tion and its probable nature and extent that I wish to discuss.

Dr. Wallace already vaguely indicated that there may have been several very distant periods of union in the past of the Southern with the Northern continent. This supposition received a startling confirmation by the discovery first by Dr. Wortmann and then by Professor Osborn15 of true armadillo remains in the middle Eocene beds of Wyoming. If he does not actually speak of a direct land connection between the two continents in early Tertiary times, Professor Osborn16 suggests as much in his remark that this discovery "adds another fact to the growing evidence that North and South America were related in the Mid-Cretaceous and perhaps early Tertiary and then separated again until the Pliocene." He does not specify in any way in what manner this relationship had been brought about. His views would be of particular interest, considering that Dr. von Ihering's extensive zoological and botanical researches have led him to believe that the South American continent itself must be of comparatively recent geological origin.

66

The latter declares that South America had arisen as a continent only since the Oligocene period. It then consisted of two parts united by a narrow strip of land in the west, which later on developed into the great mountain chain of the Andes. These two parts, which he calls 'Archiplata" and "Archiguiana," were previously separated from one another. The first embraced Chile, Argentina, Uruguay and southern Brazil, the other the highland of Venezuela and Guiana. Each of these possessed, according to Dr. von Ihering," its own characteristic fauna and flora and these were totally distinct from one another.

A somewhat similar theory as to the origin of South America, largely based on the geographical distribution of fresh-water decapods, has been advanced by Dr. Ort'Osborn, H. F., "An Armadillo from the Middle Eocene, 16 Osborn, H. F., "Mammalian Palæontology," p. 99. "Thering, H. von, "Archhelenis and Archinotis,'' p. 79.

15

pp. 163-165.

mann.18 In place of the present Southern continent he thinks that toward the end of Mesozoic times, there existed the old Brazilian land (Archiplata), an Antillean continent (including the West Indies and Venezuela) and also the Chilean coast range. These three land masses were separated from one another by wide oceans. Just before the close of the Secondary era the Antillean continent, and with it Venezuela and even the Galapagos Islands, became united with western North America, the latter being then still detached from eastern North America. When Venezuela in early Tertiary times at last became fused with the other larger South American land masses, the interoceanic connection across Central America had severed it from North America.

Under such geographical conditions the Wyoming edentates alluded to by Professor Osborn could only have been derived from Venezuela, it being the sole portion of the present Southern continent that had any relationship with North America in those remote times. No fact, however, has been brought to light, either in the recent or fossil history of the edentates to lead us to imagine that they had originated in the northern part of South America.

That South America owes its origin to the union of several independent land masses is so clearly indicated by the existing fauna of the continent, that a similar evidence should also be revealed by a study of its rocks and fossils.

The geology of South America is unfortunately as yet little known. Yet even such a cautious observer as Professor Suess19 ventured to suggest, on stratigraphical grounds, that an arm of the sea may have penetrated right across the continent in Cretaceous times. The archaic rocks of eastern Brazil and Guiana certainly were then raised above the sea, since the younger formations appear to be superimposed with great regularity further and further to the west of this ancient formation. Dr. 18 Ortmann, A. E., "Geographical Distribution of Decapods,” pp. 365– 366.

Katzer20 contends that during part of the Mesozoic era the Pacific ocean extended eastward to the shores of this land, whose rivers then drained westward into the ocean, as they even continued to do until Miocene times.

While it is therefore by no means evident from the geologist's point of view how and when the various land masses became joined to form the present South America, the geographical distribution of the living fauna, together with a study of the paleontology, has furnished most valuable hints as to the probable geological history of the continent.

Dr. Gill claims that the fishes are among the best indicators of former geographical conditions. Turning to the most recent studies on the South American fish fauna, those of Professor Eigenmann,21 we find that he also is impressed by the dissimilar elements of which it is composed. He explains this varied character of the fauna by the supposition that two independent land masses, originally separated in the region of the Amazon valley, became welded together in early Tertiary times.

