!N0. I DISTRIBUTION OF THE ONYCHOPHORA — CLARK 19 



Africa. This hypothesis, moreover, accords with what we know of 

 the distribution of many other southern types. 



As the American species of Peripatoidinse are far more specialized 

 than the American species of Peripatidse, we may assume that the 

 connection between southern South America and the Australian 

 region persisted to a much later date than that by which the Peri- 

 patidae arrived from Africa. 



Although, judging from what we know of the other elements of 

 the faunas of Australia, Tasmania, and New Zealand, it is easy to 

 understand how the Peripatoidinae entered southern South America 

 from the Australian region, it is not so easy to understand how they 

 entered South Africa, unless we are willing to assume that there has 

 been a connection between South Africa and Antarctica by way of the 

 Crozet and Kerguelen Banks, which was more or less contemporane- 

 ous with that between southern South America and Antarctica. 



The African genus Opisthopatus is very closely allied to the Amer- 

 ican Metaperipatus, the alliance being much more close than in the 

 case of the African and American genera of the Peripatidas. These 

 two genera are less specialized than are the other genera of the 

 Peripatopsidse, and the explanation at once suggests itself that, be- 

 sides, being later arrivals in America than the genera of the Peri- 

 patidae, they, like Oroperipatus, indicate the extreme limits of the area 

 over which their group (the Peripatoidinae) was at one time dominant, 

 and exist at present in localities with a physico-economically very re- 

 stricted radius which approaches the physico-economical conditions of 

 the original habitat of their subfamily more closely than does the 

 habitat of any of the more specialized Australian genera, so that they 

 have had but little incentive to change in order to meet new conditions. 



If this were so, it would suggest of itself that the Peripatoidinae 

 in the past had their headquarters in the extreme south, in contrast to 

 the primarily tropical Peripatidse. 



The sharp separation in the distribution of the Peripatoidinae and 

 the Peripatopsinae in the East Indian and Australian regions suggests 

 a long and complete separation of the land of which the Moluccas, 

 New Guinea, and New Britain (and southeastern Africa) were once 

 an integral part, from Australia (including Tasmania and New Zea- 

 land, but possibly excepting the Torres Strait region) , this separation 

 long antedating the separation of Australia from Antarctica, but 

 being subsequent to the isolation of the Malayan region from the 

 Moluccas and the islands farther east. 



