tf.DI J. PIERPONT MORGAN PUBLICATION FUND REPORTS OF THE PRINCETON UNIVERSITY EXPEDITIONS TO PATAGONIA, 1896-1899 J. B. HATCHER IN CHARGE EDITED BY WILLIAM B. SCOTT BLAIR PROFESSOR OF GEOLOGY AND PALEONTOLOGY, PRINCETON UNIVERSITY VOLUME IV PALEONTOLOGY I PRINCETON, N.-J. THE UNIVERSITY STUTTGART SCHWEIZERBART'SCHE VERLAGSHANDLUNG (E. NAGELE) 1901-6 Q l?5 P95 1903 J. PIERPONT MORGAN PUBLICATION FUND REPORTS OF THE PRINCETON UNIVERSITY EXPEDITIONS TO PATAGONIA 1896- 1899 VOLUME IV PALAEONTOLOGY I PART I THE MARINE CRETACEOUS INVERTEBRATES BY TIMOTHY WILLIAM STANTON UNITED STATES NATIONAL MUSEUM PART II TERTIARY INVERTEBRATES BY ARNOLD EDWARD ORTMANN PRINCETON UNIVERSITY PART III MAMMALIA OF THE SANTA CRUZ BEDS. MARSUPIALIA BY WILLIAM JOHN SINCLAIR PRINCETON UNIVERSITY PRINCETON, N. J. THE UNIVERSITY STUTTGART SCHWEIZERBART'SCHE VERLAGSHANDLUNG (E. NAGELE) TMf New En* pniNim,. COM IANCA8TEN, PA, TABLE OF CONTENTS, VOL. IV. PART I. THE MARINE CRETACEOUS INVERTEBRATES. BY T. W. STANTON. PAGE LETTER OF TRANSMITTAL I INTRODUCTION . 3 PELECYPODA ............ n SCAPHOPODA ............ 28 GASTROPODA ............. 29 CEPHALOPODA ............ 35 PART II. TERTIARY INVERTEBRATES. BY A. E. ORTMANN. LETTER OF TRANSMITTAL 45 CONTENTS 47 INTRODUCTION 48 SYSTEMATIC PART 51 ECHINODERMATA 51 ECHINOIDEA ............ 51 Cidarida Wright .......... 5 l Echinidce Wright .......... 52 Clypeastrida Ag. .......... 55 Cassidulidce Ag. .......... 60 Spatangidce Ag. .......... 62 VERMES 63 CH^ETOPODA ............ 63 Tubicolcz (Sedentaria) ........ 63 BRYOZOA 64 CHILOSTOMATA ............ 64 Cellariidce Hcks. ........ 64 Escharidce Johnst. ....... 67 CYCLOSTOMATA ........ 68 Ltchenoporida Sm. ........ 68 BRACHIOPODA 7° Rhynchondlidce d'Orb. • 7° Terebratididce King ....... 73 v VI CONTENTS. MOLLUSCA 80 PELECVPODA 80 Nuculidcc Ad. ........... 80 Ledidee Ad 83 Parallelodontidtz Dall 86 Ltmopsidee Dall. .......... 91 Arcidce Dall 93 Pernidce Zitt 97 Ostreidte Lamck. .......... 98 Pectinidte Lamck. .......... 114 MytilidaYtem 120 Carditida Gill 125 LitcinidcB Flem. ........... 1 29 Cardiidce Fisch. ........... 132 Veneridce Leach. . . . . . . . . . . 135 Tellinida Desh. . . . . . . . . . . . 147 Psammobiidce Dall. .......... 149 Mactrida Gray. . . . . . . . . . *. 149 Corbulida Flem. . . . . . . . . . . . 151 SaAicavida Gr. . . . . . . . . . . . 152 Pholadida Fisch. . . . . . . . . . . 155 SCAPHOPODA 157 Dentaliidtz Gray .......... 157 GASTROPODA ............ 161 Patellida Carp. ........ ..161 Fissurcllida Riss. . . . . . . . . . . 161 Delphinulidcz Fisch. .......... 162 Trocluda Ad. ........... 1 62 Pyramidellidce Gr. . . . . . . . . . . 173 Scalariida Brod. . . . . . . . . . . 175 Capnlidce Cuv. ........... 177 Naticida Forb. ........... 186 TurritcllidcE Gray. .......... 192 Vertnetidce Ad 198 Aporrtiaidce Phil. .......... 200 Strombidce d'Orb. .......... 201 DoliidcB Ad. ........... 204 Tritonidce Ad. ...... 206 Buccinida Trosch. .......... 208 Muricidce Tryon 214 Fusida Tryon ........ 221 Volutidce Gray ........... 224 Cancellariidce Ad. . . . . . . . . . . 235 Terebridce Ad. ........... 236 Pleurototnida Stol. . . . . . . . . . . 238 CONTENTS. Vii Actaonida d'Orb. . ' . . . . . . . . . 343 Bullida Pilsbry ........... 245 CRUSTACEA 247 ClRRIPEDIA 247 Lepadidce Darw. ........... 247 Verrucidce Darw. .......... 248 Balanida Darw. . . . . . . . . . 249 DECAPODA . . . . . . . . . . . . . 255 Carcinoplacidtz Ortmann . . . . . . . . . 255 LIST OF FOSSILS DESCRIBED . 256 GENERAL CONSIDERATIONS 260 THE PATAGONJAN BEDS ........... 260 IDENTITY OF PATAGONIAN AND SUPRAPATAGONIAN BEDS ...... 265 THE AGE OF THE PATAGONIAN BEDS 286 THE MAGELLANIAN BEDS ........... 303 THE CAPE FAIRWEATHER BEDS (? MARINE TEHUELCHE BEDS) .... 307 ORIGIN AND DEVELOPMENT OF THE PATAGONIAN MARINE FAUNAS . . . . 310 1. THEORY OF ANTARCTICA .......... 310 2. RELATIONS OF THE PATAGONIAN DEPOSITS TO OTHER PARTS OF SOUTH AMERICA AND TO THE REST OF THE WORLD : THEORY OF "• ARCHIPLATA " AND " ARCHHELENIS ". .......... 319 BIBLIOGRAPHY 325 PART III. MAMMALIA OF THE SANTA CRUZ BEDS. MARSUPIALIA. BY W. J. SINCLAIR. INTRODUCTION 333 CLASSIFICATION OF THE SANTA CRUZ MARSUPIALS .... 334 POLYPROTODONTIA . . 34 1 THYLACYNID^J ............ 341 Borhycena Amegh. .......... 347 Prothylacynus Amegh. ......... 362 Cladosictis Amegh. .......... 376 Amphiproviverra Amegh. ....... 394 RELATIONSHIPS OF THE THYLACYNID^E ......... 406 DlDELPHYID/E ............ Microbiothcrium Amegh. ........ DIPROTODONTIA 416 C^ENOLESTID^E ......... 4'6 CMNOLESTIN& ........ ••4I9 Halmarliiphus Amegh. . . . . . • • • 4J9 Garzonia Amegh. ...... • 422 PAL&OTHENTIN& . ....... 425 viii CONTENTS. Palaothentes (Moreno) Amegh. . 425 CaJlointmts Amegh. ...... . 434 Decastis Amegh 436 ABDERITIN& ...... . .438 Abderites Amegh. ......... 438 RELATIONSHIPS OF THE CENOLEsriaE 441 BEARING OF THE SANTA CRUZ MARSUPIALS ON ZOOGEOGRAPHY 444 MARSUPIALIA INCERT^E SEDIS 444 Borhyana Amegh. - . . . . . . . . . 444 Acrocyon Amegh. ......... 445 Conodonictis Amegh. . . . . . . . . . 445 Arclodictis Mercerat . . . . . . . . . 446 Prothylacynus Amegh. ......... 446 Naponodictis Amegh. ......... 446 Agustylus Amegh. . . . . . : . . . 446 Cladosictis Amegh. . . . . . . . . . . 447 Hattiliacynus Amegh. ......... 447 Thylacodictis Mercerat ......... 448 Amphiproi'iverra Amegh. . . . . . . . . 448 Perathereutes Amegh. . . . . . . . . . 448 Sipalocyon Amegh. ......... 449 Acyon Amegh 449 Ictioborus Amegh. . . . . . . . . . 450 Microbiotherium Amegh. . . . . . . . . 450 Stylognathus Amegh. ......... 450 Eodidelphys Amegh. ......... 450 Prodidclphys Amegh. ......... 45 1 Abderites Amegh. . . . . . . . . 451 Mannodon Amegh. ......... 452 Decastis Amegh. ......... 452 Acdestis Amegh. ......... 452 Dipilus Amegh. ......... 453 Metriodromus Amegh. ........ 453 Halmadromus Amegh. . . . . . . . 453 Callomenus Amegh. ........ 454 Palaothentes (Moreno) Amegh 454 Prepanorthus Amegh. ....... 455 Halmaselus Amegh. ....... 455 Essoprion Amegh. ....... 455 Pichipilus Amegh. ....... 455 Garzonia Amegh. ...... 456 Parhalmarhiphus Amegh. ...... 457 Halmarhiphus Amegh. ....... 457 StUotherium Amegh. ..... 41-7 Cladodinus Amegh. ..... CONTENTS. JX BIBLIOGRAPHY 458 ERRATA 460 INDEX 461 DATES OF PUBLICATION OF THE PARTS OF VOLUME IV. Dates of issue were not printed on the covers of the separate parts of this volume. The dates of actual publication are as follows : Pp. i- 44. Pll. I-X, published November 6, 1901. Pp. 45-332. Pll. XI-XXXIX, published April 19, 1902. Pp. 333-460. Pll. XL-LXV, published September 14, 1906. WASHINGTON, D. C, Jan. 14, 1901. Prof. W. B. SCOTT, Princeton University, Princeton, N. J. Dear Sir : I have the honor to transmit herewith for publication a brief paper on "the marine cretaceous invertebrates obtained by the Princeton ex- pedition to Patagonia in 1899." The collection was with your permission delivered to me for study by Mr. J. B. Hatcher in October, 1899, and its examination and descrip- tion have continued at intervals since that time as official and other duties permitted. The fossils proved to be unusually interesting from the fact that they represent a facies of the Cretaceous fauna not hitherto described from South America. There are about 40 species of Mollusca in the collection, of which 31 are sufficiently well represented to be named and described. These indicate the Lower Cretaceous age of the beds from which they were derived. Very respectfully, T. W. STANTON. THE MARINE CRETACEOUS INVERTEBRATES. BY T. W. STANTON. THE fossils discussed in the following pages were collected in March, 1899, by Mr. J. B. Hatcher, in charge of the Princeton Expedition to Patagonia, and it is through his kindness and the generous courtesy of Prof. W. B. Scott that I have had the privilege of studying the collection, which, though not large, has proved very interesting. In a paper entitled "Sedimentary rocks of southern Patagonia," Mr. Hatcher1 has described the section from which the fossils were obtained, and named the various Cretaceous horizons recognized. The entire col- lection came from two localities, only a few miles apart, in the neighbor- hood of Lake Pueyrrydon, south latitude 47° 30', west longitude 72°. One of the localities is four miles east of Lake Pueyrrydon, near the mouth of the canon of Rio Tarde, a small stream emptying into the east end of the lake. The other locality is about ten miles east of the same lake, and about two miles south of the western border of White Lake.2 The following description of part of Mr. Hatcher's section along the Rio Tarde is condensed from his account and gives only the features es- sential to the present discussion.. The series about 750 to 800 feet in thickness that includes all the fossiliferous Cretaceous horizons is called the Pueyrrydon series. In it four formations are recognized, in ascending order as follows : i. The Gio beds, consisting of 100 feet of soft green sands or marls with several harder, brown layers, each about two feet thick and full of the large Ostrea tardensis, with occasional specimens of a Lithopliagtis, form the lowest horizon exposed. 1 Am. Jo2tr. Sci., 4th Sen, Vol. IX., pp. 85-108, Feb., 1900, with a map of southern Patagonia. 2 These geographic features are all shown on the map accompanying the paper just cited. In the descriptions of species the locality labels accompanying the fossils are copied. 3 4 PATAGONIAN EXPEDITIONS PALAEONTOLOGY. 2. The I^nuer conglomerate, 20 feet in thickness, yielded occasional pieces of petrified wood filled with boring Mollusks (Turnus dubius).1 3. The Bclgrano beds, conformably overlying the lower conglomerate, consist of 300 feet of soft greenish sandstones and clays replaced toward the top by several layers of harder sandstones and impure limestones. These beds, especially in the upper portion, are rich in Cretaceous inver- tebrates and at the two localities have yielded all the species described in this paper except those just mentioned. The list of species will be given later. 4. The Upper conglomerates, forming the highest member of the Pueyr- rydon series, consist of 330 feet of hard, fine-grained, red and variegated, sandstone with occasional layers of clay, ending above in a series of very hard fine conglomerates, which have not yielded any identifiable fossils. Resting unconformably on the upper conglomerates are the Variegated sandstones, 1350 feet in thickness, referred to the upper Cretaceous on stratigraphic grounds. Overlying these is a very thick series of Tertiary beds characterized by both vertebrate and invertebrate fossils. From this brief description it is seen that all the Cretaceous fossils under discussion came from the lower 420 feet of the Pueyrrydon series, and a very large proportion of the species were collected from the upper part of the Belgrano beds about the middle of that series. All the speci- mens from the Gio beds and the lower conglomerate were obtained at the locality near the mouth of the canon of Rio Tarde. The Belgrano beds at the same locality yielded the following species : Lima sp., Solecurtus (?) limatus, Pecten (Camptonectes) pueyrrydo- Pleuromya latisulcata, nensis, Corbula crassatelloides, Pecten argentinus, Martesia argentinensis, Pecten octoplicatus, Pleurotomaria tardensis, Avicula (Oxytoma) tardensis, Lunatia pueyrrydonensis, Gervillia hatcheri, Aporrhais (?) sp., Mytilus (?) argentinus, Hatchericeras patagonense, Leda (?) corbuliformis, Hatchericeras argentinense, 1 The specimens of Tubulostium pupoidcs have the same field label as this species but Mr. Hatcher states that the lower conglomerate yielded no other fossils except the wood with boring Mollusks and it is almost certain that the Tubulostium came from either the Gio beds or the Bel- grano beds, as the appearance of the specimens would indicate. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. Trigonia subventricosa, Astarte peralta, Astarte postsulcata, Hatchcriceras (?) tardense, Hatchericeras (?) pueyrrydonense. Nine of these species together with a number of additional ones were obtained at the other locality in the Belgrano beds ten miles east of 'Lake Peuyrrydon. It is evident from an examination of the fossils that the collections from these two localities supplement each other and are from practically the same horizon as Mr. Hatcher determined in the field. The list of species from the second locality is as follows : Pecten (Camptonectes) pueyrry- donensis, Pecten argentinus, Pecten octoplicatus, Mytilus (?) argentinus, Pinna sp., Nucula pueyrrydonensis, Trigonia subventricosa, Trigonia heterosculpta, Trigonia, sp., Astarte postsulcata (?), Tapes (?) patagonica, Tapes (?) sp., Tellina sp., Solecurtus (?) limatus, -Mactra (?) sp., Corbula crassatelloides, Dentalium (Laevidentalium) lima- tum, Vanikoro (?) sp., Lunatia constricta, Aporrhais patagonica, Cinulia australis, Tornatellaea patagonica, Hatchericeras argentinense. Combining the collections from both localities, the Belgrano beds have yielded 36 species of invertebrates that are more or less fully described in the following pages, besides several additional forms represented by material too scanty and imperfect for generic identification. Although the specific names in the above lists are all new, the assem- blage of genera at once proclaims their Mesozoic character, and the ap- pended excellent illustrations, drawn by Dr. J. C. McConnell, give abundant proof of their Cretaceous age. The exact position of the beds yielding them in the Cretaceous system, as developed in Europe and other parts of the world, is more difficult to determine, because the evidence is incomplete and somewhat conflicting. Perhaps the first question to suggest itself is whether more than one great subdivision of the Cretaceous is represented in the three fossiliferous 6 PATAGONIAN EXPEDITIONS PALAEONTOLOGY. horizons that Mr. Hatcher has recognized in the Pueyrrydon series. The fact that the species found in each of the three horizons appear