22 Mr. Potter on Photometry in connexion xmth Physical Optics. 



clination to the surface, and in another experiment he found 

 703 reflected. When afterwards he is speaking more parti- 

 cularly on the subject, he says he found at 21° degrees inclina- 

 tion in two experiments 637 and 666 reflected of every 1000. 

 Sir W. Herschel (Phil. Trans, for 1800.) found that his 

 specula reflected 67262 of every 100 thousand rays at an in- 

 cidence nearly perpendicular. 



In my paper on the reflective powers of metals, published 

 in the Edinburgh Journal of Science for 1830, one speculum 

 gave 69"4'5 rays reflected of every 100 incident at 20° in- 

 cidence; another set of experiments gave 68*61 at 10° in- 

 cidence, and another 66'4'2 at the same incidence ; the de- 

 terioration in the successive sets being attributed chiefly to 

 my rubbing the speculum too severely, with the intent to pre- 

 vent being deceived at the higher incidences by any film re- 

 maining on the surface. 



In the beginning of his memorandum he refers to a paper 

 by Mr. Green, B.A., Fellow of Caius College, in this Univer- 

 sity, which appears in the forthcoming volume of the Camb. 

 Phil. Trans, which he says accords with Fresnel's formula. 

 This is an example of my statement that any prima facie case 

 is sufficient with the advocates of the undulatory theory. Mr. 

 Green finds that the intensity of the light reflected at the po- 

 larizing angle of a transparent substance, when the incident 

 light is polarized in a plane perpendicular to the plane of in- 

 cidence is represented by the formula 



for glass and /x = 5*1 this becomes -^^ very nearly, which is 

 more than one-half the light reflected at the perpendicular in- 

 cidence even taking the undulatory formulae, which give g j, 

 and if such a proportion still existed no one would say that 

 light polarized in the plane perpendicular to the plane of in- 

 cidence vanishes at the polarizing angle. 



To test this further I will use an experiment brought for- 

 ward by Mr. Airy, in the fourth volume of the Camb. Phil. 

 Trans, in a paper to which Mr. Green refers. It is there 

 stated, and easily verified, that when Newton's rings are 

 formed by placing a piece of glass on diamond, the rings dis- 

 appear at the polarizing angle for the glass when examined by 

 light polarized perpendicularly to the plane of incidence. 

 Hence there is not light reflected from the glass surface of the 

 thin plate of air, which could interfere with the light from the 

 diamond surface. 



Now in Newton's rings, formed between two glasses and 



