80 



on the square inch clehver air with a greater velocity than can 

 result from a chimney 40 feet high ; the quantity however makes up 

 for this. It has been said that air is much obstructed in its 

 passao-e through tubes, and accordingly it has been recommended to 

 make the inside of the tube as sniooth as possible, and carefully to 

 avoid any angles in its course. No doubt these directions are, 

 where it is convenient, to be attended to, but the neglect of them 

 cannot materially influence the effect. In some experiments which 

 I made on the motion of air through tubes, and which if the duties 

 of my station permit me, I will at some future period complete, I 

 found that it was not much retarded by roughening the internal sur- 

 face of the tube through which it passed. In one of these trials a 

 tube of glass was used, 30 inches long and twice bent at right an- 

 gles, it was ^ inch bore, and a considerable quantity of air was 

 forced through it from a large gasometer with a velocity of 1 18 feet in 

 a second, this required 161 half seconds ; the tube was then moistened 

 with Lac Varnish, and filled with sand which coated its interior. 

 The same quantity of air was passed through it in this state and the 

 time was 200, but the adhering sand had diminished the width of 

 the tube so that a correction was required (/j), and the corrected 

 times were as 190: 200, or in this case, the resistance was about 

 -'_ of tlie entire force, this was an extreme case, for the velocity was 

 great and the space through which it passed small, and as the retarda- 

 tion arises principally from the adhesion of air to the surfaces over 

 which it passes, in a narrow tube this must be much increased. 



{p) The correction was made by removing the sand from the interior of the tube, leaving 

 only a narrow ring where it adliered ^ of an inch broad. When llie tube was reraovetl 

 entirely, the time of discharge was 156 half seconds. 



