OF THE RETINA TO LIGHT AND COLOUR. 
159 
closely of the same size on A, and by adjusting the mirror, M", the two patches 
could be superposed. Kotating sectors with adjustable apertures were placed in the 
path of the reflected beam as shown. A rod, It, cast two shadows, which were 
arranged to touch one another. When the undiminished reflected beam fell on A 
the two shadows were made equally luminous by rotating B. The number of degrees 
from the zero point was then noted. The sectors then were set so that half the 
light of the reflected beam was cut off, and equality of illumination of the shadows 
again secured by rotating B. The number of degrees was again noted. When the 
sectors were set at 45° the same procedure was adopted. If the coefficient of 
absorption (obstruction) did not vary, the intervals between first and second readings 
and between the second and third should be the same. By again setting the angle 
of the sectors to give one-eighth part of the light, another reading could be obtained, 
and so on. By altering the angle of M' a new set of readings could be obtained, till 
every part of the annulus had been tested. By altering the position of the slit, S, in 
the spectrum, the obstruction coefficient for any colour could be arrived at. It will 
be seen that by this method one beam of light alone is utilized in a very simple 
method, and is in fact a modification of the method adopted when graduating wedges 
in white light, as described in the monthly notices of the Eoyal Astronomical Society 
four years ago. From the extreme red to beyond G no difference was found in the 
“ obstruction coefficients above that which would arise from pure error in observa¬ 
tion. The following is the adopted coefficient for the annulus mostly employed :_ 
Table I.—Scale of Wedge, the coefficient being *0086 for each degree. 
Degrees. 
Log. 
Value of the light 
penetrating. 
DegTee.s. 
Value of the light' 
penetrating. j 
0 
4-000 
10,000 
190 
2-366 
' 232 
10 
3-914 
8,200 
200 
2-280 
190 i 
zO 
3-828 
6,870 
210 
2-194 
156 : 
30 
3-742 
5,500 
220 
2-108 
128 
40 
3-606 
4.520 
230 
2-022 
105 
to 
3-570 
3,710 
240 
1-936 
86-1 , 
GO 
3-484 
3.040 
250 
1-850 
7T5 
70 
3-398 
2,500 
260 
1-764 
58-0 
80 
3-312 
2,050 
270 
1-678 
47-8 
90 
3-226 
1,680 i 
280 
1-592 
39-0 
100 
3-140 
1,380 i 
290 
1-506 
32-0 
no 
3-054 
1,130 ! 
30U 
1-420 
26-2 
120 
2-968 
925 1 
310 
1-334 
21-6 
130 
2-882 
760 i 
320 
1-248 
17-8 
140 
2-796 
625 I 
330 
1-1'62 
14-5 
150 
2-710 
512 ! 
340 
1-076 
1T9 
160 1 
2-624 
420 
350 
0-990 
9-7 
170 
2-538 
345 
360 
0-904 , 
8-0 
180 
2-452 
283 
1 
1 
i 
