ON THE MECHANICAL EQUIVALENT OF HEAT. 
319 
could have any intrinsic value, but, as the matter was exciting great interest in 
the laboratory, I carefully considered the conditions which would be necessary in 
order to render the great facilities, which this brake was thus seen to afford, available 
for an independent determination. 
The institution of an air thermometer was carefully considered and rejected. But 
it occurred to me that it might be possible to avoid the introduction of scales of the 
thermometers, just as before, and yet obtain the result. If it could be so arranged 
that the water should enter the brake at the temperature of melting ice and leave it 
at the temperature of water boiling under the standard pressure, all that would be 
required of the thermometers would be tbe identification of these temperatures. At 
first the difficulties appeared to be very formidable. But on trying, by gradually 
restricting the supply of water to the brake when it was absorbing some 60 h.p., and 
finding that it ran quite steadily with its automatic adjustment till the temperature 
of the effluent water was within 3° or 4° of 212° Fahr., I further considered the 
matter and formed preliminary designs for what seemed the most essential appliances 
to meet the altered circumstances. 
These involved— 
(1.) An artificial atmosphere, or a means of maintaining a steady air pressure 
in the air passages of the brake of something like one-third of an 
atmosphei’e above that of the atmosphere- 
(2.) A circulating pump and water cooler, by which the entering water (some 
30 lbs. a minute) could be forced through the cooler and into the 
brake, at a temperature of 32°, having been cooled by ice from the 
temperature of the town main. 
(3.) A condenser by which the effluent water leaving the brake at 212° Fahr. 
might be cooled down to atmospheric temperature before being dis¬ 
charged into the atmosphere and weighed. 
(4.) Such alteration in the manner of supporting the brake on the shaft as 
would prevent excess of leakage from the bushes in consecjuence of the 
greater pressure of the air in the brake, since not only would the leaks 
be increased, but wdren the rise of temperature of the water was 
increased to 180°, the quantity for any power would be diminished to 
one-sixth part of what it would be for 30°, so that any leakage would 
have six times the relative importance. 
(5.) Some means which would afford assurance of the elimination of the 
radiation and conduction, as, with a rise of 140° Fahr. above that of 
the laboratory, these would probably amount to two or three per cent, 
of the total heat. 
(6.) Scales for greater facility and accuracy in weighing the water, with a 
switch actuated by the counter. 
(7.) A pressure gauge or barometer, by which the standard pressure for the 
