[[p. 431]] The following communication is due to the elaborate papers of Professor F. W. Hutton "On the Origin of the Fauna and Flora of New Zealand,"¹ which were made accessible to me by the great kindness of the author. On the whole, the views contained in those papers agree with my own ideas of the former connection between the areas under consideration, which are in opposition to Mr. Wallace, from the views of whom I, with Professor Hutton, differ in very essential points.

    In my opinion, an important defect in Mr. Wallace's studies is the fact that he makes too little distinction between the different groups of the animal kingdom. Birds and mammals, whose living genera appear only in the Tertiary era, must evidently show a different geographical distribution from the Teleostei, reptiles, &c., which are represented in the Cretaceous and beginning of the Tertiary era; or to the land and fresh-water molluscs, many of which were living already during the Secondary or even the Palæozoic era. Mr. A. R. Wallace ("Darwinism," 2nd ed., London, 1889) still upholds the doctrine of the "permanence of oceanic and continental areas." I am as much convinced of the erroneousness of this doctrine, which quite arbitrarily takes the 1,000-fathom line as corresponding essentially to the limit of the ancient continents, as I am that the ideas of Darwin and Wallace on "natural selection" as the cause of the origin of species will have but a [[p. 432]] historical interest in the coming centuries. Mr. Wallace's supposition that land and fresh-water molluscs, lizards, &c., were distributed over the whole Pacific Ocean by the waves of the sea I think to be more than bold. Under such circumstances, it is but natural that one looks for a less constrained explanation. It is a fact, curiously enough hardly noticed till now, that those land and fresh-water molluscs which appear only in the Tertiary era show a very limited distribution, or else are restricted to a few regions only; whilst genera of the Palæozoic era are cosmopolitan. Thus, Ampullaria and Anodonta, appearing only towards the end of the Cretaceous period, are found neither in Chile, nor in Australia, New Zealand, or Polynesia. But the genus Unio, which we know to occur already in the Jura formation, is not wanting in the countries mentioned, and may perhaps also be found on the Fiji Islands. In Chile and New Zealand Unio is the only representative of the family Naiadæ, likewise in south-east and west Australia; but in north Australia a representative of the genus Mycetopus is found, more or less allied forms of which occur also in South America, Africa, and southern Asia, and which evidently reached Australia from the latter country.

    With still wider distribution than Unio, there are in Polynesia representatives of Limnæa, Physa, Planorbis, Ancylus, Amphipeplea, Pupa, Zonites, Succinea, almost all of these genera being known from the Carboniferous. Too little attention has hitherto been paid to this fact. To give the distribution of the Nephropneustæ (Pulmonata stylommatophora, aut.) is not possible at present, because the anatomy of the genera is not sufficiently known. A few weeks ago it happened to me that I recognised, by aid of anatomical examination, one of our Hyalinæ to belong really to the genus Microcystis, a genus which, according to its anatomy--absence of the receptaculum seminis--may be considered as one of the oldest. Yet the oldest known Nephropneust is a Zonitid from the Devonian, perhaps actually a Microcystis, which is, at any rate, a nearly-allied genus. Perhaps Patula, Streptaxis, Stenogyra, &c., belong also to this group of very ancient and cosmopolitan land-shells, whose distribution in space and time deserves a close study. This certainly can only be done after careful anatomical researches, and our knowledge may therefore progress but slowly. Further, many of these animals may have been introduced. All this is less to be feared with fresh-water shells, whose introduction may be intentional only as far as it is done with the spawn of fishes, &c.; nevertheless, to a smaller extent. On the other hand, mistakes may be more likely expected in their determination.

    [[p. 433]] All these facts induced me to take up the special study of the Naiadæ and the fresh-water fauna in general, as it proves to be the most reliable guide to the knowledge of the geography of our globe during Palæozoic and Mesozoic times. I am convinced that the importance of the fresh-water fauna in this sense will soon be generally acknowledged, and on this point I fully agree with Professor Hutton, who so clearly pointed out that the immigration of birds and frogs to Australia must have taken place by different ways and at different times. Moreover, I agree with Professor Hutton in thinking that an ancient land-communication must have existed between South America and New Guinea, Australia, &c., to account for the close relations between the two territories in flora and fauna. But, with regard to South America, my own researches, as well as some other new discoveries, have led to so many new and important points of view that some of Professor Hutton's statements require alteration, and it is with reference to this that these lines have been written.