to be peculiar to it would indicate different epochs if each horizon yielded enough species to constitute a real fauna. But the Gio beds yielded only a single species of Ostrea which is occasionally bored by a Litho- pliagns and the lower conglomerate contained only the small Turnus diibins boring in fossil wood. These are all forms that would not seem out of place if immediately associated with the fauna of the overlying Belgrano beds and probably do not differ greatly from it in age, the vertical distribution of species in this part of the section being due rather to local conditions than to great faunal changes. This conclusion is in harmony with Mr. Hatcher's observation that the beds are conformable. In attempting to correlate this series with horizons that have been es- tablished elsewhere the natural course is to begin comparisons with for- mations described in adjacent regions, or at least on the same continent, but the data for direct comparisons are almost wholly lacking. It will, doubtless, be a surprise to the reader, as it was to the writer, to find that no previously described species is recognized in this collection from southwestern Patagonia. This is the more surprising for the reason that the Cretaceous is known to be widely distributed and represented by many horizons in South America. It covers considerable areas in Brazil where it has yielded a large fauna described by White1 but the facies is entirely different from that of the Cretaceous of Patagonia, if indeed, the same horizons are represented. Along the western Cordillera, both Upper and Lower Cretaceous fossils have been described or reported from many areas extending from Venezuela and Colombia to the Strait of Magellan and as Mr. Hatcher's localities are in this western belt, it was naturally expected that many of the species would be referable to de- scribed forms. A careful examination of the literature describing South American Mesozoic fossils failed to reveal a single species with which any of the fossils here described can be positively identified. It should be remembered in this connection, however, that none of the collections previously described was obtained within several hundred miles of Lake Pueyrrydon, and that, with few exceptions, the collections were small and not really representative of the faunas. In a few cases species that I 1 Contributions to the Paleontology of Brazil. Archivos do Museu Nacional do Rio Ja- neiro, Vol. VII, 1887. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 7 have described as new may be represented in previous South American collections by forms referred to or compared with European species. Be- sides searching for similar species in the American Cretaceous faunas, comparisons with the Old World faunas have been as careful and com- plete as the facilities and time at my command would allow, and in many instances related species have been cited, but I do not consider it wise to identify a form with a species described from a region thousands of miles distant, unless the agreement is so complete as to leave no room for doubt as to their identity. Examples of such species that may possibly have been thus cited under other names in previous papers are the Ostrea, some of the Ammonites, Trigonia subventricosa, and the T^lbulosti^lm, all of which will be further discussed. In recent reviews of the geology and palaeontology of the Argentine Re- public by Valentin1 and Ameghino the data for Mesozoic invertebrates are derived almost exclusively from the observations of Steinmann2 in southern Patagonia and the work of Behrendsen3 on the collections of Bodenbender from the western Cordillera between 35° and 40° south latitude. In the neighborhood of Port Famine and the Brunswick peninsula Steinmann recognized the Lower Cretaceous, from which Darwin and others had obtained a few fossils, to which he added an Inoceraimis com- pared with /. concentricus. Two degrees north (about latitude 51°) he found well-preserved examples of an ammonite of the late Cretaceous group of "Haploceraten," which was thought to be identical with an Indian species, and with it Ananchytes, Gastropods and fossil wood. Still farther north between Lake Argentina (near latitude 50°) and Lake Rica the Cretaceous was reported to have more sandstones and more fre- quent fossils, including Inoceramus labiatus and /. brongniarti, or related forms. "From these finds [he states] it is evident that the clay shale system of the eastern slope of the Cordillera, whose thickness may be esti- mated at not less than i ,000 meters and which forms a broad zone from the Strait of Magellan to the lakes of Santa Cruz (perhaps to the latitude of Valdivia [40°] ) belongs to the older and later Cretaceous." ^egundo censo de la Republica Argentina, tomo I, pp. 63-255, Buenos Aires, 1898. 2 Reisenotizen aus Patagonien, Neues Jahrb. f. Min., 1883, Bd. II, pp. 255, 256. 3 Zur Geologic des Ostabhanges der Argentinischen Cordillere. Zeitschr. d. Deutsch. Geol. Gesellsch. Bd. 43, pp. 369-420; Bd. 44, pp. 1-42, 1891-1892. 8 PATAGONIAN EXPEDITIONS PALEONTOLOGY. It seems probable that Steinmann's clay-shale system includes the Pueyrrydon series, but of the few fossils that he cites not one is repre- sented in Mr. Hatcher's collection by similar species, nor even by the same genus, unless it is the ammonite, which is not cited with sufficient definiteness to permit comparisons. The collections studied by Behrendsen came from a region 500 to 800 miles north of Lake Pueyrrydon. From them he determined the pres- ence of several Jurassic and Cretaceous horizons, including Lias, Lower Oolite, Tithonian, Neocomian, Aptian (Gault), and Upper Cretaceous. The small number of species cited from some of the localities and horizons, however, must have made part of these determinations doubtful. At one locality on the Arroyo Pequenco, between Rio Salado and Rio Malargue, strata referred to the Upper Neocomian yielded only Exogyra con/oni, fragmentary remains of a Crustacean, a poorly preserved Trigonia that is compared with T. aliformis, a Rhynchonella, and Mytilus cumeri. The fact that some of the European forms referred to E. couloni differ so much from the typical form that they may be compared with Ostrea tardensis, suggested that the latter may be the form referred to E. couloni by Behrendsen, but his description and the figures to which he especially refers for comparison indicate a very different form. A poorly preserved specimen of Trigonia subventricosa might also reasonably be compared with T. aliformis. There is a possibility therefore that the Pequenco horizon may be the same as a part of the Pueyrrydon series. The Aptian or Gault of Behrendsen's paper is represented by a few fossils from " Portezuelo de Carqueque," some distance north of the last mentioned locality. The only forms listed from this horizon are Am- monites sp., Ostrea sp., Pecten sp., and Serpula phillipsi Roemer, the last named species evidently being the basis of the determination of the age of the bed. The Ostrea is a mere unidentifiable fragment and the de- scriptive notes on the Ammonites and the Pecten make it certain that they can not be identical with any of the Pueyrrydon species. Serpula pJril- lipsi, however, as described and figured from the Gault of England and Germany has considerable resemblance to the form I have described as Tnbnlostinm pitpoides, though for reasons pointed out in the description of the latter species it is not considered identical nor even congeneric. It is mentioned here merely as an example of possible identity that may some time be established by comparison of specimens from the two re- STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 9 gions. Even if the Pueyrrydon species should prove to be identical with the form from Portezuelo, which according to Behrendsen shows no es- sential difference from the European species, I should not consider that in itself sufficient proof of the Gault age of the beds containing it. One other of Behrendsen's horizons, " the Neocomian of the Arroyo Triuguico and of Quili Malal," affords a slight basis of comparison. Two of the Ammonites, Hoplites desori and H. neumayri, are somewhat sug- gestive of two of the forms I have described as species of Hatchericeras and they are referred to, or compared with some of the same European species mentioned in my descriptions. It is evident, however, that none of the forms figured by Behrendsen is identical with or very closely re- lated to any of the Pueyrrydon ammonites. Several other genera are represented in both this Neocomian and the Pueyrrydon fauna but the species are not identical. Comparisons made with Cretaceous fossils described from other regions along the Andes from Chili northward gave similar negative results, and there are no indications of close relationship with North American Creta- ceous faunas that are worthy of mention. The very close resemblance of one of the Pueyrrydon species of Tri- gonia to T. ventricosa Krauss from the Uitenhage beds of South Africa led to a close examination of the fauna of that formation, and while another of its species, T. van, proved to be related to a Pueyrrydon form, and there are superficial resemblances in the species of Gervillia, Astarte and Ostrea, the fauna as a whole is too different to permit definite correlation. The Uitenhage beds are now generally referred to the Lower Cretaceous, though they were formerly assigned to the Jurassic. As it is impossible to correlate the Pueyrrydon series by means of identical species, it is necessary to rely on more general comparisons, at- tempting to give due weight to the somewhat conflicting evidence. Ammonites and Trigonias are usually among the most trustworthy groups in determining the age of beds. In this case it has been thought necessary to refer all the species of ammonites to a new genus whose re- lationships are not very firmly established, but nearly all the species with which these forms are compared occur in the Lower Cretaceous, and as- suming that the genus is derived from Hoplites, or its near relatives, the stage of development observed is what one would expect to find at that period. Certainly no such forms are known later than the middle of the Cretaceous. IO PATAGONIAN EXPEDITIONS PAL/EONTOLOGY. Trigonia subventricosa belongs to a purely Cretaceous section which has similar forms in the Upper Cretaceous, but the .most closely relatep form is T. ventricosa from the Lower Cretaceous Uitenhage beds, which also yield a form similar to the other Pueyrrydon species T. heterosculpta, though another species which may belong to the same section occurs in the Upper Cretaceous Quinquina beds of Chili. The evidence of both Ammonites and Trigonias, therefore, seems to favor the Lower Cretaceous age of the series. The specific characters of some of the other groups, such as the Ostrea, the Gervillia, the Astarte species, the Pleuromya, and the Solecurtus (?) also tend to place the beds below the middle of the Cretaceous. Certain of the other forms, such as Cinulia, Tornatellaea, Lunatia, Martesia, Turnus, Mactra (?), and Pinna, have a more modern aspect and would not be out of place in an Upper Cretaceous fauna, while the remainder are mostly of types that have a greater range within the Mesozoic. After my preliminary examination of this collection the opinion was expressed1 that the horizon represented is "about the middle of the Cre- taceous, at least not lower than the Gault." This judgment was influ- enced to some extent by the supposed occurrence in the collection of the characteristic Upper Cretaceous genus Pugnellus. Further study of the material after it was better cleaned showed that this generic identification was incorrect, and the closer examination of the collection has in several cases tended to emphasize the evidence for older rather than newer Cre- taceous age. The former opinion is, therefore, now modified to this ex- tent, that the horizon is not later than the Gault. Although the evidence as above sketched and as given more in detail in the specific descriptions does not seem to me to justify the definite reference of the Pueyrrydon series to any one of the European Cretaceous horizons it is reasonably certain that it belongs within the Lower Cretaceous and is not younger than the Gault. 1 Quoted by Hatcher, Am. Jour. Sci., 4th Ser., Vol. IX, p. 90, and published in abstract of communication to Geological Soc. of Washington, Science, N. S., Vol. XI, p. 349, 1900. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. I I PELECYPODA. OSTREA TARDENSIS Sp. HOV. PI. I, Figs. I and 2, and PI. II, Figs, i and 2. Shell large, massive, subtriangular, or more or less crescentic in outline ; lower valve very thick and very convex, obscurely carinate, with the beak more or less twisted laterally, but not distinctly coiled, the lower third of the valve also laterally curved, so as to give the shell its crescentic shape ; upper valve thinner, flat or somewhat concave, with the beak more nearly straight and the other end curved to fit the lower valve ; surface of both valves with rather coarse concentric lamellae or imbrications, though on most of the specimens in the collection these are obscured by weathering. Some individuals also show obscure radiating plications. The pit or groove for the ligament is very large and broad and only slightly curved in both valves. An average specimen measures 150 mm. from beak to base and 90 mm. in greatest breadth at right angles to that line. Convexity of the two valves 53 mm. The eight specimens in the collection show some variations in form and proportion, but not more than species of Ostrea usually show. The curved form gives it the appearance of an Exogyra, but the beaks, espe- cially of the upper valve, lack the spiral form characteristic of Exogyra. The fact that Exogyra couloni Defrance has been reported from several localities in South America led to the comparison of that species with our Patagonian form. While it is true that some extreme varieties that have been figured as belonging to that very variable European Lower Creta- ceous species somewhat resemble the Patagonian shell in form and gen- eral appearance, it does not seem to me that they can be identical. O. tardensis is certainly not at all like the typical and ordinary forms of E. couloni. Behrendsen1 reports the occurrence of E. couloni at "Arroyo Pequenco between Rio Salado and Rio Malargue, not far from the Villa Beltran," Argentine Republic, and states that it is present in almost all the forms represented by d'Orbigny and Pictet, but agreeing especially with the figures given by Bayle and Coquand2 of specimens from Chili. 