CONTRACTION OF WATER BY HEAT. $51 



prifing, that the bottom of the fluid was not apparently affected rauft have beta 

 in its temperature by the ice which fo long occupied its fur- JJJ^JjL 

 face. It might be expected, though no cold currents de- 

 fended from above, that the caloric (hould be conducted from 

 below, and that the temperature fliould by that have been re- 

 duced *. I fuppofe that the caloric did pafs from the lower 

 ftrata upwards, but extremely flow, by reafon that fluids/ as 

 Count Rum ford taught us, are exceflively bad conductors of 

 heat, and fo very flowly, that the caloric entered from the 

 atmofphere with fufficient quicknefs to prevent any depreffion 

 of temperature below the 39th degree. 



This experiment, I may conclude with remarking, is very 

 well calculated to exhibit the error of the popular opinion, that 

 " heat has a tendency to afcend." 



* ANNOTATION, by the Author, f 



This experiment may perhaps be thought to give counte-The opinion »f 



nance to the opinion of the very ingenious Count Rumford, c ° un J- R um f °r* 

 1- . , ,«.,i. ..i r that fluids can- 



that fluids cannot conduct heat, and that no interchange ot not con d U cT: 



heat can take place between the particles of bodies in a fluid heat from par- 

 irate, feeing that for two days the fluid at the bottom of the tic,et0 ^^ 

 veflel never fell below 39°, though the furface was at 32°. 



From the circamftances detailed in his feventh eflay, the 

 Count concluded, that heat cannot defcend in a fluid. From 

 v the prefent, it might with equal juftice be inferred, that heat 

 cannot afcend. 



Had I not the fulleft conviction that this celebrated plulofo- appears to beh»~ 

 pher has puthed his ideas too far, I might be difpofed to con- accuracc « 

 tider this experiment as according well with the hypothec's. 



Soon after the interefting fpeculations of the Count ap- 

 peared, I began to investigate the fubjeel ; and, by a pretty 

 long train of experiments, which I have annually taken an 

 opportunity of detailing in my lectures, fatisfied myfelf that he 

 affigned to fluidity a character that does not belong to it. 

 Though fmce the date of thefe experiments, the public has 



f As this note fubjoined at the foot of the page after the words 

 temperature Jhould by that have been reduced, in the original, is of 

 fuch considerable length, I have taken the liberty of putting it in 

 die fame type, as the text.— N. 



become 



