LXIV. DISCUSSION. 
to lessen those strains by putting the steam through two or three 
cylinders as the case may be. The author had adopted a direct 
system of connection in his refrigerating machine, thus allowing 
of the reduction of the strain to a minimum, reducing friction in 
a very material degree, and enabling the various parts to be made 
lighter ; another very important item is that the. machine is so 
nearly balanced. Asa distinct and very important branch of 
engineering, refrigerating machinery cannot be overlooked ; inas- 
much as some of our principal industries depend to a very large 
extent upon the efficiency and economy of the machines which 
are used in the refrigerating process. The many cases of failure 
of refrigerating machinery resulting in the loss of cargoes of meat 
were known to all, and a potent cause of failue no doubt lay in 
the fact that the machinery was often not properly duplicated. 
Mr. W. B. Srarnam said he made in 1893 eleven tests on two 
different machines, each of twelve tons ice making capacity per 
twenty-four hours. One a Linde machine using ammonia, and 
the other acarbonic acid machine. These machines were working 
under exactly the same conditions, and all measurements were 
carefully gauged, and the results checked by an independent 
engineer. The object of these tests was to find out in the first 
place the relative efficiency of the two refrigerating mediums 
before mentioned when working with different temperatures of 
cooling water. The initial and final brine temperatures were 
kept constant as near as possible throughout the eleven tests, and 
as the duty of the machine fell off consequent on the rise in tem- 
perature of the cooling water, so the quantity of brine in circula- 
tion was proportionately reduced. No measurements were taken 
until all fluctuations in temperature had ceased. , The diagram 
marked “ A” shewed two curves, the lower one representing the 
results obtained with the carbonic acid machine, and the upper 
one the results obtained with the ammonia machine. Plotted as 
ordinates are the number of B.T.U. extracted by each machine 
per indicated horse power, and as abscissw are given the tempera 
tures of the refrigerant (ammonia or carbonic acid) measured 