1 Zeitschr. Deutsche Geol. Gesellsch., Bd. 43, p. 419, 1890. 2 Mem. Soc. Geol. de France, 2d Ser., t. IV, p. 37, pi. 7, figs. I and 2, 1851. 12 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. These Chilean specimens are distinctly exogyrate and have very little re- semblance to the Patagonian form. In a later work by Coquand * they are referred to Ostrea a<]iiila, which is also made to include Exogyra im- bricata Krauss, a similar form from the Uitenhage beds of South Africa. A species that resembles O, tardensis more closely than those above mentioned has been described by Stoliczka2 under the name Gryphaa an'aua from the Arrialoor group of the Indian Cretaceous. Judging from the figures the principal difference is in the somewhat less strongly curved form of G. ariana. It is possible, though I think not probable, that the Patagonian Ostrea here described as new is specifically identical with some of the South American forms that have been referred to Exogyra coulom, but even if that should prove to be true, the identification of that species is believed to be erroneous in those cases and a new name is necessary. Locality and position. — Mouth of canon of Rio Tarde, four miles east of Lake Pueyrrydon, from a horizon (Gio beds) 400 feet below the Ammo- nite layer. LIMA sp. Shell small, ovate, slightly oblique, convex with prominent beaks and small inconspicuous ears; surface marked by about 15 rather prominent radiating ribs about equal in width to the interspaces, each of which bears a fine line. The anterior and posterior portions of the shell not covered by the ribs also bear a number of fine radiating lines and show irregular lines of growth. The"three imperfect specimens in the collection are probably immature, the largest measuring only seven millimeters in length and five milli- meters in breadth. Locality and position. — From the Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon. PECTEN (CAMPTONECTES) PUEYRRYDONENSIS sp. nov. Pi. IV, Fig. i. Shell of medium size, ovate or subcircular in outline, moderately con- vex ; right valve with unequal ears, the anterior one much larger, trian- 1 Monographic du Genre Ostrea, p. 158. 'Cretaceous Fauna of S. India, vol. 3, p. 465, pi. 43, figs. 2, 2a, pi. 44, figs. 1-3. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 13 gular with rounded extremity and bounded below by a deep byssal sinus, posterior ear much smaller, with the outer margin oblique and broadly rounded above, both with radiating striae ; the body of the shell marked by rather conspicuous irregularly spaced concentric lines and by very fine curved radiating impressed lines. Height, 25 mm.; length of hinge line 11 mm.; greatest length (about the middle of the valve) 23 mm. This description is drawn from the type, a well preserved right valve, from the locality ten miles east of Lake Pueyrrydon. The collection from four miles east of the same lake contains some less perfect specimens be- lieved to belong to this species, and among them are two left valves hav- ing the same sculpture and general form as the type. These are slightly more convex than the right valve and the anterior ear is similar in form and only slightly larger than the posterior. The species has the typical Camptonectes sculpture and form as seen in a number of described Jurassic and Cretaceous species, but according to Dall l Camptonectes is not of generic or even subgeneric rank but should be placed as a section under Pseudamusium. Locality and position. — From the Ammonite (Belgrano) beds at the two localities above mentioned. PECTEN ARGENTINUS sp. nov. PL iv, Fig. 5. Shell of moderate size, ovate in outline, very gently convex, with me- dian pointed beaks ; ears subequal, in the form of slender right-angled triangles, projecting beyond the beak at their outer angle, separated from the body of the shell by impressed lines that form a right angle where they intersect at the beak ; basal margin forming almost a semi-circular curve ; surface smooth and polished, with very fine lines of growth, occa- sional more prominent impressed concentric lines, and very faint indica- tions of fine radiating lines. The figured specimen, which is of average size, is 23 mm. in height and 2 1 mm. in greatest length. The corresponding dimensions of the largest example in the collection are about twice as great. The convexity of sin- gle valves is not more than two to three millimeters. 'Trans. Wagner Free Institute, Vol. Ill, pt. IV, p. 697, Philadelphia, 1898. 14 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. The type is apparently a left valve and the right valve is probably very similar with subequal ears and without a byssal notch, since there are about twenty such valves in the collection, and there are no others of different form that could be right valves of this species. . The form is like that of the group for which Meek proposed the name Entolium, but it lacks the diverging "teeth" characteristic of that subgenus. Similar forms are not uncommon in the Jurassic and Cretaceous, such as Pecten operculifonnis Gabb from the Lower Cretaceous Horsetown beds of California, which, however, has broader ears and no radiating striae. Locality and position. — From the Ammonite (Belgrano) beds at mouth of cafion four miles east of Lake Pueyrrydon, and ten miles east of the lake. The figured specimen is from the latter locality. PECTEN OCTOPLICATUS sp. nov. PI. IV, Figs. 2 and 3. Shell small, ovately subtriangular, conspicuously inequivalve, with eight strong radiating ribs on each valve. Right valve very convex, with prom- inent narrowed beak ; anterior ear elongate-triangular, with a deep byssal notch beneath it ; posterior ear much smaller and inconspicuous ; ribs very prominent, about as broad as the interspaces, subangular, each bearing several obscure radiating lines and crossed by numerous fiae concentric lines. Left valve much less convex, with subequal, rather broad triangu- lar ears ; ribs more rounded, much less elevated and relatively more nar- row ; the radiating and concentric lines also less conspicuous than on the right valve. Height of an average specimen from beak to base 9 mm.; greatest length at right angles to above measurement 8 mm.; convexity of right valve about 4 mm., of left valve about 2 mm. Locality and position. — Abundant in the Ammonite (Belgrano) beds at mouth of cafion four miles east of Lake Pueyrrydon and represented by three left valves apparently belonging to the species from the same ho- rizon, ten miles east of the lake. AVICULA (OXYTOMA) TARDENSIS sp. nov. PI. IV, Figs. 6 and 7. Shell small, obliquely ovate, inequivalve ; length of hinge line not quite equal to height of shell ; beaks rather prominent, extending beyond the STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 15 hinge line. Left valve convex; anterior ear rounded, inconspicuous; posterior ear broad, flattened and more or less mucronately produced at the extremity, though this is not preserved on most of the specimens; surface of the whole valve, except the ears, with 25 to 30 fine radiating ribs, which tend to vary in size on the posterior portion. Right valve con- siderably less convex, with a deep byssal sinus under the anterior ear; posterior ear broad and flattened ; surface marked by much finer and more numerous radiating lines than those on the left valve. An average specimen measures 1 1 mm. in height and about the same in greatest length, which is below the middle of the valve. The convex- ity of the two valves united is about 5 or 6 mm. The species is similar in form and sculpture to A. nebrascana Evans and Shumard, as figured by Meek,1 from the Fort Pierre formation of the western United States. The subgenus Oxytoma ranges throughout the Mesozoic and is by some authors referred to Pseudomonotis instead of Avicula. Locality and position. — Abundant in the Ammonite (Belgrano) beds at mouth of canon of Rio Tarde four miles east of Lake Pueyrrydon. GERVILLIA HATCHERI sp. nov. PI. Ill, Figs, i and 2. Shell very large, slender, obliquely produced, rather convex, with very thick test ; beaks terminal, pointed, not conspicuous ; ventral margin nearly straight from the beaks for about one-third the length of the shell, then broadly convex to the rounded posterior end ; dorsal margin slightly concave from the hinge to the posterior end of the shell ; posterior wing not preserved on the type but evidently narrow ; hinge area almost par- allel with the longer axis of the shell, broad with seven or eight large transverse pits or grooves for the ligament, and three or four obscure, ob- liquely elongated teeth ; surface marked only by lines of growth. Length 260 mm.; greatest breadth (at the beginning of the posterior third of the shell) 64 mm.; greatest convexity of single valve about 28 mm., and thickness of test in anterior portion about the same. The only specimen in the collection is the left valve figured, which is in an excellent state of preservation, except that nearly all the posterior wing 'U. S. Geol. Surv. Terr., Vol. IX, p. 34, pi. 16, figs, ja, 3b. 1 6 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. is broken off and some of the hard, sandy matrix still clings to a part of the surface. The species may be compared with G. alpina Pictet and Roux,1 which is almost as large and somewhat similar in form, but is a straighter shell and has a more distinct anterior wing. Gervillia dentata Krauss 2 from the Uitenhage beds of South Africa is another large species that has some general resemblance to this form, but it is not so thick shelled, the beak is not so pointed and the shell is straighter. Locality and position. — From the Ammonite (Belgrano) beds at mouth of cafion four miles east of Lake Pueyrrydon. MYTILUS (?) ARGENTINUS sp. nov. PI. IV, Fig. 4. Shell small, elongate-ovate, moderately convex; beaks prominent, slightly incurved ; dorsal margin slightly convex without definite posterior angulation ; ventral margin nearly straight, posterior end regularly and broadly rounded ; surface marked only by very fine growth lines and con- centric wrinkles, which are inconspicuous except when magnified. Length of an average specimen from beak to base 8 mm.; greatest breadth 5.5 mm.; convexity of the two valves about 6 mm. This little species has almost the form of a Crenella, and some obscure radiating lines on a weathered specimen increased the resemblance and led to its reference to that genus, when the collection was first examined. Well-preserved specimens, however, show no radiating sculpture. Locality and position. — The figured type and four other specimens are from the Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon, and one specimen from the same horizon ten miles east of the lake. LlTHOPHAGUS Sp. This genus is represented in the collection by a single small flask-shaped burrow in a fragment of oyster shell from the Ostrea horizon (Gio beds) 400 feet below the Ammonite layer at the mouth of the canon four miles 'As figured by Pictet and Campiche, Terrain Cretace de Sainte Croix, pi. 155, figs. 2-4. According to these authors the species includes the Cretaceous forms figured by Sowerby (Min. Conch., pi. 511) under the name G. aviculoides but not belonging to that Jurassic species. 2 Nova Acta, Vol. 22, p. 458, pi. 50, figs, \a-\c. STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 1J east of Lake Pueyrrydon. The burrow, which is rilled with calcareous sand and probably still contains the shell, measures 10 mm. in length and 5 mm. in greatest diameter. PINNA sp. A single small Pinna in the collection is too imperfect for specific de- scription. The shell is rather slender, moderately convex, not carinate, nor distinctly angular on the median line, with the upper two thirds of the shell marked by 10 or 12 radiating lines and the rest of the surface bear- ing irregular, less conspicuous lines of growth. The specimen, which is probably not a mature shell, measures 90 mm. in length, 28 mm. in great- est breadth and 10 mm. in convexity of both valves. Compared with P. robinaldina d'Orbigny, which Behrendsen1 has re- ported from the Lower Cretaceous of the Argentine Republic, this species may be easily distinguished by its more slender and less convex form and by the absence of a median angulation. It much more closely resembles the Upper Cretaceous P. lakesi White 2 from the Fort Pierre formation of Colorado. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon. NUCULA PUEYRRYDONENSIS Sp. nOV. PI. IV, Figs. 8 and 9. Shell of medium size, elongate subovate, moderately convex; beaks nearly terminal at the posterior end, which is almost vertically truncate ; dorsal margin slightly convex ; anterior end broadly rounded, most promi- nent above, passing below into the convex ventral margin ; surface nearly smooth, bearing very fine, closely arranged lines of growth, with a few more conspicuous concentric furrows at wide intervals. Length 17 mm.; height 12 mm.; convexity of the single valve about 5 mm. The species is quite similar in form and surface to N. simplex Des- hayes, from the Neocomian of France, as figured by d'Orbigny,3 except that its anterior end is more broadly rounded. 1 Zeitschr. Deutsche Geol. Gesellsch., Bd. 44, p. 25, 1892. 