36 H. S. Williams — Devonian System of Devonshire. 



If we examine the Devonian limestone fauna of Devonshire 

 we find its generic combinations very similar to those of the 

 Corniferons limestone of onr Appalachian basin, but the same 

 genera are also seen in the calcareous strata of the Hamilton 

 formation, and the specific types of the genera running through 

 the whole system are more closely allied with those of our 

 Hamilton, and even Chemung horizons than with those of our 

 Lower Devonian. The great prominence of corals reminds us 

 of the Corniferous, but when we compare the Brachiopods we 

 find numerous forms, the representatives of which do not 

 appear in our Appalachian sections till after the Hamilton ter- 

 rane is passed. Such are Spirifera disjuncta, Rhynchonella 

 pugnus and acuminata, Spirifera curvata, Orthis striatula, 

 Rhynchonella cuboides, etc. 



Some of. these are not in the main limestone on the conti- 

 nent, and there are some indications of a separation of the 

 fauna, in the sections of the Hartz and of Russia, more nearly 

 corresponding to our division into Middle and Upper Devo- 

 nian. And even in the more western sections of Europe the 

 Frasnienne, as distinguished from the Grivetienne limestone 

 contains a decidedly later fauna than the latter. It appears 

 probable that the limestones of South Devonshire represent the 

 general interval between the close of our Corniferous and the 

 early part of our Chemung formation. 



Another j)roblem is here suggested, viz : was there any mi- 

 gration of the faunas? For the determination of this point I 

 have made a study of the Cuboides fauna, tracing it from New 

 York to England, Belgium, France, Germany, Russia, Siberia, 

 Persia and China. There seems to be good evidence that this 

 fauna whose place is at the top of the Devonian limestone 

 periods of these regions had a center of distribution nearer 

 north France than either eastern America on the one hand or 

 China on the other. If we take this as a uniform horizon, 

 homotaxially, it may be said that a considerable number of 

 species (including forms, under different names, which are 

 very closely allied modifications of the same races) appeared 

 before the " Cuboides " stage in the English, European and 

 Russian sections, but not till after that stage in the New. York 

 sections. 



In the sections of the interior of North America this partic- 

 ular fauna has not been recognized. Rhynchonella castanea 

 Meek, of the MacKenzie River basin and Nevada, although in 

 some respects resembling R. cuboides Sow., appears to be spe- 

 cifically extinct. The few species which occur in the " Cu- 

 boides zone " of Europe and range across the continent of 

 America, in their generic history appear to represent a Carbon- 

 iferous stage of development, and while abundant in the Euro- 



