GEOLOGICAL PAPERS. 117 



gypsum, or in the lower part of the basins toward the interior, where 

 the waters, deprived of their gypsum content, had retreated. 



If these conditions were true, the salt might later have been re- 

 moved by solution in downward percolating waters which dissolved 

 the more soluble sodium chloride. The gypsum now remaining does 

 show marked effects of solution, the surface being rounded and fur- 

 rowed by solution, and in places it is entirely removed. These ef- 

 fects would have been far greater in the common salt. The salt- laden 

 waters or brines would flow downward along the slope of the rocks 

 and through them, finally remaining at rest in the lower porous for- 

 mations, where it is now found. Further, the salt seems to be found 

 in greater amounts toward the interior of the basin than near the 

 edges; more at Saginaw, Ann Arbor. Lansing, etc., than at Tawas 

 and Grand Rapids, though it is found in all these places. 



Another possible explanation of the final history of this sea is to 

 be found in the great extension of the sea in the next epoch, when 

 the St. Louis limestone was found. The sea in the St. Louis epoch 

 extended its borders north and south, and passed across the interior 

 basin of Michigan to Grand Rapids on the west and to Huron county 

 on the east. Possibly this renewal of the waters took place before 

 the Michigan sea had disappeared by evaporation, or before it had 

 evaporated enough to deposit a large quantity of salt, except in cer- 

 tain smaller basins separated by the dividing ridges. 



From the evidence of sandstones and shales of the Michigan series 

 found in the well borings of the interior, it would seem that the ocean 

 flowed over the southern barrier into the interior basin a number of 

 times before the greater St. Louis inundation, and at these times de- 

 posited the sediments which are lacking in gypsum and salt contents. 

 At these times the water would be diluted, its specific gravity lowered, 

 so that precipitation of the salts would not take place. These over- 

 flowing waters, local in their occurrence, cannot be correlated with 

 other sections, unless with those of the Logan series of Ohio, whose 

 origin may be similar. 



In the deeper Michigan borings, gypsum appears to be replaced 

 by anhydrite ; but where the depth of concentrated waters is 325 feet, 

 giving a pressure of ten atmospheres, anhydrite is formed instead of 

 gypsum. 



This theory, as outlined for the Michigan gypsum deposits, is 

 based on the study of a few well borings and a comparative study of 

 the conditions in the Caspian sea of to-day and those of the Michigan 

 area as far as they can be determined. There is a wide range of 

 probability involved, and while the theory is advanced as a theory rest- 

 ing on limited data, it may be taken as representing approximately 

 the conditions of origin of these deposits. 