2 12th Ann. Rep. U. S. Geol. and Geog. Surv. Terr., pt. I, p. 17, pi. II, fig. i, 1880. •'Paleont. Franq. Terr. Cret, t. Ill, pi. 301, figs, u and 12. 1 8 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon, represented by a left valve, an internal cast, and the imprint of the exterior of another one. LEDA (?) CORBULIFORMIS sp. nov. PI. IV, Fig. ii. Shell small, elongate-ovate, rather ventricose, with prominent, tumid submedian beaks ; dorsal margin descending almost equally before and behind the beaks ; anterior end broadly rounded ; posterior end more narrow, subtruncate and very slightly upturned at the extremity ; ventral margin gently convex ; surface marked by relatively coarse, regular con- centric lines. The type measures 10 mm. in length, 7 mm. in height and 6 mm. in convexity of the two valves. The description is drawn from a single specimen which retains most of the shell of the left valve and shows the internal cast of the right. Im- pressions of small nuculoid hinge teeth are faintly shown. The reference to Leda is based on the form of the shell and is somewhat doubtful. Locality and position. — Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon. TRIGONIA SUBVENTRICOSA sp. nov. PI. IV, Figs. 19 and 20. Shell rather large, sublunate, inflated and broadly rounded in front, much contracted and considerably produced behind ; beaks prominent, near the anterior end ; ventral margin very convex and prominent in its middle third, where the strongest costae terminate, in front passing imper- ceptibly into the anterior margin by a more gentle curve, and behind passing almost straight or with a slightly concave curve toward the pos- terior end ; dorsal margin concave ; area rather narrow, convex, divided above the middle by a narrow groove, with inconspicuous lines of growth, not bounded by distinct carinae ; escutcheon also nearly smooth, large, deeply excavated, bearing only faint costellae, that near the front are very short, directed backward very obliquely from the area, while toward the posterior end they become somewhat larger, more nearly transverse, pass- ing entirely across the escutcheon and sometimes curving forward near STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 19 the margin ; remainder of the shell bearing about 22 to 24 strong costae, radiating from the margin of the area and divided into two rather distinct sets. The anterior 10 or n are very strong, distant, coarsely and irregu- larly tuberculate, curved forward near the margin of the shell, occupying the inflated anterior two thirds. Successive costae become larger and more nearly straight, the Qth or loth usually being the largest. The other costas on the contracted posterior third of the shell are much finer, more closely arranged and nearly straight or slightly irregular and sinuous, without tubercles, and directed obliquely downward and backward. The surface also bears rather conspicuous, closely arranged lines of growth. The figured type, which has lost a small portion of the posterior end, measures 73 mm. in length, 63 mm. in height, and about 57 mm. in con- vexity of the two valves. The corresponding dimensions of another specimen are 82 mm., 66 mm., and 60 mm., respectively. In each case the length is measured from the front margin to the posterior end, and the height somewhat obliquely from the beak to the most prominent part of the ventral margin. This species belongs to the section Scabrae, which is characteristic of the Cretaceous, and it is somewhat closely related to T. aliformis Park- inson. The form which most closely resembles it, however, is T. ventri- cosa (Krauss),1 from the Uitenhage beds of South Africa. Comparisons have been made with some small specimens collected by Dr. Holub on Zwartkop river, as well as with the published figures, and while the general resemblance is very great, the Patagonian form differs in being somewhat longer and less inflated, and the tubercles on the anterior ribs are coarser, less regular and more distant. T. tuberculifera Stoliczka from the Upper Cretaceous Trichinopoly beds of southern India is also similar in general form and sculpture, but it is still shorter than T. veutri- cosa and the costae on the posterior portion are coarser and not so nu- merous. In beds referred to the Upper Neocomian at Arroyo Pequenco, Argen- tine Republic, several hundred miles north of these Patagonian localities, Dr. Bodenbender collected a Trigonia, listed by Behrendsen2 as Tri- 1 Nova Acta Acad. Caes. Leopold-Carolin. Nat. Cur., Vol. 22, p. 456, pi. 49, figs 2a-2d. Bet- ter figures have been published by Lycett in Brit. Foss. Trigoniae, p. 119, and by Stoliczka in Cret. Fauna of S. India, vol. 3, pi. 15, figs. 9, 9«. 2 Zeitschr. Deutsche Geol. Gesellsch., Bd. 43, 1891, p. 418. 20 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. gonia conf. aliformis Park., which may possibly be identical with our spe- cies, but the single specimen collected was too imperfect for full description. It was associated with an oyster identified with Exogyra couloni Defr. Locality and /05///cw.— Represented by 13 specimens from the Ammo- nite (Belgrano) beds at mouth of cafion four miles east of Lake Pueyrrydon and by four specimens from" the same horizon ten miles east of the lake. TRIGONIA HETEROSCULPTA sp. nov. PI. IV, Figs. 1 6-1 8. Shell rather small, ovately trigonal, moderately convex ; anterior end broadly rounded, rather prominent ; ventral margin gently convex ; pos- terior end slightly produced, narrowly rounded at the extremity and very obliquely subtruncate above ; dorsal margin almost straight (very slightly concave) from the beak to the posterior end ; beaks not very prominent, situated about one third the length of the shell from the anterior end ; area and escutcheon narrow, not very sharply defined on the adult shell, both being destitute of sculpture other than rather prominent lines of growth and a broad furrrow above the middle of the area. The sculpture of the young shell, as seen on the beaks and on small specimens, is entirely different from that of the adult, and the form also is different. In young shells, seven millimeters or less in length, the height and length are about equal and the sculpture consists of strong concentric ribs, parallel to the growth lines and about as prominent on the area as on the anterior por- tion of the shell. At this stage the area is bounded above and below by well-marked carinae and the small escutcheon is smooth. As the shell grows, the next two or three ribs become swollen and bent downward a short distance in front of the area, then one or two form V-shaped angles there, and finally in the adult form of sculpture there are two distinct sets of costae — one set consisting of slender, smooth costae resembling those of the young shell, but ranging obliquely backward and downward from the front of the shell until they almost meet the other set, the number varying from 6 to 12 or more according to age; the posterior set consist- ing of fewer and larger, more nearly vertical, smooth costae that range downward from the margin of the area to the ventral border. The sculpture is thus seen to agree closely with that of the Undulatae section of Trigonia, except that the area is not so well defined. The STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 21 largest specimen illustrated is so badly weathered that the anterior set of costae has been almost obliterated, and the figure is therefore misleading in that respect. The other figures, though smaller, show the sculpture better. The largest specimen in the collection (PI. IV, Fig. 18) measures 51 mm. in length, 38 mm. in height and about 19 mm. in convexity of the single valves. The corresponding dimensions of another specimen are 29 mm., 22 mm. and about 9 mm., respectively. The only American Cretaceous Trigonia that resembles this one in gen- eral appearance is T. hanetiana d'Orbigny,1 from the Quinquina beds of Chili, which may be easily distinguished by differences in the sculpture, the posterior set of costae radiating from the beak instead of from the margin of the area, and tending to break up into large irregular tubercles toward the ventral margin. T. vau Sharpe2 from the Uitenhage beds of South Africa is more nearly related, as its sculpture is of the same type. It differs, however, in outline, the posterior end being more prolonged and broader, and the anterior ribs are more oblique, while the posterior set also have a different inclination. T. robinaldina d'Orbigny,3 a French Neocomian species, has somewhat similar sculpture but differs in outline and is more convex. T. heterosctilpta is not easily assigned to any of the described sections of Trigonia. The costae are similar to those of the Undulatae but the ill- defined area and escutcheon seem to prevent its reference to that group, which is said to be characteristic of the Jurassic. Locality and position. — From the Ammonite (Belgrano) beds ten miles east 'of Lake Pueyrrydon, represented by about 20 valves. TRIGONIA sp. Associated with the preceding from the locality ten miles east of Lake Pueyrrydon, are two specimens evidently belonging to a distinct species of Trigonia, but too imperfect for specific description, as they do not show the character of the posterior end, area and escutcheon. The form is rather ventricose and short and the anterior portion of the shell bears nine 'Voyage dans 1'Amerique Mend., t. Ill, pt. 4, p. 127, pi. 12, figs. 14-16. The species is also figured by Steinmann, Neues Jahrb. f. Min., etc., Beilageband X, p. 101, pi. 7, figs. 8, 9. 2 Trans. Geol. Soc. Lond., 2d Sen, Vol. VII, p. 194, pi. 22, fig. 5. 3Paleont. Franc.. Terr. Cret, t. Ill, pi. 299, figs, i, 2. 22 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. or ten large smooth ribs, that radiate from the beak and are broader than the interspaces. ASTARTE PERALTA Sp. nOV. PI. V, Figs, i and 2. Shell very large and massive, ovate, moderately convex, considerably higher than long, with prominent submedian beaks, from which the outline descends rapidly and almost equally, with a gentle convex curve behind, and concave in front to the extremity of the lunule, which is rather large and deeply excavated with abrupt walls ; ventral margin regularly rounded; surface marked by fine lines of growth and by rather prominent concen- tric ridges, which are less prominent on the middle of the valve than toward the ends, and become broader and less conspicuous toward the ventral margin of the adult; hinge broad and massive, showing in the right valve a small anterior cardinal tooth, a very large middle cardinal, two large sockets for the reception of the cardinals of the left valve, and traces of anterior and posterior laterals ; posterior cardinal not de- veloped. The free border of the shell is not crenulated on the interior. Height 125 mm., length 101 mm.; convexity of both valves about 80 mm. This description is drawn from a well-preserved right valve showing all the essential specific and generic features, though a fragment has been broken from the posterior end. There is another imperfect specimen showing the beak of a left valve. The species is distinguished by its large size and its great height, as compared with its length. In these features it recalls some of the gigan- tic species of Astarte from the Jurassic, such as Astarte damesi Boehm,1 though differing from them in outline and other details. The growth lines of A. peralta show that its height, as compared with the length, increases rapidly with age and these two dimensions are equal at least until they reach 25 mm. At that size the shell resembled A. Postsulcata, except that the sculpture was considerably coarser and the an- terior end was less prominent. Locality and position. — From the (Belgrano) Ammonite beds at mouth of cafion four miles east of Lake Pueyrrydon. 1 Palaeontographica, Supplement II, Die Bivalven der Stramberger Schichten, p. 561, pi. 63, figs. 1-3. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 23 ASTARTE POSTSULCATA Sp. HOV. PL V, Figs. 3-7. Shell of medium size, subcircular in outline, moderately convex, sub- equilateral, the anterior end of the shell being slightly more prominent than the posterior; beaks prominent, angular; lunule rather large, deeply excavated, with abrupt walls ; surface marked by rather coarse lines of growth, by occasional more prominent concentric furrows and by a broad, shallow depression or furrow extending in a curve from the beak to the postero-ventral margin : hinge with two well-developed cardinal teeth in each valve, and both anterior and posterior laterals considerably devel- oped. Margin of the shell not crenulate within. An average specimen measures 25 mm. in height, 26 mm. in length and 1 8 mm. in convexity of the two valves. Shells of this size have comparatively thick massive shells, as shown in figures 5, 6 and 7, while other specimens only slightly smaller and agreeing in all other respects, have very much thinner shells. This comparative thinness of test is be- lieved to be due to immaturity. This species belongs to the same group as A. peralta, the young of which it evidently closely resembles, but it may be distinguished even from specimens of the same size by its much finer lines of growth, by the pres- ence of the radiating posterior furrow, and by slight differences in outline, especially the greater prominence of the anterior end. Locality and position. — Represented by over 20 specimens from the Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon. TAPES (?) PATAGONICA sp. nov. PI. IV, Figs. 12 and 13. Shell small, rounded subquadrate, moderately convex ; beaks promi- nent, situated on the anterior third of the shell ; dorsal margin excavated in front of the beaks, gently convex behind them ; anterior end broadly rounded, passing imperceptibly into the convex ventral margin ; posterior end broad, obliquely subtruncate ; surface marked by rather coarse, regu- lar concentric lines and grooves, with occasional deeper furrows, the sculpture being strongest on the middle of the valve and fading out to- ward the ends. There is no distinct lunule and the narrow escutcheon is about half filled by the ligament. 