ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 341 



be ; the mirror must be turned so that its surface is at right augles to 

 the bisecting line. All this takes a little trouble at first, but after it 

 has been set up once or twice, it does not take thirty seconds to place the 

 separate parts in their proper positions. The results obtained from this 

 polarizer are, apart from special optical test, practically indistinguishable 

 from those derived from a large Nicol, v;\i\ch would cost more than the 

 whole microscope and its outfit ! For example, with a li-inch objective 

 and a f -cone from Quekett condenser, A eyepiece, and a granite section 

 on stage, flat of flame of paraffin lamp being the illuminant, the image 

 was perhaps too bright for the eye when the analyzer was rotated so as- 

 to give a dark field ; this shows how satisfactory a cheap polarizer can be. 

 Fifty years ago black glass was used, but it gives very poor results ; 

 there was, however, another form of polarizer used about that time. It 

 was made of a pile of twenty cover-glasses placed obliquely at an angle 

 of 33° in the substage. When light is transmitted through a pile of 

 plates so placed it will be polarized, but the direction will be at right 

 angles to that produced by reflexion ; this means that an analyzer, to 

 obtain a dark field, must be rotated through an angle of 90° from its 

 former position. I have never seen one of these polarizers fitted to a 

 microscope, but I am going to try it when opportunity occurs. 



" The apparatus, therefore, required for these polarizing experiments^ 

 beyond that found in an ordinary microscope outfit, is a Glan-Thompson 

 prism to fit over the eyepiece, a selenite | wave-plate, and a pile of eleven 

 plates, which takes the place of the concave mirror, as a concave mirror 

 is seldom, if ever, wanted." 



C. T. Whitmell* writes as follows : — " Having had a good deal of 

 experience with this subject, I send a few notes supplementary to 

 Mr. Nelson's interesting letter. For the polarizing reflector he recom- 

 mends about a dozen disks of thin glass kept from contact by little spots 

 of paper. But it is quite unnecessary to separate them. It is a dis- 

 advantage, for dust will get in. The disks should be in contact, and, to 

 guard against dust, it is as well to fasten a strip of paper round the 

 edges, taking care that none of the adhesive used is drawn in between 

 the disks. These are really separated by thin layers of air ; they ar^ 

 not optically in contact. For a transmission polarizer he recommends 

 a pile of twenty glass slips placed so that the light makes an angle of 33° 

 with the surfaces. But better results will be obtained, and with less 

 trouble, by using, say, half this number, and arranging for the rays to 

 pass at a smaller angle with the surfaces. It is quite remarkable how 

 considerably light may be polarized by transmission through even a 

 single glass plate if the incidence is nearly grazing. By a rough calcula- 

 tion, about 60 p.c, of the light, transmitted through a single glass plate, 

 is polarized perpendicular to the plane of incidence, when the ray makes 

 an angle of 1° with the surface. When the angle is 33°, an infinite 

 number of plates will be required to produce complete polarization. 

 Both examples are inconvenient in practice. I should advise about ten 

 slips, and make the angle as small as convenient. The polarization by 

 transmission is never perfect, as it is by reflexion. In order that the 



* English Mechanic, cv. (1917) p. 198. 