In his attempted restoration of the geographical conditions of South America during the Eocene period, Professor de Lapparent22 depicts an aspect contrasting with that of other observers, and yet he recognizes a division of South America into two parts, for he represents the continent as being dissevered by a marine channel between the Rio Negro in Argentina and Southern Peru.

Of all the maps illustrating ancient distribution of land and water, that of Dr. Arldt23 is the most striking in originality. He connects northern South America in late Cretaceous times by land with western Mexico, but not by way of Central America. He assumes that the latter was submerged at that time and that an independent land bridge extended from southwestern North America through the Galapagos Islands to Colombia. This north

Katzer, F., "Geologie des Amazonengebietes,'' p. 254.

21 Eigenmann, C. H., "Fishes of South and Middle America,' Lapparent, A. de, 'Traité de Géologie,'' p. 1455.

p. 528.

23

Arldt, Th., “Entwicklung der Kontinente" (map 19).

ern complex of land was isolated from the southern part of South America by a wide sea channel stretching right across the continent.

His conception of an extensive land having once flourished to the west of Central America, while the latter was largely submerged, is not altogether new. In alluding to the east-westward trend of the Antillean Cordillera and its abrupt termination on the Pacific coast of Guatemala, Professor Suess24 makes a suggestion as to its former westward prolongation. Precisely at the point, he says, where the arcuate continuation of this chain might be expected to meet the principal chains of South America, lie the volcanic Galapagos Islands.

Various indications in the structure of the Isthmus of Panama moreover left the impression on Dr. Hill's25 mind that large areas now covered by the Pacific, to the west of the isthmus, were once replaced by an extensive land surface.

Nothing more, however, can be deduced from geological testimony as to the presence of any land connecting North and South America at a time when Central America may have been still wholly or partially submerged. Nor can we even surmise from these suggestions at what geological period such hypothetical land may have existed. Other methods will have to be employed in order to discover the manner by which the Eocene armadillos reached North America.

If we examine the whole eastern Pacific coast line from Alaska to Cape Horn, we notice that there are two areas that have apparently remained entirely unsubmerged since Jurassic times to the present day. One of these occupies part of western Mexico and Lower California, the other a strip of the southern coast line of Chile. It is the latter coast cordillera which Dr. Burckhardt26 believed to be the remaining remnant of a mighty Pacific continent, because porphyritic conglomerates of Cretaceous age are heaped

2 Suess, E., "Antlitz der Erde," Vol. II, p. 263.

* Hill, R. T., "Geology of the Isthmus of Panama,'' p. 217. Burckhardt, C., "Traces géologiques d'un continent," pp. 12–14.

against its eastern flank, whereas still further east the latter grade into fine-grained rocks, thus indicating that the land from which they were derived lay westward, out in the ocean.

While we need not here dwell upon the theory of a former Pacific continent27 so ably supported by Professor Haug on purely geological grounds and by Professor Hutton on zoogeographical data, I should like to draw attention to some features in the geographical distribution of animals and plants which prove that southwestern North America and southwestern South America are intimately related to one another in their fauna and flora. This relationship can not be explained as the product of a similarity in soil and climatic conditions. It is not a case of mere convergence. It can be shown that it already existed in the distant past, and I venture to think that this relationship implies the presence of a former direct land connection between these two ancient areas, when the continent of South America was still in the making.

It was Dr. Wallace2s who first directed attention to the remarkable fact that many genera of insects from the north temperate regions reappeared in temperate South America, being generally absent in the intermediate stations. He explained this phenomenon by the supposition that the northern forms had traveled southward during successive glacial epochs when the mountain range of the Isthmus of Panama might have become adapted for their advance in that direction. Their southward passage was believed to have been facilitated by storms and hurricanes which carried the insects across unsuitable territories.

This interpretation of a striking feature of geographical distribution seems to have been considered satisfactory at the time. At any rate no one has raised any protest so far as I am aware. Yet I am not at all disposed to admit its correctness.

"More detailed information on the theories relating to an ancient Pacific continent will be published in my work on the geological history of the American fauna.

2 Wallace, A. R., "Distribution of Animals," Vol. II, pp. 45-47.