4 JOURNAL OF SCIENCE. 
Eastern Australia and New Zealand, many of which belong to 
tropical and sub-tropical genera. Mr. Wallace’s explanation for 
the presence of these forms is that a migration took place 
through New Zealand, South Victoria Land, South Shetland 
Islands, and Tierra del Fuego over a greater extension of 
southern lands during a warm miocene period. Now Dr. Martin 
Duncan is certainly of opinion: that the sea in this portion of 
the southern hemisphere was much warmer in the miocene 
period than at present, and he has suggested that this was due 
to an extension of the Antarctic Continent up to 50° S.,* but, on 
the other hand, Mr. Darwin considered the eocene sea of Chili 
to have been no warmer than at present, and Mr. Tenison- 
Woods says that—“ The whole evidence of the [tertiary] fossil 
corals shows a climate and isolation in the New Zealand fauna 
not very different from the conditions which exist now,” and that 
the tertiary fauna of New Zealand generally “is not that of 
a warm sea, nor like what we should find on the warmer extra- 
tropical portions of the Australian coast.” The miocene mol- 
lusca appear to me to indicate a rather warmer sea, but as 
several of the species still live as far south as Foveaux Straits,{ 
no elevation of temperature sufficient to take tropical and sub- 
tropical plants and animals to 50° S. is probable ; and in addi- 
tion to other difficulties presently to be mentioned, I shall, I 
think, be able to show that the South American connection is of 
afar older date than the miocene. Before doing so, however, 
it will be necessary to give a short review of the fauna of the 
Australian region. 
In Mr. Wallace’s opinion the deep oceans, z.¢. the Pacific, 
Atlantic, and Indian Oceans, have been in existence from the 
earliest geological times. All the principal groups of land 
animals, he thinks, have originated in the northern hemisphere, 
and have gradually migrated southwards through the continen- 
tal extensions of America, Africa, and Australia, (including the’ 
Indian Archipelago), comparatively few having subsequently 
spread east and west by means of antarctic islands now sub- 
merged. If this be true, it is evident that the fauna of Australia 
ought to be more nearly allied to that of South Africa than to 
that of South America, because the connection of the former 
with India is so much closer than the connection with the latter 
by Kamschatka and Alaska. Let us see if this is so. 
The Australian mammalia are very peculiar, and are more 
closely allied to the jurassic mammals of Europe and America 
than to any now living. The marsupials of America are related 
to the eocene marsupials of Europe and are, evidently, a younger 
branch of the family from which the Australian mammals had 
been i oar long previously. Consequently the relationship 
* Quar, Jour. Geol. Soc., 1876, p. 345. 
t Palzontology of New Zealand, Part iv. p. 4 (1880) 
t Such as Voluta pacifica, Triton spengleri, Parmophorus unguis, Chione 
stutchburyi, Tapes intermedia, Pectunculus laticostatus, Waldheimia lenticularis, 
and others. 
