524 E. O. ULRICH REVISION OF THE PALEOZOIC SYSTEMS 



tions or they are different in both respects. A satisfactory solution be- 

 tween these two interpretations is usually a very difficult matter. Strati- 

 graphic geology is full of such problems, many of them scarcely suspected 

 and a large proportion quite unsolved. 



An unusually good and difficult illustration of this principle is pre- 

 sented by the taxonomic relations of the dolomitic Niagaran limestones 

 of North America north of Saint Louis and New York to the non-dolo- 

 mitic deposits of the same series in Kentucky, Tennessee, and northern 

 Arkansas. The latter extend up the Mississippi valley by outcrop to 

 Perry County, Missouri, where they pinch out by overlap. They proba- 

 bly extend farther northward in Illinois under cover of later rocks. The 

 dolomites extend southward from Wisconsin and Iowa, growing thinner 

 but not changing in character as they proceed, to Grafton, Illinois, and 

 Lincoln County, Missouri, beyond which points this Silurian type of rock 

 is seen no more. Except locally neither crosses an irregular intermediate 

 line. In other words, it is not a case of lateral transition from the one 

 kind of rock to the other, but of an actual land barrier separating the 

 northern waters from the southern. Now, did these two seas occupy their 

 respective continental basins at the same time, or did each pulsate back 

 and forth, so that when the northern sea retreated the southern waters ad- 

 vanced ? We get some light on the question from the overlap of the two 

 types in western New York, but the problem is still an unsettled one. 

 As a matter of opinion, based chiefly on philosophical considerations dis- 

 cussed on pages 422 to 425 and 558 to 561, I favor the latter view. 



A very similar problem, now largely solved, is the age relation of the 

 Galena dolomite of the upper Mississippi Valley and the north and west 

 generally to the Mohawkian formations in the south and east. Prior to 

 1895 the Galena was usually regarded as late Trenton in age, and on 

 three occasions, the last in 1879,®^ it was correlated with the Utica shale 

 of New York. Until very recently®^ the Kimmswick limestone, which is 

 known by frequent outcrops in eastern Missouri south of Pike County 

 and in northern Arkansas and south central Tennessee, was referred to as 

 the southern equivalent of the Galena. By discovery of lower to middle 

 Galena, typical in fauna and rock, resting on upper Kimmswick lime- 

 stone in Pike and Lincoln counties, Missouri, the error of this supposi- 

 tion is now established. We know further that the lower Galena (Pros- 



«« C. D. Walcott : The Utlca slate and related formations. Trans. Albany Inst, vol. x. 



"''T. E. Savage: American Journal Science, vol. xxv, 1908, pp. 431-443, still refers to 

 the Kimmswick as "Galena Trenton," but in vol. xxviii, p. 512, of the same journal, he 

 modifies this view slightly by correlating the formation with the Fusispira bed of the 

 Minnesota section. 



