optical properties and stmcture of heated glass , &c. 7 
of the shattered drops. One of these, which was about the 
sixtieth of an inch thick, did not possess the property of de- 
polarisation, and with more than twelve fragments of different 
thicknesses, below the thirtieth of an inch, I obtained a similar 
result. A fragment, however, of a crown glass tear, which 
had burst after being dropped into the water, and which was 
about two-tenths of an inch thick, depolarised light in every 
position, but did not exhibit any coloured rings by polarised 
light. 
I have not been able to make a complete series of experi- 
ments on the effects of heat upon crystallized bodies, but it 
will appear from the following experiment that they are not 
likely to conduct us to new results. I heated to a great de- 
gree a fine crystal of spinelle ruby, which has not the pro- 
perty of double refraction, but it did not produce the least 
change upon a polarised ray. The crystal was ~ of an inch 
thick ; and a piece of crown glass of the same thickness, and 
brought to the same temperature, depolarised a considerable 
portion of light. 
The effects of heat, as indicated by the preceding experi- 
ments, are, perhaps, too imperfectly developed to authorise us 
to draw those important conclusions, to which they seem so 
well calculated to conduct us. One of these, however, is so 
palpable, and so clearly deducible from the phenomena, that 
it must already have suggested itself, namely, the production 
of a new species of crystallization by the agency of heat alone. 
When light is transmitted perpendicularly through a plate of 
glass, the glass exercises no more action upon it, than if it 
were a mass of water. When the glass, however, is heated, 
the particles not only expand, but assume a new arrange- 
