PAST AND PRESENT INTERPRETATIONS. 15 



Litholpgically the Coffee sand of Tennessee is strikingly similar to the part of the Eutaw 

 formation below the Tombigbee sand member, with which it was correlated by Safford. This 

 portion of the Eutaw (see Pis. IX, in pocket, and X, p. 20) is separated from the basal beds 

 of the Eutaw by the intervening Tombigbee sand member. It seems appropriate, therefore, 

 to give to these beds a separate designation, and for this reason Safford's term Coffee sand is 

 here revived. The type exposure is in a bluff of Tennessee River just above Coffee Landing, 

 Hardin County, Tenn. Plate III, B, shows a typical exposure of the Coffee sand member near 

 Pittsburg Landing, on Tennessee River. 



From western Alabama the Eutaw formation, including the Tombigbee sand member, 

 extends eastward entirely across the State of Alabama into Georgia. In this direction the 

 formation is underlain conformably by the Tuscaloosa formation to a line within a short dis- 

 tance west of Alabama River, where the Tuscaloosa pinches out rather abruptly between 

 Lower Cretaceous beds which make their appearance equally abruptly, and the overlying 

 Eutaw formation. From this place to its extreme eastern limit in Georgia the Eutaw is under- 

 lain unconformably by Lower Cretaceous strata. (See Pis. I, B, p. 10, and IV, A.) 



In the Chattahoochee region the Eutaw formation includes the Eutaw group of Langdon 1 

 and about 120 feet of the overlying beds which that author included in his Ripley group but 

 which, on paleontologic grounds, are here correlated with the Tombigbee sand member of the 

 Eutaw formation. The Blufftown marl of Veatch 2 included these Tombigbee representatives 

 and a part of the overlying Ripley formation. A typical exposure of the Tombigbee sand as 

 developed on Chattahoochee River is shown in Plate IV, B. 



SELMA CHALK. 



Name. — The term " Rotten limestone," as previously explained, was introduced by Alexander 

 Winchell 3 in 1857 for the terrane now known as the Selrna chalk. This descriptive name was 

 the commonly accepted designation of this great body of chalk rock until the year 1894, when 

 Smith, Langdon, and Johnson 4 proposed the geographic term Selma chalk, as a coname with 

 "Rotten limestone," and since then the geographic term has been the accepted designation. 



Extent. — In western Alabama and east-central Mississippi the Selma chalk makes up all 

 the Upper Cretaceous strata above the Tombigbee sand member of the Eutaw formation. 

 The terrane rests conformably upon the Tombigbee sand and from the eastern part of Marengo 

 County, Ala., to the northern part of Noxubee County, Miss., is overlain unconformably by 

 Eocene strata carrying characteristic fossils. 



Character. — The Selma chalk is described by Smith, Langdon, and Johnson 5 as follows: 



The rock is of comparatively uniform composition, being a gray to bluish-colored argillaceous limestone, traversed 

 at intervals by beds of purer limestone, which is at the same time a little harder in texture. In some places the 

 material is a dark-bluish clay marl. * * * 



Mr. K. M. Cunningham, of Mobile, a well-known microscopist, undertook for me the investigation of the char- 

 acteristic rocks of this formation, and the results of his examination of specimens collected from various points from 

 Montgomery westward are given in the accompanying article by him. The statement has often been made that the 

 true chalk is absent from the Cretaceous formations of North America, but we have here evidence that it is present 

 in no inconsiderable proportions in Alabama, and Hill has recently shown that it occurs in Arkansas and in Texas, 

 and we may reasonably infer that other States will yield upon closer examination very similar material. 



In western Alabama the Selma chalk has a thickness of nearly 1,000 feet. The basal Eocene 

 strata which in this region overlie the Selma chalk were formerly correlated with the Ripley 

 formation. The error was corrected in part by Harris 6 in 1S96, in part by Smith 7 in 1910, and 

 in part by the writer. 8 Typical exposures of the Selma chalk are shown in Plate V, A and B. 



1 Langdon, D. W., jr., Variations in the Cretaceous and Tertiary strata of Alabama: Bull. Geol. Soe. America, vol. 2, 1S90, pp. 5S7-G06. 



2 Veatch, Otto, Second report on the clay deposits of Georgia: Bull. Geol. Survey Georgia No. IS, 1909, pp. S2-106. 



J Winchell, Alexander, Statistics of some artesian wells of Alabama: Proc. Am. Assoc. Adv. Sci., vol. 10, pt. 2, 1S57, p. 91. 



« Smith, E. A., Langdon, D. W., jr., and Johnson, L. C., On the geology of the Coastal Plain of Alabama, Geol. Survey Alabama, 1S94, p. 255. 



5 Idem, pp. 276, 285. 



s Harris, G. D., The Midway stage: Bull. Am. Paleontology, vol. 1, No. 4, 1896, pp. 25-36 (139-150). 



7 Smith, E. A., Cretaceous-Eocene contact, Tombigbee River, Ala.: Jour. Geology, vol. 18, No. 5, 1910, pp. 430-434. 



3 Stephenson, L. W., unpublished notes on sections at Bridgeport and Old Canton landings, Alabama River, in Wilcox County, and at Liv- 

 ingston, in Sumter County. 



