252 
MR. W. CROOKES ON REPULSION RESULTING FROM RADIATION. 
These actions are of interest as showing a much stronger proportionate action 
behind a water screen than with no screen. The average of the first column with 
no screen is 55*8, whilst that of the water screen column is 6 5 '2. At the conclusion 
of this tabulation of results I shall collect together the variations in the actions and 
comment upon them. The strong action on the oxide of tungsten is probably caused 
by its dark colour, the blue appearing almost black. 
231. Table VIII. — Dyes and Colouring Matters of Organic Origin. 
No screen. 
Water screen 
interposed (5 millims.). 
Lampblack ( standard disk) 
100-0 
100-0 
Bismark brown 
52-7 
42-6 
Fluorescin 
52-2 
29-0 
Magdala rose 
47-0 
32-5 
Eosen 
43-6 
27-7 
Saffranin 
41-0 
52-5 
Product of the decomposition by acids of the 
green colouring matter of leaves (from 
Professor Stokes) . 
39-2 
47-1 
Aniline scarlet 
37-0 
21-7 
Isatin . 
34-5 
15-3 
These organic substances do not show many striking variations. The saffranin, and 
the product of the decomposition of chlorophyll show an increased ratio of action when 
the heat rays are cut off by water. From the appearance of the spectrum of the 
chlorophyll product, Professor Stokes, who kindly prepared me the specimen, was led 
to believe that it would be little affected by invisible heat rays, but affected to a 
considerable extent by light. The figures in the two columns show that this 
theoretical reasoning is quite borne out by facts. The results obtained with the other 
organic colouring matters show that when the invisible heat rays are cut off, the 
action declines in proportion to that on the lampblack. 
Leaving out saffranin and the chlorophyll product, the mean actions of the other 
substances are with no screen, 44*5 ; with a water screen interposed, 28 T. The 
results obtained for Bismark brown and aniline scarlet, behind a water screen, are 
not very satisfactory, and must only be considered approximations to the truth. It 
was difficult to get a definite position of rest for the luminous index, as after a 
certain time the oscillations increased in amplitude as the light continued to shine on 
the disk. 
