shell it is a vitally
serious matter. From all I can learn regarding European practice, it
appears that not only are the explosives made sluggish, but the
quantity seldom exceeds thirty per cent. of the weight of the shell,
and the velocities, notwithstanding, are kept very low. In the
pneumatic gun the velocity is low also, but so is the pressure in the
gun. The pressure in the firing reservoir is kept at the relatively
low figure of 1,000 pounds per square inch or less, and the air is
admitted to the chamber of the gun by a balance valve which cuts off
just the quantity of air (within a very few pounds) that is required
to make the shot. The gun is long, and advantage is taken of the
expansion of the air. In no case can the pressure rise in the gun
beyond that in the reservoir.

Up to the present time there have been no accidents in using the most
powerful explosives in their natural state, and in quantities over
fifty per cent. of the weight of the projectile. I have seen
projectiles weighing 950 pounds, and containing 500 pounds of
explosives (300 pounds of the blasting gelatine and 200 pounds of No.
1 dynamite) thrown nearly a mile and exploded after disappearing under
water. According to Gen. Abbot's formula such a projectile would have
sunk any armorclad floating within forty-seven feet of where it
struck. Apparently there is no limit to the percentage of explosive
that can be placed in the shell except the mechanical one of having
the walls thick enough to prevent being crushed by the shock of
discharge. In the large projectiles a transverse diaphragm is
introduced to strengthen the walls and to subdivide the charge.

The development of the pneumatic gun has been attended with some other
important discoveries, which may be of interest. It is well known that
mortar fire is very inaccurate, except at fixed long distances, in
consequence of the high angle, the slowness of flight of the
projectile, the variability of the powder pressure, and the inability
to change the elevation and the charge of powder rapidly. In the
pneumatic gun, the complete control of the pressure remedies the most
important of the mortar's defects and makes the fire accurate from
long ranges down to within a few yards of the gun. It is obvious that
the pressure can be usefully controlled in two ways: (1) by keeping
the elevation of the gun fixed and using a valve that can be set to
cut off any quantity of air, according to the range desire