MR. W. CROOKES OK REPULSION RESULTING FROM RADIATION. 
275 
that no observations of any value could be taken, the crystal being attracted to one 
side or the other of the glass tube. A thin plate of artificial tourmaline (herapathite 
or iodo -sulphate of quinine) was then used at i, instead of the tourmaline, and with 
this no inconvenience was observed from electrification. 
A few experiments showed me that the red and heating rays of the spectrum were 
most active in repelling this artificial tourmaline, so a platinum spiral, heated to full 
redness by a 2-cell Grove’s battery, was used as the source of radiation.* A large 
silvered reflecting mirror was placed behind the hot spiral and a lens in front of it, 
so as to form a highly luminous image of the spiral on the tourmaline. Close to the 
window c' a large slice of real tourmalinet was mounted as a polariser in a frame, so 
that it could be easily rotated. The apparatus was covered with black velvet, and 
care was taken to allow no light to get to the artificial tourmaline except what had 
passed through the polarising tourmaline. 
A luminous index, reflected from the mirror Jc to a graduated scale, showed the 
movement of the beam. The contact key was pressed down, and the spiral heated for 
time sufficient to allow the index to get to the extremity of the first oscillation. The 
number of degrees on the scale at which the index stopped was noted. Observations 
were sometimes taken with the tourmalines crossed and parallel alternately, and at 
other times a series of observations was taken with the polarising tourmaline one way, 
and then a series with the tourmaline turned the other way. The means of each set 
of observations, which were fairly concordant, are given below : — 
Artificial and natural 
tourmalines. 
Parallel. 
Crossed. 
Group 1 
18-0 
1 7*8 
„ 2 
11-5 
13-5 
Mean . . 
14-75 
15-65 
Observations were now taken by keeping the battery key down for exactly five 
seconds : — 
Artificial and natural 
tourmalines. 
Parallel. 
Crossed. 
o 
O 
Group 3 
3-5 
3-5 
„ 4 
3-4 
3-5 
„ 5 
3-5 
3-5 
Mean . . 
3-47 
3-50 
These results look as if there were no difference between the actions ; but the 
* A platinum ball, beated to redness by a gas flame, was at first tried, but tbe beat from tbe gas was 
too great. 
t I am indebted to Professor Maskelyne for the loan of several plates of tourmaline of very large size. 
2 N 2 
