704 Report of the Consulting Engineers on the Trials of 
cent, of the heat that water needs to raise it the same number of 
degrees in temperature, its " specific heat," as it is termed, being 
only 'IISS. The quantity of heat contained by the iron is 
therefore found by multiplying its weight by 'IISS, and by the 
number of degrees. The method of the whole calculation will 
probably be made clear if we go through a particular case, say 
Davey Paxman and Co.'s Simple Engine, No. 3125. Its trials 
were completed at 6.17 P.M., and at that time the pressure gauge 
marked 100 lbs., corresponding to 338 "5° temperature. The 
steam space of this engine measured 18*23 cub. feet, the water 
at working level weighed 1317 lbs., the weight of metal affected 
by heat was 8052 lbs. 
From Mr. Cotterill's tables it appears that a cubic foot 
of steam at 100 lbs. pressure, weighs '2609 lbs., and that 1 lb. 
of steam contains 1185 "1 units of heat above 32^, and 1 lb. of 
water at the temperature of the steam contains 810 "23 units. 
Hence the total units of heat above 32° in the engine were : 
c. ft. lbs. u. units. 
Steam ,. 18-23 x -2609 x 1185-1 = 5,63G 
Water .. 1317 lbs. X 310-23u. = 408,580 
Metal .. 8052 lbs. x '1138 x (338-5° -32°) = 280,850 
Total uDits of heat above 32° .. ,. 695,066 
As the engine cools, a portion of the steam is condensed, so 
that the weight of steam is constantly diminished, while the 
weight of water is increased, the sum of the two being constant 
at all times, and on this supposition Table VIII., page 728, has 
been calculated. 
It will be noticed that no observation was made when the 
steam was at 95 lbs., its working pressure, yet it is at this point 
that we must know the rate of cooling. 
To ascertain this, we construct the diagram (Fig. 3), on the 
base line of which we mark off the times at which the readings 
were taken, and on the ordinates or vertical lines, the units of 
heat which remained in the engines at each moment of obser- 
vation ; but to avoid an inconveniently large diagram, we make 
the zero-point at the last reading, or 6.50 A.M. on the day 
after the trial, by deducting the heat in the engine at that hour, 
namely, 364,600 units, from all the values above it. 
Having determined these points, a line is run through them, 
and it will be seen that they fall in very closely to a uniform 
curve, which shows that the rate of cooling decreases with the 
decrease of pressure and temperature, a fact which is in accord- 
ance with the law that the rate of cooling is proportional to the 
difference of temperature between the hot body and the surround- 
ing space. To show this more clearly, column 12 in Table VIII., 
