piece
of 3/16 or 7/32 inch tubing is joined to each end of a piece of tubing
5/8 by about 5 inches, and two constrictions made in the large tube, by
the method described on page 10. The small tubes are then bent in the
same plane, as shown, and their ends fire-polished (Fig. 9).

[Illustration: FIG. 9.--Tube for condensing sulphur dioxide.]

EXERCISE NO. 6

BULB AT THE END OF A TUBE

For this exercise tubing of 1/4 inch diameter and moderately strong
walls is selected. A tail is drawn out on one end of the tube, and a
piece of tubing about nine or ten inches long is cut off. The tail
should be carefully drawn in the axis of the tube, and in the same
straight line with it, as it is to be used as a handle in assembling the
glass for the bulb. This tail must be long enough so that it can be
conveniently held in the left hand, as described on page 13, and rotated
about the same axis as the main tube. Holding the main tube in the right
hand and the tail in the left, the tube is rotated in a large flame so
that a piece of it, beginning where the tail stops and extending about
an inch to the right, may be uniformly heated to the highest temperature
at which it can be kept in shape. As soon as this temperature is
reached, the tube is removed from the flame, continuing the rotation and
taking care not to draw out the heated part, and gently blown. The
rotation is carefully continued during the blowing, holding the tube in
approximately a horizontal position. As soon as the tube has expanded a
little the tail is pushed gently toward the main tube, continuing the
gentle blowing. If this is properly done, the heated piece of tube will
become a short bulb of about double its original diameter, and about the
same wall thickness as the original tube. It will have somewhat the
appearance of _a_, Fig. 10, when properly manipulated.

[Illustration: FIG. 10.--Blowing a bulb on the end of a tube.]

The tube is now reheated as before, taking care this time that the
heating extends over all that part of the bulb to the right of the
dotted line in the figure, as well as part of the main tube adjoining.
If this heating has been properly placed, when the operation of blowing
and pushing together is repeated the result will be to lengthen the bulb
into a uniform cylinder, as shown in _b_, Fig. 10. Otherwise the result
will be a series of bulbs, as in _c_, Fig. 10, separated by thickened
ridges which will be almost impossible of removal l