    Professor Hutton says: "Our general results, then, are that in early Mesozoic times New Zealand, eastern Australia, and India formed one biological region, land probably extending continuously from New Zealand to New South Wales and Tasmania. During the Lower Cretaceous period a large Pacific continent extended from New Guinea to Chile, sending south from the neighbourhood of Fiji a peninsula that included New Zealand. Nearly all the southern part of America was submerged. This continent supported dicotyledons and other plants, insects, land-shells, frogs, a few lizards and perhaps snakes, and a few birds, but no mammals. In the Upper Cretaceous period New Zealand became separated: the South Pacific continent divided in the middle between Samoa and the Society Islands, the eastern portion being elevated while the centre sank. It ultimately became what we know now as Chile, La Plata, and Patagonia."

    Let us examine now the relations of the faunistic and geological evidence to this hypothesis. The investigation of the fresh-water fauna shows a great agreement between Chile and South Brazil. In Chile, as well as in South Brazil, the waters, which are covered with Lemna and Confervæ, and overgrown with rushes--Typha, Sagittaria, Potamogeton, &c.--contain besides Unio, Cyrena, Pisidium, species of Planorbis, Limnæa, Ancylus, Physa, Chilina, and, in addition, several genera of frogs and fresh-water fishes, and of Crustacea Parastacus and Æglea. It is not known how far the alliance of the lower animals goes; but Æglea lævis occurs in Chile as well as in the Rio Grande do Sul, in both places with the parasite Temnocephala chilensis. This parasite is found here also living on Parastacus and Ampullaria canaliculata. In [[p. 434]] St. Catharina Fritz Mueller found a second species of it. The distribution of Parastacus is very peculiar. As well as in Chile species of this genus are found in Entrerios in Argentina (Burmeister), Uruguay (E. Berg), Rio Grande do Sul (Hensel, von Jhering), and St. Catharina (Fritz Mueller). The species of Chilina show exactly the same distribution. A short time ago I received a new species, standing between fluminea and gibbosa (Chilina muelleri, mihi), which was collected by Mr. Fritz Mueller in the Itajahy, in St. Catharina, where it is common on stones with Lithoglyphus lapidum. This district will probably extend to Parana, but it is not very likely that it will go as far as Rio de Janeiro, where many naturalists have been collecting. The rivers near Rio de Janeiro flowing to the Atlantic Ocean show a special fresh-water fauna, chiefly Naiadæ, whilst those from St. Catharina are more like the forms from Chile. I know about one dozen species of Unio from Chile, and six of them are also found in Rio Grande do Sul or in La Plata. These are the following: Unio diplodon, U. lepidior, U. montanus, U. beskeanus, U. koseritzi, U. auratus, U. rhuacoicus, U. casablancæ, U. æthiops, U. modestus, U. charmanus, U. colchaguensis, U. faba.

    Evidently we have to deal here with a coherent fresh-water area, which no doubt in ancient times formed also a geographically uniform and united territory, for which I propose the name of Archiplata. This land, low-lying and abounding with rivers and swamps, must have been in existence before the upheaval of the Andes, because this is the only way by which we can explain the distribution of forms bound and limited to fresh water. The chain of the Andes would have proved an insurmountable obstacle to them, as it really did later on for all the immigrants from north to south, such as Anodonta, Mycetopus, &c., or Ampullaria, and to the alligators and tortoises, the Characinidæ, Chromidæ, &c., and to the representatives of a new tropical fresh-water flora, such as the species of Victoria and Pontederia. All these are missing in Chile, for the only reason that the upheaval of the Andes had commenced when they began to spread into Archiplata. The Andes may not have been very high then, but they formed a sufficient barrier to prevent migration to the west.

