160 
DE. J, P. JOULE ON THE SUEEACE-COKDENSATION OP STEA3I. 
On examining the Table of results with air as the refrigerating agent, we may 
remark, — 
1st. That a film of air does not adhere to the surface of the tube so tenaciously as a 
film of water does. This is evident from a comparison of Nos. 1, 2 and 3, with 4 and 5 ; 
from which it appears that for the spaces '025 and *06 inch the pressures able to propel 
' equal quantities of air were as 7’66 to 1, or nearly as the squares of the xelocities. 
When water was employed in the same tubes, these pressures were as 18‘8 to 1. 
2ndly. That the velocity of the elastic fluid appears to exercise a much more con- 
siderable influence on the conductmty than it does in the case of water. 
3rdly. That spirals exercise a beneficial influence. This will be noted on comparing 
Nos. 6 to 15 with Nos. 16 and 17. 
The very small conductivity when Air is the refrigerating agent will probably prevent 
its being employed for the condensation of steam, except in very peculiar cases. 
I must remark, in conclusion, that the above research, however laborious, has left 
much to be accomplished. One of my chief objects was to obtain figures which might 
prove useful to practical men, and I have therefore confined myself to such tubes as 
were most likely to be generally used. In taking up the subject afresh, greater acciu’acy 
might be attained by the use of a sheaf of tubes, so as, by condensing a larger quantity 
of steam, to diminish the amount of temperature corrections. It would also be desirable 
to employ tubes of great thickness, so as to obtain the conductivity of metals after 
eliminating the resistance of the fluid film. The effect of irregularities in the water 
space might also be exactly determined, and the action of arrangements for agitating 
the refrigerating water more completely discussed than I have been able to do in the 
present memoir. 
