202 Dr. H. R. Wright on the Photometry of the 



photometer. The plate (a) (see figure) is fixed with the surface 

 in a vertical plane on the table (b), which turns round a vertical 

 axis and carries a graduated circle (c) . The source of light 

 has a fixed position and is not to be moved ; the angle of 

 incidence is varied by turning the table (b). The graduated 

 circle (d) concentric to (c) determines the angle of emission. 

 Contrary to other investigations in this line only one source 

 of light has been used instead of two. The advantage of this 

 method is self-evident inasmuch as all variations of the 

 intensity are eliminated. 



The Standard of Comparison is drawn from the source of 

 light which illuminates the plate in the following way : — A 

 mirror (/) placed at 45° to the horizon, the centre of which 

 is in the vertical axis of the graduated circle, throws the 

 parallel horizontal rays of light vertically down upon a diffuse 

 reflecting plate (a'). This little plate, of the same substance 

 as the one which is to be studied, is inclined at nearly 45° and 

 projects the light scattered to the slit of the photometer. At 

 the commencement the size and angle can be suitably varied 

 between certain limits so that one is able to gain very nearly 

 the same intensity of the standard of comparison as the plate 

 itself has. This, of course, makes the measurements more 

 exact. 



The Source of Light is an arc-lamp of about 500 candle- 

 power, which burned with great constancy after having been 

 carefully regulated. This was chosen as the best means of 

 procuring a parallel beam of light of great intensity, such as 

 the theory demands. 



As Photometer Glands * spectrophotometer was used. In 

 front of the slit 1 placed Crova's f little plate with convergent 

 edges. The width of the slit was only 0*4 mm. To be 

 prepared for all eventualities the Wollaston prism was turned 

 so that light with a vertical plane of polarization entering 

 the lower half of the slit vanishes ; because the traces of 

 reflected polarized light vibrate in a vertical plane when at all 

 visible. 



The Selection of the Parallel Pays out of the Scattered Light. 

 — The theory demands not only for the incident rays but also 

 for the reflected ones that they should be parallel ; not conical 

 as was markedly the case with other observers. As each 

 particle scatters light in all directions, we have to separate 

 one of these rays from the others and to prevent the latter 

 from reaching the photometer. For this purpose it is best 

 to make use of a cylinder-lens (</) placed at its focal distance 



• * P. Glan, i Ueber em neues Photometer,' VVied. Ann. i. p. 351 (1877). 

 t M. A. Crova, Ann. chim. et. phys, (5) xix. p. 495 (1880). 







