MECHANICS AND USEFUL ARTS. 169 



its elasticity. But wrought iron and all malleable bodies are capable of being 

 extended without fracture much beyond their power of elasticity. They 

 may, therefore, be greatly elongated without being weakened. Hence we 

 have only to form the hoops small in excess, and they will accommodate 

 themselves under the strain without the least injury. It will be found best 

 in practice, therefore, to make the difference between the diameters of the 

 hoops and the parts which they surround, considerably more than y^yoth 

 part of a diameter. The fixing the hoops in their places by the screw, or 

 some equivalent, is absolutely necessary, not merely to reinforce the body 

 against cross fracture, but to prevent them from starting with every shock of 

 the recoil. I know, by experiment, that the screw thread will fix them 

 effectually. The trunnions must, of course, be w r elded upon one .of the hoops, 

 and this hoop must be splined, to prevent its turning by the recoil. Small 

 splines should likewise be inserted under every hoop. It will, moreover, be 

 advantageous to make the tlireads of the female screws sensibly finer than, 

 those of the male, to draw, by the shrink, the inner rings together endwise. 



It will be seen that with a gun made in this way, we must depend upon the 

 cast iron body to resist the strain tending to produce cross fracture, though this 

 resistance will be in some degree supported by the outer rings breaking joint 

 over the inner rings. But if the body be made to constitute half the thickness 

 of the walls, it will be found sufficient for the purpose without any reinforce- 

 ment from the rings. This results from a principle or law, winch, so far as I 

 know, was first published by me in the year 1845, ha a pamphlet on wrought 

 iron and steel cannon. As I cannot put this matter in a better form than that 

 in which I have there given it, I will here quote the statement as then made. 



" Let us suppose we have a hollow cylinder, say twelve inches long, the 

 calibre being one inch in diameter, and the walls one inch thick, giving an 

 external diameter of three inches. Suppose tin's cylinder to be perfectly and 

 firmly closed, at its ends, by screw plugs or any other sufficient means. Let 

 this be filled with gunpowder and fired. The fluid will exert an equal pres- 

 sure in every direction, upon equal surfaces of the sides and ends of the hol- 

 low cylinder. Let us next examine the resisting power of a portion of this 

 cylinder, say one inch long, situate in the middle, or equally distant from the 

 ends, so that it shall not be strengthened by the iron which is beyond the 

 action of the powder. The fluid inclosed by this ring of one inch long con 

 tanas an area of one square inch, if a section be made through it in the direc- 

 tion of its axis ; and the section of the ring itself, made in the same direction, 

 will measure two square inches. "We have then the tenacity or cohesive 

 force of two square niches of iron hi opposition to an area of the fluid measur- 

 ing one square inch; and if we take the tenacity of the iron at 65,000 pounds, 

 the cylinder will not be burst, in the direction of its length, unless the expan- 

 sive force of the fluid exceed 130,000 pounds to each inch. Next, let us sup- 

 pose a section made through the cylinder and fluid, transversely. The area of 

 the fluid equal to the square of the diameter of the hollow cylinder, is one cir- 

 cular inch, and the area of the whole section, the diameter being three inches, 

 is nine inches. Deduct from this the area of the calibre, and we have eight cir- 

 cular inches. That is, the section of the iron is eight times greater than that of 



8 



