346 
Dr.  Gundrv  on  the  Asymmetrical  Action  of  an 
NI 
It  is   seen  that  only  for  high  frequencies  is  the  relation 
=  const,  in  any  measure  satisfied. 
The  sudden  change  in  the  magnitude  of  the  effect  between 
N  =  400  and  1ST  =  600  appeared  so  striking,  that  I  repeated 
the  measurements  on  a  new  solution.  The  results  are  shown 
in  Tables  IV.  and  V.  In  this  series  of  experiments,  the 
resistance  W  was  30,000  ohms.  For  the  sake  of  comparison, 
I  have  reduced  the  observed  current  I  to  the  value  V  that 
it  would  have  if  the  external  resistance  were  100,000  ohms, 
as  in  the  former  experiments. 
Table  IV. 
Altera. 
Direct 
r 
Altera. 
Direct 
V 
Freq. 
Current 
Current 
|  Freq. 
Current. 
Current 
V 
r. 
!    V 
>o- 
I'. 
1 
V 
10-4  amp. 
10— S  amp. 
10-4  amp. 
i  10-8  amp. 
332 
0-205 
5-27 
;  1251 
1362 
0-563 
657 
20-7 
0-304 
11-85 
i  1281 
0-835 
14-8 
21-2 
0-400 
159 
!    99-4 
1-099 
20-6 
17-08 
518 
0-268 
5-75 
i    80 
2000 
0-689 
6-91 
14-58 
0-398 
12-47 
1    78-9 
1-02 
14-35 
13-80 
0-523 
18-26 
;    669 
1-341 
214 
11-92 
523 
0-319 
5-96 
;    58-6  ! 
1  2930 
0-835 
744 
10-67 
0-475 
12-45 
'    55-2 
1-24 
|     17-1 
111 
0-625 
18-9 
1    48-4 
1-63 
24-2 
9-12 
758 
0-399 
0-591 
0-779 
605 
133 
191 
1    38-0 
,    38-1 
31-6 
1 
1006 
0-466 
6-48 
i    29-8 
534 
0-311 
6  75 
70 
0692 
13-8 
!    28  8 
- 
0462 
13-91 
65-4 
0-91 
20  2 
!    24-4 
i 
0-608 
20-2 
54-7 
Table  V. 
I 
I' 
NT 
N. 
V 
V 
Vs 
332 
1266 
38-0 
12600 
518 
79-4 
23-8 
12300 
523 
56-9 
17-08 
!        8940 
758 
38-0 
11-4 
8940 
1006 
29-3 
8-8 
8810 
1362 
20-9 
6-27 
8550 
2000 
14-2 
4-26 
8510 
2930 
10-9 
3-27 
9550 
534 
67-7 
20-3 
10850         | 
