THE ELECTRICAL RESISTANCE OF THIN LIQUID FILMS. 
483 
Table VIII. 
I. 
II. 
III. 
IV. 
V. 
VI. 
VII. 
VIII. 
IX. 
X. 
Sept. 27, Ii. 
Sept. 27, III. 
Sept. 28, I. 
Sept.28,111. 
Sept. 30,1. 
Sept. 30, II. 
Hygrometer 
53-0 53-1 
53-1 53-2 
53-2 53-1 
53T 53-2 
51-1 51-2 
51-2 51-3 
Mean. 
Thermometer 
20-9 20-6 
20-6 20-4 
18-8 18-9 
18-9 18-7 
17-8 17-9 
17-9 18-1 
16-0 
R (6, 3) 
13-30 
1-014 
1-013 
1-012 
1-013 
15-0 
G (6, 6) 
12-47 
1-002 
1-012 
1-006 
1-008 
1-003 
1-006 
14-0 
R (5, 9) 
11-64 
0-999 
1-006 
1-005 
1-010 
1-001 
1-004 
1-004 
13-0 
R (5, 3) 
10-80 
1-001 
1-006 
1-003 
1-016 
1-001 
1-007 
1-006 
12-0 
G (5, 6) 
9-97 
1-000 
1-000 
1-000 
1-000 
1-000 
1-000 
1-000 
11-0 
R (4, 9) 
9-14 
0-994 
0-994 
0-997 
0-992 
1-001 
1-000 
0-996 
10-0 
R(4,l) 
8-31 
0-990 
0-987 
0-989 
0-986 
0-998 
1-001 
0-992 
9-0 
G (b 6) 
7-48 
0-992 
0-976 
0-987 
0-989 
0-996 
1-011 
0-992 
8-5 
G (4, 1) 
7-06 
0-994 
0-983 
0-988 
1-015 
0-995 
8-0 
BR (3, 2) 
6-65 
0-994 
0-988 
0-984 
1-010 
0-994 
7-5 
R(3, 2) 
6-23 
0-983 
0-987 
0-988 
1-005 
0-991 
7-0 
Y (3, 3) 
5-82 
0-992 
0-985 
1-006 
1-004 
0-997 
6-5 
G (3, 4) 
5-40 
0-999 
0-983 
1-011 
0-998 
6-0 
B (3, 4) 
4-99 
0-999 
0-992 
1-002 
0-998 
5-5 
F (3, 2) 
4-57 
1-010 
1-010 
1-010 
5-0 
O (2, 9) 
4-16 
1-004 
1-007 
1-005 
4-5 
Y (2, 4) 
3-74 
1-026 
1-007 
1-016 
Mean 
1-001 
The constancy of the numbers in the last column of this table is extremely satis¬ 
factory. In no case do they differ from their mean value by more than 1'5 per cent., 
and this result is certainly as good as could be expected when it is remembered that 
the thickness is measured by a colour estimation only. It is, however, worth while to 
investigate the outstanding discrepancies a little more closely. The numbers in 
column X. are a little less for medium than for large or small thicknesses. The reality 
of such an alternate decrease and increase in the specific resistance is very improbable, 
while on the other hand the method of correcting the table of colours by the quotients 
is likely to produce small periodic errors. Each observation on Newton’s rings is, 
however, independent of the rest, and another table (Table IX.) was therefore 
prepared as follows :— 
Column I. gives in terms of the corrected scale the thicknesses corresponding to the 
colours measured directly by means of Newton’s rings. 
Column II. gives numbers proportional to the mean specific resistances of the six 
films in Class I. at those thicknesses ; these are deduced approximately from column X. 
in the above table. 
3 R 
MDCCCLXXXI. 