    This Tertiary influx has changed the flora and fauna east of the Andes to an enormous extent. The fresh-water fishes, two only of which are known to me as common to Chile and Rio Grande do Sul--Gobius and Trichomycterus--were retreating southwards, or perished in the struggle with the new immigrants. It is the task of future investigation and research to discover the complete flora and fauna of this Archiplata area. Many things will be understood through it: thus the distribution of the penguins (Spheniscus magellanicus) found [[p. 435]] by me, and of many of the Lacertidæ on the coast of Rio Grande do Sul. The species of Liolæmus, Saccodeira, and Urostrophus which I found in the Rio Grande do Sul belong to genera represented chiefly in Chile and Patagonia. Of Patagonian shrubs and trees many are found in the south of Rio Grande do Sul.

    In comparing the old fauna of Archiplata with that of New Zealand and Australia many points of contact are found. The perception that the species of Unio from New Zealand and many adjoining countries have their near allies only in Chile and South Brazil formed for me the starting-point for the pursuit of the questions here dealt with. I hope that, through the kind aid of Professor Hutton and other scientists in New Zealand and Australia, I shall be enabled to study this question thoroughly, especially with regard to the animals. Günther also unites New Zealand with Chile and Patagonia into one region for the fresh-water fishes. The Parastacidæ show the same, and many botanical facts point to a former land-communication between these countries.

    Before we enter further on this subject we must try to form an idea of the geology of Archiplata. Relying on the distribution of the Jurassic strata in the Andes, Professor Hutton concludes that the whole of this area has been submerged by the sea during the Jurassic period. No doubt a deep bay occupied then the place of the Andes, but on its sides low-lying land might exist. In the old Archiplata fauna we meet with Mollusca and Crustacea whose genera are already found directly, or in nearly allied forms, in the Jurassic. Besides the very old character of this fauna, which was in existence before the upheaval of the Andes, there is geological evidence to show that parts of Archiplata belong to a very old continent. Outside the Andes the Jura formation has been found neither in La Plata nor in Brazil. In Rio Grande do Sul only the Carboniferous formation has been found. Many times it has been taken for a younger formation, but I have seen myself a beautiful trunk of Lepidodendron in the mines of Jeronymo. Liais² says that it has been described by Carruthers as Flemingites pedroanus, and that, besides it, Noeggerathia, Glossopteris, Odontopteris, and Calamites are found. We leave it undecided whether this flora belongs to the Carboniferous or, as Liais thinks, to the Trias; at all events, it proves that land with a rich flora existed here at the beginning of the Mesozoic era.

    The Cretaceous formation is not known in this district outside the Andes, but is well developed between Pernambuco and the Amazon, and has been studied by White in a most [[p. 436]] admirable way. This scientist also made us acquainted with a fresh-water Cretaceous formation which he discovered near Bahia. During the Tertiary period the sea extended to the Andes, where near Pebas a brackish-water fauna was deposited, which has been the subject of very careful study by Dr. Boettger. But lately Ochsenius³ has shown that considerable parts of the Andes were upheaved to their present height towards or after the end of the Tertiary era. In Chile (lat. 37° S.), as well as in Bolivia (Potosi), Tertiary plants have been found, which have been studied by Engelhard. Those from Chile have not been published yet, but they belong to genera living now in tropical South America, eastward of the Andes, indicating a warm, humid climate, and are now extinct on the western side of the Andes. The same applies to the species of Cassia, Sweetia, Leptolobium, Myrica, &c., found near Potosi (lat. 19° S.), at an altitude of 4,200 metres, which cannot have been grown at such an elevation as they are found at now.

    We therefore have to consider the present flora of Chile, which is adapted to the more rough climate of the Recent period, as having in a great measure arrived from the south. It was preceded during the Tertiary era, when the Andes had only a small elevation, by a tropical flora, which spread to Brazil and the adjoining tropical countries whilst the Andes were being upheaved. How many changes in the animal kingdom may have been going on hand-in-hand with the alterations in the flora? I only wish to point out the absence of Hyla in Chile, which genus probably was represented in the Archiplataic area as well as the Cystignathidæ, both of which are found in Australia, and missing in Africa--facts of importance awaiting interpretation. Chile, in my opinion, must have had species of Hyla during Tertiary times, although no representatives of the Hylidæ maintained themselves in the country.

