192 Proceedings qf'the Royal Society ()f' Edinburgh. 



1st, They are confined to certain specimens of glass, but not to 

 ■perfectly parallel plates. 



2d, They are exhibitec^ while the plates are at a considerable dis- 

 tance from each other, provided their surfaces are preserved as nearly 

 parallel as possible ; and they are not alFected even by the interposi- 

 tion of another plate between those by which they are formed. 



3d, They are destroyed by the application of Canada balsam or 

 oil of turpentine to any one of the four surfaces. 



4th, They are of uniform breadth and appearance, so long as the 

 disposition of the plates remains the same, and are not affected by 

 pressure in whatever manner it may be applied. The Newtonian 

 fringes, on the other hand, are affected, both in breadth and direction, 

 by the manner in which the plates are pressed together, so that they 

 can be produced at right angles, or in any other position, with re- 

 spect to the new bands. 



5th, The fringes under consideration are produced by the light 

 which is returned inwards upon the plates by reflection from their 

 anterior surfaces, so that the rays suffer three reflections and four re- 

 fractions before reaching the eye. 



(yth. They present phases of revolution which follow a different or- 

 der according to the surfaces that are placed together, and varying, 

 as the revolution is made, from right to left or from left to right. 

 The bands revolve only at half the rate at which the plates move 

 during the first semi-revolution ; and during the last quarter of revo- 

 lution, when certain faces are together, there is a complete breaking 

 up of the rectilinear fringes, which spread themselves in curvilinear 

 forms over the whole surface of the plates. 



'Jth, When viewed by homogeneous light, the fringes appear as 

 light and dark stripes, covering the entire surface of the plates, and 

 of uniform breadth which cannot be altered, except by changing the 

 arrangement of the plates ; but by turning one of these, both the 

 breadth and direction of the stripes are changed. Their number varies 

 from 10 or 12 to nearly 2000. 



Sth, The plates which exhibit the fringes do not display any symp- 

 toms of possessing the doubly refracting structure, when viewed by 

 polarized light. The appearance of the bands, however, is the same 

 as that of the fringes, produced by crossing wedge-shaped plates of 

 sulphate of lime, and passing polarized light through them, as de- 

 scribed by Dr Brewster. It is therefore probable, that the new 

 fringes are occasioned by the intersection of oppositely polarized 

 pencils of light, whose polarity is induced by the repeated reflections 

 and refractions which they undergo in passing and repassing through 

 the plates — and it would seem as if each surface exerted an inde- 

 pendent and peculiar polarizing effect on the rays, — a hypothesis 

 which appears necessary to account for the phenomena attending a 

 change in the disposition of the surfaces. 



9th, They possess a peculiar property, which the author conceives 

 will render them available for the purposes of micrometry. When 

 the surfaces of the plates are parallel, two of the bands are united 

 into one at the centre ; but if a film be introduced between the plates, 

 so as to cause them slightly to diverge, the two bands in the centre 

 will be separated, and move laterally from each other, still preserving 



