172 FROGS OF AUSTRALIA. [CH. Ill 



remarkable that oceanic islands which happen to be within 

 the districts visited by icebergs have not also been stocked 

 with fish. This criticism it will be seen cuts both ways. 

 We cannot apply it to continents, if we are forbidden to 

 consider it in connection with oceanic islands. It has 

 already been pointed out in connection with the same 

 subject of oceanic islands that they are invariably desti- 

 tute of frogs or of amphibians of any kind. These crea- 

 tures are quite powerless to migrate across the sea. Now 

 a family of frogs, the Gystignathidce, are common to South 

 Australia and to South America. I do not think it 

 prudent to lay any particular stress upon the likeness 

 in the Coleopterous fauna of the antarctic parts of the 

 world; for beetles or their larvae can with comparative 

 ease be carried about by that convenient deus ex machina, 

 the floating log; more remarkable is the negative fact 

 that the land shells do not support the idea of a former 

 extended antarctic continent. Precisely the same facilities 

 are enjoyed by the terrestrial mollusca for dispersal as by 

 insects. It is however difficult to extract anything of 

 value out of this fact either in favour of or against the 

 view that the antarctic current had formerly a greater 

 extent to the northward. It is perhaps to be correlated 

 with the fact that there is only a single land shell in Ker- 

 guelen, which would seem to be some indication that in 

 this quarter of the globe some hindrance exists to the 

 migration of these animals. The matter must remain 

 unexplained for the present. 



The distribution of the Isopod genus Serolis may be 

 cited as further evidence in the same direction ; by some 



