Behind the engine compartment, but completely shut off from it, is another almost empty division occupying a considerable part of the length of the torpedo, and known as the “buoyancy chamber.” But it contains, attached to the bottom of it, a certain amount of ballasting, so adjusted to balance the weights of the other parts that the whole floats horizontally, and at the same time preserving the tube in one vertical position as regards its transverse diameter, _i.e._, so that the horizontal rudder is always horizontal. The shaft from the engine passes through this compartment, as also the rod from the small motor that moves the horizontal rudder. These, of course, pass through water-tight bearings.
At the tail of the torpedo, behind the rudders, are _two_ three-bladed screw propellers, of which the anterior one is mounted on a tubular shaft having a common axis with the other, but made to revolve in the opposite direction by means of a bevel wheel mounted on each independent shaft, with a third such wheel connecting them. The object of the double screw is to obviate “slip,” that is, ineffective motion of the blades through the water, and by this means the full power of the engines can be developed; while any tendency to _deviation_ to right or left, due to the rotation, is reduced to a minimum. We have spoken of one horizontal and one vertical rudder, although externally there appear to be two of each kind, right and left, above and below, on the tail of the torpedo. These pairs, however, are so connected as to be always in the same respective planes. The controlling mechanism acting in two different ways on the horizontal rudder has been already indicated, but nothing has yet been said about the vertical rudder. It is not moveable by anything within the torpedo, but is commonly fixed by clamping screws in or about the same vertical plane as the axis of the torpedo, and it performs the same function as a kind of back fin, which, in the earlier forms, extended nearly the whole length of the tube; and that is obviating any tendency of the torpedo to roll about its axis. The vertical rudder can also be fixed at a considerable inclination to the axis should occasion require, and the effect of that would be to cause the torpedo to pursue a circular course of greater or less radius, according to the less or greater degree of inclination. Very rarely, however, would this be required, and the vertical rudder may be considered as fixed in the axial plane, or having such slight inclination as may, on trial, have been found necessary to counteract any tendency to lateral deviation.