REPORT OF THE DIRECTOR I914 1 23 



barrier between these deposits. But the faunas in these two approx- 

 imate points are still so unlike that to infer a junction of the waters 

 across the isthmus is not yet justified. Should it so turn out that 

 the Oriskany strand was here unbroken, we should then have the 

 fauna completing an entire circuit in Appalachia. The St Genevieve 

 basin at the west seems to present the duplication of the conditions 

 in the northeastern lime basins of Gaspe, an essential exclusion of 

 the sand from the contemporaneous deeper water. But we know 

 too little yet of this Missouri basin to institute any satisfactory 

 comparison between its Devonic elements and those of the northeast. 



To return then to the Brazilian fauna of the Maecurti river sand- 

 stones, we may feel reasonably secure that its distance from the 

 overspread of the northern sands, the opportunities for variations 

 from those faunas by development under conditions of isolation, 

 are the responsible factors for actual and apparent differences in 

 these faunas from those of the north. 



It will not do, however, to intensify these differences by statement. 

 The affinities are obvious and they are distinctive in generic char- 

 acters. In a certain sense there is in comparison with the northern 

 Oriskany, a later tinge to the fauna ; species which carry suggestions 

 of the stage next succeeding in the New York succession. This 

 is entirely in accordance with such expectations as we should derive 

 from our knowledge of the Brazilian sections, for the next term 

 above the Maecuru is the sandstone of the Rio Erere, and its 

 fossils, as described by their discoverers, Hartt and Rathbun, and 

 confirmed by the evidence brought out by Derby and myself, indicate 

 their Middle Devonic age and their rather intimate relations with the 

 Hamilton of New York. The intermediate limestone term of the 

 series present in the New York succession is then missing here, 

 with our present knowledge, but the open and freer connection of 

 this later shallow sea with the northern seas of the Middle Devonic 

 is a very pronounced fact. In speaking of the Gaspe sandstone 

 fauna, I have referred to Professor Williams's intin.ation that its 

 pelecypods might be construed as Lower Devonic (Oriskany- 

 Coblentzian), and I may here refer to Doctor Katzer's contention 

 that the Erere sandstone fauna is likewise Lower Devonic. The 

 two suggestions are not alike in quality nor based on like argumenta- 

 tion, though similar in purport. Yet with close analysis of these 

 faunas we are not prepared to concede these alternative propositions, 

 even though at the south, the later Erere sandstone is in direct 

 continuity with the earlier or Maecuru. At all events, the specific 

 and generic similitudes of these two Amazon faunas with those at 



