—— Or 
IN PLATES, TUBES, AND CYLINDERS OF GLASS. 365 
5. On the Conversion of Plates of Glass with One Axis into 
Plates with Two Axes. 
From the different experiments which I have described, 
both in this and in other papers, it appears, that in almost 
every case where polarising forces are developed, two negative 
structures are separated by a positive structure, or two positive 
structures by a negative structure, both of which are simulta- 
neously produced, like the two opposite polarities in electrici- 
ty and magnetism. When the two structures are disposed in 
a different manner, or when only one structure is developed, 
the regular arrangement of the polarising forces may be effec- 
ted by a slight change in the form or in the mechanical condi- 
tion of the plate. 
If we take a piece of unannealed glass perfectly circular, we 
have only one axis, or one structure, as shewn in Fig. 10., where 
AB, CD is the black cross, which preserves the same appearance 
by turning the plate round its centre. But if we grind the 
smallest quantity from any two opposite sides CB, AD, so as to 
induce a slight degree of ellipticity, a new axis is created; the 
tint which it produces appears at the centre of the plate, and the 
system of rings has the form shewn in Fig. 8., where the inter- 
nal structure within the black fringes AD, CB, is negative, and 
the two external structures without AD, CB, positive. 
If, on the contrary, we now grind a small quantity from 
the sides AC, BD, so as to reconvert the elliptical plate into a 
circular plate, we shall find that the new axis which was gene- 
rated by the change of form, has entirely disappeared, so that 
the plate exhibits the figure shewn in Fig. 10. 
This singular experiment, in which one of the axes may be 
extinguished and revived at pleasure, is worthy of the most 
attentive 
