“A forced draft in the furnaces,” explains the _Marine Engineer_ of September, 1887, “can be generated in two ways: first, by exhausting the uptakes and funnels of the products of combustion, when a greater flow of air will necessarily take place through the fire-bars; and secondly, by increasing the pressure of the air in the furnaces beyond that of the atmosphere. The steam-blast in marine boilers is well known to engineers as a means of quickly getting up the steam after its pressure has dropped; but the locomotives on our railways afford a very good illustration of how boilers may be continuously worked under forced combustion through a jet of steam exhausting the smoke-box and funnel of the products of combustion. This system of creating a draft involves a very large expenditure of steam and water, and as it is a _sine qua non_ in these days of high pressure that only fresh water should be used in boilers, and also as only a limited supply of this element can be carried in a ship, it follows that the plan of inducing a forced draft by means of a steam jet in the funnel cannot be well adopted in marine boilers. “Mr. Martin, the inventor of the well-known furnace doors, substitutes a fan in the uptake for the steam jet, and so arranges his funnel that in the event of the forced draft not being required the gases of combustion arising from natural draft will not be impeded in their exit to the atmosphere. He claims for his invention that it does away with all necessity for closing in the stoke-holds or furnaces, and that in war-ships funnels could be dispensed with, as the gases and smoke could be discharged anywhere from the fans. He also claims that by his plan of producing a draft the boiler-tubes become much more efficient as heating surfaces, and that the ends of the tubes in the fire-box are not so liable to be burned away, and that therefore there will be less chance of the boiler leaking round the tubes. There appears to be some grounds for these latter assumptions, for it is a well-known fact that the tubes of locomotive boilers, which are worked, as we have seen, on the exhaust principle, do very much more work than those of marine boilers before they are ferruled or rolled. It can also be shown by a very simple experiment that when the gases are sucked or drawn through the tubes the flame extends a much greater distance along the tube than when the gases are driven through the tubes. In this latter case the flame impinges on the tube-plates before separating into tongues and entering the tubes; but when sucked through the tongues of flame commence at some little distance from the plate before penetrating the tubes, and the ends are not therefore burned as when the flame impinges directly on them. It may be urged, however, against Martin’s system that owing to the greatly increased volume of the products of combustion due to their temperature, fans of from three to four times the size of those