ME. W. CEOOKES ON EEPTJLSION EESULTING EEOM EADIATION. 
359 
Table (continued). 
Date. 
Portion of spectrum 3 
thrown on to pith. 
Deflection of 
index ray 
of light. 
Observations. 
millims. 
Aug. 17. 
just below A 
90 1 
65 ' 
C 
just above C 
64 
D 
63 
Sun very bright, but small hazy clouds 
b 
48, 42 ' 
> 
floating about. 
F 
35 
„ 
between F and G 
17 
„ 
G 
li 
H 
9 J 
31. 
just below A 
205 1 
C 
between D and E 
E 
155 1 
110 
85 
67 i 
► 
White clouds in patches. Sun occa- 
sionally obscured. Could not work 
after noon. 
„ 
F 
40 
„ 
between F and G 
13 J 
Sept. 1 . 
ultra-red 
35 
133 
„ 
170 
„ 
190 
„ 
127 
„ 
A 
113 
B 
102 
’’ 
between C and D 
79 
Sun very bright, but large white clouds 
floating about. Observations taken 
77, 75 
59 
> 
” 
between D and E 
during clear intervals. 
„ 
b 
34 
„ 
F 
23 
„ 
between F and G 
14 
„ 
„ „ 
13 
„ 
„ „ 
12 
9 J 
„ 
H 
1 
10. 
ultra-red 
21 - 
100 
” 
” 
169 
215 
Sun quite bright up to 12.30 p.m., then 
” 
C 
149 
the sky became rather hazy. 
„ 
E 
82 
„ 
F 
29 
16. 
E 
88 1 
I 
„ 
F 
47 
I 
„ 
G 
18 
> 
Perfectly clear sky. 
„ 
H 
13 
” 
ultra-violet 
11 > 
1 
190. In fig. 14, Plate 35, which is reduced to half the linear scale, I have attempted 
to show the above results graphically. The horizontal band, crossed by vertical lines, 
at the lower part of the figure represents the solar spectrum. The vertical lines to 
which letters are attached are the Fraunhofer lines. The other lines do not repre- 
sent spectrum-lines, but are put there to mark the parts of the spectrum at which 
observations were taken. The spectrum is of half the length which I used in the 
experiments, and the lines are also to scale. The square black portion at the left extre- 
mity shows the relative size of the pith surface in the torsion-apparatus. 
3 d 2 
