Crosby.] 168 [March 5, 
It is well known that petrosilex sometimes passes gradually into 
rocks which are wholly silicious, so that we are warranted in saying 
that some jaspers and quartzites have been formed under substantially 
the same conditions as the petrosilicious rocks; and in some cases 
the structure and general relations of the jaspers point to the same 
conclusion. 
Now one of the most interesting rocks in the Marquette iron dis- 
trict, in the Upper Peninsula of Michigan, and the one most closely 
associated with the iron ore, is a brownish or reddish jasper; it some- 
times becomes chloritic or micaceous, passing into chlorite schist, etc., 
but for the most part it is a distinct and beautiful stratified jasper. 
This Lake Superior jasper, like all the petrosilicious rocks so far 
as known, belongs to the Huronian formation, and may, apparently, 
be fairly taken to represent the petrosilex and felsite characterising 
many other Huronian areas, but apparently wanting here. Its as- 
sociation with the iron ore is usually very intimate; the two sub- 
stances being interlaminated in such a manner as to give rise toa 
banded structure which matches in all important particulars the 
banding of the petrosilex of eastern Massachusetts and other regions, 
the hematite simply taking the place of the feldspar. The extreme 
irregularity of the banding in many cases makes it not only proper 
but necessary for us to conclude that, as in the case of the petrosilex, 
it is largely the result of a segregating process, the two constituents, 
hematite and jasper, having been originally more intimately mixed. 
With very few exceptions this Marquette ore always contains some 
oxide of manganese, usually from one to two per cent., though the 
ore from one bed contains nearly seven per cent. 
Here, then, so far as chemical composition is concerned, we have a 
formation almost identical with some of the silicious oozes of the 
deep sea; while the chief structural distinction consists in the differ- 
ent forms of the segregated masses of the iron and manganese 
oxides, lenticular layers taking the place of irregularly rounded 
nodules, certainly a distinction of no great importance. 
The iron ores of the Iron Mountain district in Missouri are also 
Huronian; and here the associated rock is a true petrosilex. The 
mingling of rock and ore is less perfect than in the Marquette 
district, but the general plan of the structure is the same. The red 
petrosilex is distinctly stratified, and the hematite is interstratified 
with it in a way to show that they are cotemporaneous deposits ; 
although the hematite has experienced considerable segregation sub- 
