ting waters. The ores at Shepherd Mountain are similar vein- 

 deposits, but the porphyry is here seen in an undecayed state. 



As regards the very unlike deposits of Pilot Knob, Dr. Schmidt 

 accepts the first hypothesis of Prof. Pumpelly. and supposes that 

 solutions, similar to those which deposited the ore in the fissures of 

 the porphyries elsewhere, have here effected the conversion of the 

 porphyry into ore. It is, as he admits, difficult to explain in this 

 view, the removal of the resulting silicate of alumina, and not 

 less difficult to explain the removal or replacement of the quartz, 

 as supposed by Pumpelly. When we consider that iron oxyds are 

 frequent elements in gneissic and other crystalline rocks, and that 

 they have been directly deposited in later sedimentary formations, 

 it will seem to many simpler to accept the hypothesis that these 

 iron and manganese oxyds in the porphyries and conglomerate 

 beds, instead of having come from the replacement either of feld- 

 spar and quartz or of carbonate of lime, may have been deposited 



Besides these ores associated with the Eozoic rocks, Dr. Schmidt 

 describes several other classes of iron-ore deposits, one of the 

 most interesting of which occurs in the sandstones immediately 

 above the 3d Magnesian limestone above named, and often fills 

 Miiall basins or excavations in this sandstone, nearly vertical walls 

 of which are seen to limit the ore-deposit. The ore in these is 

 stratified, and is often both overlaid and underlaid by beds of el ay. 

 flint and broken sandstone, and, it is suggested, may have been de- 

 posited in cavities produced by a subsidence of the strata into 

 caverns in the limestone beneath. The ore is sometimes specular 

 red hematite, and at other times limonite, occasionally also mag- 

 netite, and sometimes includes rounded masses of ferruginous lime- 

 stone with crystals of iron-carbonate. This association leads Dr. 

 Schmidt to suggest as an alternative hypothesis, that these deposits 

 may have been formed by the transmutation of limestone deposits 

 previously occupying these basins. To this class belong the ores 

 of the Merramec district. 



In the Carboniferous series again, deposits of red hematite ore 

 occur in sandstone, forming nodular or concretionary masses or 

 regular bods. In one locality also, we have here described a large 

 cavern or sink in the Receptaculite limestone at the summit of the 

 Trenton, in which occur stratified layers of hematite and limonite, 



