general near the seat of the
projectiles. But, as it is not always found at precisely the same
point of the circle of the bore, a convenient mode of searching for it
is desirable. This is supplied by a disk for circular measurements,
which may also be considered as belonging to the star-gauge. It is
made of composition, and is divided into halves, with a hole through
the centre to receive the staff of the star-gauge.

It is turned so as to fit into the muzzle of the gun closely, with a
projecting lip two or three inches broad to hold it in place, and with
cleats overlapping the edges, to keep it from going in too far. The
face is a plane surface. The circumference is divided into as many
equal parts as may be thought desirable, which are numbered in regular
order. The centre hole is reinforced on the inside by a projection
which is turned to receive a collar that fits closely around it, and
holds the two halves together when they are placed on the staff.

When ready for use, the face is in the plane of the muzzle-face. Its
zero-mark is made to coincide with a light punch-mark on the
muzzle-face, directly below the line of sight.

On the staff of the star-gauge a brass slide is fitted, having a
thumb-screw to hold it in any position; from the inner end a point
extends at a right angle to its axis, of sufficient length to meet the
points at the circumference of the disk; a centre line extends from
the base to the apex. The slide being moved so as to make its inner
end coincide with any mark upon the staff, at which a circular
measurement is required, and the centre line of the point being made
to coincide with that of the staff, it is secured by the thumb-screw.
The point of the slide is then in the same plane with the
perpendicular measuring-points, and its direction always indicates
them; a series of measurements, made before proof, may thus be
compared with another made at the same points after proof.

It is obvious that the determinations will not be absolutely accurate,
for when the gun is worn, should the stationary points be
perpendicular, the movable points, being then horizontal, would fall
below the true horizontal diameter, and the measurements would be more
in error than it would be with the points in any other direction.
Still, if care is taken to preserve the points at the greatest length
possible, a very tolerable degree of accuracy may be attained. In the
inspection of guns arranged on skids, the gun i