24 DISCOVERY REPORTS 



Conversely, with few exceptions, forms described from those areas have not extended 

 their range into the Weddell Sea. In short, that, apart from cosmopolitan species, the 

 Weddell Sea fauna has few points in common with the fauna found to the west of a line 

 running from Graham Land through the South Orkneys to the South Sandwich Islands. 



What is the reason? There are physical differences in temperature and salinity be- 

 tween the waters of the Weddell and Scotia Seas, but so far as I understand them they 

 are not sufficient to account for the differences in fauna, the Foraminifera on the whole 

 being very tolerant of such slight changes. A more feasible explanation is that the two 

 areas were separated by a land barrier until comparatively recent geological times, and 

 this theory would be supported by the Discovery soundings. These appear to indicate 

 clearly that there was once such a barrier — the Scotia Arc — running in a curve from 

 Tierra del Fuego to the northern point of Graham Land, via the Burdwood Bank, 

 South Georgia, the Gierke Rocks and the South Sandwich, South Orkney and South 

 Shetland groups of islands. Since the disappearance of this barrier (? in Tertiary times), 

 the Weddell Sea current sweeping up the east coast of Graham Land and outwards into 

 the South Atlantic has apparently sufficed to prevent the incursion of Pacific forms into 

 the Weddell Sea. Evidence in support of this may perhaps be found in the absence of 

 Miliammina from the Weddell Sea and also from the Gauss area. The genus, which is 

 common right across the Ross-Bellingshausen-Scotia Seas, has extended its range thence, 

 in the line of the West Wind Drift, across the South Atlantic to Tristan d'Acunha and 

 the Cape of Good Hope. There are other forms which are found in the same line of 

 migration, but which have not penetrated into the Weddell Sea. 



Eastwards from the Weddell Sea we have no useful records of the foraminiferal 

 fauna across the gap C between io° W and 72° E, where the Gauss records begin. The 

 fauna described by Wiesner is considerably richer, in its 268 species and varieties, than 

 that of the Weddell Sea, to which however it presents more points of resemblance than 

 it does to the fauna of the Scotia-Bellingshausen-Ross Seas. The absence of Miliam- 

 mina, already alluded to, is a marked point of distinction and others could be quoted. 

 On the other hand, the presence oi Keramosphaera in the Weddell Sea and in the Gauss 

 material certainly indicates a common origin for the Weddell Sea and Kaiser Wilhelm's 

 Land faunas, the only other record for the genus being the original Challenger dis- 

 covery intermediate between the two. But we must await information as to the fauna 

 of the gap C, which represents nearly one-quarter of the Antarctic circumference before 

 the identity of the Weddell Sea and Kaiser Wilhelm's Land faunas can be considered 

 to be proved. 



I think that I have already supplied sufficient evidence to establish the identity of the 

 foraminiferal fauna over the circumpolar area between 10° W and 158° E going west- 

 ward, and the principal interest will now lie in determining how much farther to the 

 westward this fauna continues. Somewhere between 158° E, the most westerly point of 

 the Terra Nova material, and 95° E, the most easterly point of the ' Gauss ' — ^or perhaps 

 rather 90° E the most easterly point of the ' Gauss ' on the Antarctic continental shelf — 

 I think we shall find evidence of a barrier separating the Kaiser Wilhelm's Land fauna 



