190 Astronomical and Nautical Collections, 



completely polarised by reflection before their introduction 

 into the crystallized plates, as in the apparatus already 

 described : and let us suppose that the axes of the two plates 

 are perpendicular to each other, and each make an angle of 

 45° with the plane of reflection. According to the theory 

 of moveable polarisation, all the emerging rays must b6 

 polarised in a direction parallel or perpendicular to this 

 plane, which is that of the primitive polarisation : thus each 

 of the two groups of fringes, which are observed to the right 

 and left, results from the interference of the two pencils 

 polarised both in this plane, or both in the direction per- 

 pendicular to it : consequently, if the two groups of fringes 

 could exhibit signs of polarisation, it could only be in one or 

 the other of these orthogonal directions : now the experiment 

 is as opposite as possible to this consequence, since it is pre- 

 cisely when we place the principal section of the rhomboid 

 in one or the other of these directions, that the two images 

 of each group possess the same intensity : and in order that 

 one of them may vanish, it is necessary, on the contrary, that 

 the principal section of the rhomboid should make an angle 

 of 45° with these directions, that is to say, that it should be 

 parallel or perpendicular to the principal sections of the two 

 plates. When it is parallel to the left hand plate, it is the 

 left hand group that disappears from the ordinary image, 

 and the reverse. It is obvious that the direction of the 

 polarisation is the same as in the experiment last related, in 

 which the incident light had not undergone any previous 

 polarisation, before it passed through the crystallized plates. 



Thus, whether we employ direct or polarised light, the 

 ordinary and extraordinary pencils into which it is divided 

 in passing through a crystallized plate, are always polarised, 

 the one in the plane of the principal section, and the other in 

 a direction perpendicular to it. 



We have hitherto employed plates not less than a twentieth 

 of an inch in thickness, and we have constantly found, in the 

 ordinary and extraordinary rays, the same direction of pola- 

 risation as they manifest when they are separated into distinct 

 pencils. It was, however, interesting to ascertain also, by 

 means of interferences, whether the same mode of polarisation 

 was also to be found in much thinner plates, such as those 



