94 The Middle Devoxian Deposits of Maryland 



and that they are faunally most closely related to the Marcellus. It is 

 further evident that they are older than the zone of Liorhynchus limitare 

 in Maryland. 



The determination of the exact age of these beds is difficult since the 

 principles of correlation of deposits of this type are not well established. 

 It is evident that faunas that are successive in one area must frequently 

 be contemporaneous elsewhere, since the advent of highly developed 

 species immediately succeeding another fauna in one region indicates 

 that both were in existence earlier in some other basin and hence were, 

 for a time at least, coexistent. When therefore a mingling of the species 

 of two such faunas is observed in a third area it is at times impossible to 

 affirm their precise age with certainty. 



The following methods of solution appear possible under such circum- 

 stances; the study of the direction of migration of the species and their 

 relations to sediments and physical conditions, the emphasis of certain 

 species rather than others in correlation, the determination of the rela- 

 tions of the fauna as a whole and correlation by some recognizable horizon. 



The first of these methods appears the most decisive. Unfortunately 

 knowledge of the necessar}^ data is rarely obtainable in an early stage of 

 the investigation so that while this method has been used with much 

 success in the sti;dy of the Upper Devonian it does not appear to be con- 

 clusive here. Mr. Kindle has contributed a valuable discussion of this 

 aspect of the problem in another part of the present volume. 



The second method appears to lead to very different results accordingly 

 as certain elements of the fauna are emphasized rather than others. If 

 we assume that the Marcellus of Maryland may represent but a part 

 of the Marcellus of New York, a possible assumption in view of the 

 varying limits of that formation, then the beds under consideration may 

 be of early Marcellus age. This view would accord not only with the fact, 

 60 often observed, that species of an earlier fauna may persist and become 

 mingled with those of a later; but also with the presence of so large a 

 number of species restricted to the Marcellus and later beds in New 

 York. Indeed many students would not hcstitate to assume that species 

 observed in later faunas elsewhere are, in general, the latest immigrants 



