An interval then occurs, and the next formation probably belongs to the jurassic period. In this we find Belemnites aucklandicus, which can hardly be distinguished from B. canaliculatus, and Astarte wollumbillaensis. The ferns, too, found so plentifully near Port Waikato, in the Clent hills, at the Mataura, and at Waikawa harbour, are considered by Professor McCoy to be identical with Australian ferns from the same formation. At the close of this period movements on an extensive scale commenced in New Zealand, the land was upheaved, and an extensive mountain chain formed. Á long blank now occurs in our geological record (see Geo. Reports, 1872, p. 105), the next formation belonging to quite the uppermost part of the secondary epoch, later I believe than the white chalk of England. In it we find remains of dicotyledonous plants and large Saurians belonging to the genera Crocodilus? and Plesiosaurus. Here also we find three fossil shells (Dentalium majus, Lucina americana and Cucullœa alta), similar to those found in South America, one of which, Lucina americana, is found in the lower cretaceous rocks of Tierra del Fuego, and the other two in the miocene formations of Patagonia and Chile; thus showing that during this blank in our geological record an intimate connection had existed between New Zealand and South America. The disposition, however, of these beds shows that the New Zealand Alps were not submerged. A long interval now follows, during which New Zealand was again upheaved, and the next rocks that we find are of upper eocene date (Geo. Rep., 1872, p. 182). From that time until the close of the miocene period New Zealand was greatly depressed, and divided into several islands, but at the close of the miocene period it was once more upheaved. During this period we find several South American miocene shells not met with in the older formation, as well as several Australian ones. During the newer pliocene period it again subsided, and the Wanganui beds were deposited. From that time I can see no evidence of the land having ever stood at a higher level than it does at present, but as the later changes in the physical geography of New Zealand have a most important bearing on the present condition of its fauna, beyond the scope of my present inquiry, I propose treating the subject in a separate paper. ^* The geological evidence is, therefore, that since the jurassic period there have been three principal upheavals in New Zealand, in the lower cretaceous, lower eocene, and older pliocene periods respectively, and that these, were divided by two insular periods, viz., during the upper secondary (Danien), and from the commencement of the upper eocene to the close of the miocene, thus agreeing completely with the zoological evidence.

The dates assigned by the geological evidence also agree well with those derived from zoology. We have seen that it is necessary to suppose that the first great antarctic continental period was anterior to the date of the spread

of the Mammals southwards. Now a few Marsupials are known in the triassic period, but it is quite possible either that they spread very slowly, or that barriers existed that prevented any southward migration. In the eocene period, however, some placental Mammals were in existence, although Marsupials, not of Australian types however, still formed in Europe the principal mammalian life; and if the supposed barriers to a southward migration were still in existence, we know, from what happened in the Northern Hemisphere, that the whole, or nearly the whole, of the Marsupials would have been exterminated. The Marsupials, therefore, must have migrated south not later than the eocene period, and as we know that our connection with Australia and South America must have been before that migration, it follows that the first, or lower cretaceous period of upheaval, must have been the time of the antarctic continent. This is rendered still more probable by the fact that our jurassic fossils show a connection with Australia only, while our upper secondary fossils show for the first time a relation to South America. The fact, too, of the cretaceo-oolitic rocks of Tierra del Fuego having been largely disturbed, metamorphosed, and broken through by dykes of green-stone, shows that extensive elevatory movements have taken place there, also, since they were deposited. It is therefore to the lower cretaceous period that we must probably look for the time of the dispersion of the Struthious birds. With regard to the date of the second, or Polynesian continental period, the only zoological evidence we have is that it probably preceded the wide dispersion of the Hemiptera, and the butterfly section of the Lepidoptera. This, therefore, could not have been later than the eocene, for a fossil butterfly (Vanessa pluto) has been found in the lower miocene deposits of Radaboj in Croatia, and fossil Heteroptera in the miocene beds of Œningen in Switzerland. The elevation during the lower eocene period was therefore probably the one which formed the continent that I have described as including New Caledonia and some of the Pacific Islands. At this period probably Northern Australia was submerged, and the southern portions of Australia and Tasmania formed one large island, while New Guinea, including the Solomon Islands and New Hebrides on the south, and the Molucca Islands on the north, formed another large island, divided from the New Zealand island, or continent, by the straits between New Caledonia and the New Hebrides.

This was the time of the migration from China southwards, and it is worthy of notice that at the same time a large ocean existed from southern Europe to China, in which the nummulitie limestone was being deposited. Would it be too bold to speculate that it was along the shores of this ocean that those fish, crustaceans, and shells migrated, which are now found in the North Atlantic or Mediterranean on the one hand, and in China or Japan on the other, but not on the southern shores of Asia; and that the anomalous

distribution of European forms of fish, shells, etc., in New Zealand may be traced to the same route? This same period of sea communication between Europe and Japan will also probably have been the time of the land connection that once existed between India, Madagascar, and Africa (the Lemuria of Dr. Sclater), as proved by the recent fresh-water fish, and birds, as well as by the miocene Mammalia, ^* and to this period we may also refer the origin of the curious affinity between some of the birds of Celebes and Africa. The long insular period during the upper eocene and miocene times will, therefore, be the period of specific change in the moas, while the older pliocene upheaval will be the time of the mingling of the various species in New Zealand, and the peopling of the Chatham and Auckland Islands. The newer pliocene was the time when the two islands of New Zealand were divided, and also the period when the Chatham and Auckland Islands were separated from them, but the latter occurrence probably preceded the former by a long interval.

Such appears to me to be the hypothesis most capable of accounting for the present fauna of New Zealand.

The objection, however, may be fairly raised that, if it is true, evidence of its truth ought to be also found in the flora of the country, which is not the case. I fully acknowledge the force of this argument, but think that some slight evidence can be found in the phænogamic flora. The distribution of Eucalyptus for instance, is somewhat parallel to that of the Marsupials, and can be only explained in the same way. Stilbocarpa polaris has its nearest allies in China and the Himalaya Mountains; while the distribution of Metrosideros, Ligusticum, Angelica, and perhaps Veronica, implies a connection between New Zealand and Asia not by way of Australia. This connection is obscured by the great preponderance of Australian and South American forms, but still furnishes an indistinct copy of the bolder outline sketched out by the fauna. This is owing to the wider distribution of genera among plants than among animals, and to me it appears to prove that the flora of a country, as a whole, is of a more ancient date than its fauna. Among the cryptogamic plants no trace of this outline can be discerned, as also is the case with the lower classes of the animal kingdom, owing to the genera having been, so to say, universally spread before the last migration from Asia took place.

That the facies of a fauna and flora should date back from so long a period as I suppose, is certainly at variance with ordinarily received opinion, but from a study of the fauna and geology of New Zealand I do not see how we can escape from the conclusions that I have arrived at. I am well aware,