412 Dean — Formation of Missouri Cherts. 



must have been laid down with the rock and the argument 

 seems well founded ; the conclusion is however not with- 

 out an alternative as will be discussed later. There are 

 also many difficulties which confront a sea-bottom depo- 

 sition. It is rather hard to conceive of so dilute a solu- 

 tion of silica being precipitated in such large masses and 

 remaining intact sufficiently long to be transformed into 

 chert (if indeed it ever would be). The experiments 

 given by Tarr to show the precipitation of dilute silica 

 hydrosol by the salts of the sea would have been 

 more convincing if made with a pure dialized silica 

 hydrosol which differs markedly in properties from the 

 impure hydrosols used by Tarr and which more nearly 

 approaches the natural waters. If, however, the silica 

 is precipitated by the salts of the sea these salts or at 

 least their cations should be the principal materials to be 

 adsorbed by silica and would hence form the principal 

 impurity in the chert. Since sodium chloride is the salt 

 present in largest amount in the sea water we should 

 expect to find a predominance of sodium in the chert; 

 as a matter of fact the analyses show very little sodium 

 and a decided predominance of calcium and magnesium 

 carbonates. 



The silica required for the epigenetic origin of the 

 cherts may have been from different sources in dif- 

 ferent localities but its immediate source is unques- 

 tionably the ground water. In the rocks of Mississip- 

 pian and Cambro-Ordovician age the silica was probably 

 derived from the overlying carboniferous shales by the 

 decomposition of silicates by carbon dioxide and water. 

 The ground waters at present circulating in the Cambro- 

 Ordovician limestones contain a notable quantity of 

 silica as is evidenced by the replacement of certain lime- 

 stone fragments in the top member of the Roubidoux 

 Sandstone when this member is exposed to the ground 

 waters, as for example in stream beds. Accordingly 

 the formation of chert is probably going on at the 

 present time. 



The ground waters containing the silica hydrosol and 

 carbon dioxide do not deposit the silica since carbonic acid 

 is a weak electrolyte, i. e., undergoes slight dissociation in 

 water. But when water containing carbon dioxide comes 

 in contact with carbonate rocks ordinarily limestone and 



