NATURE 
[FeBruary I1, 1897 
346 
Distance Time required 
Ultra-violet- between to come to 
Jan. 28. light-zero. surfaces. steady reading. 
fsc. divs. from) : : 
12.20p.m. + 15° metallic zero | 4°3 cms. 4 mins. 
2.0 55 +134 ” ” 3°90 45 9 4 
2.10 ,, +121 “e er 2-0 us Sua 
2.20 45 +102 oS Bs TeO 56 Ses 
2.40 4, + 86 - = 0:6 “45 Seco 
2.50 ;, am 169 ” ” 40 » IO 55 
3-0 35 aP UO 5p » Se Ss 
3-20 5, +199 ? ) 770 55 5 ” 
> 
(Sensibility of electrometer 140 sc. divs. per volt. ] 
The fact that in experiments (2) and (6) a longer time 
was required before a steady reading was obtained, 
probably depended on the way the light fell on the sur- 
faces and on variations in intensity of the light. 
In this table we see that the steady electrometer 
reading (which we have called the ultra-violet-light-zero) 
is largely influenced by the distance between the plates, 
being greater the greater the distance. This isa very 
remarkable result. It was first discovered by Righi, and 
very clearly described in papers of his to which we have 
referred. It may be contrasted with the non-difference 
of electrometer readings for different distances between 
the plates in a volta-zinc-copper and single fluid cell. 
Added February 6. [We have also made an exactly 
similar series of experiments with Rontgen rays. The 
same insulated oxidised copper plate was placed inside 
the same tinfoil box, and the Réntgen rays shone in 
between the two metals so as to shine on both. The 
following results were obtained with the oxidised copper 
at different distances. 
February 5, 11.30 a.m. 
Rayszero: Distance between 
surfaces. 
+23°5 sc. divs. from metallic zero I‘2 cms. 
+250 ” 2 23 se 272 yy 
ae eo) ” ” ” woe 5 Oman 
+ 23'0 ” ” ” : (GRO) =, 
We next removed the oxidised copper plate, and sub- 
stituted a polished zinc disc. With it we obtained the 
following results. 
Distance between 
Rays-zero. 
< A surfaces. 
—82sc. divs. from metallic zero ar i CL: 
a 79 ” ” ” I 5 ” 
= 81 ” 22 ” 3 O 45 
— 90 ” ” oe) 79 45 
— 90 ” ” ? 7 5 ” 
7 we 56 
The steady reading of the rays-zero was very nearly 
reached in each case in about 15 secs., but the observa- 
tion was continued for one or two minutes till we found 
the reading steady. 
Thus we see that, as previously found by Mr. Erskine 
Murray, the rays-zero is independent, or nearly inde- 
pendent, of the distance between the opposed metallic 
surfaces. ] 
Towards realising the case of an insulated metal sur- 
rounded by metal of identical surface-quality connected 
to sheaths, we covered over the oxidised copper with 
tinfoil. The tinfoil wall facing it was very rough, and 
not so well polished. The insulated tinfoil was 4 cms. 
distant from the end of the box to which its surface was 
parallel. 
When the ultra-violet light fell on the insulated 
metal alone through a slit, the ultra-violet-light-zero was 
+ 53 scale divisions from the metallic zero. A charge 
given to it, whether positive or negative, was discharged 
slowly. After making these experiments, we again ob- 
served the difference of zeros, and found that now the 
ultra-violet-light reading was at the end of the first four 
minutes +2 scale divisions from the metallic zero ; at 
the end of the next four minutes it was — 8 scale divisions 
from it. 
When the ultra-violet light fell on the disinsulated 
NO. 1424, VOL. 55] 
metal and not on the insulated, the insulated when 
charged retained its charge. 
With the light shining on both through a window 7 
cms. broad, 13 cms. high, both positive and negative 
charges given to the insulated metal were discharged, 
and the ultra-violet-light-zero deviated from the metallic 
zero by — 152 scale divisions. 
This difference was reduced to about — 30 scale 
divisions when the experiments were repeated after the 
apparatus had been left to itself for a night. 
To make similar experiments with the Roéntgen rays, 
it was found necessary to cover the window near the 
lamp with tinfoil gauze connected to sheaths, and the 
window on the opposite side was covered with non- 
perforated tinfoil. In this way direct electrostatic 
induction was avoided. We had also a thin sheet 
aluminium window between the tinfoil gauze and the 
Rontgen lamp. 
When the Réntgen rays fell on both insulated and 
disinsulated metal the rays-zero was —5 scale divisions 
from the metallic zero, and both positive and negative 
charges fell to this zero in a few seconds. 
With the rays shining only on the insulated metal the 
same small difference of zeros was obtained, and both 
positive and negative charges fell to the rays-zero, though 
much more slowly than before—in about four minutes. 
With the Réntgen rays shining on the insulated tinfoil 
through the disinsulated tinfoil gauze, the rays-zero was 
—g scale divisions from the metallic zero, and both 
positive and negative charges were removed in about a 
minute. 
On substituting an aluminium gauze for the tinfoil 
gauze, and sending rays through it on the insulated 
tinfoil, the rays-zero was + 25 scale divisions from the 
metallic zero. 
Added February 6. With a polished zinc disc as the 
insulated metal, and with the same windows to the tin- 
foil box, the Réntgen rays were shed in between the 
insulated zinc and the opposite wall of tinfoil from a slit 
in a lead screen outside. This slit was 4 cms. long by 
Icm. broad. The distance between the two metals was 
7 cms. . The rays illuminated only part of the air space 
between the two, and also a part of the tinfoil covering 
the two windows. 
The following are some of the results obtained :— 
[Sensibility of electrometer 140 sc. divs. per volt. ] 
February 5, 1897. Zinc charged negatively to 285 scale 
divisions from the metallic zero. 
Reading from metallic zero with Rontgen lamp acting :-— 
Time 
— 276 scale divisions after I min. 
— 265 rr, "6 as a aes 
= 255 Se eo 
— 243 ” oun One » 4 
— 22 99 eee ase 73) Sees 
— 214 - a an AO ae 
a 184 8 a” 
g we ies a 
Discharge still continued. 
The zinc was then discharged by metallic connection. 
The readings, with the Rontgen light shining, and the 
two pairs of electrometer quadrants again disconnected, 
were :— 
— 4 sc. divs. from metallic zero after 4 min. 
= 1S ” ”? ” Ik ” 
— 41 ” »” ” 23 ” 
= Sey} ” ” 2 3h ” 
— 61 ” ” 2” 43 ” 
= 67 ” ” 3 55 ” 
SOS) ” oe ” 6s ” 
SS ” ” oe} 7 ” 
The difference between the rays-zero and the metallic 
zero is thus found to be —71 sc. divs., or —0'5 of a volt. 
Immediately after this experiment, we removed the lead 
window and allowed the Réntgen light to shine on both 
