SOMETHING ABOUT ROCK-SALT AND GYPSUM. 391 



dition that it was underlaid by an impervious floor. This 

 is generally the case with the soluble matters of the Salina 

 group. If, however, the saliferous formation were under 

 laid by a porous sandstone, this would become the reser 

 voir in which .the leachings of the saliferous formation 

 would be preserved. Thus the Conglomerate becomes in 

 Ohio and Michigan the reservoir for the Coal-measures 

 (Fig. 91). Borings for salt must necessarily extend to the 

 formation in which the brine is accumulated. This is com 

 monly designated the salt-rock; but it is not necessarily 

 the mother-rock of the brine. 



Such I believe to be a true account of the natural his 

 tory of rock-salt and native brines. The phenomena of 

 gypseo-saliferous formations seem incompatible with any 

 other explanation. 1. The rocks composing these forma 

 tions are regularly stratified, and furnish the usual indica 

 tions of sedimentary origin. The beds of gypsum and of 

 rock-salt, when existing, are entirely conformable with thd 

 argillaceous strata, and approximately coextensive with 

 them. On this theory, having ascertained the existence of 

 a brine formation on the west side of the State of Michigan, 

 I successfully predicted its discovery on the east side. The 

 extensive gypsum beds, also, of the east side were brought 

 to light by a similar prediction based on the same theory ; 

 and I have evidence that the gypsum formation of Grand 

 Rapids and Alabaster, on opposite sides of the state, is 

 absolutely continuous beneath all the intervening region. 

 2. Gypseo-saliferous formations contain all the well-known 

 constituents of sea-water. I do not consider it likely that 

 these constituents would be associated in the same way in 

 both cases, unless the one were the historical consequent 

 of the other. 3. The order of arrangement of these con 

 stituents is the order of their solubility. When natural 

 brines are operated upon for salt, the least soluble constit- 



