294 
SOCKET SHELLS, AND ARTILLERY OE THE EUTURE. 
We thus arrive at the primary idea of firing a rocket from a rifled 
gun, and this is the general outline of the manner in which I would 
apply it 
The figure represents an iron shell with studs, filled with rocket 
composition, and hollowed out inside, as in the ordinary rocket, with 
this difference: the coned hollow penetrates to the head of the projectile, 
so causing very rapid combustion. There are three vents in the base, 
closed by rivet-shaped plugs fitting easily, and a steel plate over the 
vents, closed round the edges with lead, hammered or cast in between 
it and the base of the shell. The usual “ half-shields” of rockets will, 
I imagine, be found superfluous. The difficulty of ensuring the three 
vents being blown out simultaneously, at first _suggested having one 
vent in the centre of the base; but I do not think this would answer, as 
the equilibrium of forces acting on the shell, when burning, would not 
be so well established, and the flight would therefore be more irregular. 
The shell should be increased in length to make up the weight to 
that of the ordinary shell for the particular gun, and to enable the 
ordinary service charge to be used. 
Sufficient rocket composition to burn from 3 to 5 secs, will answer 
all requirements. 
As regards the head of the rocket shell, two plans suggest them¬ 
selves :—(1) to have a solid plug screwed in over the fuze; (2) to have 
a fourth vent in the head. Having gained all the advantages of a flat 
trajectory, it may possibly be found useful, once the projectile has 
reached the zone in which it is intended to act, to reduce its velocity 
to some extent, or even to delay it as long as possible while burning. 
It is difficult from theory alone to determine the precise merits of these 
alternative plans. Possibly they may both be found useful, and then it 
would be left to the battery commanders in the field to decide, accord¬ 
ing to requirements, whether solid or hollow plugs should be used. 
While wavering between these different plans, the time fuze question 
must be considered for each particular case. 
First, suppose shells with head-vents to be alone employed. The 
head-vent must be restricted to a certain size (to be determined by 
experiment), in order to ensure sufficient internal pressure to blow off 
the vents and base plate. The ordinary R.M.L. fuze (9 secs.) will 