24 PATAGONIAN EXPEDITIONS \ PAL/EONTOLOGY. Length of the type, 17 mm.; height, 15 mm.; convexity of the two valves, 8 mm. Besides the well-preserved type, the species is represented by a cast of one valve and possibly by two other imperfect valves, that are referred with some doubt to the species. The hinge and other internal characters are unknown and the generic reference is, therefore, very uncertain. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon. TAPES (?) sp. Associated with the preceding species at the same place and horizon is another form of about the same size, but considerably more elongate, with the beaks nearer the anterior end, less convex and with smoother surface. The pallial sinus is deep, rather broad and horizontal. Represented by a nearly complete internal cast retaining small portions of the shell and by the imprint of a part of the surface of the same individual. TELLINA sp. A single small internal cast retaining small portions of the shell is re- ferred to this genus on account of its form and general aspect. The valves are slightly twisted laterally ; anterior end broadly rounded ; pos- terior end obliquely subtruncate, much more narrow and somewhat shorter, than the front end ; ventral margin slightly convex ; beaks rather prominent ; surface apparently smooth. Length, 21 mm.; height, 13 mm.; convexity of the two valves, 5 mm. Locality and position. — Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon. SOLECURTUS (?) LIMATUS Sp. nOV. PL IV, Fig. 10. Shell small, elongate oval, subequilateral, with very small, inconspicu- ous beaks ; dorsal and ventral margins very slightly convex ; anterior end regularly rounded ; posterior end slightly contracted, broadly rounded into the ventral margin below, more abruptly rounded or subangular above ; surface with a polished appearance but marked by numerous fine growth lines that vary somewhat in size. Internal casts show imprints of two very small teeth and subequal STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 2$ ovate adductor muscle impressions. The pallial line is very faint and the sinus not distinctly seen. Length of an average specimen 20 mm.; height 9 mm.; convexity of both valves about 4 mm. This species closely resembles Solen cequalis d'Orbigny 1 which the same author afterward referred to Solecurtus. It is smaller than that species and not quite so slender. Locality and position. — The figured specimen is from the Ammonite (Belgrano) beds 10 miles east of Lake Pueyrrydon. It also occurs at the locality four miles east of the lake, and is represented by about a dozen specimens from the two places. PLEUROMYA LATISULCATA sp. nov. PI. VI, Figs. I and 2. Shell of medium size, oblong ovate, or subcuneate in form, the great- est convexity near the front end midway between the beak and the ven- tral margin ; anterior end broadly rounded, posterior end more narrowly rounded and slightly gaping ; ventral margin very slightly convex ; dor- sal margin descending abruptly in front of the beaks and rather rapidly behind them ; beaks prominent, approximate, somewhat flattened, situated near the anterior end of the shell ; a rather prominent broad ridge, on which the sculpture is most strongly marked, extending from the beak al- most vertically to the antero-ventral angle, a much more obscure pos- terior umbonal ridge running obliquely to the posterior end, and the side between these two ridges flattened ; surface marked by very prominent concentric ribs separated by broader furrows, and the whole covered with finer growth lines. Length, 38 mm. ; height, 25 mm. ; convexity of the two valves, 20 mm. Locality and position. — From the Ammonite. (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon. Represented by a single specimen. MACTRA (?) sp. Shell rather small, elongate, moderately convex, with prominent beaks slightly in advance of the middle ; anterior end broadly rounded ; posterior end narrow, obliquely subtruncate; ventral margin gently convex: an 'Pal. Fr. Terr. Cret, t. 3, p. 321, pi. 350, figs. 5-7. 26 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. angular umbonal ridge extending near and almost parallel with the dorsal margin from the beak to the posterior end ; surface marked by regular, comparatively rather coarse, concentric lines and furrows ; hinge and other internal features unknown. Length of the best preserved specimen 28 mm.; height 16 mm.; con- vexity of single valve about 5 mm. Another fragment indicates a speci- men with corresponding measurements about one third greater. The material is too imperfect and fragmentary for a satisfactory specific description or figure and as the hinge is unknown no definite generic ref- erence can be made. The form, sculpture, and other features observed in- dicates that it belongs to the Mactridae though probably not to Mactra in the restricted sense as defined by Dall. It has a greater resemblance to Spisula. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon. CORBULA CRASSITELLOIDES Sp. nOV. PI. IV, Figs. 1 4 and 15. Shell small, subtriangular, rather convex ; beaks prominent, broad, sit- uated slightly in advance of the middle ; dorsal margin descending almost equally forward and backward from the beaks ; ventral margin convex ; anterior end broadly rounded ; posterior end obliquely subtruncate ; um- bonal ridge distinct, angular, descending obliquely to the postero-ventral angle ; surface marked by regular fine concentric lines. Length of an average specimen, 5.5 mm.; height, 4 mm.; convexity of the two valves, 4 mm. This abundant little species resembles C. bodenbenderi Behrendsen1 from the Neocomian of Arroyo Triuguico, Argentine Republic. It is somewhat smaller than Behrendsen's species, more inequilateral, more convex, and the posterior end is relatively broader. In general form it resembles the young of some species of Crassatella. Locality and position. — Represented by numerous specimens from the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon and at mouth of canon four miles east of the same lake. 'Zeitschr. deutsch. geol. Gesellschaft, Bd. 44, 1892, p. 19, pi. 3, figs. 6a-6d. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 27 MARTESIA ARGENTINENSIS sp. nov. PI. VI, Figs. 3 and 4. Shell small to medium in size, elongate ovate or cuneate, very inflated, with a spherical aspect from the front, regularly tapering to the posterior end, which gapes rather widely, anterior hiatus shield-shaped, closed by a callum ; a single broad accessory umbonal valve present (form not accu- rately determined), other accessory valves unknown ; surface marked by fine regular lines of growth, parallel with the margins of the shell, and by two faint umbonal furrows that diverge slightly in passing obliquely from the beaks to the ventral margin just behind its- most prominent portion, and are more prominent on internal casts than on the shell itself; the burrows in fossil wood, in which the shells are found, not lined with a calcarous shell. An average specimen measures 13 mm. in length; 7.5 mm. in height, and 8 mm. in greatest convexity. The two umbonal furrows in this species suggest its reference to Para- pholas, which is represented in the Cretaceous by somewhat similar forms, but that genus has paired umbonal valves, or a single one formed of two fused pieces. Locality and position. — From the Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon, represented by a dozen specimens, most of which are immature. TURNUS DUBIUS sp. nov. PI. VI, Figs. 5-8. Shell small, subglobose, or broadly ovate, gaping widely behind, and in front with abroad, shield-shaped hiatus, which in some specimens seems to be filled by a callum, its upper corners almost rectangular; beaks prominent, approximate, in front of the middle of the shell ; dorsal margin nearly straight, posterior end and venter very broadly rounded ; umbonal groove slightly oblique, narrow, inconspicuous, somewhat more prominent on internal casts than on the exterior, extending from the beak to the most prominent part of the ventral margin ; surface also marked by very fine, closely arranged, regular lines of growth that cross the umbonal fur- row obliquely and are sharply bent upward in front of it, parallel with the margin of the anterior gape. 28 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. The internal cast also shows a deep groove, corresponding to a heavy internal rib, extending back from the beak to the posterior margin above the umbonal ridge, and some specimens show a much fainter furrow in the same position on the exterior of the shell. On one specimen near the beak there is a small subtriangular structure that appears to be an accessory valve (protoplax) and indicates by its form that there were two of them, as in Xylophaga. The animal burrowed in wood, forming long more or less tortuous shelly tubes like those of Teredo, the surface of the tubes bearing irregular annular wrinkles, or lines of growth. Length of a medium sized specimen, 7 mm.; height, 6.5 mm.; convexity of both valves, 6 mm. The larger tubes measure 7 mm. in diameter and some of them, though broken, are over 30 mm. long. This species is quite similar in habit and general form to some species of Teredo, such as T. torulosa Stoliczka from the Cretaceous of southern India, but the apparent presence of a callum and of accessory valves and the strong internal rib prevent its reference to Teredo. In the presence of a callum closing the anterior hiatus, it differs also from the type of Turnus, but in other characters, including the supposed " protoplax," it agrees with that gemis, for although described as without accessory valves, a specimen of the type species ( T. plenus] from the Cretaceous of Cotton- wood Creek, California, shows a structure precisely like that described as a probable protoplax in this species. The presence or absence of a cal- lum in the adult is considered less important than the other features de- scribed. Several fragments of fossil wood in the collection are filled with the tubes and these have yielded 19 more or less perfect specimens of the shells. Locality and position. — From mouth of cafion of Rio Tarde, four miles east of Lake Pueyrrydon, Lower conglomerate, 300 feet below Ammonite bed. SCAPHOPODA. DENTALIUM (L^EVIDENTALIUM) LIMATUM sp. nov. PI. VI, Fig. 9. Shell rather large, slightly arcuate, with circular cross-section; surface appearing smooth and highly polished, but showing when magnified very STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 29 fine, closely arranged lines of growth that pass directly around the shell. The figured specimen measures 55 mm. in length and 7 mm. in diam- eter at the larger end. The test itself rather thick and the surface is not well preserved on the specimen figured but another fragment shows it per- fectly. None of the specimens are well enough preserved to show the apertures unbroken, and the subgeneric reference is therefore uncertain. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon, represented by four fragmentary specimens. GASTROPODA. PLEUROTOMARIA TARDENSIS sp. nov. PL VII, Figs, i and 2. Shell large, broadly conical, consisting of not more than seven or eight convex whorls ; apical angle about 90° ; base broadly rounded, not um- bilicated ; slit rather broad (5 mm. in the type), extending back over about one-fifth of the last whorl, situated above its middle, so that the slit-band on the whorls of the spire is near the middle of their visible por- tion ; aperture obliquely subovate ; outer lip simple, acute ; inner lip rounded below and forming a distinct callus above, which is especially prominent and thick over the umbilical region where it spreads out in a broadly crescentic form ; surface marked by numerous inconspicuous spiral lines, by an obscure furrow a short distance below the smooth slit- band and another a little farther above it, and by rather coarse, irregular lines of growth. The type measures 110 mm. in height (with apex of spire restored) and 127 mm. in greatest breadth. The species is based on a single specimen, which, though lacking the apex of the spire and a part of the test, is otherwise in an excellent state of preservation. It probably should be referred to the section Pero- trochus, which Fischer established for P. quoyana Fischer and Bernardi and to which he provisionally referred a number of Jurassic species which have the same general features as this shell, though none of them is so stout in form. No Cretaceous species known to me is so closely related as to require detailed comparison. 30 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. The specific name is derived from the Rio Tarde, near which the fossil was found. Locality and position. — From the Ammonite (Belgrano) beds at mouth of cafton four miles east of Lake Pueyrrydon. TUBULOSTIUM PUPOIDES Sp. nOV. Shell of medium size, dextral, subglobose, or very stout pupiform, with a very broad, rounded apex, umbilicated, with the umbilicus broader in the young than in the adult, when it is almost closed ; whorls four or five, slightly flattened on the sides, convex above and below, the last one much contracted and produced in a short free tube near the circular aper- ture ; surface marked by obscure irregular transverse wrinkles and by a small spiral furrow near the middle of the whorl. On the best preserved specimen the sutures are linear and inconspicuous, but two other speci- mens, believed to belong to the same species that have lost the outer layer of shell, show rounded whorls and deep sutures. The type measures 10 mm. in height and 9 mm. in greatest breadth. The contracted aperture is 3 mm. in diameter. This species was overlooked until after the drawings were all made and arranged in plates and for that reason a figure is not given. It is evi- dently congeneric with Tubulostimn callosmn Stoliczka,1 from which it differs in its more nearly pupoid form, in its rounded base and in the ab- sence of the "external callosity." Stoliczka refers the genus to the Ver- metidae. Somewhat similar forms have been described as Annelids, and one such, Serpula phillipsi Roemer, is mentioned by Behrendsen 2 as oc- curring in the Aptian of Portzuelo de Carqueque, Argentine Republic. In fact the determination of the Aptian or Gault at that place seems to be based on the presence of that species. As figured by Phillips3 under the name Vermicularia sowerbii it is somewhat larger than our species, its apex is more conical, the umbilicus is broader, and the last whorl is not narrowed and produced in a free tube. These differences are certainly sufficient to sep- erate the Patagonian form from Roemer' s species, and yet the general re- semblance is close enough to suggest a possibility that Behrendsen may have had the Patagonian species. He states, however, that he had nu- 1 Cretaceous Fauna of S. India, Gastropoda, p. 241, pi. 18, figs. 26-32. 