l8 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



types in South Africa and the least specialized types in New Britain, 

 New Guinea, and Ceram. This would appear to indicate that the 

 headquarters of the group was originally somewhere between New 

 Guinea and South Africa, and that New Guinea and the adjacent 

 islands became very early detached and separated by a water barrier 

 so that the endemic onychophores were protected from the intrusion 

 of later and more efficient types, exactly as were the species of the 

 genus Oroperipatus west of the Andes. If this view is correct, 

 Madagascar should support a more specialized type of this subfamily 

 than either South Africa or New Guinea and the adjacent islands. 



We do not know any onychophores from the Cape York peninsula 

 in Australia ; it is probable that such forms as occur there belong to 

 the Peripatopsinse, and are related to the forms in New Guinea and 

 the adjacent islands. 



The distribution of the subfamily Peripatoidinse is very interesting ; 

 this subfamily occurs in New Zealand, Tasmania, southern and 

 western Australia, South Africa, and southern South America. 



The forms occurring in Australia, New Zealand, and Tasmania 

 collectively make up a very closely knit faunal unit, indicating the 

 fundamental faunal homogeneity of these areas ; Tasmania, however, 

 lacks the less specialized component of the fauna of Australia and 

 New Zealand, a fact which may or may not be significant. It is most 

 probable that this type will eventually be found there. 



There are two possible explanations for this distribution ; ( i ) the 

 species of this subfamily may have been extirpated from all the more 

 desirable localities by more efficient and more aggressive species of 

 the other groups, or (2) the subfamily may have attained its present 

 distribution through following a more southern route. 



The first of these alternatives seems untenable, for if it were so 

 Ave should expect to find species of Peripatoidinse north as well as 

 south of the species of the other groups, and also occurring in isolated 

 situations, such as mountain tops, where the other species could not 

 penetrate. But nothing of the kind occurs. Moreover the species of 

 the Peripatoidinse are very highly specialized, so much so that if they 

 came into competition with the species of the other families they 

 probably would, other things being equal, prove themselves dominant. 



Therefore we must tentatively accept the second alternative, 

 namely, that the Peripatoidinse attained their present distribution 

 through originally having been widely spread over a southern land 

 which at one time or another included within its boundaries New 

 Zealand, Tasmania and southern South America, as well as South 



