392 W. D. MATTHEW THE CRETACEOUS-TERTIARY PROBLEM 



Eiver indicate a fauna of tht same facies as the Lance, doubtfully more 

 primitive in stage. 



Comparing the Lance with the Puerco and Torrejon, we find a fauna 

 of very different facies, but indicating no very wide gap in time. It is 

 not very clear whether the great faunal difference is due to diverse local 

 environments or to a great movement of faunal migration, but a combi- 

 nation of both seems to fit the data most exactly. 



Comparing tlie Puerto and Torrejoii witli tlie Wasatch, we find fauna* 

 which appear to represent similar facies, but a very marked change in 

 the sudden appearance of new orders and families of mammals and rep- 

 tiles which can best be accounted for as immigrants. The lapse of time, 

 as measured by the change in phyla which pass through, is not very great 

 between Torrejon and Wasatch and ver\- slight between Clark Fork beds 

 and Wasatch. But the close of the Paleocene is marked by a great migra- 

 tion movement. 



Comparing the mammal fauna of the Upper Fort Union with that of 

 the Torrejon, we find that it appears to be of the same age, as indicated 

 by the identity- of a part of the fauna. But it apparently represents a 

 somewhat different facies, with certain points of analogy to the Lance. 



Comparing the Paskapoo fauna with the Lance, it appears by the same 

 criteria to be equivalent or only slightly later in age, while apparently 

 older than the Puerco and presumptively older than the Upper Fort 

 Union; but it represents a fa4:-ies very different from that of the Lance, 

 corresponding more nearly with that of the Puerco and Torrejon, and per- 

 haps still more closely with that of the Fort Union. It thi*ows some light 

 on the interpretation of the break ]>etween Lance and Puerco, for it con- 

 tains an element that may be regarded as ancestral to a part of the Paleo- 

 cene placentals, but does not appear to l)e related to the major and more 

 progressive part of tliem. This would indicate tliat tlie absence from the 

 Lance of the more primitive and arcliaic groups of the Puerco-Torrej«n 

 fauna is a matter of facies ; but tliat the absence of the larger, more pro- 

 gressive and abundant Paleocene placentals from the Lance is to be 

 ascribed to a migration movement at its close. The evidence on this 

 point is, however, too scanty to be of any considerable weight. 



By similar methods of correlation Brown has shown that the Edmon- 

 ton formation underlying the Paskapoo and overlying the Pierre and the 

 Ojo Alamo beds, which lie unconformably beneath tlie Puerco, are older 

 than the Lance and equivalent to "the Belly River, and that the Hell 

 Creek beds of Montana are equivalent in age to the typical Lance, all 

 representing iiearly the same facies, but the Canadian formations some- 

 what more aecessilde to the marine fauna, as indicated by the finding of 