2Zcitschr. Deutsch. Geol. Gesellsch., Bd. 43, p. 418, 1891. 3 Geology of Yorkshire, pi. 2, fig. 29. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 3! merous well-preserved examples and he could find no essential difference between them and the European forms. Stoliczka refers Serpula phillipsi" to Burtinella, a Gastropod genus related to Tubulostium. Locality and position. — From mouth of canon four miles east of Lake Pueyrrydon, 300 feet below the Ammonite beds, according to the labels. The material has more the appearance of specimens from the underlying Gio beds or the overlying Belgrano beds. VANIKORO (?) sp. A single, small, probably immature, specimen has the form and sculp- ture of this genus. The form is stout, consisting of three rapidly increas- ing whorls, with the surface marked by rather prominent transverse lines, or small costae crossed by numerous much finer spiral lines; aperture broadly ovate ; umbilicus rather narrow. Height, 6 mm.; breadth, 5.5 mm.; height of aperture, 4 mm.; breadth of same, 3.5 mm. Locality and position. — From the Ammonite (Belgrano) beds ten miles east of Lake Pueyrrydon. LUNATIA CONSTRICTA Sp. nOV. PI. VI, Figs. 10 and 1 1. Shell rather small, ovate, consisting of about four convex whorls that are slightly flattened or compressed on the upper third; suture deeply impressed ; aperture ovate, broadly rounded below ; inner lip forming a moderately heavy callus, reflected below, so as to partially cover the nar- row umbilicus ; surface marked by numerous coarse lines of growth and by sharply marked narrow constrictions, or furrows, of which there are four, parallel with the growth lines on the last whorl of the type. Height of type specimen, 24 mm.; greatest breadth, 21 mm.; height of aperture, 19 mm.; breadth of same, 12 mm. The species is represented by four fairly well-preserved specimens, the best of which is figured. Associated with these are about a dozen more imperfect specimens, some of which seem to have a smoother surface and less elevated form and may belong to a distinct species. Locality and position. — From the Ammonite (Belgrano) beds, ten miles east of Lake Pueyrrydon. 32 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. LUNATIA PUEYRRYDONENSIS. PI. VI, Fig. 12. Shell of moderate size, broadly subovate, consisting of about four rap- idly increasing convex whorls ; suture impressed, bordered by a narrow, flattened shoulder; aperture subovate, narrow above, broadly rounded below; callus of the inner lip rather narrow and thin, slightly reflected over the narrow umbilicus below ; surface marked by numerous distinct, crowded lines of growth. Height, 33 mm.; greatest breadth, 34 mm.; height of aperture, 28 mm.; breadth of same, 1 8 mm. This species is easily distinguished from the preceding by its stouter form, shouldered whorls and different sculpture. It is possible that these two species may prove to be synonyms of European forms, but in the absence of actual specimens for comparison I consider it safer to treat them as distinct species, rather than to attempt to identify such simple forms by descriptions and figures only, especially when the associated faunas are different. Locality and position. — Represented by a single specimen from the Ammonite (Belgrano) beds at mouth of canon four miles east of Lake Pueyrrydon. APORRHAIS PROTUBERATUS sp. nov. PL VI, Figs. 13-15. Shell of medium size, rather stout, consisting of six or seven convex whorls, of which the last on approaching the aperture becomes carinate above the middle and flattened below, the carina extending in a curve to the upper extremity of the wing ; aperture rather narrow and elongate ; outer lip prolonged upward beyond the preceding whorl and produced in a broad, very thick, subquadrate wing, whose outer margin is broadly rounded below and extended above in a short blunt process ; inner lip with a heavy callus that forms a subspherical protuberance just above the anterior canal and extends in a thinner deposit over a large part of the spire ; anterior canal short, broad and nearly straight, with a slight notch or emargination at the extremity ; posterior canal not distinctly developed, but apparently represented by the callus extending up the spire ; surface STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 33 of the spire marked by numerous fine, thread-like, spiral lines, and by rather prominent, slightly curved transverse ribs, with a tendency to form blunt tubercles on the middle of the whorl, giving it a subangular ap- pearance. The transverse ribs nearly or quite disappear from the back of the last whorl when it becomes carinate, and on the front aspect of the shell the sculpture is almost entirely concealed by the callus. Height of the largest type specimen, with apex of spire restored, 27 mm.; greatest breadth, 21 mm.; breadth of last whorl, exclusive of wing, 1 1 mm. In the general form of the aperture, the excessive thickening of the outer lip and the heavy deposits of callus, this little shell resembles some forms of Pugnellus, and in the preliminary examination of this collection it was referred to that genus. Pugnellus manubriatus Gabb,1 on which the subgenus Gymnarus was based, especially resembles it in the form of the wing-like expansion of the outer lip and in the callus restricted to the front of the shell. On cleaning some specimens of the Patagonian species more thoroughly, however, it was found that the anterior canal is much shorter and straighter than in any species of Pugnellus, that it lacks the well-developed anterior notch or sinus and is not bent inward toward the aperture at the extremity, and the affinities of the species seem to be with Aporrhaidae rather than Strombidae. It is not a typical Apor- rhais. It has many features in common with the recent A. occidentalis Beck, for which Gabb proposed the subgenus Arrhoges, though the heavy callus on the inner lip and spire and the greater development of an anterior canal prevent its reference to that subgenus. The peculiar rounded boss at the lower end of the callus is not duplicated in that posi- tion in any other Aporrhaid species known to me. Locality and position. — Abundant in the Ammonite (Belgrano) beds, ten miles east of Lake Pueyrrydon. Represented in the collection by over 30 individuals, most of which are very imperfect. APORRHAIS (?) sp. A larger species, apparently belonging to the Aporrhaidae, is represented by a single specimen consisting of four whorls of the spire from the locality four miles east of Lake Pueyrrydon. The whorls are convex and each 1 Palaeont. of California, Vol. I, p. 125, pi. 29, fig. 229. 34 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. bears about fifteen prominent transverse costae, crossed by numerous fine, thread-like, spiral lines. There is also a prominent spiral ridge, which is just covered by the succeeding whorl and below which the surface bears only the spiral lines. Species with similar form of whorl and sculpture are common in Anchura, Aporrhais and other genera of this family. TORNATELL/EA PATAGONICA Sp. nOV. PI. VI, Figs. 1 8 and 19. Shell of medium size, ovate, consisting of four or five rapidly increas- ing convex whorls, of which the last constitutes about five-sevenths of the total height ; apex of spire obtuse, not prominent ; aperture elongate, narrow above, rounded and apparently somewhat sinuous or emarginate below ; outer lip slightly thickened and smooth within ; inner lip forming a moderate callus and bearing two distinct folds, one of which is near the lower end and the other below the middle of the aperture ; surface marked by rather coarse spiral furrows, of which there are about 25 on the last whorl. Height of the larger specimen, 14 mm.; breadth, 10 mm.; height- of aperture, 10 mm.; breadth of aperture, 4 mm. The species is represented by only two specimens, both of which are figured. Similar forms have frequently been described as Actaeon and Solidula, but according to Cossmann's J revised descriptions of those groups such forms should be referred to Conrad's genus Tornatellaea, which ranges from the lower Jura to the Miocene. It should be stated, however, that the types of this species have the outer lip and lower part of the aperture broken and the generic reference is therefore not absolutely certain. Locality and position. — From the Ammonite (Belgrano) beds, ten miles east of Lake Pueyrrydon. ClNULIA AUSTRALIS Sp. nOV. PI. VI, Figs. 1 6 and 17. Shell small, subglobose, consisting of about four convex whorls, of which the last forms three-fourths of the total height ; suture slightly impressed ; surface marked by numerous inconspicuous, impressed, spiral lines (about 'Essais de Paleoconchologie comparee, Liv. I, pp. 45-50, Paris, 1895. STANTON I THE MARINE CRETACEOUS INVERTEBRATES. 35 30 on last whorl); aperture not perfectly preserved, but evidently entire and rounded below, with outer lip thickened and reflected, so as to form a smooth band two millimeters wide externally, and columella with two strong folds. Height, 10 mm.; greatest breadth, 9 mm.; height of aperture, 7.5 mm.; breadth of aperture, 4 mm. The type specimen is very well preserved, but the outer lip is broken, so that the form of the upper part of the aperture cannot be determined accurately, and it is not certain whether the inside of the outer lip is crenulated. There are two other imperfect specimens in the collection. Behrendsen1 reports imperfectly preserved specimens of a Cinulia from the Neocomian at Arroyo Triuguico near its junction with the Rio Neuquen, but it is not possible to determine from his descriptive note whether they are identical with the present form or not. Locality and position. — From the Ammonite (Belgrano) beds, ten miles east of Lake Pueyrrydon. CEPHALOPODA. Genus HATCHERICERAS gen. nov. Shells attaining a large size, compressed, involute (the umbilicus usually forming about one-fifth of the diameter of the shell), with rounded venter and very slightly convex sides, which in the adult may be nearly or quite smooth, but in the young are marked by low, curved, branching ribs that sometimes tend to form tubercles around the umbilical margin and on either side of the venter. In the early stages the ribs cross the venter but later they are interrupted more or less distinctly there before they disappear from the flanks. The relatively narrow umbilicus is funnel-shaped, its slopes becoming smooth and somewhat concave in the later stages but the inner whorls have rounded margins on which the ends of the ribs are seen within the umbilicus. Lobes and saddles of the suture not very complex, nor deeply incised, and characterized generally by their great breadth. The ventral lobe with lZeitschr. deutsche geol. Gescllschaft, Bd. 44, 1892, p. 18. 36 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. its two robust branches on each side is almost as long as the first lateral lobe, which is irregularly tripartite and very broad at the base. The sec- ond lateral lobe is similar to the first, but very much smaller and more slender. There is only one well developed auxiliary lobe outside the umbilicus. (The details of the suture on the umbilical slopes have not been seen.) The siphonal saddle is oblong, rather broad, tripartite at the extremity and serrated on the sides in adult sutures. The other sad- dles are more or less distinctly bipartite but not symmetrically divided. The external and first lateral saddles have about the proportions of the first and second lateral lobes, respectively, and the second lateral saddle is relatively much broader and stouter than the first. The internal, or dorsal, portion of the suture has not been studied in the type species, but in H. argentinense the antisiphonal, or dorsal, lobe is slender, pointed and dentate and there are two rather simple, paired dorsal saddles of which the outer one is the broader. The type is Hatchericeras patagonense sp. nov., described below. The few ammonites collected by Mr. Hatcher in Patagonia have proved to be very interesting, and at the same time troublesome, in that they show features suggesting relationship with several described generic groups, while they do not possess all the essential characteristics of any one of them. It seems necessary, therefore, to state the facts, so far as they are determined, and to propose a new generic name for them. The forms thus grouped together, vary considerably in sculpture and general appear- ance, but in at least two of them, the type and H. argentinense, these differences are superficial only, and there is general agreement in all essen- tial features. The other two forms, each represented by a single small specimen, are placed here with some doubt, for reasons that will be given in connection with the specific descriptions. In the generic comparisons that follow, the data for Hatchericeras are taken from H. patagonense, unless otherwise stated. The adult in general form and surface has some resemblance to Haplo- ceras, but the sutures are very different in all their details. In its smooth adult whorls, the amount of involution, the form of the umbilicus and, to a less extent, the character of the suture, it suggests Ammonites cleon d'Orbigny, which has been made the type of Cleoniceras by Parona and Bonarelli,1 but is referred with the other members of the group to Desmo- 1 Palaeontographia Italica, Vol. II, 1896, p. 83. