ZOOLOGICAL EEGIONS. 61 



The result of these causes is that the locality on the west coasts of South Africa and 

 South America where the mean temperature of July is 77° is more than a thousand miles 

 further north than it is on the east coasts of the two continents. 



In the annexed map the upper dotted line represents the isothermal of 60° for the 

 month of July ; the white space above it consequently represents the Arctic Region of the 

 Charadriidae, where the mean temperature of the month of July is 60° or less. 



The dark space below this line represents the North Temperate Region of these birds, 

 where the mean temperature of the month of July ranges from 60° to 77°. 



The middle dotted line represents the isothermal of 77° for the month of July, and the 

 lower dotted line the same isothermal for the month of January. Between these two lines 

 is a white space which represents the Tropical Region of the Charadriidae, where the mean 

 temperature of the month of July north of the Equator, and that of the month of January 

 south of the Equator, ranges from 77° to 90°. The dark spaces below the lower dotted 

 line represent the South Temperate Region of the Charadriidae, where the mean temperature 

 of the month of January ranges from 77° to 60° at the Cape, but down to 45° at Cape 

 Horn and Kerguelen Island. The latter locality is not visited by any species belonging to 

 the Charadriidae; but it contains a Chionis, which also belongs to the Limicolae, and is 

 represented by allied species in the Crozets and the Falkland Islands. Possibly a 

 more logical way of treating the subject would be to regard Kerguelen Island, the Antarctic 

 South Island of New Zealand, Patagonia, the Falkland Islands, Tristan d'Acunha, and Kegion. 

 the Crozets, where the mean temperature of the month of January is below 60°, as an 

 Antarctic Region. My reason for not doing so is that the Antarctic Region is only 

 visited for breeding-purposes by half a dozen species of Charadriidae, belonging to genera 

 essentially temperate in their range. 



If, however, we were to extend our range to the whole of the suborder Limicolae, we 

 should be compelled to recognize an Antarctic Region, having characteristic genera 

 apparently so distinct from those comprising the Charadriidae as to warrant their being 

 placed in separate families, Chionidae and Thinocoridse. The Antarctic Region appears to 

 be also the centre of dispersal of two suborders — the Procellaridae and the Impennes ; so that 

 it must be regarded as a very important Region when the distribution of the whole order 

 Charadriformes is considered. 



If I am right in supposing that the post-pliocene glacial epoch was the main factor in 

 the dispei'sal and consequent isolation which resulted in the differentiation of the present 

 species of Charadriidae, and that to find the common ancestors of the species now 

 belonging to different genera it is necessary to go beyond the prae-pliocene glacial epoch, 

 it is only fair to assume that as the Antarctic Region is characterized by Limicoline 

 forms belonging to different families, the few species of Charadriidae which encroach 

 on its limits are stragglers from the north polar basin, and that their ancestors were 

 isolated from the ancestors of the Antarctic Limicolae possibly so long ago as Eocene 

 times. 



