In this process there is required a different kind of care and accuracy. Up until this time the care has been to provide a metal of proper consistency and quality. From this point forward the manufacture of a gun requires the machining and fitting of this metal into a shape and form so accurate that the full strength of the gun and the best accuracy of fire may be attained. To explain how and why hoops are placed upon the gun tubing and how the various hoops are shrunk from the outside diameter of the gun will require a few lines. Cannon are made of concentric cylinders shrunk one upon another. The object of this method of construction is twofold. The distinctly practical object is the attainment throughout the wall of each cylinder of the soundness and uniformity of metal which is more certainly to be had in thin pieces than in thick ones; the other object is more closely connected with the theory of gun construction. When a hollow cylinder is subjected to an interior pressure the walls of the cylinder are not uniformly strained throughout their thickness, but the layer at the bore is much more severely strained than that at the outside. This can be readily seen if we consider a cylinder of rubber, for example, with a bore of 1 inch and an exterior diameter of 3 inches, which are about the proportions of many guns. If we put an interior air pressure on the cylinder until we expand the bore to 2 inches, the exterior diameter will not thereby be increased 1 inch. But supposing that it were increased as much as the bore, that is, 1 inch, we would have the diameter, and therefore the circumference, of the bore increased 100 per cent, and the circumference of the exterior increased 33-1/3 per cent. That is, the layer at the bore would be strained three times as much as that at the exterior, and the interior layer would commence to tear before that at the exterior would reach anything like its limit of strength. The whole wall of the cylinder therefore would not be contributing its full strength toward resisting the interior pressure, and there would be a waste of material as well as a loss of strength. Let us now consider, instead of our simple cylinder, a built-up cylinder composed of two concentric ones, the inner one of a bore originally a little greater than 1 inch, and the outer one of exterior diameter a little less than three inches, originally; so that when the outer one is pressed over the inner one (its inner diameter being originally too small for it to go over the inner one without stretching) the bore of the inner one is brought to 1 inch, and the exterior of the outer one to 3 inches. We now have a cylinder of the same dimensions as our simple one, but in a different state; the layers of the inner one being compressed and those of the outer one extended.