REVISION OF STRATIGRAPHIC CLASSIFICATION 299 



ment. In citing examples of such periodic emigrants the most striking 

 doubtless occur among the Brachiopoda. Varieties of Dalmanella testu- 

 dinaria, for instance, appeared in the Mohawkian and Cincinnatian rocks 

 of Kentucky and Tennessee no less than five times; Plectambonites 

 sericeus is represented by barely distinguishable varieties in the Stones 

 River, the late Black River, in late Trenton, in the Eden, and finally by 

 two mutations in the Richmond; Atrypa reticularis first appears in the 

 Clinton, and then continued, with occasional intermissions, to hold its 

 own to near the close of the Devonian; Leptcena rhomhoidalis, the Methu- 

 salah of fossils, enjoyed two rather brief trips to the interior in Ordo- 

 vician ages, lived there rather generally during the whole of the Silurian, 

 and reappeared again and again during Devonian and early Mississippian 

 times. 



If the several invasions of such broadly conceived species can be dis- 

 criminated — as a rule it is not difficult to do so — they constitute excellent 

 datum planes in correlation. If, however, the paleontologist will not 

 take the trouble to learn the often very slight modifications of the specific 

 type that distinguishes each appearance from the other, then they are 

 more likely to prove stumbling blocks than aids in determining equiva- 

 lence. 



Moreover, as such long-ranged species are generally very abundant, 

 when they occur at all, they will, of course, always figure in lists of 

 standard dominant species, prepared according to the suggestions of 

 Prof. H. S. Williams. The listing of names of species like Leptcena 

 rhomhoidalis, Atrypa reticulaiis, Plectambonites sericeus, and Dalmanella 

 testudinaria, without designating exactly the particular mutation of each 

 intended, can not possibly tend to accuracy in correlation. Indeed, they 

 may lead to positive error, since the less common though more character- 

 istic species of a formation may be crowded out of the list by their 

 ubiquitous but not diagnostic competitors. 



Recurrence of the Catheys fauna. — Three excellent and instructive ex- 

 amples of recurrent faunas deserve mention here. The first of these 

 cases is displayed in the Ordovician rocks of central Kentucky and mid- 

 dle Tennessee. The Catheys formation in Tennessee is the uppermost of 

 three formations which together are regarded as representing the Trenton 

 limestone of New York. The fauna of the Catheys really embraces two 

 faunules, both of which reappear in more or less modified forms in subse- 

 quent deposits. The first consists chiefly of conical and massive corals 

 belonging to the genera Streptelasma, Stromatocerium or Lahechia, Tetror 

 dium, and Columnaria. These conspicuous fossils are confined to the 

 lower half of the Catheys, but reappear in a well marked zone well up in 



k 



