STRATIGRAPHIC CLASSIFICATION DIASTROPHIC CRITERIA 423 



vale fauna could have been overlooked in Canada if it invaded from the 

 north. 



In Ohio and Kentucky the Arnheim is followed by the Waynesville 

 and Liberty formations, but north of Gallatin, Tennessee, the Fernvale 

 wedges in between the Arnheim and the Waynesville. Evidently dif- 

 ferential movements had again occurred, bringing about conditions in the 

 Waynesville-Liberty stage simulating those prevailing during the 

 Arnheim. 



In eastern Missouri and southern Illinois the Fernvale is succeeded un- 

 conformably by a southward extension of the Maquoketa shale of Iowa. 

 North of Sainte Genevieve, Missouri, this extension, save that it is thin- 

 ner, is in every respect typical of the Iowa formation. Further south, 

 however, at Cape Girardeau, Missouri, and Thebes, Illinois, the basal part 

 only consists of dark shale, the middle and upper parts of the zone being 

 of arenaceous shale -and sandstone — a lithological difference Justifying 

 the local designation "Thebes formation." No Richmond, nor indeed 

 Silurian deposits of any age, are found on the north, south, and west 

 sides of Ozarkia.^^ 



The geographic derivation of the fauna of the typical dark Maquoketa' 

 shale is not positively determined. Though both the formation and its 

 fauna are best develop'ed in eastern Iowa, neither has been identified north 

 of the United States, east of Wisconsin and Illinois, nor to the west of 

 the 100th meridian. But as the fauna has its nearest relatives in the 

 Utica, an unquestionable north Atlantic fauna, it is quite possible that 

 the Maquoketa is represented in the supposed Utica on Georgian Bay and 

 Hudson Strait, and that it invaded the Mississippi Valley from that 

 direction. 



The Whitewater and Saluda divisions of the Eichmondian in Indiana, 

 Kentucky, and Ohio, succeed the Liberty, but their relations to the 



38 Possibly such rocks occur farther west beneath the cover of overlapping late Paleo- 

 zoics, the eastern edge of which as a rule rests on Ozarkian dolomites ; but they must be 

 very far away, since no evidence of such a condition is even suggested by the records of 

 deep wells in southwestern Missouri. In northwestern Missouri, however, as shown by a 

 deep well at Forest City, the strata dip into a Paleozoic basin containing Mississippian 

 and Devonian deposits that are wholly absent where they should outcrop on the margin 

 of the area of Ozarkian rocks. Though the well at Forest City does not reach the hori- 

 zon, the probable existence of the Maquoketa shale in this basin and of an arm of this 

 Richmond gea extending far southward from Iowa, is strongly indicated by the lithologic 

 character and the fauna of the Sylvan shale in the Arbuckle uplift of Oklahoma. A 

 similar connection between Oklahoma and the Iowa-Minnesota region is no less clearly 

 suggested during the deposition of the Decorah shale and the Prosser limestone In the 

 northern States by fossils found In the upper part of the Bromide formation and In the 

 lower part of the Viola limestone In the southern area. 



