IMPARTED TO A VACUUM BY HOT CONDUCTORS. 
525 
In order to illustrate the magnitude of these changes, the numbers in columns I. 
and III. have been plotted in the accompanying curve (fig. 12). 
The chief objects of these experiments on the relation between the negative 
current from and the resistance of, a carbon filament, was to determine the 
dependence of the former in the temperature, 
method, has given numbers connecting the 
temperature of a carbon filament with its 
resistance. If we plot a curve from these 
numbers between the temperature of the 
filament and the ratio of its resistance at that 
temperature to its resistance at 15° C., we can 
use this to obtain the temperature of any other 
filament from its resistance. In this case we 
have again to face the uncertainty caused by 
the permanent change in the resistance of the 
filaments when heated. I have attempted to 
correct for this in the same manner as has been described above. Le Chatelier 
states that in his experiments there was a permanent lowering of the resistance of 
the filament amounting to about 10 per cent. This change has been distiibuted 
among the observations in such a way that the observations at the highest tempera¬ 
tures are responsible for the greater part of the alteration. The corrected curve thus 
coincides with the original one up to about 1000° G., after wliich it branches off, the 
divergence between the two becoming gradually greater, until finally at about 2000° 
it ends 10 per cent, higher than the one plotted from Le Chatelier’s numbers. 
The numbers in the table on jj. 523, when treated in this manner, yield the 
following :— 
I. 
II. 
III. 
1 
•39 
•610 
1 
1250 I 
•943 
•606 
1265 
3-25 
•602 
1285 
8-55 
•599 
1305 
19-04 
•595 
1325 
43-2 
•592 
1345 
133-4 
•589 
1365 i 
353 
•586 
1380 
780 
•582 
1400 ; 
2475 
•577 
1430 
3950 
•572 
1460 i 
9050 
•567 
1490 i 
Saturation current, unit being 10“'^ ampere. 
Ratio of resistance to corrected resistance at 11° C. 
Temperature in degrees Centigrade. 
* ‘ Journal de Rhys.,’ loc. cit. 
I. = 
II. = 
III. = 
Le Ch atelier,"'" using an optical 
Fig. 12. 
