94 
AMERICAN GEOLOGY. 
limits are unknown. The degree of oxidation of the parti¬ 
cles composing the masses of this ore seem to he unequal, and 
hence there is more difficulty than usual attending its reduction. 
This species of iron in northern New York, the Highlands, 
and New Jersey, is associated with quartz, hornblende, and 
feldspar; and two very extensive bodies of ore are intermixed 
almost exclusively with phosphate of lime. But the magnetic 
oxide is usually grouped with one of the three first named. 
Quartz is the most favorable rock for reduction, in an econo¬ 
mical point of view. The specific gravity of octahedral iron is 
5*09; the rocks which are grouped with it scarcely exceed 3'00. 
If, then, these minerals were fused together, the iron, from its 
greater specific gravity, would sink through the molten mass, 
and be found at a lower level. In the majority of mines of this 
ore, the leanest part is at the surface. Particles of ore are 
scattered sparsely through the rock at the outcrop of the vein; 
but at the depth of twenty feet, and perhaps less than twenty, 
there is a perceptible increase of metal. The gravity of the 
iron ores may therefore explain the fact of their comparative 
absence as a rock at the surface; and it may be reasonably 
inferred from this and other facts, that the veins of ore are con¬ 
nected with much larger masses beneath than any which have 
found their way to the surface. Serpentine is very common 
among the beds and veins of this ore in northern New York. 
It is not in large masses, neither have I always found it in the 
beds of octahedral iron. It also accompanies the specular oxide 
of iron in St. Lawrence county, New York. 
The great extent of iron ores of these two species, the mag¬ 
netic and specular oxides of iron, seem to require that, in this 
country at least, they should be embraced in the rocks. They 
have hitherto been described as minerals only; but as they 
occur in mountain masses, occupying positions analogous to 
to serpentines, limestones, and granites, there can be no objec¬ 
tion to ranking them with the subordinate rocks of the globe. 
Magnetic iron occurs in masses and veins. In the hypersthene 
rock it is in masses subordinate to that rock, while in the gneiss 
