on Steam-Boats. 18] 
also very good; that is an inverted siphon, with a column of 
mercury proportioned to the purposes for which it is to be em- 
ployed. 
Do you apprehend much danger to arise, in case of explosion, 
from that mercury if it was employed ?—No, because the tube is 
always perpendicular, and if the mercury shoots out, it goes away 
and falls down in rain; I am of opinion, a boiler may be made 
safe either of wrought or cast iron, but for great strain 1 would 
prefer cast iron, contrary to the opinion of many people, and the 
reason I would prefer it is the same for which it is preferred in 
making canon. It is not possible to get thick plates of wrought 
jron perfect throughout, and you trust at last to rivets in joining 
them, but cast-iron boilers can be made of any strength you 
please; instead. of having a boiler that will stand sixty, it may 
be made to stand six hundred, of either wrought or cast iron. 
Another reason why I would prefer cast iron is, that the sheet 
jron corrodes much quicker and destroys by oxidation, so that a 
boiler may be safe when first set up and stand its proof, but very 
~ soon become unserviceable, or at least comparatively so. Boilers 
should always be cylindrical tubes, and for an obvious reason, 
capacity should be got by length and number rather than by 
diameter. There is no more danger to be apprehended from 
steam as to bursting, than from the employment of condensed 
air, only that the water may scald ; but as to the danger of the 
fragments being scattered about, it is the same with air as with 
steam, and yet all the engineers constantly employ cast-iron re- 
ceivers, condensers, or air-vessels where pressure is wanted. 
Is not cast iron liable to suffer some material injury from the 
contraction and expansion by heat and subsequent cooling ?— 
Whether a boiler be made of wrought or of cast iron the metal 
expands and contracts, and expansion or contraction is more or 
less injurious in proportion as it is often repeated, but it does not 
‘prejudice a boiler made of cast more than one made of wrought 
jron. 
Is not it more injurious to cast than wrought-iron boilers ?— 
No, I do not think it is. z 
In case of accident by explosion in a cast and wrought-iron 
boiler, which, in your opinion, would be attended with the 
greatest mischief to the persons about it ?—If an actual explo- 
sion takes place, 1 should think from the cast iron; but I con- 
ceive that a properly constructed cast-iron boiler would be 
stronger, and therefore would not explode so soon. A boiler 
should be proved with cold water, if it is to be applied to high 
pressure, 
Are you not aware that cast iron, notwithstanding the greatest 
possible attention of the founder, is liable to cavities in the in- 
M3 terior 
