BIPOLARITY 65 



It would seem that hypotheses (1) and (2), which are basically the same, adequately explain this 

 apparent phenomenon in general, and that (3) is scarcely justified, since present anomalies in distribu- 

 tion probably result only from incomplete knowledge of past conditions. For instance, did the periods 

 of glaciation in the Pleistocene bring the cold-water faunas of both hemispheres sufficiently close 

 together to allow of a more efficient interchange of species than at present? 



In Figs. A and B I have plotted the known ranges of the molluscan genera Aforia, Fusitriton and 

 Acanthina. These genera occur both in northern and southern high latitudes ; they are not cosmopolitan, 

 yet they have preserved a connected range that is obviously resultant from the continuity of the western 

 coastline of the Americas plus their southern extension as the Scotia Arc. Aforia is stenothermic and 

 has achieved continuity by going deep beneath the surface warm zones, but Acanthina is eurythermic 

 and maintains a connected distribution over the shallow warm zones. It is best developed in the central 

 warm-water portion of its range, but has developed a cold-water tolerance also. Fusitriton is apparently 



140 160 



Fig. B. Distribution of Acanthina (shallow water) and Fusitriton and Aforia (continental shelf and deeper). 



stenothermic, with a considerable depth range, to as much as 1800 fathoms in one instance, but it is 

 restricted for the most part to the continental shelves. Its almost complete continuity along the western 

 coastline of the Americas is undoubtedly due in some measure to upwelling of cold currents such as 

 the Peru coastal current. 



The littoral southern molluscs of the genera Kerguelenella and Pachysiphonaria have their northern 

 counterpart in Liriola, but connecting forms are absent from warm seas, since these stenothermic 

 animals apparently have not adapted themselves to a benthic habitat. The question is, under what 

 former conditions did they achieve a connected range? 



Stephen (1941, p. 243), in his report on the Echiuridae, Siphimculidae and Priapulidae, lists twenty- 

 three Antarctic and Subantarctic species, of which ten are identical with and two close to Arctic 

 species. A remark, applicable also to the molluscs, is that ' most of the bipolar species are confined to 

 the South American Quadrant '. 



The bulk of the Antarctic and Subantarctic molluscs are derivatives of a few cosmopolitan families : 

 notably the Patellidae, Fissurellidae, Trochidae, Littorinidae, Naticidae, Muricidae, Neptuniidae, Volutidae 



and Turridae. 



3-2 



