FAUNAL RELATIONS OF EARLY VERTEBRATES 401 
America was closed during the same interval, though of this we cannot 
be at all sure till we know more of the South American Permian fauna, 
which, so far, lacks every distinctive form peculiar to North America. 
Whether or not the communication between North America and 
the eastern continents was by way of the north Atlantic, it is quite 
evident that there must have been free communication during part 
or all of the Mesozoic time between North and South America, proof 
of which is seen in the dinosaurs, mosasaurs, and crocodiles, some of 
them, according to competent observers, identical generically even 
with North American forms. We have yet much to learn about the 
Mesozoic fauna of South America, but, so far as our knowledge yet 
goes, there is a close relationship between them. ‘This similarity, 
of course, may have been the result of a westward migration from 
Africa to South America by the way of a southern land communication, 
and a concurrent intermigration of the same types from Africa north- 
ward to Europe and thence by the north Atlantic to North America. 
But a simpler explanation would be that of a land communication 
between North and South America, and a single trans-Atlantic bridge, 
which, in my opinion, was the southern one. 
It is very true that such hypotheses as I have offered are largely 
based upon negative evidence. Future discoveries may bring to 
light, both in Europe and America, types which now appear to have 
a more restricted geographical distribution; especially may future 
discoveries in South America and Africa show more distinctive types, 
or, on the other hand, more common forms. JI do believe, however, 
that the long-continued exploitation of the Mesozoic rocks of North 
America is gradually converting negative into positive evidence; 
that we may say with tolerable certainty that certain types of land 
vertebrates, such as the Proterosauria, Proganosauria, Pareiosauria, 
Therodontia, etc., have never existed in North America. 
In the accompanying table I have given, as fully and as accurately 
as the present state of our knowledge will permit, the geological range 
and distribution of the larger groups of air-breathing vertebrates, 
with especial reference to North America. In not a few instances 
precise stratigraphical data are wanting, so that groups must be 
recorded throughout a division of the chart, which later may be found 
to have a more restricted range. An attempt has been made to indi- 
