Tourmaline } fyc. within Mica and other Minerals. 271 



5. In the most remarkable symmetrical forms like sceptres or 

 maces, resembling some of those symmetrical cavities which I 

 had previously found in the white topazes of New Holland*. 

 See fig. 7. 



6. In some specimens the plates of titanium are actually bent, 

 as in fig. 8. 



7. In little groups of transparent circular plates of a scarlet 

 colour, and having concentric rings. 



When light is reflected from the separating faces of the tita- 

 nium and topaz, it is almost completely polarized ; and at greater 

 angles than that of maximum polarization, colours of singular 

 brilliancy cross the reflected images. These colours are doubt- 

 less connected with the fact, that at some of these faces there 

 are three images of a luminous object seen by reflexion, one of 

 the two outer ones being polarized oppositely to one of the 

 double middle images, as in the case of the multiplication of 

 images in composite crystals of calcareous sparf. 



6. On the Crystals and Cavities in Garnet. 



In the greater number of the crystals of garnet which I have 

 had occasion to examine, I have found many crystals and cavities, 

 and much amorphous matter. In one specimen, in particular, 

 the included crystals form a larger mass than the garnet, which 

 is merely a cement for holding them together. These crystals 

 have various crystalline forms, while some are amorphous, though 

 regularly crystallized in their interior. All these crystals are 

 doubly refracting, and give the colours of polarized light from 

 their small size. 



In another specimen, many of the crystals, in the form of 

 hexagons and rhombic plates, are opake, and exhibit by polarized 

 light the remarkable phenomenon, which I had never before 

 seen, of having luminous edges ; so that when the rest of the 

 crystal and all the field of view is dark, we observe hexagons and 

 rhombs, and other geometrical figures, depicted in lines of red 

 light It is not easy to ascertain the cause of this singular ap- 

 pearance, because we cannot see the form of the crystals where 

 the light exists ; but I have no doubt that the luminous lines 

 consist of light depolarized by reflexion from the sides of the 

 hexagonal and rhombic plates, because the illuminating pencil 

 is much larger than the crystals, and the crystals much smaller 

 than the pupil of the eye, so that light must be reflected from 

 the prismatic faces of the hexagons and rhombic plates if they 

 have sufficiently broad faces, and that light so reflected must 

 enter the pupil of the eye. 



* See Edinburgh Transactions, 1826, vol. x. plate 20. 

 t See Phil. Trans. 1815, plate 15. fig. 2. 



