GLASS JACKSON. 243 
and not merely single substances as in the previous illustrations. 
Solutions of sodium acetate and Rochelle salt, obtained by adding to 
hot water as much of the salts as will dissolve and cooled so that no 
unfiltered air can enter the containing vessels, and carry nuclei to start 
crystallization, remain clear and fluid at the ordinary temperature. 
If such a '' supersaturated " solution of sodium acetate be cooled in 
liquid air it is converted into a vitreous solid quite clear and trans- 
parent. On removal from the liquid air its temperature rises and very 
soon crystallization starts and proceeds right through the vitreous 
mass. The solution of Rochelle salt treated in the same way yields a 
similar vitreous mass, but as it warms up no crystallization takes 
place. It slowly goes back to the original liquid condition. The 
cold vitreous sodium acetate solution may be taken as analogous to a 
glass from which crystals readily separate on warming up, while 
the Rochelle salt is analogous to a glass which shows no tendency 
to crystallize through the whole range of temperature from the 
solid form to the point at which it is a mobile liquid. Glasses are 
known which tend to crystallize in all degrees of readiness. As a 
simple glassy substance, zinc silicate maj^ be taken. It can be ob- 
tained, by moderately quick cooling of the molten mass, in a vitreous 
form which is stable for a number of years — at least, some has re- 
mained with no sign of crystallization at the ordinary temperature 
for 22 years. By heating to a few degrees above its softening point, 
it changes to a translucent crystalline mass. Taking a more complex 
glass, but still a moderately simple one, we may study the behavior 
of heat on a lime soda silicate glass. The specimen of greenish glass 
with some large and many smaller opaque nodules in it was given 
to me in the early part of the war by Mr. Frank Wood as an example 
of glass taken out of a tank furnace. The nodules are calcium 
silicate, which is the least fusible silicate potential in the glass. 
They have been formed through the glass in the particular part of 
the furnace from which it was taken being at a temperature too low 
to keep this silicate in solution or combination, and it has separated 
out in the form shown. To get this glass back to a complete vitreous 
state again would require a somewhat higher temperature than is 
used in the manufacture of the glass, since the calcium silicate is 
itself very infusible, and the rest of the glass has to be made very 
fluid before such masses of this silicate can be dissolved in any 
reasonable time. 
This glass, then, in a molten state, is an example of a solution out 
of which a slightly soluble constituent has separated, when the condi- 
tions were suitable for that separation. It may be interesting to 
turn for a moment to a consideration of conditions for crystallization 
and to go back to simple glassy bodies such as zinc silicate and some 
borates. With all the vitreous substances which have been tried and 