    Ochsenius recalls to mind that, according to Conte, the upheaval of the Sierra Nevada, in California, also took place in Post-tertiary times; that in Lake Titicaca a relic-fauna of Crustacea from the Pacific Ocean is living, and that Agassiz found in its neighbourhood, 900 metres altitude, fossil corals corresponding with recent forms of the Pacific.

    The upheaval of the Cretaceous formations in Peru and Bolivia is, according to Ochsenius, essentially due to events during the Quaternary period, whilst the Chilean Tertiary coal-bearing littoral sank at the same time. To such a relative [[p. 437]] late subsidence of the Chilean coast points likewise the great similarity of the flora and fauna of Chiloë with those of Chile. The Unios, for instance, from Chiloë are identical with those from Chile, and perhaps also with those from New Zealand.

    The opinion of Ochsenius is shared by many renowned geologists, but disputed by others. The facts we have mentioned may well be in accordance with it, inasmuch as our zoogeographical reflections lead to the conclusion that with the beginning of the Tertiary era, or soon afterwards, an upheaval, inconsiderable perhaps, must have been going on in the district of the present Andes, which explains why the successively-arriving immigrants of the fresh water could not pass over this barrier. On the other hand, this low central mountain-chain offered a way to the placental mammals of the Argentine Tertiary by which they could accomplish their entrance, probably in the Oligocene, certainly not later than early in the Miocene.

    If we inquire after the aspect of the orographical conditions in early Tertiary times, there are, I think, sufficient facts to give a satisfactory answer. The valley of the Amazon was covered by the sea. Deposits of the Cretaceous sea, in some parts perhaps rather belonging to the Eocene, cover large areas south of the Amazon, eastwards in Para, Paranahyba, Sergipe, &c., as far as Bahia, and westwards in Peru and Bolivia. At the end of the Cretaceous period, and during the Eocene, South America was broken through from ocean to ocean by the sea in the direction of the Amazon. In all probability the upheaval of the Andes had begun already in the Eocene period, and soon afterwards the Pebas strata must have been deposited in brackish water on the foot of the still but little elevated chain of the Andes. According to Boettger the Pebas deposits are of Oligocene age. It is to be supposed that during the Oligocene period a land-communication was existent in the district of the present Andes, between the high land of Guyana and Venezuela (Archiguyana) and Archiplata, by way of which the old fauna of placental mammals would arrive in Argentina.

    The mammal fauna has its nearest allies in that of the European Oligocene, which was probably represented in Africa by more or less similar forms. It is true that the supposed Anoplotheria of Argentina have been recognised as belonging to another genus (Proterotherium), but they are, together with the numerous Rodentia, more nearly allied to the European Oligocene fauna than to any other. Schlosser has pointed out that the Theridomydæ of the European Oligocene passed over to South America as Chinchillidæ, Echimyidæ, and Caviadæ. Of these Rodentia scanty remains are found on the Antilles and in Africa, and none, either recent or fossil, in [[p. 438]] North America. They therefore cannot have immigrated by way of North America, but during different geographical conditions.

    Between the Oligocene and the end of the Pliocene period South America was completely isolated, and therefrom it derives its peculiarity in flora and fauna. It was only at the end of the Pliocene, or perhaps even later, that North American mammals appeared in South America, whilst in return South American ones advanced to Mexico. There are no differences of opinion on this point, and the geology of Central America tells us that the Isthmus of Panama did not come into existence before the end of the Tertiary era. But it seems that many scientists think that a land-communication between North and South America might have existed in the beginning of the Tertiary era. No doubt many circumstances have here to be taken into consideration, as, for example, the communication of Florida with the West Indies, as proved by the Miocene land-shells of Florida, elaborately described by Dall, and the many changes which have taken place within the area of the West Indian Islands, a reliable judgment on which we are unable to give at present; but, as for the older period, we must, in my opinion, admit a sharp separation between the territories of North and South America. The fresh-water fauna of both shows enormous differences, in the Naiadæ; and other molluscs, as well as in the fishes, tortoises, amphibians, &c. North America, with its Cyprinidæ, Urodela, &c., is in all these points more nearly related to the palæarctic region, while South America--that is to say, Archiguyana--shows close alliances only to Africa. With regard to the fresh-water fishes this is so well known that I need not dwell on it further, but I would point out that the old common features have been obscured through later relations to the Mediterranean and Indian faunas, wherefrom many elements have been absorbed.

