involved in the Construction of Artillery. 143 



ing the mouth of the gun) is the element upon which alone the chief practical 

 difficulty of increasing the caliber of cannon depends. (Note D.) 



4. Now this increases enormously with every increment of caliber. For 

 supposing the shot spherical in all cases, its weight increases as D"^, and as 

 the " work done" by the powder in giving it its final velocity is proportionate 

 to, 



5 being the specific gravity of the shot, and the length of very large guns not 

 being very much more than those of the smaller ones in use, so that the space 

 through which the force acts is miich the same, it follows, that the maximum 

 strain per square inch is greatly augmented. Add to this, that the lineal 

 windage being the same, the proportionate loss of effort by windage will be 

 less in the larger gun in about the inverse ratio of the square of the windage, 

 and finally, when, as is often now the case, cylindrical or cylindro-conoidal shot 

 are substituted for spherical, and so the ratio of the weight to the bore of the 

 gun further increased, while the windage is again lessened by the contraction 

 of the passage of escape due to the elongated form of shot ; all these circum- 

 stances so increase the pressure per square inch, that the utmost resources of 

 metallurgic skill are barely able to cope with it. And, besides the above, there 

 are other causes of increased strain upon the gun; thus the charge of powder 

 must be augmented largely, but its mass increases much more rapidly than the 

 internal surface of the cylinder of the gun, which is recipient of the heat of the 

 inflamed powder, and which carries off heat from its inflamed gases to the cold 

 metal of the gun ; hence the actual heat available for the expansion and in- 

 creased tension of the gases is greater as the size of the gun is greater, so that 

 the pressure due to this cause rises very considerably in large charges ; in other 

 words, a large mass of powder inflamed in a comparatively cold metallic recep- 

 tacle will produce an effect more than proportional to that of a much smaller 

 mass, so that the theoretic and the actual work done by difierent charges shall 

 sensibly differ. 



5. Thus the strain on the gun increases faster than D"'. To meet all this, the 

 thickness of the gun for any given material must be increased largely ; but in 



v2 



