1904.] HATCHER — MARINE AND NON-MARINE FORMATIONS. 351 



coast to advance from X. to K, represented at Y. by a vertical 

 thickness of deposits of about 2000 feet. 



With the overlying brackish water and littoral formations rep- 

 resented by Nos. 2 and 3, the stratigraphic position of any horizon 

 at any locality within either would be of minor importance in 

 determining its age relative to the formation as a whole. Since the 

 deposition of these beds would not take place simultaneously over 

 the entire area of their distribution, but would be restricted to a 

 narrow belt along the coast which would move eastward as the 

 coast advanced, it is evident that these deposits would be oldest to 

 the west and newest toward the east. This is an instance where 

 geographical position is of far more importance than stratigraphical 

 in determining the relative age of a given stratum within a forma- 

 tion. 



From the above observations it appears to the present writer that 

 other elements than their present stratigraphic position relative to 

 one another must be considered in determining the age and in 

 correlating any series of formations deposited along a receding or 

 advancing coast. There is evidence that several of the Mesozoic 

 formations of our Middle West were deposited under such condi- 

 tions. It should first be determined whether the sea was advancing 

 or receding during the deposition of any given series. Next the 

 direction taken by such advance or recession of the sea should be 

 ascertained, in order to determine in what geographical posidon 

 the older beds of the series are to be found. Then the rate of de- 

 position as compared with the rate of advance or recession of the 

 sea should be determined, so as to be able to correlate with one 

 another, approximately at least, strata within the same formation 

 but occupying different geographic and stratigraphic positions. In 

 the following attempt to correlate various formations in the region 

 referred to the methods and principles outlined above have been 

 applied only in a very general way, and the results can only be con- 

 sidered as suggestive of what might be accomplished by more ex- 

 tended observations and careful determinations in the same region. 



The Atlantcsaurus Beds and Dakota Sandsto7ies considered as the 

 possible equivalents of the Upper Jurassic and Loiver Cretaceous. 



There is a considerable thickness of non-marine sandstones and 

 shales with a wide geographical distribution in the region under 

 question, concerning the age of certain members of which there has 

 been much difference of opinion. The lower member of these was 