    After the upheaval of the Isthmus of Panama, mammals, birds, and Lepidoptera of the adjoining countries rapidly became mixed, and this was also the case with all the other rapidly-wandering creatures. Thus we see Cinosternum and other tortoises immigrate to the north of South America; even two species of Urodela (Spelerpes) advance as far as Ecuador; and even more noticeable is the exchange of Batrachia and Lacertidæ. Nobody will deny the importance of studying separately the two elements of South America if an account is to be given of the geographical distribution of the frogs. Ranidæ are missing in Archiplata just as completely as Aglossa, Urodela, and others, but are found in Ecuador and Columbia. The Aglossa and Dendrobatidæ are restricted to Archiguyana, Africa, and Madagascar. Even some genera of [[p. 439]] the fresh-water fishes, such as Pimelodus, have species in South America and Africa, the common area of the Chromidæ and Characinidæ; and, in the genus Pontederia of the water-plants, actually one and the same species, Eichhornia natans, occurs in the interior of Africa and South America.

    No one studying carefully the fresh-water fauna can fail to perceive the enormous difference between North and South America, while Central America joins to Mexico. The marine deposits of the Cretaceous and Tertiary sea in Central America give us the key. I am of opinion that facts pointing in this direction can hardly be disputed, in spite of their being obscured by the Pleistocene intermixture of both faunas; but since Wallace's severe condemnation great prejudice exists against the assumption of the "Atlantis" (the subsided land-communication between Archiguyana and Africa) having existed up to the Oligocene period. The reason given by Wallace is the very considerable depth of the sea; as if a subsidence of 5,000 metres was more miraculous in itself than an elevation to the same amount! Also, Wallace is not even consistent. If he concedes "Lemuria," and even thinks of a land-communication in Miocene times between New Zealand and South America, then he should not oppose the Atlantis. It will hardly be possible to-day to say where this bridge was situated. The scanty remains of it may have lost, to a great extent, their old fauna, through alternating upheaval and subsidence, like New Zealand. But the subfossil Bulimus of St. Helena shows this island to be part of this old bridge across the Atlantic Ocean, and it explains the occurrence of numerous identical species of marine Mollusca in Brazil, the West Indies, and the Atlantic coast of Africa.

    In consequence of what has been said hitherto, there can be no doubt about Archiplata being an old continent, existing since the Trias formation. I arrived at this conclusion through the study of the fresh-water fauna,⁴ and it has quite unexpectedly, and almost at the same time, been confirmed from another side. A scientist who deserves great merit for his knowledge of the Tertiary mammals of Argentina--Florentino Ameghino⁵--has published a treatise on the old mammalian fauna of Patagonia, this fauna belonging partly to the Cretaceous partly to the Eocene formation, but I fancy that the whole of it may be of Cretaceous age. Ameghino includes under the name of Diprotodonta most of the Australian Recent marsupials, the Plagiaulacida, and the Mesozoic genera of [[p. 440]] marsupials, which are characterized by the syndactyly of the second and third toe, a pair of well-developed upper incisors, accompanied by two pairs of smaller ones, and one pair of strongly-developed lower incisors, and small or missing canines. It now appears that in Patagonia deposits are existent whose fossil mammals are entirely composed of Plagiaulacida. To my knowledge, no Tertiary fauna of a similar composition is known, but Marsh has lately described mammals from the Cretaceous formation of North America which are likewise exclusively composed of Plagiaulacida, and this is the reason why I think it very probable that the Plagiaulacida of the riverbanks of the Rio St. Cruz, in Patagonia, belong to a Cretaceous fauna.

