GLASS — JACKSON. 249 
are being worked in the flame, does also appear to hasten devitri- 
fication. 
Mention was made above of the influence of boric anhydride to 
retard the crystallization of glasses in which it is an ingredient. 
Alumina is another substance the presence of which confers upon a 
glass the property of working well in the flame without devitrification. 
More striking, perhaps, as a vitrifying agent is titanic oxide. A soda- 
lime silicate glass was made which could not be worked in a flame 
at all so readily did it devitrify. The substitution of a small amount 
of its silica by titanic oxide converted the glass into one which could 
be heated and worked in the flame almost indefinitely without visible 
change. The statement " without visible change " is true of the 
behavior of this glass ; but some, and notably very soft glasses con- 
taining titanic oxide, become colored in the ordinary blowpipe flame 
through reduction of this oxide to a lower state of oxidation. Zir- 
conium, tin, and thorium oxides have been found to promote the sta- 
bility of the vitreous state in a number of glasses prone to devitrifi- 
cation in the flame. They are mentioned as being of the same chemi- 
cal family as silica and titanic oxide ; but to deal with the effect of 
a number of rarer compounds not generally employed in glass-making 
would take up too much of the rest of the time at my disposal. Arse- 
nic and antimony oxides may also be put among substances which 
render glasses less liable to devitrification; but glasses containing 
these oxides are not suitable for ordinary working in the blowpipe, 
since they darken in the reducing area of the flame. Tin oxide, men- 
tioned above, is also not a generally suitable ingredient, since some 
glasses containing it darken badly through reduction in the flame, 
though others can be made which are quite workable except in the 
hottest kind of blowpipe flame. 
I must leave out of consideration the relation of general composi- 
tion and of varying proportions of ingredients to the tendency of 
glasses to devitrify, and content myself with the remark that for 
glasses of comparable composition those containing soda only as the 
alkali, are usually found to devitrify more readily in the flame than 
those in which the alkali is potash, or a good proportion of potash 
with soda. 
In this lecture it will only be possible to deal more or less briefly 
with opal and colored glasses. Many vitreous bodies which crystal- 
lize fairly readily when heated can be seen to pass through a stage 
in which the material segregating from them appears first as an opal- 
escence increasing in density as the heating is continued and finally 
passing into a visibly crystalline form. A glass approximating in 
composition to NagO-CaO-gSiOo shows this opalescence well before 
small crystalline nodules appear similar to those in the tank glass 
