whether there were defcending currents, tmt whether 

 water be a condu£Vor or not. Now, if water be a non-condu£lor, I aik how the hot 

 water was cooled down to 40° ? Not at the furface ; for the Count himfelf tells us, that there 

 the temperature was never under 108 ° ; not by the fides of the veflel ; for the defcending 

 current in one experiment was exadly in the axis ; and it follows irrefiftibly from the ex- 

 periment with the flips of wood, that thefe defcending currents fell equally upon every 

 part of the furface of the ice ; which would have been impolTible, if thefe currents had 

 been cooled by the fide of the veflel. The hot "water, then, mufl: have been cooled down to 

 40° by the cold water below it ; confequenfly, it muft have imparted caloric to this cold 

 water. If fo, one particle of water is capable of abforbing caloric from another ; that is, 

 water is a conduElor of caloric. After the hot water had flood an hour over the ice, its. 

 temperature was as follows : 



At the furface of the ice - - - - 40" 



1 inch above the ice - - _ _ _ 80 



2 inches -----.- itg 



3 inches -.-..-. 128 



4 inches ----_.. 130 

 7 inches - - - - - - - 131 



Ho\Y is it poCTible to account for this gradual diminution of heat as we approach the 

 i<e, if water be a non-condudor ? The water, it may be faid, gives out caloric at its furface, 

 falls down and arranges itfclf according to its fpecific gravity. If fo, how comes it, that 

 there is only one degree of difference between the temperature at 4 and at 7 inches above 

 the ice ? Thus it appears that the Count's experiment, inftead of demonftrating that water 

 is a non-conduftor, rather favours the fuppofition that it is a conduftor. 



The Count drew, as a corollary, from this experiment, that water at 41° will melt as 

 much ice in a given time as boiling water, when both ftand over the ice. He found this 

 aftually to hold ; or rather he concluded from his iexperiments, aflifted, however, by a 

 little calculation, that water at 41° will melt more ice in a given time than boiling water. 

 It would not be difficult, perhaps, to find fome flaws in the calculation ; but granting the 

 truth of the fa£l:, I do not perceive how it contributes in the fmalleft degree to prove that 

 water is a non-conduftor. For water at 41° (being denfer than water at 32") melts the 

 ice, not by its condu£ting power ; but by a£lually travelling to it particle after particle, 

 and giving out its caloric: whereas, boiling water caa only aiSl by its conducing power, 

 3 till 



