Epiperitnenis on ihe Conducing Powers of Fluids tuit?) Regard to Heat, ^99 



inches. A plate of fleel I, four inches in diameter, and 0.3 inches thick, was ufed as the 

 cover for the veflel ; the remaining half inch of the excavation being left to receive the 

 fluid, which would rife by expanfion when heated. D is a tin veflel, of which the dimen- 

 fions may be fufficiently known from the figure. It has a loofe cover or valve at E, and is 

 clofed every where elfe. At H a branch proceeds, which ends in a cylinder F ; and at G 

 is a fcreen, to prevent the heat of the lamp K from affefting C ; in addition to which, 

 during the experiment, there was placed a pafteboard fcreen upon the ftage near the letter 

 K. It will now be eafily feen that when the veflel C was correftly filled (except the ex- 

 cavation at top) with any fluid, and the fteel cover laid in contaft with the furface, and the 

 veflel D partly filled with boiling water, and then placed in its fitnation here fhewn, 

 and, laftly, the fpirit lamp K placed beneath the outer end of the branch H, — the tem- 

 perature of 212° would be confl;antly applied to the fteel, and the furface of the fluid ac- 

 cordingly heated. Whence if any heat proceeded downwards, it would be fhewn by the 

 thermometers ; and if flow currents of difl^erent temperatures prevailed fo as to produce the 

 variation before ftated, thefe thermometers would differ. It was thought that fuch a dif- 

 ference would prove the exiftence of currents, and the probability that the wooden vefl'el 

 was the chief agent of the conduct ; but it was not fo clearly admitted that their agreement 

 would prove the contrary pofition. 



Brevity prevents my relating our arguments and views, and the nature of unpremeditated 

 converfation, as well as other motives, fotbid me the honour of mentioning the names 06 

 the philofophers to whom I refer in this difcourfe. I fhall likewife, for the fake of con- 

 cifenefs, omit feveral experiments which were made with other apparatus of lefs fimplicity, 

 or leading to the fame fa£ls. One, however, with water I may curfortly notice. It was 

 made in a glafs veflel, with a cork bottom, nearly of the dimenfions of the wooden veflel 

 here defcribed. During the time the boiling veflel E was kept in contadl with the furface 

 of the water, minute bubbles afcended flowly from the feveral parts of the water, which 

 appeared to Indicate that there was no perceptible current. A quantity of elaftic fluid in 

 bubbles occupied the furface of the liquid, and prevented, its contaft with the veflel. The 

 thermometers were raifed confiderably, but flowly, as was the cafe in m.any other ex- 

 periments with water; but as thefe require to be repeated, I omit the refults. 



Several trials were made with mercury and the apparatus in the Plate. The following 

 was made with much care r 



Pure diftilled mercury was put into the veflel, and left undifturbed, with the fteel cover 

 upon it, for upwards of an hour. The veflel D was then partly filled with boiling water,, 

 and carefully placed on the fteel cover, and the lamp K lighted. This was at thirty-two> 

 rninutes part eight. The conclufion of the experiment was efi^edted by lifting off" the veflel. 

 D. by the hand covered with a glove. 



Time.. 



