192 



Sir J. Conroy. 



[Jan. 10, 



Table VI. — Speculum metal, with Light polarised perpendicularly to 

 the Plane of Incidence. 



A n crl p nf 

 In oi rl pn pp 



A. 



B. 



C. 



D. 



Mean. 



o 



30 



59-31 



57-86 



59 -83 



59 -63 



59-16 



40 



53 -30 



54-01 



56-41 



54-29 



54-50 



50 



49 -47 



51 -44 



49-61 



49-69 



50-05 



60 



41 -50 



43 -36 



44-02 



43 -83 



43-18 



65 



39-95 



39-12 



40-50 



40-85 



40-10 



70 



38-27 



35-84 



37-42 



38-29 



37 '45 



75 



36 -20 



34-45 



36-84 



35 -88 



35-84 



80 



40-51 



38-67 



41-22 



41-15 



40*39 



The principal incidences and azimuths for both plates were deter- 

 mined in the manner described in " Proc. Roy. Soc," vol. 31, p. 486, 

 and vol. 35, p. 32. Pour observations were made in each position of 

 the retarding plate, two with the principal section of the polarising 

 Nicol on the right, and two with it on the left of the plane of inci- 

 dence. The means of the several sets of eight observations are given 

 in Table VII ; the probable error of the mean result, calculated by 



the ordinary formula 0*6745 a / ^ - is also given. 



V n{n — 1) 



Steel 



Speculum metal. 



Principal 

 incidence. 



Table VII. 



Mean. 



Principal 

 incidence. 



28° 31' 



Mean. 



Prob. error. 

 20' + 5' 



31 + 7' 



75 39 

 75 32 



Prob. error. 



29' +r. 



+ 6'. 



33 23 

 33 01 



A beam of ordinary light being equivalent to two beams of half its 

 intensity polarised at right angles to each other, the percentage of 

 light reflected by the plates, if ordinary unpolarised light were inci- 

 dent upon them, is given by half the sum of the intensities of the 

 light polarised in, and perpendicular to, the planes of incidence. 



