There were no means of regulating the time of burning of these pipes, which were generally metallic. The composition had a constant (and unalterable) length, corresponding roughly to that required for the _maximum_ range at which shell could be fired. Whatever the range, the oblong bombs were fired with a fuze that burned some fourteen seconds, answering to about 1000 yards range; the spherical grenados with a fuze that burned some twenty seconds, answering to about 2100 yards range.[614] In firing against works, &c., it was essential that mortar shell—and until the siege of Gibraltar, 1779, all shell were mortar or howitzer shell[615]—should not burst before impact. A shell which burst in flight was a shell wasted, but it mattered little whether it burst on impact or a few seconds afterwards.[616] But it was occasionally necessary to use shell against troops, and it was then that the radical defect of the tube became fully manifest. In this case it was desirable that the fuze should fire the bursting charge the instant the shell touched the ground,[617] and this was impracticable with the primitive pipe. While the long fuze was burning down to its end, the explosion might be prevented, or its effects might be neutralised in many ways. The shell might be thrown bodily into a pond or the sea;[618] the fuze might be extracted;[619] or it might be extinguished with water.[620] But in the vast majority of cases the explosion of the shell was neutralised in a much homelier and less heroic way; those near whom it fell waited for no command to quit its neighbourhood in all haste. It was long before any real progress was made in the adjustment of fuzes to burn a certain time, chiefly because the early gunners had no timekeeper. A striking illustration of this fact is found in the _Artis Magnæ Artilleriæ_, &c., of Siemienowicz, published in 1650. Wishing to give his readers an idea of the action of a certain fireball, he explains that it burned in the time one takes to recite deliberately the Apostles’ Creed.[621]