1>RINCIPLES OF STRATlGRAPHiC CORRELATIONS 553 



lationship may after all be remote, so far as geological time is concerned. 

 Besides, comparison with the Canadian faunas in Pennsylvania and the 

 valley of Lake Champlain seems to antagonize rather than to support 

 this suggestion, the oldest Canadian fauna there being very different, 

 while alliances with the Yellville fauna are apparent only in the Cassin 

 fauna of Division D. Under the circumstances the final solution of the 

 problem is yet a matter of opinion, mine being that the emergence of the 

 Appalachian Valley basins at the close of the Chepultepec continued to 

 the beginning of Stonehenge deposition^ and that when the restricted 

 Jefferson City sea was withdrawn the emergence involved the whole 

 continent. Further evidence supporting this opinion will be given in 

 Part III. (See page 673.) 



The Canadian-Ordovician emergence. — The evidence on which exten- 

 sive emergence is inferred at the close of the Canadian in America is 

 similarly intricate, but on the whole more conclusive. The solution of 

 the problem again involves the disposition of a series of deposits in the 

 Mississippi Valley — the Saint Peter sandstone and associated limestones 

 (see page 479) — that can not be identified in the Appalachian Valley. 

 The case differs in that the missing beds are the first instead of the last 

 of the system to which they belong; hence that the Ordovician submer- 

 gence began in the Mississippi Valley and not in the Appalachian basins. 



The Bellefonte dolomite, the last of the Canadian deposits in central 

 Pennsylvania, is succeeded at Bellefonte by upper Stones Eiver. This 

 gap is readily diminished to the extent possible in the Appalachian 

 Valley by comparison with sections at Chambersburg, Pennsylvania, and 

 Mai-tinsburg, West Virginia. In these sections the full Stones Eiver 

 group is represented in a maximum thickness of about 1,200 feet of solid 

 limestone. Its base is in contact with the Beekmantown limestone, the 

 top of which is here very late Canadian in age. That the contact is un- 

 conformable and marks a stratigraphic hiatus is indicated by various 

 signs, among them occasionally a thin layer of quartz conglomerate. 



The Beekmantown- Stones River hiatus is thought to be represented in 

 part by the Saint Peter series in the Mississippi Valley. In north 

 Arkansas this sandstone series rests with unquestionable unconformity on 

 the Yellville or, where that formation is absent, on the Jefferson City 

 dolomite. There is also an unconformity at the top of the Saint Peter 

 sandstone and of its seaward representative, the Joachim dolomite; but, 

 despite the fact that the overlying beds are either late Stones Eiver or of 

 Lowville age, this unconformity does not impress one as so important as 

 the break at the base. As must be apparent, the evidence is involved, 



XXXVII— Bull. Gkol. Soc. Am., Vol. 22, 1910... 



