sosman: problems of the oxides of iron 71 



cipitation of hydrated ferric oxide, and the number of forms which 

 are known to be associated in some way with the precipitation 

 is constantly increasing. It is not yet clear in all cases whether 

 the organism actually brings about the reaction through its 

 own life processes, whether the precipitation is due to the de- 

 composition of the dead organism, or whether the organism 

 acts merely as a nucleus for the collection of agglomerating colloidal 

 oxide. The origin of the immense and very pure iron ores of 

 Minas Geraes, Brazil, is ascribed by Harder and Chamberlin 30 

 to bacterial activity, apparently for the indirect reason that 

 there are objections to any other mode of formation. The 

 reader is left with the uneasy feeling that perhaps only our lack 

 of knowledge prevents our objecting to the bacterial explanation 

 as well. 



There is one possibility which deserves attention in this con- 

 nection. Experiments with closed tubes show that a compara- 

 tively moderate increase of temperature will hydrolyze various 

 ferric salt solutions, and precipitate ferric oxide, which does not 

 redissolve. These ferric solutions seem relatively unstable and 

 easily hydrolyzed, and the colloidal ferric oxide only needs to 

 be agglomerated to form a precipitate that will settle. Are there 

 any possible variations in the composition, concentration, or 

 temperature of sea water which would cause it to precipitate 

 the ferric oxide from a hydrolyzed, even very dilute, ferric 

 solution in much the same way as the Mississippi silt is precipi- 

 tated by the waters of the Gulf of Mexico? Bacteria, able. and 

 willing to take advantage of the conditions favorable to the 

 reaction, need not be excluded, but, as may be the case 

 with calcium carbonate, 31 inorganic causes may be primarily 

 responsible. 



The intimate association of silica with many ores, amounting 

 even to chemical combination, as in some of the siliceous ores 

 of the Lake Superior region, may represent a simultaneous 

 agglomeration of colloidal silica and ferric oxide, or an adsorp- 

 tion of ferrous and ferric iron by agglomerated silica. It is to be 

 noted, also, that such precipitated colloids can sometimes be 



;, ° Jour. Geol., 23:385-404. 1915. 

 See Johnston and Williamson. Jour. Geol., 24: 729-750. 1910. 



