58 AUSTRALASIAN ANTARCTIC EXPEDITION. 
Excluding “‘ Magellania” kerguelensis of the Marion Island group, which has 
in all probability been recently derived from the Kerguelen group, and the doubtfully 
localised Terebratella rubiginosa of the Cape, the Marion Island and South African 
faunas have not a typically southern fauna of the circum-Pacific type since they do 
not contain Crania, Hemithyris, or the higher members of the Magellaninw, while the 
species of Terebratulina and “ Terebratula” may be of southern or of northern facies. 
The Gondwana-land element exists in four species of Avaussina at the Cape, and 
connections with the Mediterranean are shown by the presence of Platidia anomiodes 
in Marion Island and of Agulhasia in South Africa. It seems safe to conclude not 
only that these two districts have been isolated from the rest of the southern lands 
since the Miocene, but that they did not share in the still earlier means of communica- 
tion which permitted the southern Pacific Oligocene-Miocene fauna to attain the 
Kerguelen area. 
The distribution of southern Recent Brachiopods, then is satisfactorily explained 
by an ancestral distribution in the Miocene, and not only does it not call for any land 
bridges or shallow submarine connections between the various southern continents 
and islands since that date, but is distinctly opposed to any such means of intercom- 
munication except between South America and the Antarctic. 
The generic similarities between the four southern Oligocene-Miocene faunas, 
on the other hand, are of such a nature as to demand at some earlier date much 
greater means of intercommunication between the lands bordering the South Pacific 
Ocean than exist at the present day. The fact that the Gondwana-land element 
occurs both in New Zealand and Australia, but not in the Antarctic and South America 
suggests that the intercommunication between New Zealand and Australia did not 
occur at the same time as that between New Zealand and the Antarctic, but that the 
latter was the earlier, or we may suppose a land barrier from New Zealand to New 
California separating a sea south of Australia from the Pacific Ocean and not breaking 
down till the Oligocene. A Tertiary fauna from New Caledonia would help materially 
in picturing the former connections and is likely to be found. Unfortunately little 
is known of the ancestral forms of the peculiarly southern Oligocene-Miocene genera, 
so that it is impossible in this way to form an estimate of the dates of the necessary 
intercommunications. So far as the genera are concerned they might have occurred 
as far back as the Cretaceous. 
Certain specific similarities between the Oligocene-Miocene faunas, however, 
demand a considerably later date for the intercommunications. The following species, 
or perhaps more correctly nearly related forms, are common to two or more of the 
four districts :— 
Hemithyris squamosa (Hutton).—New Zealand, Australia, and Antarctic. 
Murravia catinuliformis (Tate).—Australia and New Zealand. 
Terebratulina suessi (Hutton).—New Zealand and Australia, 
