Although, apparently in vindication of some conclusions of his which had been criticised by Dr. Hartlaub, Wallace subsequently denied the necessity of postulating the existence of such a continent, his general recognition of the facts of subsidences and upheavals of great portions of the earth's surface, as well as the inferences which he draws from the acknowledged relations of living and extinct faunas as above stated, remain of course unaltered.
The following extracts from Mr. H. F. Blandford's most interesting paper read before a meeting of the Geological Society deals with the subject in still greater detail:—[8]
"The affinities between the fossils of both animals and plants of the Beaufort group of Africa and those of the Indian Panchets and Kathmis are such as to suggest the former existence of a land connexion between the two areas. But the resemblance of the African and Indian fossil faunas does not cease with Permian and Triassic times. The plant beds of the Uitenhage group have furnished eleven forms of plants, two of which Mr. Tate has identified with Indian Rájmahál plants. The Indian Jurassic fossils have yet to be described (with a few exceptions), but it has been stated that Dr. Stoliezka was much struck with the affinities of certain of the Cutch fossils to African forms; and Dr. Stoliezka and Mr. Griesbach have shown that of the Cretaceous fossils of the Umtafuni river in Natal, the majority (22 out of 35 described forms) are identical with species from Southern India. Now the plant-bearing series of India and the Karoo and part of the Uitenhage formation of Africa are in all probability of fresh-water origin, both indicating the existence of a large land area around, from the waste of which these deposits are derived. Was this land continuous between the two regions? And is there anything in the present physical geography of the Indian Ocean which would suggest its probable position? Further, what was the connexion between this land and Australia which we must equally assume to have existed in Permian times? And, lastly, are there any peculiarities in the existing fauna and flora of India, Africa and the intervening islands which would lend support to the idea of a former connexion more direct than that which now exists between Africa and South India and the Malay peninsula? The speculation here put forward is no new one. It has long been a subject of thought in the minds of some Indian and European naturalists, among the former of whom I may mention my brother [Mr. Blandford] and Dr. Stoliezka, their speculations being grounded on the relationship and partial identity of the faunas and floras of past times, not less than on that existing community of forms which has led Mr. Andrew Murray, Mr. Searles, V. Wood, jun., and Professor Huxley to infer the existence of a Miocene continent occupying a part of the Indian Ocean. Indeed, all that I can pretend to aim at in this paper is to endeavour to give some additional definition and extension to the conception of its geological aspect.
"With regard to the geographical evidence, a glance at the map will show that from the neighbourhood of the West Coast of India to that of the Seychelles, Madagascar, and the Mauritius, extends a line of coral atolls and banks, including Adas bank, the Laccadives, Maldives, the Chagos group and the Saya de Mulha, all indicating the existence of a submerged mountain range or ranges. The Seychelles, too, are mentioned by Mr. Darwin as rising from an extensive and tolerably level bank having a depth of between 30 and 40 fathoms; so that, although now partly encircled by fringing reefs, they may be regarded as a virtual extension of the same submerged axis. Further west the Cosmoledo and Comoro Islands consist of atolls and islands surrounded by barrier reefs; and these bring us pretty close to the present shores of Africa and Madagascar. It seems at least probable that in this chain of atolls, banks, and barrier reefs we have indicated the position of an ancient mountain chain, which possibly formed the back-bone of a tract of later Palæozoic, Mesozoic, and early Tertiary land, being related to it much as the Alpine and Himálayan system is to the Europæo-Asiatic continent, and the Rocky Mountains and Andes to the two Americas. As it is desirable to designate this Mesozoic land by a name, I would propose that of Indo-Oceana. [The name given to it by Mr. Sclater, viz., Lemuria, is, however, the one which has been most generally adopted.] Professor Huxley has suggested on palæontological grounds that a land connexion existed in this region (or rather between Abyssinia and India) during the Miocene epoch. From what has been said above it will be seen that I infer its existence from a far earlier date.[9] With regard to its depression, the only present evidence relates to its northern extremity, and shows that it was in this region, later than the great trap-flows of the Dakhan. These enormous sheets of volcanic rock are remarkably horizontal to the east of the Gháts and the Sakyádri range, but to the west of this they begin to dip seawards, so that the island of Bombay is composed of the higher parts of the formation. This indicates only that the depression to the westward has taken place in Tertiary times; and to that extent Professor Huxley's inference, that it was after the Miocene period, is quite consistent with the geological evidence."
