66 ANNUAL OP SCIENTIFIC DISCOVERT. 



five luintlrod 2:i";i»ini('s of very inf'nsilile cast-iron in thirty min- 

 utes. — Intdledual Observer, March, 180G. 



PETROLEUM AS A FUEL. 



Mr. C. J. Ricliarilson has so Hir succeeded in ntilizinj? petrole- 

 um as a steam fuel for marine enjjines, that at a recent trial of 

 his iini)roved petroleum boiler, at Woolwich Dockyard, the most 

 favoral)le results are said to Iiave accrued. It is re])orted that 

 tlie boiler vaporized aljout three thousand pounds of water, at the 

 rate of thirteen and a half ixnmds to one pound of fuel, in about 

 three hours, the lowest class of English coal beino; used. Petro- 

 leum is tlie ijxact ojjposite of coal ; it is slow burnin<r, permitting 

 little waste, requiring a small fire-box and no ash-pit. An ash, 

 the petroleum coke, forms itself on the surface of the grate, and 

 is of great service to the combustion. After a few tons of the oil 

 an; burned, this would Ijecome several inches in thickness, and 

 form a porous grate l)etter than any that could Ije uianulactiu'ed 

 for the i)urpose. — Mechanics' Magazine, Jan., 1866. 



SIEMENS' REGENERATIVE GAS FURNACES. 



Although this furnace has been described in the " Annual of 

 Scientific Discover}' " for 1804, the facts elicited are so important 

 and suggestive, that attention may be called to them again. 

 The points of special intc^rest are, 1st, the extremely high tem- 

 perature which can be obtained, and which, in fact, is limited 

 only by the nature of the materials employed in the construction 

 of the furnace ; and 2d, the possibility of emjjloying at Avill 

 either an oxydizing or a reducing atmosphere. The furnaces 

 have been applied to puddling and re-heating, and, no doubt, will 

 soon be extensively used in metallurgical processes. It is well 

 worth while to determine by direct experiment, on a large scale, 

 whether the rich iron ores of Lake Champlain, Lake Su2>erior, 

 and Missouri, cannot be directly reduced to tlie metallic state by 

 heating them to a sulliciently high temperature in the chamber 

 of a Siemens' furnace, and then clianging tlie gaseous mixture iu 

 the furnace to a reducing condition. This would, in fact, be 

 blooming upon a large scale, and would perhaps avoid the incon- 

 venience and expense of blooming in the small way, which, in 

 spite of the superior quality of the ii'on produced, has been almost 

 wholly superseded by the cheaper j^i-ocess of puddling. Exper- 

 iment only can determine whether fluxes can be used with ad- 

 vantage in blooming in this manner, when poorer ores are em- 

 ployed. Ores of co^jper could doubtless be roasted and reduced 

 in furnaces of this construction, and, with some additions to the 

 original plan, the sulphurous acid formed during the roasting 

 might be directly conveited into sulphuric acid in leaden cham- 

 bers. But it is for the metallin-gy of iron that the new furnaces 

 will probably be found most advantageous. As the temperature 

 attainable is exti'emely high, it may even be found practicable to 

 melt the malleable ii'on formed by the direct reduction of the ore, 



