//. S. Williams — Devonian System oj Devonshire, 35 



layers represent the condition of a more northern area, and 

 doubtless when fossils are obtainable will be found to repre- 

 sent distinct (at least local) faunas. 



It must be evident that the marine invertebrate faunas of 

 the whole Devonian are hut the equivalents of the vertebrate 

 fauna and the flora of the Old Red Sandstone. This striking 

 law is easier to accept and practically understand than the 

 other, viz : that marine invertebrate faunas of very different 

 species with few and possibly in many eases no common forms 

 actually co-existed in the same ocean at the same time. With 

 this second law in mind it is clear to see that the shifting of 

 currents in the ocean — the oscillation upward and downward 

 of the land in relation to the sea-level, the many changes in the 

 relations of land and sea, of which there are unmistakable 

 evidence — all these events must have produced mixing and 

 changing of the faunas over any particular spot, not only con- 

 stantly but to an extent we can scarcely conjecture from the 

 very slight evidences preserved. 



A s* the off-shore and deeper sea faunas of to-day differ from 

 those living between tides on the coast, so we must believe 

 they differed in the Devonian age. As the fauna off the 

 Florida coast differs from that of Labrador now, so we must 

 believe there were striking differences between the faunas of 

 the warm ocean of the equatorial regions and the faunas of the 

 colder polar regions of the same sea for a time as far back as 

 such differences in climate existed. 



When, therefore, we attempt to draw parallel lines to con- 

 nect the stratigraphical series of Xew York or other parts of 

 America, with those of England or the continent of Europe, 

 the mere identity of species in their numerical relations is 

 an unsatisfactory guide. 



The species that are found identical both sides the sea are 

 likely to be species whose vertical range is as long as a whole 



tern, and the closely related forms may be either («), one the 

 successor of the other, or (7>), one the modified migrant of the 

 other. As to the relationship between two separate faunas, the 

 one following the other, the difference in species and genera is 

 often greater than that between either of the faunas and the 

 one next below it in a like kind of deposit. As an example, 

 we may cite our Genesee shale and Marcel lus shale whose 

 respective faunas more closely resemble each other than either 

 of them does that of the Hamilton fauna between. To elim- 

 inate errors of this kind we should compare faunas of like 

 deposits with each other, not because the terrane of the one 

 region has in its beginning and ending any necessary relation- 

 ship with the corresponding one of the other region, but 

 because conditions of life are likely to have been more nearly 

 uniform where the deposits are alike. 



