178 The Petrography and Genesis of Sediments 



of pyrite as noted above in this summary. Whether this is due to the pres- 

 ence of an excess of H 2 S as he mentions on page 171, or to the humus com- 

 pounds (the existence of humic acids is now generally discredited) as in 

 lake deposits 1 has not been proved ; recent observations tend to show 

 that certain special bacteria are factors both in the precipitation of FeS 

 and in its oxidation to FeS,; but whatever the process the fact may 

 be accepted that in the presence of abundant organic matter in fairly 

 quiet waters FeS 2 is formed. Collet presents for the steps of the process 

 of glauconite formation an explanation, 2 somewhat simplified from that of 

 Murray and Kenard, based on elaborate and extended studies of his own. 

 In both theories, to start with, a colloid is assumed. Murray and Renard 

 conceive of the production of colloidal silica by the action of sulphuric 

 acid derived from the oxidation of the FeS present, while Collet starts 

 merely with the colloidal matter of clay. This, through the processes of 

 sedimentation, has naturally come to fill the forminiferal shells present. 

 The Al of the clay is first exchanged with Fe, and this new compound 

 combines with potassium present in the sea water, and also with some 

 water, to form the glauconite. In support of this theory Collet finds 

 many intermediate stages from grains having the appearance of fresh 

 clay to grains turned increasingly deep brown by taking up iron. The 

 writer's observation of grains having the form of glauconite, the appear- 

 ance of clay, but an aggregate polarization, was made without any knowl- 

 edge of Collet's observations and is therefore independent testimony in 

 support of this view. 



The occurrence of similar material in sample i 1 (p. 160, above) , which 

 contains FeS 2 (marcasite ?), is perhaps more questionable. Moreover, on 

 reviewing the sediments as a whole, the writer is not inclined to consider 

 the little clay accretions or nodules in the sulphide-bearing samples 1 and 

 2 as related to the glauconite. On the contrary, in view of the impregna- 

 tion of organic fragments with some iron salt (probably marcasite) that is 

 shown to have taken place there, it seems more probable that this same 

 mineral is responsible for the clay nodules. In fact, these questions can be 



'Collet, L. W., Op. cit, pp. 178, 179. 

 2 Collet, L. W., Op. cit., p. 176. 



