518 
PROFESSORS A. W. REIROLD AND A. ^Y. RtlCKER OR THE 
oleate of soda, upon which it acts, might be very considerable, we have tested the 
matter by direct experiment. A series of observations were made when the air which 
surrounded, and that which was used to blow the films had been carefullv freed from 
C0.2 by passage through three tubes containing caustic alkali. As it would have 
been extremely inconvenient to have made the apparatus air-tight, a current of the 
purified air was passed through it continuously during the whole of the experiment. 
Although, therefore, the proportion of CO^ in the air around the films was—if not 
absolutely nil —very much reduced below its normal value, the results as regards the 
electrical thickness were practically identical with those obtained when no such 
precautions were taken. 
Table XI, 
Proportion 
Electrical thickness of black in /i./x. 
Date. 
of oleate 
to water. 
Highest. 
Lowest. 
Mean of all. 
Number of 
films. 
June and July, 1890 . 
T 
7 0 
240 
79 
148 
14 
Taken without any special precaution. Linen 
band not used. 
November and December, 1890 . 
1 
7 0 
243 
73 
154 
7 
Linen band used. Films blown with air freed 
from CO 2 , but no continuous flow of purified 
air through the film box. 
i 
July, 1891. 
4 0 
250 
84 
184 
6 
See Table VII. Every known precaution used. 
Continuous flow of air freed from COo through 
the film box. 
Although the mean values differ somewhat, it will be observed that the highest 
and lowest values in each set are about the same. There is thus no indication that 
the presence or absence of COo in the air surrounding the film has any marked 
influence on its electrical thickness. This argument against CO 3 being the cause of 
tlie high conductivity of the thin films is much strengthened if we consider the ratio 
of the electrical to the optical thicknesses, i.e., the ratio of the specific conductivity of 
the film to that of the liquid in bulk. Dividing each of the electrical thicknesses in 
the above table by the corresponding optical thickness given in Table IT,, w^e get the 
three values, 5*2, 5'4, 8'3. Hence the specific conductivity of the film was about 
50 per cent, greater when every precaution was taken to exclude CO 3 than when no 