After proceeding at some length to instance the close relationship of many of the fauna in the lands under consideration (Lion, Hyæna, Jackal, Leopard, Antelope, Gazelle, Sand-grouse, Indian Bustard, many Land Molusca, and notably the Lemur and the Scaly Anteater) the writer proceeds as follows:—
"Palæontology, physical geography and geology, equally with the ascertained distribution of living animals and plants, offer thus their concurrent testimony to the former close connexion of Africa and India, including the tropical islands of the Indian Ocean. This Indo-Oceanic land appears to have existed from at least early Permian times, probably (as Professor Huxley has pointed out) up to the close of the Miocene epoch;[10] and South Africa and Peninsular India are the existing remnants of that ancient land. It may not have been absolutely continuous during the whole of this long period. Indeed, the Cretaceous rocks of Southern India and Southern Africa, and the marine Jurassic beds of the same regions, prove that some portions of it were, for longer or shorter periods, invaded by the sea; but any break of continuity was probably not prolonged; for Mr. Wallace's investigations in the Eastern Archipelago have shown how narrow a sea may offer an insuperable barrier to the migration of land animals. In Palæozoic times this land must have been connected with Australia, and in Tertiary times with Malayana, since the Malayan forms with African alliances are in several cases distinct from those of India. We know as yet too little of the geology of the eastern peninsula to say from what epoch dates its connexion with Indo-Oceanic land. Mr. Theobald has ascertained the existence of Triassic, Cretaceous, and Nummulitic rocks in the Arabian coast range; and Carboniferous limestone is known to occur from Moulmein southward, while the range east of the Irrawadi is formed of younger Tertiary rocks. From this it would appear that a considerable part of the Malay peninsula must have been occupied by the sea during the greater part of the Mesozoic and Eocene periods. Plant-bearing rocks of Rániganj age have been identified as forming the outer spurs of the Sikkim Himálaya; the ancient land must therefore have extended some distance to the north of the present Gangetic delta. Coal both of Cretaceous and Tertiary age occurs in the Khasi hills, and also in Upper Assam, but in both cases associated with marine beds; so that it would appear that in this region the boundaries of land and sea oscillated somewhat during Cretaceous and Eocene times. To the north-west of India the existence of great formations of Cretaceous and Nummulitic age, stretching far through Baluchistán and Persia, and entering into the structure of the north-west Himálaya, prove that in the later Mesozoic and Eocene ages India had no direct communication with western Asia; while the Jurassic rocks of Cutch, the Salt range, and the northern Himálaya, show that in the preceding period the sea covered a large part of the present Indus basin; and the Triassic, Carboniferous, and still more recent marine formations of the Himálaya, indicate that from very early times till the upheaval of that great chain, much of its present site was for ages covered by the sea.
"To sum up the views advanced in this paper.
"1st. The plant-bearing series of India ranges from early Permian to the latest Jurassic times, indicating (except in a few cases and locally) the uninterrupted continuity of land and fresh water conditions. These may have prevailed from much earlier times.
"2nd. In the early Permian, as in the Postpliocene age, a cold climate prevailed down to low latitudes, and I am inclined to believe in both hemispheres simultaneously. With the decrease of cold the flora and reptilian fauna of Permian times were diffused to Africa, India, and possibly Australia; or the flora may have existed in Australia somewhat earlier, and have been diffused thence.