1893-94.] S. Kimura on Electrical Properties of Gases. 209 
Tempera- 
ture of 
Hotter 
JunctioD. 
Galvano- 
meter 
Deflection. 
Deflection 
Reduced to 
Microvolts. 
Tempera- 
ture of 
Hotter 
Junction. 
Galvano- 
meter 
Deflection. 
Deflection 
Reduced to 
Microvolts. 
Heating. 
26-00 
4-00 
8-28 
201-00 
9-50 
19-50 
30-00 
4-10 
8-48 
210-75 
9-60 
19-71 
44-75 
5-00 
10-28 
220-75 
10-75 
22-10 
67-75 
5-25 
10-88 
230-50 
11-25 
23-10 
79-75 
5-80 
11-99 
241-00 
13-25 
27-16 
99-00 
6-75 
13-87 
250-40 
14-00 
28-77 
112-25 
7-00 
14-33 
124-75 
7-00 
14-33 
128-00 
7-00 
14-33 
133-00 
6-85 
14-06 
Cooling. 
138-00 
6-95 
14-13 
251-50 
12-75 
26-12 
143-75 
7-20 
14-75 
249-20 
12-75 
26-12 
148-75 
7-10 
14-55 
245-50 
12-15 
24-90 
153-50 
7-20 
14-75 
241-00 
12-15 
24-90 
157-25 
7-25 
14*88 
231-50 
11-40 
23-40 
163-50 
7-75 
15-90 
218-50 
11-00 
22-60 
168-50 
7-85 
16-10 
189-50 
10-00 
20-55 
173-50 
8-30 
17-05 
162-00 
8-75 
19-93 
181-75 
8-60 
17-67 
120-50 
8-25 
16-93 
191-50 
9-00 
18-47 
83-50 
7-85 
16-10 
The direction of the thermo-electric current is from normal iron to 
that charged with hydrogen, through hotter junction. Looking at 
the curve (PL II.), the march of the thermo-electric current is 
almost a straight line from the temperature of the colder junction 
up to 95°. Prom 95° to 140° the curve becomes almost horizontal, 
indicating an escape of hydrogen at this stage of the heating. Prom 
140° up to 250° the curve is regularly convex towards the tempera- 
ture ordinate. During cooling the curve has at first a greater and 
then a less inclination to the temperature ordinate than the curve of 
heating, and cuts the latter at about 230°. Thus the thermo-electric 
current between the heating and cooling wires changes its direction 
at 230°. The curve has a strong resemblance to those of iron charged 
with carbonic oxide or acid ; but the difference lies in this, that the 
curves during cooling and heating do not intersect at the tempera- 
ture of rapid escape of gas, but at a temperature much higher than 
this. The direction of the current was the same as that of the 
indirect preliminary experiments. 
VOL. XX. 19/4/94 O 
