GLASS — JACKSON. 241 
differ in the readiness with which they show this flow under stress; 
but not any glass is so perfectly solid as to give no indication of move- 
ment under stress if tested by sufficiently delicate means. This ques- 
tion of permanent stability of form in glasses has some bearing on the 
choice of glass for the manufacture of large lenses and prisms, and 
the flowing of the surface of glass under mechanical pressure comes 
in as a very important matter in the explanation of the mechanism 
of polishing glass surfaces. A great deal has already been written 
on this subject, and it would take too long to deal with it here. An- 
other reason why I refer to it but intend to leave it is that I hope it 
will not be long before the published papers and other work of Sir 
George Beilby on the influence of mechanical disturbance on the 
physical state of a very large number of substances will be brought 
together into one connected story, when it will be seen that this sub- 
ject of polishing glass has been dealt with in a comprehensive man- 
ner, and that the principles underlying it are shown to have very 
wide application. 
The question of the relative plasticity of various glasses has two 
important bearings which are of some interest. For many indus- 
trial and scientific purposes it is necessary to be able to seal metallic 
wires into glass, and early in the war some difficulties were experi- 
enced in obtaining suitable glasses. To obtain successful joints be- 
tween the metal and glass without fear of the latter cracking, it 
was generally considered that the glass aimed at should be one which 
had a coefficient of expansion as close as possible to that of the 
metal intended to be used, and there is no doubt that this question 
of expansion has to be taken into account. In making a large num- 
ber of glasses and in experimenting with them there did not appear 
to be that close connection between the coefficient of expansion of 
the glass and its behavior with metal wires which was at first ex- 
pected. The coefficient of expansion of copper is about double that 
of platinum, and the coefficient of iron is about midway between the 
tw^o. Glasses were made which gave successful joints with platinum 
and copper wires, but which cracked inevitably with iron wire. It 
did not appear, therefore, that the coefficient of expansion was the 
only factor to be taken into consideration. It is an important factor, 
but not the only one, and it soon became clear from the study of 
various glasses that the plasticity of the glass had a great deal to 
do with its utility. A careful examination showed that there was 
evidence that in the case of soft metals, like copper and platinum, 
the glass in setting could pull and deform the metal wires so that 
no great strain was permanently left in the glass. With hard metals 
like iron and tungsten, it was necessary to devise a glass which had 
marked plasticity over such a long range of temperature that when 
