Of WATER by BEAT. 387 



would indicate that water follows the ufual law, and is expand- 

 ed like other bodies by heat. 



Or if I fhould obferve that warm water, in cooling to the 

 freezing point, had the coldefl portion uniformly at the bottom, 

 the fame conclufion would follow ; while a different inference, 

 and the exiflence of the fuppofed anomaly, would be deducible 

 fliould the event prove different. The only circumflance, I can 

 figure to myfelf as tending in any meafure to render this mode 

 of examining the point doubtful, is, that water near its congeal- 

 ing point may have fo little change of denfity occafioned by a 

 fmall variation of temperature, that its particles may be prevent- 

 ed by their inertia, or by the tenacity of the circumfluent mafs, 

 from affuming that fituation which their fpecific gravity would 

 allot to them. 



It will appear, however, very clear, from the circumflances of . 

 the experiments which I fhall immediately detail, that no ob- 

 ftacle to the fuccefs and precilion of the experiments proceeded 

 from this fource. 



It is not neceffary for me to relate all the experiments I have 

 inade. I fhall reftridl myfelf to the detail of fix, which prefent. 

 varieties in the modes of procedui-e, and which afford the moft. 

 ftriking refults. 



Experiment I. . 



1 TILLED a cylindrical jar of glafs 8j inches deep, and 44- in 

 diameter, with water of temperature 32°, and placed it on a table, 

 interpofmg a confiderable thicknefs of matter poffeffed of little 

 power of conducing heat. , I fufpended two thermometers in 

 the fluid, nearly in the axis, of the jar, one with its ball about 

 half an inch from the bottom, the other at the fame dillance, 



below 



