MSS intended for publication and books, etc., intended for review should be One hundrea reprints of contributions are supplied to authors free of charge. Subscriptions and advertisements should be sent to the publishers. The Lancaster, Pa. THE SCIENCE PRESS NEW YORK: Sub-Station 84 Garrison, N. Y. Entered as second-class matter, April 2, 1908, at the Post Office at Lancaster, Pa., under the Act of Important New Scientific Books Botany PUBLISHED BY THE MACMILLAN COMPANY The Origin of a Land Flora. A Theory based upon the facts of Alternation. Cloth, gilt top, xi+727 pp., illus., index, 8vo, $5.50 net. NOTE.-A profound study in the morphology of the lowest forms of plants, with special reference to the development of their reproductive systems. The author endeavors to show that the present land flora has originated from an aquatic ancestor, and traces the methods of specialization to the land habit, and the establishment of the forms of the higher plants. A book of the highest importance not only to botanists but to biologists in general. NEW AND STANDARD HAND-BOOKS ON AGRICULTURE, ETC. Cloth, xvi+329 pp., illus., appendix, glossary, index, 12mo, $.75 net. NOTE. This is a thoroughly modern treatise on farming as practiced in the Gulf States, from the point of view of the scientific agriculturist. It is adapted to school use, but is of practical service to any farmer in that part of the country. The author is vice-director of the Texas Experiment Station. Re-issues of the highly valued Hand-Books known as the Garden Craft and Rural Science Series, edited by L. H. BAILEY, of Cornell University, Editor of "Cyclopedia of American Horticulture," "Cyclopedia of American Agriculture," etc. Published The Principles of Fruit Growing. By L. H. BAILEY. Tenth Edition, $1.50 net, by mail $1.66. Fifth Edition, $1.50 net, by mail $1.68. Bush Fruits. By FRED W. Card. Tenth Edition, $1.25 net, by mail $1.38 Farm Poultry. By GEO. C. WATSON. Sixth Edition, $1.25 net, by mail $1.42 The Macmillan Company 64-66 5th Ave., New York THE AMERICAN NATURALIST VOL. XLIII September, 1909 No. 513 ON AN EARLY TERTIARY LAND-CONNECTION BETWEEN NORTH AND SOUTH AMERICA R. F. SCHARFF, PH.D., B.Sc. NATIONAL MUSEUM OF IRELAND, DUBLIN THAT the two great continents-North and South America-have only become joined to one another by land in later Tertiary times, is a widely accepted assumption. Yet the various authorities who have made this problem a subject of special investigation have not all come to precisely the same conclusions as to the geological age during which this union of the two continents was brought about. A study of the marine fishes on both sides of the isthmus of Central America, for example, convinced Dr. Günther1 that up to a recent geological period the latter was only represented by a chain of islands similar to that of the Antilles. But the number of species of fishes on the Atlantic and Pacific coasts of Central America that were supposed to be identical has been considerably reduced during more recent surveys. They no longer amount to more than 4.3 per cent. of the total number of fishes known to occur in both areas. Professor Jordan2 on that account maintains that the isthmus has not been depressed during the lifetime of most of the existing species. The submergence, he argues, must have supervened at a more remote time. In the belief that the p. 280. 1 Günther, A. C. L. G., Study of Fishes, 2 Jordan, D. S., "Study of Fishes,'' Vol. I, pp. 274-280. Miocene may be taken as the date of origin of the modern genera of marine fishes, he contends that an open communication between the two oceans may have existed during that geological period. It is important to note that the sea currents seem at that time to have set westward, thus favoring the transfer of Atlantic rather than Pacific types across the isthmian area. Mr. Regan3 is inclined to put the date still a little further back in urging that the marine connection between the two oceans ceased to exist at the beginning of the Miocene. An investigation of the Crustacea and their distribution led Dr. Ortmann to the conclusion that at the dawn of the Tertiary era an oceanic connection was in actual existence between the Atlantic and the Pacific in the isthmian region. This communication, he thinks, persisted until the Miocene. In the commencement of that period the isthmus was elevated, thus joining North and South America. The Mexican amphibians and reptiles have been utilized by Dr. Gadow in the solution of the same interesting problem with the result that he assumes the establishment of land-continuity between North and South America in either late Oligocene or early Miocene times. The whole character of neotropical zoology, remarks Dr. Wallace, whether as regards its deficiencies or its specialties, points to a long continuance of isolation of South America from the rest of the world, with a very few distant periods of union with the northern continent. 