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 37 ceras by Sarasin,1 who has thoroughly studied them more recently. The entire absence of periodic constrictions, and the reduced number and dif- ferent form of the auxiliary lobes, in the species now referred to Hatch- ericeras, prevent their reference to Desmoceras. Another group that suggests comparison, both in general form and in sutures, is that typified by the South Indian Ammonites telinga Stoliczka, for which Kossmat2 proposed the name Neoptychites, but this is somewhat more involute than Hatchericeras, the whorls are more inflated, the aper- ture is greatly contracted laterally, and the sutures show many differences in details, especially in the proportions of the dorsal and external saddles, and the first lateral lobe and the form and posture of the accessory lobes and saddles. It is true that no examples showing the complete aperture of Hatchericeras have been seen, but one showing a small part of the body chamber gives no evidence of lateral contraction. The suture of Ammonites leopoldinus d'Orbigny, which is generally re- ferred to Hoplites, closely resembles that of H. patagonense and the gen- eral form of the adult shell is somewhat similar, though the sides are less convex, the umbilicus is broader and has not the funnel shape character- istic of Hatchericeras. Young shells show much greater differences in both form and sculpture, Amm. leopoldinus having a row of distinct elongated tubercles on either side of the somewhat flattened venter, ob- scure ribs that are not developed on the middle of the flanks ending ab- ruptly at these tubercles, and another row of tubercles around the umbil- ical margin. The young of our new genus, which have been studied only in H. argentinense, do not have the venter distinctly flattened, nor bor- dered by well-developed tubercles elongated parallel with the venter. As to the similarity of the sutures, Sarasin, in the paper above cited, has called attention to the fact that similar sutures, suggesting a transition to Placenticeras, are developed in both Hoplites and Desmoceras, and that such resemblances in sutures do not necessarily mean generic identity. In this connection mention should be made of the striking general resemblance in the sutures of the Patagonian Ammonite and those of Amm. clypeiformis d'Orbigny, which is referred by most authors to 1 Quelques considerations sur les genres Hoplites, Sonneratia, Desmoceras et Puzosia. Bull. Soc. Geol. de France, 3d Sen, Vol. XXV, 1897, pp. 760-799. 2 Untersuchungen uber die siidindische Kreideformation, Beitrage zur Palaont. und Geol. Osterreich-Ungarns und des Orients, Bd. IX, p. 165, Wien, 1895. 38 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Placenticeras, but is regarded by Kossmat as representing a distinct group related to his Neoptychites and also showing some affiliation with Hop- litcs leofwldinus. But, although there is also similarity in the amount of involution and in the form of the umbilicus, the differences in the gen- eral form of the shells, especially in the ventral region, and in the details of the sutures, show that the species referred to Hatchericeras are not congeneric with Amm. clypeiformis. The conclusion drawn from all these comparisons is that the Patagonian ammonites described herein have an assemblage of characters not found in any described group, and they are therefore assigned to a new genus, whose affinities are closest with the Hoplitidae, though Hyatt's1 definition of that restricted family is hardly broad enough to include it. HATCHERICERAS PATAGONENSE sp. nov. PI. VIII, Figs, i and 2, PL IX, Fig. i. Shell large, involute, compressed, with very slightly convex sides that show an obscure flattening between the middle of the whorl and the umbilicus ; venter regularly rounded ; umbilicus about one fifth the diam- eter of the shell, funnelform, the shoulder broadly rounded and the umbil- ical wall of each whorl slightly impressed or concave in the middle ; surface of adult whorls marked only by rather coarse irregular growth lines that are seen when the shell itself is preserved but leave no trace on the internal cast. These growth lines show slightly in the umbilicus and are most conspicuous on the middle of the flank, where they form rather coarse, slightly curved wrinkles that may indicate the position of a lateral ear or lappet, such as are frequently seen in Hoplites and other groups of ammo- nites. The figured type gives no indication of the sculpture of the young, as its umbilicus could be cleaned only enough to show about one and a half whorls, but the smaller specimen, showing about three whorls in the umbilicus, bears on the umbilical margin of the inner whorl the ends of rather distant prominent ribs (or a row of elongated tubercles) closely resembling those seen in the umbilicus of H. tardense. The suture has been sufficiently characterized in the generic description and is figured natural size on plate IX. The other figures of the type specimen are one half natural size. 1 InZittel's Text-book of Palaeontology, Vol. I, pt. II, p. 584, Macmillan & Co., 1900. STANTON : THE MARINE CRETACEOUS INVERTEBRATES. 39 There are only three specimens belonging to the species in the collec- tion and they agree very closely except in size and state of preservation. Of the dimensions given below those under I belong to the figured type. I II III Diameter 250 mm. 210 mm. 300 mm. Umbilicus 53 45 ±62 Height of aperture 148 113 ^i/S Breadth of aperture 72 63 . 94 The comparisons with described species have already been suggested in the discussion of the genus. It seems remarkable that no young speci- mens of the species were collected and every small ammonite in the col- lection has been critically examined in the hope of finding the sculptured young of this form. The only one that at all suggests such an immature stage is the specimen described beyond as H. tardense, which shows similar sculpture in the umbilicus and even at the diameter of 74 mm. begins to show the disappearance of the ribs, especially on the venter and the umbil- ical wall, but the umbilicus is relatively broader and the sculpture seems too pronounced to permit its reference to this species. Locality and position. — "Mouth of canon four miles east of Lake Pueyr- rydon ; Ammonite (Belgrano) beds." HATCHERICERAS ARGENTINENSE sp. nov. PI. IX, Figs. 2-5. Shell of about the same form and proportions as H. patagonense but probably not attaining so great a size, the flattened band between the umbilicus and the middle of the whorl somewhat more distinct than in that species ; venter regularly rounded, and smooth on mature whorls ; umbilicus funnelform, nearly one-fourth the diameter of the shell, with rather narrowly rounded or subangular shoulder and smooth walls on adult whorls, but showing distinct, closely arranged ribs in the earlier stages ; surface marked by rather prominent, closely arranged ribs, part of which begin in the umbilicus of young shells up to a diameter of 50 mm., but on later whorls originate on the umbilical shoulder, cross the sides in a gentle sigmoid curve and become swollen but not distinctly tuberculate near the rounded venter, where they are interrupted by a smooth band, 4 P- 33^) appar- ently does not belong to this species, since he says that individuals from the Bay of San Jorge in his possession differ in the number of tubercles in the interambulacra. Possibly they belong to the next species. Record of specimens : San Julian, Darwin Station, i sp. Distribution : l So far only known from San Julian (d'Orb.). Affinities: The chief character of this genus is found (according to Desor) in the primary tubercles of the ambulacra, which are, on the actinal side, about as large as those of the interambulacra, but decrease rapidly in size toward the abactinal system, becoming much smaller than the latter. Although this character does not seem to warrant the generic separation of this species from Echinus, it is not observed in any other species, and so it is impossible to point out any particular relations to any known spe- cies of Echinus. Gen. TOXOPNEUSTES Ag. 3. TOXOPNEUSTES PR/ECURSOR Ortmann. PI. XI, Fig. 3«. ». ? 1897 Hypechinus patagonicus v. Ihering, in: Rev. Mus. Paul., v. 2, p. 336 (non H. patagonensis d'Orb.). 1900 Toxopnetistes prtectirsor Ortmann, in: Amer. Journ. Sci., v. 10, p. 369- Test suborbicular, slightly subpentagonal, subconical. Ambulacral and interambulacral spaces with 4-8 vertical rows of tubercles, of subequal size, those of the ambulacral spaces being a little smaller. Poriferous zone moderately broad, about half as broad as the median part of the ambu- lacrum. The latter with 4 vertical rows of tubercles, the two outermost the largest, and only these extending to the abactinal system. Pores in three pairs, the two outer vertical rows of pores placed slightly closer together, and separated from the inner row by a small tubercle, these 1 Under this head I shall give the data of distribution previously reported. 54 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. tubercles forming a vertical row of secondaries outside of the main row of the ambulacrum. Interambulacral space with 6 vertical rows of sub- equal tubercles, to which are added, on the ambitus, two irregular vertical median rows of smaller tubercles. Of these six rows the outer and inner ones disappear toward the abactinal system. All the tubercles, in both the ambulacra and interambulacra, are surrounded by small secondaries and miliaries. The median space of the interambulacra appears compara- tively smooth toward the abactinal system (as is the case in the recent species of the genus). Abactinal system unknown. Actinostome sunken, and lower surface concave ; the actinal cuts are comparatively slight. Diameter, 48; height, 25 mm. Diameter, 1 7 ; height, 9 mm. Remarks: The large number of vertical rows of tubercles in the inter- ambulacral spaces places this species with the genus Toxopneustes. The cuts of the actinostome are but slightly developed, but these vary consid- erably even in the recent species (see: Agassiz, 1873, p. 498). Record of specimens: San Julian, Oven Poin, 3 sp. ; Shell Gap, lower horizon, 3 sp. Distribution: v. Ihering's specimens from the bay of S. Jorge, recorded by him under the name of Hypechimts patagonicus, may perhaps belong to this species. Affinities: This genus has not been recorded previously as a fossil one, except from subrecent deposits. The present species differs from the re- cent species in the larger number of vertical rows of tubercles in the am- bulacra, which are much more crowded, and in the shallower cuts of the actinostome. The most closely allied species seems to be T. pileohis (Lmck.) from the tropical Pacific (see : Agassiz, 1873, p. 497, pi. 8b, f. i, 2). I have compared two young individuals of a Toxopneiistes, without local- ity, in the Princeton Museum : they seem to belong to T. pileolus. These, being of about the same size as the best preserved individual of our fossil material, differ only in having 6 vertical rows of tubercles in the interam- bulacra, and in the tubercles being a little smaller and not so much crowded. The cuts of the actinostome are only slightly narrower and deeper in these recent specimens than in our fossil species. ORTMANN : TERTIARY INVERTEBRATES. 55 Fam. CLYPEASTRIDsE Ag. Gen. SCUTELLA Lmck. 4. SCUTELLA PATAGONENSIS Desor. PI. XI, Fig. 4"-'. 1846 Scutella patagonensis Desor, in : Bull. Soc. geol. France, ser. 2,v. 4, 287. 1846 Echinamchnius juliensis Desor, ibid. 1847 Scut. pat. and Ech. jul. Agassiz and Desor, in: Ann. Sci. nat. ser. 3, v. 6, p. 135 and 134. 1858 Scut. pat. and Ech. jul. Desor, Synops. Ech. foss., p. 234 and 231. 1897 Scut. pat. and Ech. jul. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 337 and 336. 