Steam Engine. 115 
’fiiew quantity is evaporated. By this means tlie supply 
of water is very gradual, however, and not by fits and 
starts, as here described for the sake of illustration. 
It is found by experience, to be a much better method 
than a ball-cock, and the regular supplying of the boiler 
■^vith tvater is of the first importance. As a check upon 
this, and to know perfectly the height of the water in the 
boiler, there are two cocks, g and h, one of which reaches 
nearly to the surface of the water when at its proper height, 
and the other enters a little below the surface. 
It is evident, that if the water be at the just height, and 
you open g, steam will issue ; and if h be opened, water 
will be driven out by the pressure of the steam. But if 
water come out from g, then the water must be too high 
in the boiler ; and if steam issue from h, then the water is 
too low. By this means, it is easy to know at all times 
the exact height of the water in the boiler. 
i is a safety-valve, to prevent the bursting of the boiler 
by the steam growing too strong ; k is the pipe which 
conveys the steam to the engine. 
Fig. 6 is Mr. CartwTight’s steam-engine, the construc- 
tion of which evinces much ingenuity, a is the cylinder, 
which is supplied with steam from the boiler through the 
pipe b ; c is the piston in the act of going up ; d is the 
pipe that conducts the steam into the condenser e, which 
consists of two cylinders^ one xvithin the other ^ leaving a 
small space heUveen them^ into which the steam is admits 
ted; vdiile the inner cylinder is filled with cold water, and 
also the external cylinder surrounded by the same ; so 
that, by this means, a very large surface of steam is ex- 
posed, though no water is suffered to come into actual 
contact with it. 
To the bottom of the piston c, is attached a rod, witli 
another piston e, working in the pipe d. When the pis- 
