296 On the Artificial Formation of Minerals. 
minerals, endeavoured to obtain further evidence by synthe- 
tical experiments, and, first, to find whether the anhydrous 
phosphate of iron had the same colour as the hydrate. 
With this object, he heated protosulphate of iron with 
ordinary phosphate of soda, and an additional equivalent of 
‘soda, to obviate the formation of pyrophosphate, but was 
unable to effect the formation of phosphate of iron by double 
decomposition, as the mixture proved to be infusible at a 
temperature which rendered cast-iron perfectly liquid. He 
therefore had recourse to chloride of sodium, which he added 
in large excess, and exposed the mixture to the influence of 
an intense white heat for half an hour. When the crucible 
was thoroughly closed, and the heat had not been maintained 
so long as to cause considerable volatilization of the chloride 
of sodium, the melted mass was quite homogeneous, and gene- 
rally colourless, while its surface was covered with deep red 
micaceous lamine. By treatment with water a fine crystal- 
line residue was obtained. That portion of the melted mass 
which was in contact with the sides of the crucible had ac- 
quired a dark violet colour to the depth of two or three lines, 
and very similar to the colour of some varieties of fluorite. 
The principal part of the mass, however, was colourless 
protophosphate of iron; at the surface, where oxidation was 
comparatively easy, phosphate of iron had been formed. 
When, on the contrary, there was a crack in the crucible, 
through which a portion of the chloride of sodium ran into 
the fire, or the temperature was sufficiently high and pro- 
tracted to effect the volatilization of greater part of the 
chloride of sodium, the appearance of the melted mass was 
quite different. The cavities in the chloride of sodium were 
then filled with black micaceous crystals, which had the | 
greatest similarity to micaceous hematite. This substance 
consisted of protoxide and oxide of iron and phosphoric acid, 
and, rubbed to powder with water, presented a deep blue 
colour 
It is remarkable that this phosphate of iron, when oxidized, 
while washed with water, does not, like the phosphate of iron 
occurring in peat-bogs, and in some masses of lava, assume a 
blue colour when exposed to air; but becomes brown gra- 
