168 ANNUAL OF SCIENTIFIC DISCOVERY. 



upon each square inch of the bores will be 3.35 times greater with the large 

 than with the small gun ; when at the same time, if the walls of both have 

 a thickness proportional to the diameters of the calibres hi each, the large 

 gun will be incapable of sustaining a greater pressure per inch than the 

 small one. Even with a charge of 12 pounds of powder, the stress upon the 

 large gun must be more than double that upon the small gun when charged 

 with one-third the weight of its ball. 



The preceding examination does not, I think, present the difficulties to be 

 overcome hi increasing the size of cannon as greater than they really are, and 

 although the results that I have arrived at are from extreme cases, and may 

 be said to be mere deductions, yet they are deductions legitimately drawn 

 from the most reliable experiments that have been made. How then can the 

 necessary strength be obtained? Will it be answered, by an increased 

 thickness? It is not necessary to examine the obvious objections of the 

 great increase of size and weight that this implies, because no increase that 

 can be given to the thickness will increase the strength to a sufficient degree 

 to resist the force required.' To prove this, I must ask attention to a further 

 and somewhat elaborate examination. 



About thirty years ago, Mr. Peter Barlow, of "Woolwich, published a paper 

 in the Transactions of the Society of Civil Engineers, on the hydrostatic press, 

 in which he showed that hollow cylinders of the same materials do not 

 increase in strength hi the ratio of increase in thickness, but that the ratio of 

 increase of strength is such, that, where they become of considerable thick- 

 ness, the strength falls enormously below that given by the ratio of 

 thickness. 



Now, to obviate the great cause of weakness arising from the conditions 

 before recited, and to obtain, as far as may be, the strength of wrought iron 

 instead of that of cast iron, for cannon, I propose the following mode of con- 

 struction. I propose to form a body for the gun, containing the calibre and 

 breech as now .formed of cast iron, but with walls of only about half the 

 thickness of the diameter of the bore. Upon this body I place rings or 

 hoops of wrought iron, in one, two, or more layers. Every hoop is formed 

 with a screw or thread upon its inside, to fit to a corresponding screw or 

 thread formed upon the body of the gun first, and afterwards upon each layer 

 that is embraced by another layer. These hoops are made a little, say 

 ToVo^ 1 P ar t f their diameters, less upon their insides than the parts that 

 they inclose. They are then expanded by heat, and being turned on to 

 their places, suffered to cool, when they shrink and compress, first the body 

 of the gun, and, afterwards, each successive layer all that it incloses. This 

 compression must be made such, that, when the gun is subjected to the 

 greatest force, the body of the gun and the several layers of rings will be 

 distended to the fracturing point at the same time, and thus all take a portion 

 of the strain up to its bearing capacity. 



There may, at the first view, seem to be a great practical difficulty in 

 making the hoops of the exact size required to produce the necessary com- 

 pression. This would be true if the hoops were made of cast iron, or any 

 body which fractures when extended in the least de^roo beyond the limit of 



