MAXIMUM DENSITY OF WATER. 229 



direflly over the cork cup, and fixed immoveably in that 

 fituation, I forefaw that the defcending current oF warm water 

 muft neceffarily fall into that cup, and at length fill it, and 

 that the prefence of this warm water in the cup would be 

 announced by the rifing of the thermometer. 



The refult of this very interefting experiment was juft what^t did, in faft, 

 I expeded : the conical metallic point had not been in contact ^^^jncter in the 

 with the ice-cold water more than 20 feconds, when the cup j 

 mercury in the thermometer began to rife, and in three minutes 

 it had rifen three degrees and a half, namely, from 32® to 35^°, 

 when five minutes had elapfed, it had rifen to 36®, when an 

 end was put to the experiment. 



Another fmall thermometer, placed juft below the fur- 1>« tit did not 

 face ot the ice-cold water, and only -f-^ or an men from ^^j. near the 

 the upper part of the conical point, on okc lide of it, did not iur^ace of the 

 appear to be fenfibly affeded by the vicinity of that warm * 



body. 



A third thermometer, the bulb of which was placed in the nor another near 

 brafs cup, juft on the outllde of the cork cup, and on a level [j^g^up.^ 

 with its brim, ftiewed that the water which immediately 

 furrounded the cork cup, remained conftantly at the tempera- 

 ture of freezing, during the whole time that the experiment 

 lafted. 



As I well knew, from the refults of the experiments on the The metaKat 

 propagation of heat in a folid bar of metal, of which an ac- t'jfg\^°tgr ^o^^the 

 count has been given in a memoir prefented to the firft clafs fame temp, 

 of the National Inftituteof France, on the 7 th of May 1801-, 

 that the ice-cold water in this experiment could not poffibly 

 acquire from a contaiSl with the conical metallic point, a tem- 

 perature fo high as that of 42®, I was by no means furpriled 

 to find that the thermometer belonging to the cork cup rofe 

 no higher. 



In order to fee if it could not be made to rife not only A fomewhat 

 higher, but alfo more rapidly, by employing the metallic ball^l^^.f/^^j^^^' P'^' 

 heated to fuch a temperature as it might by fiippofed would applied, 

 be fufficient to heat thofe particles of ice-cold water which 

 (liould come into contact with its conical point, to the tem- 

 perature at which the denfity of water is fuppofed to be a 

 maximijm, I rpade the following experiments. 



[^ E Xpert) lie nt. No. 2. 



Having removed the ball, I gently bruflied away the warm Experiment ^. 



water, which, in the laft experiment had been lodged in the'^^^ former ex- 

 ^ *=* . periment was 



cavity 



