te sufficient.

The diagram, of the tube shows that the main tube has been expanded
round the edges of the cathode. This is to reduce the heating
consequent on the projection of cathode rays from the edges of the
disc against the glass tube.

The anode is inserted into its bulb in a quite similar manner. If
desired it may be made considerably smaller, and does not need the
careful adjustment requisite in sealing-in the cathode, nor does the
glass near the entry wire require to be so thick.

More intense effects are often got by making the cathode slightly
concave, but in this case the risk of melting the thin glass is
considerably increased. No doubt, Bohemian glass might be used
throughout instead of soft soda glass, and this would not melt so
easily; the difficulties of manipulating the glass are, however, more
pronounced.

It will be shown directly that the best Roentgen effects are got with
a high vacuum, and it is for this reason that the glass near the
cathode seal requires to be strong. The potential right up to the
cathode is strongly positive inside the tube, and this causes the
glass to be exposed to a strong electric stress in the neighbourhood
of the seal.

Although the glass-blowing involved in the making of a so-called focus
tube is rather more difficult than in the case just described, there
is no reason why such a difficulty should not be overcome; I will
therefore explain how a focus tube may be made.

Fig. 41.

A bulb about 3 inches in diameter is blown from a bit of tube of a
little more than 1 inch diameter. Unless the walls of the tube are
about one-eighth of an inch in thickness, this will involve a
preliminary thickening up of the glass. This is not difficult if care
be taken to avoid making the glass too hot. The larger gas jet
described in connection with the soda-glass-blowing table must be
employed. In blowing a bulb of this size it must not be forgotten
that draughts exercise a very injurious influence by causing the glass
to cool unequally; this leads to bulbs of irregular shape.

In the method of construction shown in Fig. 41, the anode is put in
first. This anode simply consists of a square bit of platinum or
platinum-iridium foil, measuring about 0.75 inch by 1 inch, and
riveted on to a bent aluminium wire stem.

As soon as the anode is fused in, and while the glass is still hot,
the side tube is put on. The whole of the anode end is then carefully
annealed. Whe