MAXIMUM DENSITY OF WATER. Q3J^ 



This experiment was repeated three times the fame day, 

 (the I3th of June 1805,) and always with nearly the fame 

 refiilts. 



The mean refults of thefe four experiments were as follows : Tabutatpd re- 

 fa its of the ex- 

 _. , - , ^ - , periments re- 



Timeclapfed Temperature of the peated, with the 



reckoned from water in the cork cup, the cone at a low 



the beginning of as fiiewn by thethcr- hcU, 



the experiment. mometcr. 



JMin. Sec, 



- - - . - - - 32^ 



At 10 began to rife - - - - 32-f- 



At 23 had rifen to .... 33 



28 34. 



35 * - - - - - - 35 



48 36 



13 37 



1 35 38 



- 2 32 39 



3 41 391 



4 48 39| 



6 5 ------- 39f. 



As I had found by fome of my experiments made in the Whether amuch 

 ^car 1797, of which an account is given in my feventh eflay, g^e^'^erheat m 

 part I. that water at the temperature of about 42? F. and affecl the water 

 confequenlly what we (hould call very cold, melted confider-^'^^^^'^^P^^^^* 

 ably more ice, when (landing on it, than an equal quantity of 

 boiling hot water, in the fame fituation, I was very curioas 

 to fee whether the thermometer, the bulb of which lay in the 

 cork cup^ would not alfo be lefs heated by the ball when it 

 fiiould be applied rery hot, to the furface of the water, than 

 when its temperature was much lower. 



To determine that point I made the following experiment, 



Experiment, No. 3. 

 The cylindrical veffel with its contents having been once Experiment 3. 

 more reduced to the uniform temperature of freezino- water, '^^^ experiment 



,,.,,, , ,.,.,. , r • repeated with the 



the metallic ball was heated m boiling water, and bemg as cone heated to 

 expeditioufly as poflible taken out of that hot liquid, its*i2.«. ^ 



projecting conical point was fuddenly fubmerged in the ice- 

 cold water, as in the former experiments. 



The 



