414 Mr, Sylvester on the Motions produced hy the [JuneJ^ 



caused by the descending current, as may be observed by the . 

 direction of the particles of amber. If, instead of putting liot 

 water into the vessel, cold water were employed, and the bottom 

 of the vessel heated, the heated part becoming specifically lighter^, 

 immediately gives place to the heavier fluid which is colder, and . 

 the same appearance will take place as before ; the currents '^ 

 remaining permanently uniform, so long as there exists the same^ 

 difference of temperature between the different parts of the 

 fluid.. 



This is what takes place in the atmosphere when partial heat; 

 is applied ; the heated column in proportion to the rarefaction is 

 pushed upwards by the surrounding dense column. We have 

 lio instance of this kind so familiar as the ascension of the 

 smoke and heated air in chimneys. It has been always found 

 that the velocity of the ascending current was greater as the 

 chimney v/as higher, and in proportion to the heat applied at 

 the bottom. 



Many years ago having occasion to write an article on furnaces 

 for a popular work, I found that the power of draft in chimneys 

 was not strictly as the height, bujt as the square root of the same ; 

 and as the ratio of the difference of density between the rarified 

 column and the outer air to the density of the outer air, I also 

 found that under the greatest rarefaction that could be given, the 

 velocity was always less than a heavy body would acquire by 

 fallino' through the height of the chimney. 



If V be the velocity which a body would acquire by falling 

 through the height of the chimney, D, the density of the outer 

 air, d that of the air in the chimney, and v the velocity of the 



heated current ; then i? = V x — ^, as has been shown withv* 

 bodies rising or falling in a fluid. If A = the height of the 

 chimney, or any other column of heated air, then ^ = 2 o-^ /r 



D-d 



^ -r- 



An essay on this subject has lately appeared in a respectable 

 periodical work in which the principle of air rushing into a 

 vacuum is assumed to calculate the relative changes of heated air 

 as apphed to the ventilation of buildings. We do not require any 

 stronger proofs of the incorrectness of the principle on which the 

 calculations are founded than 4he results given, in which the 

 velocity of a current is made about five times greater than a 

 body would acquire by faUing through the height of the chimney. 

 The formulae above given agree very nearly with practice, and 

 may be made more useful by getting the value of d in terms of 

 the temperature, D being always that density which the temper- 

 ature of the atmosphere would give. For this purpose, let T = 

 the temperature of the atmosphere by Fahrenheit's scale, t that 

 of the heated column, and let e = the expansion or the increase 



