4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 8l 



the Old World (excej)! Australasia) and in the New World is south- 

 ern, having spread only as far north as the Gulf Coast of the United 

 States (Metcalf 1923a). The parasites of the southern frogs indicate 

 seemingly beyond question that the Australian and American southern 

 frogs are related and also that they arose in the Southern Hemisphere 

 and passed by some southern route from one to the other of their 

 southern habitats. It might be possible, however unlikely, that the 

 southern frogs of Australia evolved from very ancient ancestors ' in 

 a way parallel to that of the South American southern frogs, though 

 almost always in cases of parallel evolution there are found some 

 criteria to distinguish such resemblance from that due to genetic 

 relationship. But no one can for a moment believe that, along with 

 the parallel evolution of the American and Australian hosts, there 

 was also a parallel evolution of their opalinids, parallel to such a 

 degree that almost or cjuite identical species of parasites are found 

 in these frogs in South America and in Australia. The old hypoth- 

 esis of parallel evolution, put forward by Gadow (1909) and others 

 before the evidence from the parasites was known, could not now be 

 seriously entertained and Gadow himself gave up the hypothesis. 



If then the southern frogs of Australia are close relatives of those 

 of South America, how can we account for the present dispersal of 

 this family? Nearly all zoogeographers, because of evidence from the 

 Anura and from many other groups, both vertebrates and inverte- 

 brates, believe in a former land connection between South America 

 and Australia by way of an Antarctic continent, and a number of 

 the more prominent students of the subject have emphasized also, in 

 ih'is connection, the former existence of large connected areas of land 

 in the Pacific, especially the Southern Pacific, Ocean. Phytogeog- 

 raphers also have added much important evidence. The evidence 

 from plants is, however, less convincing, since many seeds and spores 

 may often be carried great distances by winds and by ocean currents. 



On the other hand a few students of dispersal have accepted 

 (Noble 1922, 1925; Dunn 1923) or been favorably inclined toward 

 (Schenck 1905, 1905a, 1907; Cheesman 1906, 1909) the hypothesis 

 of origin in a northern land mass, Arctogea, and a southward dispersal 

 via the Isthmus of Panama to South America, via Malaysia to Papua, 



^ Probably archaic toads, for archaic genera of Bufonidae are still found in 

 Australia as well as in South America. Noble (1922, 1925) calls these forms 

 leptodactylids. Archaic bufonids and archaic leptodactylids were probably 

 very similar. Herpetologists in general class these ancient genera in Austi-alia, 

 Africa and §.outh America as bufonids. From one of them in South America 

 the leptodactylids apparently arose. 



