GLASS — JACKSON. 255 
Some consideration of such a question, however, is helpful in sug- 
gesting experiments, and some instances ma}'^ be given. One is in 
connection with the composition of a glass to give the full possible 
color wdth copper. The notion of something like chemical combina- 
tion of the copper leads to the study of the effect of varying the 
relative amounts of the basic and acid ingredients of the glass. More 
of the basic part, such as the alkalies, might be expected to turn the 
copper out of combination, and more of the acid part to keep it in. 
It would be tedious to describe in detail the results of numerous 
trials with glasses, and the point can be equally well illustrated by 
simple experiments with borax beads. Copper oxide mixed with 
about twice its weight of tin oxide can be dissolved in molten borax 
in an oxidizing flame and then reduced in a reducing flame. On 
cooling, the bead is either colorless or " strikes " red, according to 
the concentration of the copper. If colorless, it can, with suitable 
concentration, be made to " strike " by reheating. Now, if to the 
bead which " struck " red on cooling, more boric anhydride be added, 
and the bead again fused, it will remain colorless on cooling; but 
unless too much boric anhydride has been added it will " strike " 
red on reheating. Addition now of more alkali, in the form of 
sodium carbonate, will restore the property of striking red at once 
on -cooling. Similarly it follows that a bead which remains color- 
less on cooling, but " strikes " on reheating, can be prevented from 
giving any color of copper at all by more boric anhydride, and the 
property of " striking " red on reheating can be restored by the fur- 
ther addition of alkali. Of course, in making these various addi- 
tions of alkali and acid to the bead there must be a change in the 
concentration of the copper; but a bead can be got in so sensitive a 
condition that a mere trace of alkali will determine whether a red 
color is developed or not, and a number of experiments on glasses 
and glazes do confirm the notion of chemical action playing a part 
on lines which would be expected from general chemical experience. 
When manganese dioxide is added to a glass as a so-called de- 
colorizing agent, it is intended to be left in an oxidized condition, 
so as to give a violet color which will disguise the green color due to 
iron and produce only a slight darkening of a neutral tint, scarcely 
visible except in thick pieces of the glass. Sometimes the violet tint 
is overdone and can easil}^ be seen, and sometimes so much of 
the manganese dioxide has been reduced that the green due 
to iron is fully visible, the lower oxide of manganese giving 
no color to the glass. In many instances of glasses in which 
one would be inclined from mere inspection to say that all the 
manganese dioxide had been reduced in the furnace, it has been found 
that a strong violet color can be developed by exposure to radium 
or by cathode discharge in vacuum tubes. In parenthesis it may be 
