ON THE TRANSMISSION OF HEAT.' 245. 



Suppofe the cylinder of ice to be filled with water at 32°, and: near 32°, and 

 .,/•,-,,,•<• , , • • , • r • .would defcend 



a heated loud to be lulpended in it, the portion ot wateF in t0 tne thernw- 



contaft with the folid, will immediately have its temperature meter. 



raifed perhaps two or three degrees. But by this increafe of 



temperature it is not expanded, but is contracted in volume, 



it becomes heavier, and muft therefore link in the fluid towards 



the bulb of the thermometer; it will be fuccecded by another 



heated portion, adefcending current muft thus be formed, and 



the thermometer will have its temperature raifed by the contact 



of particles which had been directly heated by the fufpended 



folid. 



Oil and quickfilver are not liable to this fource of fallacy, oil and mercury 

 It has been afcertained by Crawford and De Luc, that the do not c °« r j^ tf 

 diminutions of volume in thefe fluids from decrements of tern- a £ out th ' e tem . 

 perature are nearly uniform, or at leaft at any temperature perature here 

 neceflary for performing experiments of this kind they. con- mentlone * 

 tract with fufficient regularity, as is indeed fuffictently fliewn 

 by their ufe as thermometries! fluids — With thefe fluids there- 

 fore the following experiments were made : 



Experiment I. A quantity of almond oil was poured into Experiment I. 

 the ice veflel, fo as to cover the bulb of the thermometer one Almond oil at 

 quarter of an inch. A fmall iron cup C, cylindrical, two vefieTwas heated 

 inches in diameter, and flat in the bottom, capable of holding by water poured 

 two ounces by meafure, was fufpended fo as merely to touch ^etdYKveffS* 

 the furface of the oil, and was filled with boiling water. This flighttyimmerfed 

 was preferred to a folid ball, as prefenting a larger furface to "V!** . J^* 6 *' 

 the oil, and as it was more eafy to alcertain the precife depth thermouietsr* 

 at which it was immerfed. At the beginning of the experi- 

 ment the thermometer flood at 32 p . In a minute and a half 

 it had rifen to 32|, in three minutes to 34§, in five minutes 

 to 36±, in feven minutes to 37 f. At this point it became 

 stationary, having rifen 5\ degrees of Fahrenheit's fcale in 

 feven minutes. The temperature of the water in the cup had 

 in this time fallen to 96?. The thermometer, after remaining 

 Stationary at 39| for fix minutes began to fall, and it conti- 

 nued to defcend at the rate nearly of a degree in a minute 

 and a half, till it returned to 32. At the conclufion of the Excavation •£ 

 experiment the fides of the cylinder at the upper part were the ice. 

 partly excavated, the excavation was greater* at the furface of 

 the oil, it became lefs as it defcended, but it had taken place 

 in a flight degree, even lower than the bulb of the thermo- 

 meter, 



