DE. J. P. JOULE ON THE SUEE ACE - CONDENSATION OE STEAM. 
157 
The cause of the inferiority of the plain tubes may be attributed in some measure to 
a want of perfect concentricity and truth in the pipes, resulting in an irregular action 
of the refrigerating water, the greatest quantity of which would thus be transmitted 
through the mdest part of the water space. In the arrangement with spirals, the width 
of the water space was too great for any such circumstance to have a sensible influence. 
I think, however, that the imperfections of the tubes and of their concentricity were not 
such as to account for the great advantage which appeared to be produced by the spirals 
in my experiments, and I therefore attribute it to the continuous intermixture of the 
particles of water favoured by that arrangement. 
The following is a summary of the principal foregoing results : — 
1st. The pressure in the vacuous space is sensibly equal in all parts. 
2nd. In the arrangement in which the steam is introduced into a tube whilst the 
refrigerating water is transmitted along a concentric space between the steam-tube and 
a larger tube in which it is placed, it is a matter of indiflerence in which direction the 
water is transmitted. Hence, 
3rd. The temperature of the vacuous space is sensibly equal in all parts. 
4th. The resistance to conduction is to be attributed almost entirely to the film of 
water in immediate contact with the outside and inside surfaces of the tube, and is little 
influenced by the kind of metal of which the tube is cpmposed, or by its thickness in the 
limits of ordinary tubes, or even by the state of its surface as to greasiness or oxidation. 
5th. The nan-oulng of the steam space by placing a rod in the axis of the steam-tube 
does not produce any sensible effect. 
6th. The conductmty increases as the rapidity of the stream of water is augmented. 
In the circumstances of my experiments, the conduction was nearly proportional to the 
cube root of the velocity of the water ; but at very low velocities it evidently increases 
more rapidly than according to this law, whilst at high velocities it increases less and less 
rapidly as it gradually approaches a limit determined by the resistance of the metal and 
of the film of water adhering to the inside surface of the tube. 
7th. The conductivity increases so slowly in relation to the height of the head of 
water, that the limit to the economical increase of the latter is soon attained. 
8th. By means of a contrivance for the automatical agitation of the particles of the 
refrigerating stream, such as the spirals I have employed, an improvement in the con- 
ductmty for a given head of water takes place. 
9th. The total heat of steam above 0° Cent., determined by the average of the 161 
experiments, is 644°'28 for a pressure of 47’042 inches. 
The experiments in which air was employed as the refrigerating agent were made in 
a similar manner to those in which water was used. At high pressures the air was 
propelled by the condensing pump used by Professor Thomson and myself in our 
experiments, and at low pressures a large organ-bellows was employed. The tempera- 
ture of the air at its exit was obtained by placing the thermometer immediately over 
the concentric space between the tubes, varying its position from time to time so as to 
obtain an average result for the entire section of the channel. 
MDCCCLXI. z 
