MB. W. CBOOKES ON EEPELSION EESTTLTING EEOM EADIATION. 
341 
measure the amount of radiation falling upon it by the velocity Avith Avhich it re- 
volves. It may also be called the Light-Mill. The rapidity of revolution is directly 
proportional to the intensity of the incident rays. Several radiometers, of various con- 
structions, Avere exhibited at the Soiree of the Eoyal Society on April 7th, 1875. The 
folloAving experiments have been tried Avith radiometers of different kinds and very 
varying sensitiveness. I could easily have obtained better curves and closer accordance 
Avith theory by repeating some of the observations Avith more recent instruments ; but 
the results already obtained are sufficient to prove the laAVS, and I could do no more than 
this Avere I to repeat the experiments. 
As soon as the radiometer Avas seen to revolve it Avas apparent that the stronger the 
light the more rapid Avas the movement. The second instrument Avhich Avas made, the 
vacuum being very imperfect, the moving parts (straAV arms and pith surfaces) heavy, and 
the instrument accordingly comparatively insensitive, Avas mounted for the purpose of 
testing its action at different distances. 
The radiometer Avas covered Avith a thin glass shade, and in front of this Avas a large 
sheet of plate glass. The Avhole Avas covered on three sides and the top and bottom Avith 
black velvet, the fourth side admitting the light from the lamp. The source of light 
Avas one of Dietz’s paragon lamps, burning paraffin-oil. This lamp I find gives the 
brightest light and steadiest flame of any I have tried. Black velvet screens Avere put 
round the lamp, except on the side facing the radiometer. The room was darkened, 
and the temperature Avas kept uniform. The radiometer Avas kept fixed, and the lamp 
Avas moved backAvards and forwards along a graduated scale, the number of seconds 
required for the radiometer to make one revolution being recorded by a chronograph 
watch. The following Table gives the results : — 
Distance between 
Radiometer and 
centre of lamp-flame, 
in millimetres. 
Number of seconds 
of time required for the 
revolution of the 
Radiometer. 
150 millims. 
6 seconds. 
200 „ 
8, 9 „ 
250 „ 
11 „ 
300 „ 
15, 16 
350 „ 
20 
375 „ 
21 
400 „ 
23, 24 
450 „ 
29, 31 
500 „ 
34, 36 
550 „ 
40, 42, 44 „ 
600 „ 
52 „ 
650 ,, 
60 „ 
700 „ 
65 ,, 
750 „ 
74 „ 
800 
82, 84 
850 „ 
93, 95 „ 
900 „ 
100, 102 „ 
950 „ 
116 „ 
1000 „ 
129 „ 
1050 
140 
1100 „ 
158 „ 
1150 „ 
170 „ 
1200 
184, 188 
MDCCCLXXVI. 
3 B 