7 a Geologists have discussed this subject mostly from paleontological evidence. Professor Gregory' clearly demonstrated that the idea of an interoceanic connection as late as the Pleistocene period, as suggested by Dr. Spencer, could no longer be entertained. In arriving Regan, C. T., "Fishes of Central America," p. xxx. Ortmann, A. E., "Geographical Distribution of Decapods,' Gadow, H., "Mexican Amphibians and Reptiles," p. 236. Wallace, A. R., Distribution of Animals," Vol. II, p. 80. Gregory, J. W., "Palæontology of the West Indies,' 3 7 p. 305. p. 359. Ta Spencer, J. W., "Reconstruction of an Antillean Continent," p. 134. at a final decision as to the approximate date of orgin of the Central American land bridge, he was mainly influenced by Professor Scott's reference to the occurrence of Caryoderma a supposed glyptodont edentate in the Miocene Loup-Fork deposits. He urged, therefore, that the waterway across Central America was in all likelihood finally closed in the lower Miocene or possibly even in the upper Oligocene. As Caryoderma, however, is now believed to be a reptile and not an edentate, this argument no longer holds good. Less definite are the results obtained by Dr. Hills after a careful study of the rocks near the Isthmus of Panama. The only geological periods, he thinks, since the Mesozoic era, during which the Pacific and Atlantic Oceans could have been in communication with one another, would be the Eocene or Oligocene. It is important to note that Dr. Hill's conclusions were based entirely on his observations at the Isthmus of Panama. The geology of the remainder of Central America is as yet too imperfectly known to form the basis for similar speculations. The most important pronouncement perhaps which has yet been made on the subject under discussion is that by Professor Osborn." His intimate knowledge of the fossil terrestrial mammals of North America enabled him to affirm that North and South America were joined to one another more than once, as Wallace had suggested. The first union occurred in Mid-Cretaceous and perhaps early Tertiary times. Hereafter the continents separated once more until the Pliocene period. Dr. Smith Woodward,10 Mr. Lydekker and also Professor Depéret11 hold similar views with regard to the more recent junction of the two continents. The evidence on which Professor Osborn based his belief in the first and much earlier land connection between North and South America was unknown when Mr. Lydek* Hill, R. T., "Geological History of the Isthmus of Panama,' Osborn, H. F., "Mammalian Palæontology," p. 99. 10 Woodward, A. S., "Palæontology," p. 429. "Depéret, C., "Transformations of the Animal World," p. 282. p. 269. ker12 wrote his work on the Geographical History of Mammals. He was under the impression, therefore, that the mammalian fauna of the South American region had been totally isolated from that of North America up to about the end of the Miocene. I shall shortly return to Professor Osborn's views as soon as I have completed my brief historical review of the first problem. Professor Lapparent1s concurs with Mr. Lydekker's opinion that the interchange of waters between the Atlantic and Pacific oceans across Central America could only have ceased to exist at quite the end of the Miocene period. Finally, in his treatise on the development of continents, Dr. Arldt1 maintains that the Central American land bridge must have originated at the commencement of the Pliocene period, and with this view I fully agree. Part of Central America no doubt had already risen above the ocean at a much earlier period, but in its present outlines and extent it must be regarded as a geologically recent development. All those in fact who have seriously considered the problem, either from the standpoint of a marine or a terrestrial zoologist or from that of a paleontologist concur in the opinion that North and South America were separated from one another by a marine channel or by wide seas during part of the Teritary era. This, however, is the only point in which there is a general agreement. While some contend that the junction between the two continents had only been effected in comparatively recent geological times, others hold that within the life history of the great class of mammals, either in the early Tertiary or late in the Secondary era, a land bridge between North and South America had once before existed, by means of which an interchange of the faunas could have been brought about. It is this supposed earlier land connec"Lydekker, R., “Geographical History of Mammals,” p. 119. |