    These Patagonian marsupials are not more nearly related to any other group of living or fossil marsupials than to many of the recent Australian genera. This is specially seen in the simple quadritubercular structure of the teeth, whilst the European and North American representatives show numerous denticules, arranged in two or three rows, on their molar teeth. This is found neither in Australia nor in Patagonia. On this point I fully agree with Ameghino, but not that the Plagiaulacida came from Patagonia into North America by way of an Eocene migration, which was geographically impossible. Moreover, according to quite recent discoveries made by Marsh, which were not then known to Ameghino, polymastodont and quadritubercular Diprotodonta lived together during the Cretaceous period in North America, and there can be no doubt but that both types were already co-existent during the Jura formation. But the genera with the serieswise disposition of the mammiform tubercles in two or three longitudinal rows do not seem to have reached the Australian Archiplataic territory, or, at least, disappeared again at a very early period. If, as Ameghino thinks, a land-communication between North America and Argentina existed during the Cretaceous epoch, this distribution would be inexplicable. The genus Didelphys, which, however, occurs Tertiary in North America, is also found in Europe; and Didelphys, as the descendant of its European precursors, the Peratheria, may have spread to both Americas without having been compelled to reach South America by way of North America. Didelphys, which is wanting in Australia, must therefore be attributed to immigration from the Old World.

    When Ameghino from this draws the conclusion that a continent between Australia and Argentina must have been in existence in the Mesozoic era, he only proclaims what Professor Hutton, I myself, and many other scientists who have studied the flora and fauna of both countries have done before, but differing from Professor Hutton so far as to give this [[p. 441]] continent a higher age. The relative sentence runs as follows: "The recent Australian Diprotodonta therefore must date back in this continent as far as the beginning of the Eocene period at least, whilst the ancestors of the Diprotodonta common to Australia and Argentina must reach to a far more remote epoch, during which they were spreading over a very wide continent uniting Australia with South America in a more or less continuous connection." Ameghino thinks this continent was situated in the Pacific Ocean, and dated back to the Trias formation.

    It is evident that these results must alter in many respects the hitherto-prevailing ideas. The theory mentioned by Professor Hutton, which allowed a Tertiary migration of northern types over the Andes into the Antarctic area, fails, for during the Cretaceous and in the beginning of the Tertiary era there was no chain of the Andes existent⁶; and when it began to rise, though to no considerable elevation, it was covered with a tropical, not alpine, flora. The reasons given allow much less an immigration of northern plants from North America over the whole length of the Andes. So far as an exchange of plants between Australia, &c., and South America took place, it must have gone on over the same land-bridge on which, in other latitudes or at other times, the temporary migration of the Antarctic flora went on. Whether such a rigorous distinction between a South American and an Antarctic flora as has been hitherto made will be possible to the same extent in future, seems to me very doubtful. Those who do not agree with me herein must indeed first study the phytogeographical condition of Rio Grande do Sul, especially the southern part of it, where, besides plants common in Argentina and Uruguay, many of central Brazil are found also: thus Cedrela, Erythroxylon, Tecoma, Erythrina, &c., with Sentio, Duvana, Celtis, Jodina, Phyllanthus, Lacuma, &c., even the Patagonian Berberis spinescens, and, as I believe, Colletia also.

    It therefore follows that the relations of the South American to other fauna-areas, as I have represented it here, are in strong opposition to Mr. Wallace's views; but, with regard to its relations to Australia and New Zealand (though variously modified in respect of South America), it is in harmony with the theory at which Professor Hutton, and other scientists who have studied the flora and fauna of the Australian region, have arrived.

    To conclude, the results arrived at may be summed up as follows: South America was separated from North America from the Cretaceous to the end of the Pliocene period. A [[p. 442]] South American continent has existed only since the Oligocene period. It then consisted of two parts, united by the narrow isthmus of the Andes, which were completely separated from one another before the Oligocene period. These two parts form Archiplata (the area occupied at present by Chile, Argentina, Uruguay, and South Brazil) and Archiguyana (comprising the high plateau of Venezuela and Guyana). Each of these territories had its own fauna and flora, which differed as much from one another as those of interior Africa and North America do at the present time. Archiguyana must have been united with Africa by a land-bridge, a remnant of which is represented by St. Helena, whilst Archiplata extended to the south in a South Pacific Antarctic continent, which during the whole of Mesozoic times kept this area in communication with the Pacific continent, from which first a number of Polynesian Islands, then New Zealand, and finally Australia and New Guinea became separated. Reserved for further investigation remain the questions: Did the Atlantis reach only Archiguyana, or was a southern arm directed to that part of Brazil which extends between Rio, Bahia, and the Rio S. Francisco? What was its relation to the West Indies? What age is to be attributed to these parts of Brazil and Archiguyana? Were the mountains of the Brazilian coast also upheaved only during Tertiary times?--as I presume. All these questions have yet to be examined, as well as the original participation of each of these parts in that mixed fauna and flora which we call now "South American."

