272 JOURNAL OF SCIENCE. 
was always a series of islands, as it must have been in its earlier 
and its later stages, or whether it once was nearly continuous, is 
a matter of speculation. Of the 21 species of flowering plants of 
Kerguelen’s Land, three—or 14 per cent.—are found there only ; 
while eleven—or 50 per cent.—are confined to Kerguelen, the 
Crozets, Marion Island, and Heard Island. I should therefore 
judge, from what we know of the flora of New Zealand, that 
this group of islands separated from Fuegia in the miocene and 
that the islands themselves were not separated from each other 
until late in the pliocene. The distribution of the petrels also 
points to the ancient date of the present oceanic conditions of 
the southern hemisphere. It is the only group of birds which 
has originated in the south and spread to the north. The Alba- 
tross, Fulmar, and Shearwater of the north are all represen- 
tatives of southern species, while the south has several genera 
not represented at all in the north, ¢g¢., Ossifraga, Pterodroma, 
Daption, Prion, Pelacanoids. The only genus better developed 
in the north than in the south is that of the Shearwaters 
Puffinus), which is hardly ever seen out of sight ofland. All the 
truly oceanic petrels are of southern origin. + From this it seems 
probable that an antarctic continent south of Africa, and in- 
cluding Tristan d’Acunha and Kerguelen’s Land, may have 
existed from the eocene to the pliocene period, that it was sub- 
merged during or before the pleistocene, and that we now see rem- 
nants of it in Graham’s Land, Enderby Land, and Victoria Land. 
During the time this land existed it is possible that colder and 
warmer periods may have occurred when the eccentricity of the 
earth’s orbit was great; but this I consider a mere speculation 
unsupported by any evidence, for changes of climate are not 
required to account for any of the phenomena. 
It is evident from what has been said, that the north tem- 
perate plants came to New Zealand before the pleistocene 
period, and consequently they could not have migrated along 
the Andes during the glacial epoch of Europe. Either there 
have been many other glacial epochs, or else glacial epochs are 
not necessary for this meridional migration. I believe the latter 
to be the more correct view, because there is no evidence of 
glacial epochs in the southern hemisphere, and because the 
physical changes necessary to elevate or depress a chain of 
mountains for a few thousand feet are far less than those which 
are now acknowledged by nearly all geologists to be necessary 
for bringing about profound alterations in climate over immense 
districts of the globe. No one has as yet been sufficiently bold 
to advocate a glacial climate in New Guinea and Borneo, and 
yet the evidence of plant migration from Asia into Australia is 
as strong as that for a migration along the Andes; and, as it is 
very unlikely that an elevation of the Indian Archipelago co- 
incided in time with the glacial epoch of Europe, so it is very 
unlikely that glacial epochs are necessary for the meridional 
migration of plants. It follows that if plants have travelled from 
+ Hutton, [bis, 1865. 
