328 
DR. E. H. GRIFFITHS AND MR. EZER GRIFFITHS ON THE 
The observations were continued at intervals, as long as the level of the oxygen 
was sufficient for total immersion of the coils and portion of the stem, with the object 
of ascertaining how far the constancy was affected by the presence of traces of 
nitrogen in the liquid oxygen. 
Three different values of the battery current were used, and Table III. below 
summarises one series for thermometer Q, with a battery current giving 0'003 amperes 
through the wire of the coil. 
The quantities in brackets [ ] are the corrections necessary to reduce the resistance 
to mean box units ; temperature of box, 17° C., and barometer 760 mm. 
The value of for liquid oxygen was taken as 77 mm. Hg per degree Centigrade. 
ct 
The abnormally high value of R, indicated by the asterisk ( # ) is probably due to 
superheating of the liquid and was also shown by another thermometer standardised 
at the same time. 
Table III.—Thermometer Q in Oxygen. 
Time. 
Plugs. 
Bridge wire. 
Temperature 
box. 
Barometer. 
Resistance 
(after correction). 
h. m. 
Date : O 
° C. 
ctober 14th. 
12 32 
330-116 
3-212 
[-0-029] 
18-09 
[ + 0-078] 
767-41 
[-0-526] 
332-851 
3 21 
3 22 
5-447 
5-510 
[-0-043] 
18-485 
18-49 
[-0-109] 
766-54 
[-0-464] 
335-196* 
8 43 
8 45 
8 50 
3-209 
3-186 
3-176 
[-0-029] 
18-83 
18-85 
18-85 
[ + 0-138] 
765-81 
[-0-412] 
333-004 
Date : October 15th. 
10 37 
10 39 
10 41 
10 43 
10 46 
330-116 
3-430 
3-515 
3-630 
3-593 
3-550 
[-0-032] 
17-97 
17-98 
17-985 
[ + 0-069] 
767-22 
[-0-513] 
333-184 
