SOUDAN FORMATION. 229 



liberated, aud would join the descending waters. Thus carbonated waters free from 

 oxygen would be produced. Such waters are capable of taking a considerable 

 amount of iron carbonate and some iron silicate into solution. Large quantities of 

 these solutions would be converged upon the sides or at the bottom of the pitching 

 troughs, or in other places where there were trunk channels for water circulation. 



After an iron-bearing foraiation was exposed to descending waters for a consid- 

 erable time, a large part of the iron carbonate adjacent to the surface would be 

 transformed to ferruginous slates and ferruginous cherts. This change would take 

 place most extensively where waters were abundant and a somewhat direct course led 

 to the trunk channels. After this process was completed at such places, the waters 

 now following this direct route would pass onl}^ through the ferruginous slates 

 and ferruginous cherts and would reach the trunk channels charged with oxygen. 

 There the solutions bearing iron carbonate and those bearing oxygen would be com- 

 mingled. Iron sesquioxide would be precipitated. Therefore the iron oxide of an 

 ore body consists in part of iron compounds originally deposited in situ and in part 

 of iron brought in b}^ underground waters. The material deposited in situ may 

 have been originally detrital iron oxide or it may have been dei'ived from iron car- 

 bonate, iron sulphide, or iron silicate, which was oxidized in place, or from two or 

 all of these sources. It has been assumed that the part brought in by underground 

 wafers was mainly transported as carbonate, although a portion may have been 

 transported in some other form. Of the two sources of iron ores, the original material 

 and that added hx underground water, the latter is upon the average probabi}^ more 

 abundant. But in some exceptional cases, where there is a large amount of detrital 

 iron oxide, the material added by underground waters may be subordinate. However, 

 in all cases it may be said that were it not for the secondary enrichment by under- 

 ground waters, through the addition of iron oxide, the material would not be iron 

 ore. The evidence of this lies in the fact that the ore bodies are universallj^ confined 

 to the places where underground waters have been converged into trunk channels. 

 The ore deposits contain upon the average a less quantity of silica than does 

 the average of the iron-bearing formations. It follows therefore that silica must 

 have been dissolved. This doubtless was largely the work of the great volume of 

 water converged into the trunk channels. It has been seen that the waters which 

 carried iron carbonate to the ore deposits were carbonated. The precipitation of 

 iron oxide from carbonate liberated more carbon dioxide, so that the waters were 

 very heavih' charged Avith carbonic acid. In some of the districts basic igneous 

 rocks occur within the iron-ore deposits or as basements to them. In all such cases 

 these basic rocks are found to have lost a large part or all of their alkalies. These 

 must have passed into the solutions. Hence the waters moving along the trunk 

 channels would in some cases contain alkalies besides being lich in carbon dioxide. 

 It is well known that such solutions are capable of dissolving silica. Therefore the 

 conditions which result in the precipitation of iron oxide also furnish conditions 

 favorable to the solution of the silica. Silica is thus largely dissolved from the ore 



