XLII. NORMAN SELFE. 
perature, the losses from friction and conduction being proportion- 
ately much less on large machines, as would be supposed ; the 
results with these machines are also much affected by the moisture 
in the air and other causes. 
Both in theory and in practice compressed air machines require 
very much more power (from four to six times) than other machines 
fora given abstraction of heat, they are therefore rapidly going 
out of use except for special purposes. It is possible in compress- 
ing air to reach very high and low relative temperatures without 
much difficulty, and it occurred to the author some sixteen years 
ago, that some of the heat or energy which is dissipated to the 
condensing water in these machines, and which is equivalent to 
the whole amount of the engine power, might be utilised by com- 
bining a compressed air refrigerator with a modification of the 
Du Tremblay ether engine, and he took out a patent in April 
1880 (No. 812) for a refrigerating machine which had an ether 
engine as well as a steam engine to supply the power. In this 
- machine the heat was to be abstracted from the compressed air 
by ether sprays on the condenser tubes, and the vapour thus pro- 
duced was to be utilised to assist the steam engine and reduce the 
power. Although this machine has never been made, and in 
actual practice a very large percentage of the power thus saved 
would be required to overcome the extra friction resulting from 
the additional number of parts, still it appears absolutely certain 
that it is only in this direction by utilising the heat which is now 
thrown away in the condensers of refrigerating machines that any 
great improvement in the future of artificial refrigeration is 
possible. 
In referring to the second or more complex system of mechanical 
refrigeration it was stated that a volatile medium such as ether, 
sulphurous acid, ammonia, and carbonic acid, was employed instead 
of a permanent gas as in the air machines ; before considering the 
machines therefore, it will be well to consider some of the 
: _ PROPERTIES OF GASES. 
- liquefaction o at gases nes pressure and cold has a special 