    A great deal remains unsolved, but the results obtained regarding Archiplata and its connection with the Antarctic continent seem to me to be an important acquisition. That Archiplata owned an old common flora and fresh-water fauna; that the upheaval of the Andes, beginning at the end of the Cretaceous or commencement of the Eocene period, ending at and after the Tertiary era (in the place of a Jura sea entering with a gulf southwards into Archiplata), formed a barrier which the Tertiary fresh-water fauna migrating to Archiplata could not cross--this essential result of my studies of the fresh-water fauna gives, in my opinion, a very simple and new explanation of a large number of facts, otherwise unintelligible, regarding the geographical distribution of South America's flora and fauna.

    These being the results to which, relative to South America, zoogeography and geology are leading, I, on the other side, believe that the North American scientists are able to bring in accord with this their own experiences. Heilprin⁷ has [[p. 443]] investigated the history of the Laramie region and its lacustrine deposits, which attain sometimes a thickness of 4,000 ft. to 5,000 ft., and show in the beginning an entirely marine character, but change later on to the deposits from fresh water. The drainage was into a continental arm of the sea, which projected completely across the United States during the Cretaceous period, and which prevented an exchange of fresh-water animals between North and South America, as well as the migration of marsupials from Patagonia to the United States.

    The theory of Wallace is not only to be rejected for our special territory of South America, but it is insufficient also for Australia and Polynesia. Wallace thinks that for Polynesia the birds are the only group of the animal kingdom "on which we could rely." We should be more in the right if we reverse his sentence and say that the mammals and the birds are the only groups on which no reliance can be put if we wish to unravel the whole history of Polynesia, for birds and placental mammals belong in their recent representatives entirely to the Tertiary era, and therefore cannot be taken into consideration for the means of distribution of organisms during the Secondary era. Moreover, birds are almost useless for the discovery of old geographical land-communications, on account of their power of flight, and on account of the passive migrations to which they may be subject when driven away by the wind.

    Mr. Wallace's explanation of the distribution of the Lacertidæ through Polynesia as far as the Sandwich Islands by means of a migration across the ocean is just as bold a hypothesis as his attempt to explain the occurrence of identical fresh-water fishes in New Zealand and Patagonia by the transport of their fry on icebergs.⁸ To such theories may those adhere who wish to save Wallace's hypothesis of the stability of the continents and depths of the seas; but one cannot ask unprejudiced scientists to accept such incredible explanations.

    If we look at the circumstances as they really are we perceive that the fauna of Polynesia impoverishes from west to east. A wide distribution is seen only in those groups of animals whose origin does not reach beyond the Tertiary era, which are provided with the power of flight, as birds and bats. Active and passive migration would bring them from island to island, whereby the birds show a wider distribution than the bats. If birds fly over the Atlantic Ocean, and accomplish many other distant migrations, then we are really standing [[p. 444]] on the ground of facts.⁹ Abstracting this Tertiary fauna, we have first to consider those types which undoubtedly have their representatives in the Cretaceous period, such as snakes, serpents, and Anura. Their distribution is pretty wide, and especially on the Fiji Islands they are well represented. The Lacertidæ, whose remains are the geologically oldest of the now-living groups of reptiles, reach further on to the east. Is this chance? If floating trees and icebergs were the means of their transport, why, then, do they transport only such old forms? And has the transport of fresh-water molluscs, such as Limnæa, Physa,¹⁰ &c., also been going on across the sea? This is simply impossible, for salt water kills these inhabitants of the fresh water immediately. The land molluscs of Polynesia also belong to a very old group of the fauna. Widely distributed are species of Pupa, a genus already represented in the Palæozoic era, and all the other Nephropneusta have the simple generative apparatus without the dart of the Heliceæ. Genera which, like Unio, are known to have existed already during the Jurassic epoch, reach as far as New Zealand, and the Parastacida of the fresh water, which, according to Huxley, must also be attributed to a Jurassic age, to Fiji.

