]34 0N FLUIDS. 



and of Murray, we are indebted for an inyeftigation of the 

 Count's opinions, the refult of which feeras to be, that fluids 

 are not abfolute non-condu6tors of caloric, 

 Experiments of The experiments of Dalton and Thomfon have proved, that 

 ftn^NicSfolJl the a PP earances of currents, fuch as defcribed by Rumford, 

 and Murray, may be often illufory; and from thofe of Nicholfon, and from. 

 confidered by Murray's firit experiments, we have flrong reafons for fap- 

 pofing, that the temperature was affected by the conducting 

 power of the fluids employed; but in my opinion the experi- 

 ments of Murray with a cylinder of ice, are the moft com- 

 plete demonftration of this contefted point* In a late paper, 

 inferted in the Tranfactions of the Royal Society of London, 

 Count Rumford endeavours to obviate thefe objections to his 

 hypothecs, in his ufual ingenious manner. 

 It it not appre- Even admitting that in your experiments the caloric was 



heeded that the tranfnaitled folely by the containing veiTel (an opinion by no 

 experiment of J J ° v r j 



Murray could be means probable), and that currents, fuch as Rumford defcribes, 

 affefled by cur- j, ave all the efTed he attributes to them in certain cafes ; ftill 

 the experiments of Mr. Murray appear to me incontrovert- 

 ible. It was not, therefore, without furprife, that I ob- 

 ferved him ufe the following argument to invalidate their re- 

 fults : " When that veflel is conftrucled of ice, the flowing 

 •'• down of the water, refulting from the thawing of the ice, 

 " will caufe motions in the liquid, and confequently inaccu- 

 " racies of ftill greater moment ;" viz. than thofe produced 

 by the conducting powers of the fides of the vefiel. Now the 

 melting of the ice could affect the thermometer only by being 

 itfelf heated, and then trickling down the fides of the cy- 

 linder of ice. But I apprehend, the water refulting from the 

 melting of the ice could not gain a higher temperature than 

 32° F. while it remained in contact with ice. li we mix even 

 equal parts of water at 172° and ice, we do not find that the 

 temperature of the mixture is above 32°. If fuch a large 

 quantity of water cannot maintain its temperature in contact 

 with ice, can we fuppofe that fuch a fmall quantity as was 

 formed could rife to a higher while trickling down the fides 

 of a thick cylinder of ice. 

 and certa'tniynot But even this explanation of the phenomenon advanced by 

 carvinaveffcl Count Rumford, is entirely inapplicable to the experiment 

 of ice. with mercury; for the drops of water formed could not pof- 



fi'jly fink in a fluid fo much more denfe, nor throw it into 

 currents which could reach the thermometer. 



3 Befides 



