OF THE RETINA TO LIGHT AND COLOUR. 
Ifi9 
just necessary to fail to produce the sensation of colour takes the same form, the 
exponential coefficient being changed. The connection between star magnitudes and 
the intrinsic brightness of a measurable disc of the same brightness and colour will 
have the same form. 
8. The Extinction Dependent on the Least Diameter of the Aperture. 
There is a peculiarity in the result, in that it looks as if the diameter of the 
aperture and not its area, determined the intensity of the light required to be 
extinguished. A very ready way of ascertaining this was to use an adjustable slit, 
and to take the extinctions with varying apertures. In such a case the length 
of the slit would remain constant, whilst the width alone would vary. If it were 
the latter that determined the extinction there can be but little doubt that the area 
of the object played at all events a secondary part; for in the case of the circular 
apertures the areas would be as the squares of the diameters, whilst in the latter 
they would be as the width of the aperture. A slit from a spectroscope was 
employed which increased in aperture inch for each complete revolution of the 
screw. The head of the screw was divided into twenty-five divisions, so that five 
divisions on the screwhead opened the aperture *004 inch, the smallest aperture 
employed was *00155 inch and the largest *06555. Intermediate apertures were 
also used, and extinction of white light and red light noted. 
The following' table gives the results of the measures, the apertures being put 
in powers of 2 :— 
Table VII. 
White light. 
Red light. 
Absolute 
width 
Width in 
Readings. 
Absolute 
width 
Width in 
Reac 
ings. 
po-v\^ers of 2. 
in inches. 
powers ot 2. 
Degrees. 
Logs. 
in inches. 
Degrees. 
Logs. 
0-06155 
-4-02 
350 
0-99 
0-05355 
— 4-22 
192 
2-35 
0-05355 
-4-22 
342 
1-06 
0-04555 
-4-46 
176 
2-45 
0-04555 
-4-46 
323 
1 23 
0-037.55 
-4-73 
154 
2-68 
0-03355 
-4-90 
298 
1-44 
0-02955 
-5-08 
135 
2-84 
0-02955 
-5-08 
285 
1-55 
0-02555 
-5-29 
124 
2-94 
0-02555 
-5-29 
275 
1-63 
0-02155 
-5-53 
110 
3-05 
0-021.55 
-5-53 
267 
1-72 
0-01755 
-5-83 
98 
3-16 
0-01755 
-5-83 
250 
1-85 
0-01355 
-6-20 
76 
3-35 
0-01355 
-6-20 
228 
2-04 
0-00955 
-6-67 
50 
3-57 
0-00955 
-6-67 
202 
2-26 
0-00555 
-7-49 
156 
2-67 
0-00155 
- 9-33 
54 
3-53 
The results are plotted diagrammatically in fig. 6, from which it will be gathered 
that again the curve is a straight line in each case, and that the ordinates bear the 
VOL. CXC.—A. 
z 
