Portable Agricultural Steam Engines at Newcastle. 703 
results of his analyses of the waste gases. This table shows 
very clearly the amount of air allowed to pass through (or over) 
the fire, and the extent of the excess above that theoretically 
needed, which is one source of loss, the exact value of which 
we shall deal with hereafter, and shall contrast it with another 
loss which arises from a insufficiency of air at some period of 
the run. 
We will now consider a source of loss which renders 
the protection of steam boilers and engines from escape of 
heat by radiation and by convection a matter of considerable 
importance. 
Loss from Radiation and Convection. 
As observations on this subject have not been attempted 
at previous " Trials," it may be well to describe the process by 
which the loss of heat per hour at the working pressure and 
temperature has been determined. 
As soon as the engines had finished their trials they were left 
to cool, the ash-pans having been closed as far as practicable, so 
as to avoid draught through the boilers. At certain intervals 
the steam pressure, or the temperature of the water in the boilers, 
was noted, and from the records so obtained, the rate of cooling 
is arrived at in the following manner : 
The weight of water in the boilers at the working level, and 
the volume of the steam space of each boiler, had been ascer- 
tained by weighing the engines when their boilers were quite 
full of water, when they were emptied partially, that is to the 
working level, and finally when they were quite empty ; and 
the weight of such parts as were not sensibly heated when the 
engines were at work, as, for example, the wheels, axles, shafts, 
&c., was furnished by the makers. From these data it is 
easy to calculate the number of units of heat contained by 
a cooling engine with the pressures existing at the times when 
the observations were made. From the experiments of Regnault 
and of others, the number of units of heat contained in a pound 
of water or of steam under varying pressures and temperatures 
is known. The quantity of heat in water rises a little faster 
than the increase of temperature, and the quantity of heat in a 
pound of steam increases also with the temperature and pressure. 
(The precise amounts are to be found very conveniently arranged 
in the tables at the end of Mr. Cotterill's work on the steam 
engine.) The metals of which the boilers and machinery are 
composed were also heated, and parted with their heat on cooling. 
By far the greater proportion of the metal is iron or steel, and it 
may for our purpose all be assumed to be so, without sensible 
error. Now iron or steel requires only a little more than 11 per 
