276 ]Mr. Mallet on the Physical Conditions 



transverse section of metal throughout its length, and its elasticity perfect, being 

 given by equations 19-27, it follows that the longitudinal strain upon the 

 metal of the gun, due to recoil, will be a minimum, if the trunnions be placed at 

 the farthest point to the rear (as in mortars) ; or moves completely, if there be 

 no trunnions, and the gun be firmly and rigidly supported against the recoil at 

 the breech. 



262. But in every case there is a certain amount of end-on strain ; the metal 

 of a gun is, therefore, at the moment succeeding explosion, subjected simul- 

 taneously to three diiferent strains, acting at right angles to each other — the 

 tangential or bursting, which is exteusional, and accompanied by compressive 

 or radial strains, which are normal, and the longitudinal re-active strains of the 

 recoil, chiefly extensional, in line of the axis. There are good grounds for 

 presuming that the existence of the two latter tend to a certain extent to 

 weaken the resistance of the metal to the former. 



263. The author is not aware that any direct experiments have as yet been 

 made with a view to ascertain what effect would be produced upon the tenacity 

 of a prism or bar already strained in the direction of its length by the applica- 

 tion of new forces of extension or of compression along its whole or part of 

 its length, and perpendicular to the former. Whatever hypothesis be made as 

 to the law of aggregation, or of lateral adhesion of the ultimate molecules, it 

 would seem to follow inevitably, that lateral extending forces must reduce the 

 tenacity of the bar, and probably that lateral compressive forces might increase 

 it, dependent much, however, as regards the latter case, upon the relation 

 between the ductility and the ultimate cohesion of the material. If we assume 

 the molecules arranged equidistantly in parallel equidistant lines throughout 

 the bar (like strings of beads), whether opposite each other in the same trans- 

 verse section, or (quincunx) each falling into the space betwixt two of the 

 adjacent ones, and attracted mutually by a force varying by whatever lav/' with 

 respect to distance, it would seem likely that a force acting transversely to the 

 bar by extension, and equal to that extending the bar in length, might diminish 

 its strength by a function of -/2 : 1. Mr. P. W. Barlow's " Experiments on the 

 Existence of an Element of Strength in Beams," &c., arising from lateral action 

 of the particles (Proceedings of Royal Society, vol. vii., p. 319), as well as Vicat's 

 (Ann. de Chim.), bear upon this obscure subject, upon which experiments are 



