556 TRANSACTIONS OF THE AMERICAN INSTITUTE. 



secondly, that it is homogenous ; and thirdly, that it has fibre transverse 

 around it, in the direction best disposed to resist the pressure of the pow- 

 der. It is made by welding disks or cross sections together by end blows 

 of the steam hammer. Guns were made in this manner as long ago as the 

 year 1716, by M. Villons, of Porte de Marie, France, but they failed, and 

 the process was abandoned. With regard to the iron it is not as good as 

 the iron used by Armstrong for his large guns; the quality of toughness or 

 ductility is the one sought for by English gunmakers, and undoubtedly 

 obtained. But the Armstrong 110-pounder Avill endure but about forty 

 rounds before it has to be repaired if fired rapidly, from the distension 

 caused by unequal heating, although the same gun will stand very heavy 

 charges without injury, if fired at long intervals so that the heat becomes 

 uniformly distributed. 



90. The Salisbury iron used by Ames cannot compare with Krupp's 

 steel for ductility, a cylinder of which four inches in length and two in 

 diameter was pressed down cold in a hydraulic press to a lozenge of half 

 of its original length, without cracking in the slightest degree. Its tensile 

 strength is much greater than the Salisbury iron, yet guns made of this 

 fine Krupp's steel burst when fired rapidly, but do not explode from exces- 

 sive charges of powder. 



91. In large masses of wrought-iron, as in cast-iron, the strongest 

 metal does not make the strongest gun, as seen from the quotation already 

 made, from Hodman's work, pages 137 and 138. The reasons being from 

 want of uniformity in cooling; one part of the mass is straining either to 

 compress or rupture another part. And the more dense the metal the more 

 severe the strain. 



92. Fibre in wrought-iron is only possible, or is only attainable by 

 extending a block or bloom to from ten to sixty times its original length, 

 under pressure as between rolls. If a bloom four inches by four inches, 

 and twelve inches long were heated to a welding heat, and by the rolling 

 process drawn out into a half-inch square bar, it would be near sixty feet 

 in length, and would exhibit well defined fibre. If sufficient quantity of 

 such bars were provided, cut into lengths of five feet each, and, after being 

 polished on the surface to remove all oxide, made into a pile eighteen 

 inches square, being five feet long, for which it would require 1,296 bars, 

 held together with a strong band shrunk on each end ; upon raising the 

 temperature of the pile to a welding heat, in a most careful manner, it 

 would be found firmly united throughout, without a defective weld, with- 

 out either hammering or pressure, but upon breaking the block so made in 

 two, it will be found to have no fibre. 



93. A new ci'ystalization having taken place, under a law not well 

 understood, regulated by the size of the mass and the outside forn>, fibre 

 could be re-established in this metal only by rolling it out again to fifty or 

 sixty times its original length. Fibre could not be established under the 

 hammer without extending its length at least twice as much as with rolls. 



94. As this is an experiment I have conducted many times, I speak of 

 the facts with confidence. I have now twelve wrought-iron guns made in 

 this manner, from bars a hall-inch square, each block liaving been extended 

 under the hammer, continuing the square form (i. e. without roundiiig the 



