ORIGIN OF THE DOLOMITES. 193 



IV. Dead coral taken from the beach. Darwin: Strnctnre anil Distribution of 

 Coral Reefs, 18S<), pa,t:e IS. The fiLrures iriven above are estimated. Darwin says 

 that the iiuantity of car])onate of magnesia present in fresh coral is usually less than 

 one per cent. 



V. An average of the four precediiiLr analyses. 



Again, taking the results of tlie chemical analyses of live coijuina gravels 

 from Florida, analyzed by the United States (ieological Survey, we see 

 an average i)ercentage of 3.82 magnesium carbonate given.'*^ Fourteen 

 specimens of coral rocks from the Hawaiian islands show an average of 

 4.52 per cent of magnesium carl)onate f 



So far as can be judged, the carl)onate deposits of Paleozoic and Mesozoic 

 times originally must liave been essentially the same as the above. There 

 is nothing in the anatomy of polyps, molluskoids and mollusks to lead one 

 to believe that the i)rogenitors of living forms were very ditTerent from their 

 descendants whose structure and secretions are very well understood 

 by zoc)logists. It is therefore not an easy thing to imagine the forma- 

 tion of dolomite througli any such i)rocess as that of modern limestone- 

 building. Holding in view the Magnesian series, we are practically con- 

 fined to the conviction that the accumulations of car])onate rocks in the 

 central portion of the continent during early Paleozoic time were lime- 

 stones of substantially the same constitution as those now forming within 

 ocean areas. But they have become, quite universally, dolomites and 

 dolomitic limestones. The transformation from limestones to dolomites 

 must therefore be a subsequent process. Since there is always a per- 

 centage of magnesium carbonate in every limestone thus far known, there 

 are two ways in which a dolomite may be formed : (1) By the deposition 

 of more and more magnesium carbonate subsequent to the formation of 

 the rock, and (2) by the gradual removal of calcium carbonate from the 

 rock mass. J 



There are many reasons for believing that the first method did not 

 obtain in the region under consideration. The corroded condition of 

 the beds at many localities where they are well exposed, the porous 

 state so generally seen in all three of the dolomitic formations, the readi- 

 ness with which the dills along the river gorges retreat from the streams 

 in vertical, castellated walls until valleys miles in width are formed, are 

 phenomena opposed to the view that the beds were increased in bulk. 

 These conditions were all seen and described for other regions by Ijcopold 

 V. Bach more than seventy years ago,>^ altliough tin; real cause of the 

 ])henoraena was not clearly understood. 



• Roport of Work : Chemistry and PhysIcH, Bulletin no. 60, 1890, p. 1(52. 



t n.id., p. \(^t. 



{Compiiro IrvioK: Chemiral :inil F'liyslcrtI StmlieM in tho Motttmorphi.stu of Rocks, 1889, p. 71. 



I Soe Tojichenbuch fOr .MineraloKie, 1S24, pp. 244, 257, 283, etc. 



