me should be
just large enough to heat the whole of the cone. As the latter shrinks,
the lump at the joint is brought into the edge of the flame, and it and
a very little of the small tube allowed to shrink with the cone.

When well shrunk and heated to blowing temperature the joint is removed
from the flame and blown gently with careful rotation, pushing the tubes
together a little when the blowing is about finished, so that the cone
becomes a short thick half-bulb, as shown in _d_, Fig. 8. This
corresponds to the first thick bulb in the first method (_d_, Fig. 6),
and is treated similarly. It is again heated and shrunk, taking care not
to involve either the large tube or the small one in the shrinking,
blown quickly to about the same shape as before, (_d_, Fig. 8), and then
gently drawn out into a smooth cone (_e_), exactly as in the first
exercise. Care should be taken not to draw too rapidly or too far, as
then the resulting cone (_f_) is weaker than it should be, and does not
look well.

=Discussion.=--The beginner will find that this operation is best
learned on two tubes which are not too nearly of the same diameter. A
tube about 5/8 inch in diameter and one a little less than 1/4 inch will
be suitable. Both should have moderately heavy walls (1/16 inch or a
trifle over for the large tube, and a trifle less for the small one) but
the large tube should not be too heavy or else it will be hard to
prevent melting down too much of the small tube, and getting this drawn
out too thin during the process. One of the troublesome features of this
exercise is the difficulty of rotating two tubes of different diameters
with the same angular velocity, so as not to twist the joint. Another
difficulty is found in getting the cone uniformly heated to blowing
temperature without overheating and overshrinking the small tube. The
reason for this is obviously the much greater circumference of the cone,
especially at its large end, so that relatively much less of it is being
heated at any time. The beginner is also inclined to start with too long
a cone, or else heat so much of the large tube that part of its glass is
included in the cone, with the result that in order to get the right
wall-thickness the cone must be made too long (_g_, Fig. 8). This does
not look well, and usually will be irregular in shape.

EXERCISE NO. 5

TUBE FOR CONDENSING SULPHUR DIOXIDE

This is useful as a test of mastery of the preceding exercise. A