    The fauna of East Polynesia has such a well-pronounced Mesozoic character that the supposition of a very old Pacific continent, breaking up in pieces more and more during the Mesozoic era, may give us a natural explanation. And the further we go westwards the more frequently we meet with recent types--in New Guinea the genus Sus and Muridæ; in Australia, besides Muridæ, also Canis. The craving only to make of Australia a land completely without placental mammals accounts for Canis dingo being considered as a race of the domestic dog. This error has been settled definitely by Nehring.¹¹ Canis no doubt belongs to the oldest Carnivora. Species of Canis are found in India and Sumatra, and C. dingo is as well a domesticated sporting-dog as Canis latrans of the North American Indians, or C. ingæ of the old Peruvians.

    The reason why more recent mammals¹² did not migrate [[p. 445]] to New Guinea and Australia is evidently that those countries were already occupied by a fauna of marsupials perfectly adapted to the most varied conditions of life, and therefore later comers were unable to maintain themselves. We shall have to give up the theory of the floating trees with Sus, Canis, Muridæ, Lacertidæ, fresh-water Molluscs, &c., in their branches, as well as many other things of Wallace's hypothesis; and the botanists will not tarry to reduce the transports by currents and wind to a moderate measure.

    Opposed to Wallace's theory of the immutability of the continents and depths of the sea is another which, distinguishing strictly between the distribution of the different groups of animals, endeavours to find the ways of distribution of the genera and families existent on our globe already during Palæozoic and Mesozoic times, along which the organisms spread during the Mesozoic era. May the senior master of zoogeography not be angry with this view, but be cheerful in the thought that, however far opinions on certain questions may be divided, it was he who showed us the way by which progress can be made; that it was he who gave, in an exemplary manner, to zoogeography its modern solid basis, its new method of procedure, its ways of interrogation, as well as its aims and its problems.

_________________________

Notes Appearing in the Original Work

¹Annals and Magazine of Natural History, ser. 5, vol. xiii., p. 425, and vol. xv., p. 77.
²Liais, "Climats, Géologie, Faune du Brésil," Paris (1872), pp. 201, ff.
³Zeitschrift der deutschen Geologischen Gesellschaft, 1886, pp. 766-772; and "Die Natur," Halle, Jahrg. 361, 887, No. 40, 41; and Jahrg. 39, 1890, No. 38.
⁴H. von Jhering, "Die geographische Verbreitung der Flussmuscheln," im Ausland, 1890, Nos. 48, 49.
⁵ Florentino Ameghino, "Los Plagiaulacideos Argentinos," Buenos-Ayres, 1890.
⁶I suppose this migration to have taken place in the Pliocene. --F.W.H.
⁷A. Heilprin, "The Geographical and Geological Distribution of Animals," London (1887), p. 210.
⁸It seems much more likely to me that the enormously-wide distribution of pelagic fresh-water Crustacea is due to transportation by birds. But many of them may belong to very old types, living already in Mesozoic lakes, as Limnæa and Physa did.
⁹With regard to the bats, it is quite possible that they kept pace with the Muridæ, which, according to Wallace, have also been represented in New Zealand.
¹⁰Species of Physa occur on the Fiji and Tonga Islands, besides numerous species of Succinea, but the small islands seem to have been rather unfavourable to the conservation of the fresh-water fauna.
¹¹Nehring, Sitzungsber. d. Ges. naturf. Freunde; Berlin (1882), p. 67. Zoolog. Garten (1885), p. 164.
¹²Another very surprising immigration is that of Mycetopus rugatus of the Victoria River, in North Australia. In my opinion it originates from Asia, and not from South America, where Mycetopus originally was missing in Archiplata. My Australian colleagues should see that the anatomy and embryology of this species be studied, as it is still questionable whether it belongs to Mycetopus s.-str. or not.

*                 *                 *                 *                 *

Return to Home