tably.

=Precautions.=--The following points must be carefully kept in mind,
both during the making of the first tube and afterwards:--

(1) The hottest spot in the oxy-gas flame is at a point very near the
tip of the inner cone of the flame, and silica can be softened best at
this hot spot. The excellence of a burner does not depend on the size of
its flame, so much as on the temperature of its "hot spot," and the
success of the worker depends on his skill in bringing his work exactly
to this part of the flame. Comparatively large masses of silica may be
softened in a comparatively small jet if the hot spot is properly
utilised.

(2) Silica is very apt to exhibit a phenomenon resembling
devitrification during working. It becomes covered with a white
incrustation, which seems to be comparatively rich in alkali.[27] This
incrustation is very easily removed by re-heating the whitened surface,
provided that the material has been kept scrupulously clean. If the
silica has been brought into the flame when dusty, or even after much
contact with the hands of the operator, its surface is very apt to be
permanently injured. _Too much attention cannot be given to cleanliness
by the workman._

[27] The rock crystal exhibits a yellow flame when first heated in the
oxy-gas flame, and most samples contain spectroscopic quantities of
lithium.

(3) When a heated tube or bulb of silica is to be expanded by blowing,
it is best not to remove it from the flame, for if that is done it will
lose its plasticity quickly unless it be large. The better plan is to
move it slightly from the "hot spot" into the surrounding parts of the
flame at the moment of blowing.

It is best to blow the bulb through an india-rubber tube attached to the
open end of the silica tube. At first one frequently bursts the bulbs
when doing this, but holes are easily repaired by stopping them with
plastic silica applied by the softened end of a fine rod of silica and
expanding the lump, after re-heating it, by blowing. After a few hours'
practice these mishaps gradually become rare.

I find it a good plan to interpose a glass tube packed with granulated
potash between the mouth and the silica tube. This prevents the interior
of the tube from being soiled. The purifying material must not be packed
so closely in the tube as to prevent air from passing freely through it
under a very low pressure.

It may be mentioned here that a finished tube usually contains a