280 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VoL. 13, No. 13 
ferrous oxide charges are such that its formation may have been a 
disturbing factor in Hilpert’s work. The compositions of his iron-free 
ferrous oxide phases are probably richer in ferric iron than the phases 
which would be stable and in equilibrium with metallic iron at the 
various temperatures. 
According to the hypothesis just advanced, the ferrous oxide solu- 
tion saturated with ferro-ferric oxide would be the only one which would 
not partially decompose at temperatures above the invariant point. 
Matsubara’s experiments indicate that in the range 863°-1175°C. the 
solubility is practically independent of the temperature. His saturated 
solution contains a little less than 76 per cent total iron. Sample I, 
which behaves as one would expect the saturated solution to do, con- 
tains 76.03 per cent iron. 
The temperature of the invariant point must lie below 577°C. and 
above 526°C. Probably it lies much closer to the upper temperature 
rather than to the lower. More will be said regarding these tempera- 
tures later. 
TABLE ITI.—ExperimMents To DETERMINE THE EQUILIBRIUM CONSTANT AT 750°C. 
IRON CONTENT 
poe qin nours | RaT10 H20/Hz REMARKS 
MATERIAL PRODUCT 
Pure iron 6 0.476 99.98 No oxidation 
Pure iron 6 0.540 99.90 Oxidation slight if any 
Pure iron 6 0.568 98 .0 Oxidation started 
98.8 Analyses from different parts of 
the boat 
Pure iron 6 0.595 93.6 More oxidation 
Ill. THE EQUILIBRIUM CONSTANTS 
A series of experiments was carried out at 750°C. and atmospheric’ 
pressure with the view of checking by the stream method the diverse 
values which have been obtained for the equilibrium constant by 
static methods. A similar apparatus to that used for the iron phase 
investigation was employed. Blank experiments were carried out in 
order to test the apparatus and method. The results are given in 
Table U1. 
These results indicate that oxidation starts with a mixture having 
a ratio probably slightly less than 0.54 and certainly less than 0.57. 
Chaudron’s latest work leads to a value of 0.53 for this temperature. 
Eastman’s calculated value would be 0.441. That of Schriener and 
Grimnes would be approximately 0.66. 
