Chemistry and Physics. 279 



rnents with rods of different diameter showed that the larger ones 

 give higher values for the estimates of the duration of impact. 



Having thus satisfied himself that the experimental method 

 gives approximately correct results, Hopkinson next perfoi'med 

 some similar experiments on gun-cotton. Cylinders of dry gun- 

 cotton l - 25 in. X 1'25 in. and weighing about 1 oz., were detonated 

 with fulminate at a distance of about - 75 in. from the end of the 

 steel rod. The average value of the pressure during an interval 

 of 10 -6 sees., in the neighborhood of the maximum, was found to 

 be about 30 tons per square inch. The absolute maximum is of 

 course considerably higher. The pressure had practically disap- 

 peared in 1/50,000 sec, so that at least 80 per cent, of the impulse 

 of the blow had been delivered within that time. When the gun- 

 cotton was exploded in contact with the rod the blow was dead- 

 ened by the deformation of the steel. Nevertheless, the conclusion 

 was reached that the maximum pressure at the surface of contact 

 is at least double what it is when an air gap of # 75 in. is inter- 

 posed. The results obtained for gun-cotton throw some light on 

 the nature of the fracture which is produced by the detonation of 

 this explosive in contact with a mild steel plate. The effect of 

 the explosion is to give speed to the parts of the plate with which 

 the gun-cotton was in contact, the remainder being left at rest. 

 For example, in a plate 1 inch thick the speed imparted by a slab 

 of gun-cotton of about the same thickness is roughly 200 feet per 

 second. In static tests on mild steel the metal flows when the 

 shearing stress is of the order of 10 tons per square inch, and no 

 materially greater stress can exist. On the other hand, if the 

 rate of straining is sufficient, the viscosity of the flowing metal 

 becomes important, and the shearing stress may approximate to 

 the value corresponding to perfect elasticity. Assuming this con- 

 dition to be fulfilled in the case of the mild steel plate 1 inch 

 thick, shearing stresses of the order of 100 tons per square inch 

 may be produced. The shearing stress is accompanied by tension 

 which may be sufficient, under the given conditions, to overcome 

 the forces of cohesion. Thus the steel is cracked in spite of its 

 ductility, just as cool pitch may be broken by the blow of a ham- 

 mer. From the measured duration of the pressure produced by 

 gun-cotton it may be inferred that the speed of shear necessary 

 to fracture mild steel is of the order of 1000 radians per second. 

 The shattering of the plate probably occurs within two or three 

 hundred-thousandths' of a second, that is, while the pressure is 

 acting and before appreciable deformation has had time to set in. 

 The observed bending of the steel fragments occurs subsequently 

 and is due to the relative velocities which remain in the different 

 parts of each piece of the plate after the plate has been broken 

 and the pressure has ceased to act. — Proc. Roy. Soc, vol. Ixxxix 

 (A), p. 411. h. s. u. 



