CfN THE TRANSMISSION OF HEA?* 24-7 



Mehins of the ice floats on its furface, and hence from the 



o 



enlargement of the diameter of the veflcl, the height of the 

 column of mercury muft be diminiihed. It was therefore he- 

 ceffary to afcertain with accuracy, whether this diminution 

 had proceeded fo far, as that the particles of mercury which 

 had been in contact with the bottom of the iron cup, and di- 

 rectly heated by it, had come in contact, with the bulb of the 

 thermometer. This was found not to be the cafe. At the Proofs that It 

 end of the experiment the bulb was ftill covered with fo much dld noU 

 mercury that the cup floated freely ; and it was afcertained in 

 particular both by actual meafurement, and by a gage frozen 

 in the fide of the cylinder, that the quantity of mercury co- 

 vering it was ftill one-eighth of an inch deep. As therefore! 

 the cup merely floated on the furface, the part of the fluid 

 direftly heated by it could not have come in contact with the 

 bulb. This was likewife clearly eftablifhed by the pheno- 

 mena of the experiment itfelf; the thermometer rofe rapidly 

 at the commencement of it before the quickfilver could have 

 funk fenfibly, and it remained ftationary at the end when the 

 full defcent had taken place, although the water in the cup 

 was ftill warm. On pufhing it down fo as to caufe it to touch 

 the bulb the thermometer rofe rapidly, a proof that it had not 

 previoufly been in contacl with it. In this point Of view there- 

 fore the experiment was unexceptionable, and thofe which 

 follow were ftill lefs liable to any fallacy of this kind, as a 

 larger quantity of fluid was interpofed between the heated 

 body and thfc thermometer. 



2d. As much mercury was poured into the veflel of ice as to Exp. IV. Repe- 

 cover the bulb of the thermometer half an inch, and water at tJt,on with * 

 212° was poured into the iron cup floating on its furface. The mercury, 

 thermometer flood at 32, in one minute it rofe to 32f , in two 

 minutes to 33 J, in three minutes to 33^, it then became fta- 

 tionary, and in fix minutes more began to fall. The quick- 

 filver in the veffel was found to have funk rather lefs than one 

 quarter of an inch. The thermometer therefore at the end of 

 the experiment ftill remained Covered with one quarter of an 

 inch of fluid. 



3d. When the experiment was made a third time, the bulb Exp. V. aa<f 

 of the thermometer being covered with one inch of quickfil- ftlli S r€aleu 

 ver, the rife of temperature amounted to three-fourths of a 

 degree* . . 



All 



