the Bering Sea at times in the past. Others probably arrived over seas, 

 like the waifs and strays that became estabUshed as the fauna of the 

 Hawaiian Islands. Each major land mass — North America, South 

 America, Eurasia, Africa, and Australia — had a greater or less amount of 

 isolation, decreasing at times with establishment of broader and closer 

 land connections, increasing at times with their disappearance. In this 

 partial isolation, each land mass gradually evolved its own assemblage 

 of birds — its bird fauna — from the various immigrants that came pro- 

 gressively and are still coming to it. There was shifting back and forth, 

 with re-invasion and re-immigration, but the isolation has been long 

 enough and strict enough to produce a characteristic bird life in each 

 major land area. Certain groups of water birds became adapted to cer- 

 tain conditions — penguins to Antarctic waters, frigate birds to tropical 

 seas, and auks to northern waters. Certain types of birds became so 

 dominant and adaptable that they were able to spread to most parts 

 of the world or to all the oceans. 



In spite of this immigration and emigration, these exoduses and in- 

 vasions, the faunas of various parts of the world are so distinct that if a 

 traveller tells you what birds he has seen, you can tell what part of the 

 world he has been in. If he speaks of nesting emperor geese, he must 

 have been in the Bering Sea area; if of nesting emperor penguins, on the 

 Antarctic ice shelf; if of rifle birds, in New Guinea or North Australia; 

 if of kiwis, in New Zealand. 



It is useful to give names to the areas that are occupied by various 

 faunas of birds, or to regions that are the centers of distribution of mem- 

 bers of related species. The continents may not be completely occupied 

 by the faunas essentially characteristic for them. North Africa, for ex- 

 ample, does not have an African fauna, but a Eurasian one; the Sahara, 

 and not the Mediterranean, is the barrier. The break between two 

 faunas is not always explainable from present topography. From Ma- 

 laya through Sumatra, Java, and Bali, the fauna remains south Asiatic 

 in character; but across the narrow straits, on the island of Lombok, the 

 bird fauna is suddenly Australian. This is only understandable on the 

 basis of former land changes now amply documented by geologists. 

 There have been various sets of divisions of the surface by zoogeogra- 

 phers, many with points of disagreement, but the divisions presented 

 below, based on the study of animal distribudon, seem to represent the 

 most useful set of concepts. The regions, characterized by families or 

 groups of genera restricted to them, must be interpreted in the light of 

 the fact that there are also groups of birds that are more widespread. 



1141 



