174 ZUULUGICAL GEOGRAPHY. [part hi. 



South America, Auvstralia, and South Africa with Madaoascar — 

 have been more or less completely isolated, during long periods, 

 both from the northern continent and from each other. These 

 hitter countries have, however, been subject to more or less im- 

 migration from the north during rare epochs of approximation 

 to, or partial union with it. In the northern, more extensive, and 

 probably more ancient land, the process of development has 

 been more raj^id, and has resulted in more varied and higher 

 types ; while the southern lands, for the most part, seem to have 

 produced numerous diverging modifications of the lower grades 

 of organization, the original types of which they derived either 

 i'rom the north, or from some of the ancient continents in Meso- 

 zoic or PaliTeozoic times. Hence those curious resemblances in 

 the fauna of South America, Australia, and, to a less extent, 

 Madagascar, which have led to a somewhat general belief that 

 these distant countries must at one time or other have been 

 united ; a belief which, after a careful examination of all the 

 facts, does not seem to the author of this work to be well 

 founded. On the other hand, there is the most satisfactory 

 evidence that each southern region lias been more or less 

 closely united (during the tertiary or later secondary epocli) 

 with the great northern continents, leading to numerous resem- 

 blances and affinities in their productions. 



In endeavouring to present at a glance in the most convenient 

 manner, the distribution of the families in the several regions 

 and sub-regions, it was necessary to arrange them, so that those 

 whose relations to each other were closest should stand side by 

 side ; the first and last being those between which the relations 

 were least numerous and least important. Influenced by the 

 usual opinions as to the relations between Australia and South 

 America, the series was at first begun with the Nearctic, and 

 terminated with the Australian and Neotropical regions ; and it 

 was not till the wliole of the vertebrate families had been gone 

 through, and their distribution carefully studied, that these last 

 two regions were seen to be really wider apart than any others 

 of the series. It was therefore decided to alter the arrangement, 

 beginning witli the Neotropical, and ending witli the Australian 



