570 PLACENTAL MAMMALS xxi. 2- 



more profound than those of the Pleistocene and it has been suggested 

 that this may have been responsible for the production of mammal-like 

 characteristics among the synapsids (p. 538). It is tempting to make 

 similar suggestions about the development of placental mammals at 

 the end of the Cretaceous. The presence of cold uplands may indeed 

 have been responsible for such success as the multituberculates and 

 other Mesozoic mammals achieved during the Cretaceous. But this 

 slow revolution does not explain, by itself, the success of the placen- 

 tal, because they only became differentiated into varied groups 

 towards the end of the Cretaceous. There is some evidence that this 

 latter time was not very cold, and it will be remembered that some 

 dinosaurs persisted to the very close of the Mesozoic. On the other 

 hand, pterodactyls had died out earlier and so had some groups of the 

 dinosaurs. 



Although there may well have been large climatic changes at the 

 end of the Cretaceous period it is unwise to make simple statements 

 about their relation to the evolution of the mammals. The whole 

 Cretaceous period lasted for more than 60 million years and we can- 

 not trace in detail the numerous changes of climate that must have 

 taken place, probably in different directions in different parts of the 

 world. Even the marine faunas were affected; for instance, the ich- 

 thyosaurs and plesiosaurs died out before the end of the Cretaceous 

 and were followed by the mosasaurs (p. 409), which had a sudden 

 period of success lasting for some millions of years. 



The change of fauna at the end of the Cretaceous was as remarkable 

 for the animals that appeared as for those that were lost. The remain- 

 ing dinosaurs died out on land, as did the mosasaurs, and also am- 

 monites and belemnites, in the sea. At the same time there appeared 

 not only numerous placental mammals, but also a great variety of true 

 birds, and in the seas teleostean fishes and cephalopods of modern 

 type. 



The apparent suddenness of the change may be deceptive. Our 

 knowledge of the conditions at the beginning of the Cenozoic period 

 is fragmentary. It was a time when the land stood high, at least in the 

 regions we know best. A great part of the continental shelf was above 

 water, and therefore producing few fossils. Most of our information 

 about the fossils laid down at this time comes not from marine beds, 

 but from the Talaeocene' continental deposits, known chiefly in 

 America. These deposits lie on top of undoubted Cretaceous, dino- 

 saur-containing beds, and they contain a variety of placental mammals. 

 At the upper end these Palaeocene deposits are continuous with beds 



