Dr. S. P. Thompson on Photometry. 121 



to its own thickness between the two illuminated surfaces. 

 In that case the remedy, applied by Foacault in the modified 

 instrument which bears his name, was to shorten the partition 

 so as to permit the two illuminated parts to come just into 

 optical contact. In the case of the oblique opaque screens the 

 author tried to remedy the defect by several modifications, 

 which, though not described at the time, proved useful. In 

 one of these each surface was extended so as partially to over- 

 lap the adjacent surface, as indicated in figs. 2 and 3. In 



Fitr. 2. 



Fig. 3. 





another instrument the overlap was given as shown in figs. 4 

 and 5. 



Fig. 5. 



i 



The principle of overlap in photometers operating by the 

 diffuse illumination of opaque inclined screens has since been 

 used by Sir John Conroy *, who employed two pieces of white 

 writing-paper, inclined at about 60° to one another. He 

 found, as I did, that 90° is too large a dihedral angle for 

 exact work. Materials such as card and paper are never 

 entirely devoid of specular reflexion. At an incidence of 45° 

 on each surface there is so much specular reflexion of the 

 light as quite to vitiate the observations. 



* Phil. Mag. 1883, vol. xv. p. 423. 



