530 
MR. W. CROOKES ON REPULSION RESULTING PROM RADIATION. 
Table I. 
Tension of enclosed 
air, in millims. of 
mercury. 
Temp. = 16° C. 
Bar. =772-55 millims. 
Amplitude of half 
oscillation, in millims., 
at end of 40" obser- 
vation. 
772-55 
+ 0-46 
557-50 
+ 0-54 
472-00 
+ 0-49 
372-00 
+ 0-39 
322-00 
+0-41 
272-00 
+ 0-28 
242-00 
+ 0-18 
22200 
+ 0-15 
201-00 
+ 0-11 
167-00 
+ 0-12 
140-00 
0-07 ? 
114-50 
0-08 ? 
89-50 
0-12? 
70-50 
0-03? 
54-00 
1-02 ? 
48-00 
+ 0-12' 
37-00 
+ 0-14 
29-00 
+ 0-14 
20-00 
+ 0-18 
14-00 
+ 0-30 
9-15 
+ 0-46 
6-55 
+ 0-66 
4-65 
+ 1-00 
3-15 
+ 1-40 
2-25 
+ 1-48 
1-15 
+ 1-72 
0-75 
+ 1-70 
0-65 
+ 1-46 
0-55 
+ 1-04 
0-35 
+ 0-64 
0-25 
-0-60 
0-15 
-1-16 
-0-05 
-5-90 
minimum between a tension of 50 millims. and 150 millims., then rises as the pressure 
diminishes, until, at a tension of 1T5 millim., the attraction is nearly four times what 
it was in dense air. Above this exhaustion the attraction suddenly drops and changes 
to repulsion, which at the best vacuum I could get was nearly thirteen times stronger 
than the attraction in air. 
The last figure in the first column requires explanation. All the others are obtained 
by subtracting the height of the gauge from that of the barometer, and are positive. At 
the highest rarefactions, however, I get the gauge about 0-05 millim. above the baro- 
meter (85, note ) ; the sign, therefore, becomes negative. 
Table II. agrees in the main with Table I. The sign changes to repulsion at pres- 
sures corresponding to those queried in Table I. ; the repulsion, though slight, was 
unmistakable. At 102 millims. pressure the observation has a positive sign. This 
looks like an error ; but as it is so recorded in my notebook, and as I was at that time 
specially looking for repulsions, I do not feel justified in altering it. What I have called 