1898 Iheringia patagonensis Lahille, in: Rev. Mus. La Plata, v. 8, p. 437, pi. i and 2. Test depressed, margin thin. Outline circular, oval, subpentagonal, or more or less transversely dilated (alate), sometimes almost semi-circular, with the posterior margin truncated. Margins often undulated and the posterior one with a more or less distinct median incision. Apex sub- central. Ambulacral petals of uniform size, lanceolate, more or less open, sometimes lyrate, with a few scattered pores diverging from the extrem- ity of the petals. Interambulacral plates of upper side increasing in width from the apical system to near the end of the petals ; then they decrease suddenly in width, so that the interambulacral space narrows considerably toward the periphery. On the periphery the interambulacral space is con- siderably less broad than the ambulacral space. Ambulacral furrows of lower side dividing into two branches at a short distance from the mouth, each branch subdividing again near the periphery of the test into 2 to 4 branchlets. Anus submarginal, distant from the posterior edge of the test about i—2 times its diameter (in young individuals it is marginal, with a slight inclination toward the lower surface). Diameter of largest complete individual of rounded form : 62 mm; of a fragment : 70 mm ; alate form : longitudinal diameter : 76, transverse diameter, 94 mm. Remarks: The two forms of Scutellids, described as two different spe- cies belonging to the genera Echinarachnius and Sctdella respectively, 56 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. have been united by Lahille in his monograph on these forms into one species, for which he creates the new genus Iheringia. This generic name, being preoccupied by Keyserling in 1891, has been changed by Berg (1898, p. 1 6) into Ilieringiella (non Iheringella Pilsbry, 1893), and again into Ihcringiana (Berg, 1898, p. 41), and finally Lahille himself (1899, P- 5 of separate copy) has changed it into Iheringina. After a careful study of our rich material I am prepared to accept Lahille's view as to the identity of these supposed two species, as well as his views on the respective value of the genera Echinarachnius and Scutella, but I do not think that the Patagonian fossil ought to be placed in a separate genus ( Ilio'ingia = Iheringiana Berg) ; I prefer to leave it with the genus Scute/la. In discussing the differences of Echinaracknius and Scutella, Lahille has overlooked the fact that A. Agassiz (1872, p. 315) has given a charac- ter, by which the subgenus Echinarachnius may be distinguished from the true Scutella, viz., the arrangement of the pillars in the interior of the test. In Echinaracknius (as well as in Dendraster 2cs\& Scaphec limits] the pillars are more or less concentric with the edge of the test, while in Scutella they recall more the stellate arrangement of Mellita. This is said to be the only ground on which Echinarachnius might be separated from Scntella. I have tried to verify this character, but did not meet with much success. Failing to find any good figures representing the interior of the test of Scutella, I have compared that of Mellita as given by Agassiz (1872, pi. I2a, f. 1—4) with those of the interior of Echinarachimis (Agassiz, pi. I3a and i id, f. 45) and of Iheringia (Lahille, 1898, pi. 2, f. 11, 12), and do not find any essential differences, except that in Mellita this system of pillars is more complex, but it shows nevertheless distinctly a concentric arrangement near the edge (especially fig. 4 on pi. i2a of Agassiz). As to Scutella, I have chiefly compared the account and figures of Scut, sub- rotunda given by Quenstedt (1875, p. 544, pi. 83, f. 2, 4), and also do not find any differences from Echinarachnius; indeed, the different sections given by Quenstedt in fig. 4 render it beyond doubt that the concentric arrangement of the pillars in Scutella agrees with that of Echinarachnitis. Thus it appears that even this character does not permit a separation of Echinarachnius from Scutella, both genera (or subgenera) being practically identical ; all the characters given as distinctive (outline, ambulacral fur- rows of lower side, position of anus, shape of ambulacral petals) are only of specific value. ORTMANN : TERTIARY INVERTEBRATES. 57 As to the genus Dieringia = Iheringiana, there remains only a single character, by which it may be recognized : the considerable narrowing of the interambulacral spaces toward the margin of the test (see Lahille's generic diagnosis on p. 14). It is true, this character distinguishes the Patagonian fossil from all the known forms of Echinarachnius as well as Sctitella. Nevertheless I do not believe that it is of generic value, since this narrowing of the interambulacra is exhibited by several other species, only in a less pronounced way. Especially this is true of Echinarachnius Parma of the Atlantic coast of the United States, of which I have several hundred individuals at my disposal. Since this decrease of width of the interambulacra is brought about by an increase of width of the ambulacra, and the latter is shown in all species of Scutella and Echinarachnius, we may put it this way, that in Sciitella the ambulacral plates increase sud- denly in width from the end of the petals toward the margin, and in the Patagonian form this increase is most pronounced, so as to render the interambulacra very narrow on the margin, while in other species this in- crease goes only so far as to keep the interambulacra at the same width from the end of the petals to the margin. Sometimes it causes even a slight decrease in width: I have found a slight narrowing of the inter- ambulacra in Ech. parma, Ech. excentricus (California), and very slightly in specimens of Scut, interlineata Stps. (Gabb, 1869, p. no) from the Pliocene beds of California. Lahille compares in this respect Iheringia with Monophora, and says (p. 6 of separate copy) that in both this star-like form of the five inter- ambulacra, with five sharp points of the pentagram, is very striking. But comparing his figures of Monophora (Lahille, 1896, pi. 1-4, especially pi. 3, f. 36), there is no such close resemblance, Monophora being like Ech. parma in this respect. I was able to confirm this fact by comparison of an individual of Monophora darwini from the Territory of Chubut, sent to us by v. Ihering. As Lahille points out, there can be distinguished, in Scutella patagonica, two series of forms, one more regularly circular in outline ("mode rotun- datus"), the other more dilated and transverse ("mode alatus"). Our material also shows these two series, and I should like to make a few remarks on them. We possess altogether 87 individuals, in which the outline is distinctly recognizable. Out of this number only about 16 may be said to belong 58 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. to the "alate" form, and of these only the six largest (over 55 mm long) show this character distinctly developed. In the young ones the alate form is brought about chiefly by a comparative narrowing of the anterior end, not by an increase of the width as compared with the longitudinal diameter, although such forms are always a few millimeters broader than long. Thus, in very young individuals, the alate and rotundate forms are not so very strikingly different from each other, and, indeed, in some cases it is hard to say whether a particular individual should be classed with the one or the other. With increasing age this difference becomes more striking, and at an average length of about 50-55 mm both forms may be easily distinguished at a glance. The form "rotundatus" in young specimens is distinctly pentagonal, and also the young of "alatus" are irregularly pentagonal. But the pentagonal shape disappears with age, becoming sometimes "cordiform," when there is a posterior median emargination. The rotundate form never attains the size of the alate. Lahille's largest rotundatus is 65 mm long, 67 mm broad. We have a fragment, appar- ently belonging to the rotundate form, that has a diameter of 70 mm, while the largest complete individual is only 62 mm in diameter. Some of our alate specimens are very much larger than Lahille's (his largest is 66 by 72). I give here the measurements of our six largest individuals: Long Diameter. Transverse Diameter. Locality. 56 68 Lake Pueyrredon 59 65 Shell Gap 62 75 Salt Lake 64 ca. 71 Salt Lake 76 94 Salt Lake 8 1 89 Lake Pueyrredon These measurements show that the relation between length and width is extremely variable, some of our specimens being much more alate than any of Lahille's. As to the meaning of the existence of two such forms (rotundatus and alatus) within this species, I can only refer to Lahille's opinion : he com- pares this fact with the analogous case in Monophora, in which he believes (1896, p. 10), that these forms represent the female (rotundatus) and the male (alatus) of the same species. I cannot offer any further evidence for this theory, with only the exception, that the fact that the alate or * ORTMANN : TERTIARY INVERTEBRATES. 59 male form is more frequent and most pronounced in the largest individ- uals, seems to furnish some support for this assumption. On the other hand, the comparatively much rarer occurrence of the alate form does not favor this view, since it is hard to believe that males were so very much less in number than females, that even in some localities they have not been found at all. The actually small number of the alate individuals in our collection is not accidental, since Mr. Hatcher informs me that he has picked up every single one that he found, and that it is really very rare as compared with the rotundate form. The final settlement of this question depends on the demonstration, that in other Scutellids the male and female sexes show analogous differences in form. In this respect I may point out here a case I have noticed: Bazin (1884, p. 38, pi. 2, f. 1-5) describes from the Miocene beds of Saint Juvat, Bretagne, two species of Sctitella, S. faujasi Defr. and S. circularis Baz. The latter differs from the former just in these two characters, more circular form and smaller size, and per- haps J>. circularis is nothing but the female of S. faujasi. The same may be the case in S. subtetragona Grat. and S. striatula M. de S. (see Agassiz and Desor). Our young individuals show a position of the anus that has not been observed before. It is distinctly marginal. Although lying on the lower side of the test, its posterior margin coincides with the posterior margin of the test. This condition prevails in all (seven) of our specimens of less than 27 mm length ; the first two that show the anus a little distant from the margin (% or }4 its diameter) are 28 mm in length, but again in 4 specimens of 29, 29, 32, and 34 mm in length the same marginal position is to be seen. From 35 mm in length upward the anus is always removed from the margin, and the distance increases slightly with age, although there are variations. The smallest specimen in which it is dis- tant its own diameter is 36 mm long, the smallest in which it is distant twice its diameter is 57 mm. In the largest the distance varies from one- half to twice its diameter. Record of specimens : San Julian, Oven Point; 20 sp., 9 of which are young, i of them showing traces of the alate form, the rest are rotundate. Shore of Salt Lake ; 9 sp., 5 young ones show traces of the alate form, of the rest i is rotundatus, the rest alatus. Upper Rio Chalia; 18 more or less complete specimens, numerous fragments. i of medium size is slightly alate, the rest rotundate. Thirty miles north of Rio Chalia ; 6 sp., 60 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. all rotundate. Shell Gap, upper horizon ; 7 sp., 3 small, i large, are alate, the rest rotundate. Lake Pueyrredon, base; 27 sp., 3 of them (i medium, 2 large) are alate, the rest is rotundate. Distribiition : San Julian (Des., v. Ih., Lah.); Port Desire, coll. by Dar- win (Des.). Darwin (1846) does not mention a Scutella from Port Desire, but mentions two forms (p. no and 112) from New Bay and San Julian. But the occurrence of this species at Port Desire is confirmed by Lahille. Jegua quemada (v. Ih.); Lake Buenos Aires (Lah.). v. Ihering gives for Echinarachnius the Patagonian formation, for Scu- tella the Suprapatagonian ; but, according to our material, both are found associated in the same beds. Affinities: The most closely allied form among the recent species is Echmarachnius mirabilis (Barn.) from Japan (Agassiz, 1873, p. 526, pi. I3a, f. 5, 6) ; it resembles closely the rotundate form, but the anus is mar- ginal, and the interambulacra are broader. The alate form resembles much the Miocene Scutellcz of Europe, especially S. subrotunda Lmck. from the lower Miocene of Bordeaux (Agassiz, 1841, p. 76, pi. 17). The presence of this Scntella in the Patagonian beds points strongly to their Neogene age. Fam. CASSIDULIDA1 Ag. Gen. CYRTOMA McCl. 5. CYRTOMA POSTHUMUM Ortmann. PI. XII, Fig. i"'". 1900 Cyrtoma posthumum Ortmann, in: Amer. Journ. Sci., v. 10, p. 369. Test subcircular-elliptic. Apex central. Upper side much depressed, only slightly convex, covered with very fine and crowded tubercles. Am- bulacra petaloid, open, lanceolate, subequal, extending about two-thirds from the apex toward the periphery, the two posterior ones are closer to- gether than the others. Pores jugate. Anus situated on the upper sur- face, in a tieep depression, which commences about half way between the apex and the posterior extremity by a narrow slit, widening suddenly to- ward the margin and giving the anal depression a pyriform shape. Lower surface of the test concave, covered with larger, more distant and some- what irregular tubercles, a comparatively smooth band running from the ORTMANN I TERTIARY INVERTEBRATES. 6 1 mouth toward the posterior periphery. Mouth opening subcentral, sur- rounded by a floscelle. Diameter about no mm. Height about 28 mm. Remarks: All three specimens are imbedded in a hard, reddish, coarse grained sandstone, and are poorly preserved, which refers especially to the apical system, the details of the ambulacra, and the mouth. The anal de- pression, however, is seen in all three individuals, and forms the most striking character of the genus Cyrtoma of McClelland (1840, p. 185), al- though the original diagnosis of the genus does not bring out this char- acter sufficiently. It runs thus : " Disc oval and thin, arched to the apex ; ambulacra petaloid, and either broad and flat, or more elevated, and placed on narrow ridges radiating from the apex to the disc. The two posterior ambulacra are closer together than the others, with an interme- diate dorsal ridge leading to a dentate anus, and a depression or hollow between the latter and the disc. Inferior surface flat, mouth small and central, with five clavate ambulacra prolonged to margin." It would be impossible to recognize the genus from this diagnosis ; but the figures given by McClelland clearly establish its position with refer- ence to the genera Echinobrissus, Cassidulus, etc., and there is no doubt, that it is identical (see Zittel, 1880, p. 529) with Stigmatopygus of d'Or- bigny (1860, p. 331). According to d'Orbigny this genus comes near Cassiduhis, and differs chiefly in the peculiar shape of the anal depression. Desor (1858, p. 296) compares it with Echinanthtis. Of the floscelle sur- rounding the mouth a few traces are preserved in one of our specimens. Record of specimens : Lake Pueyrredon, base; 3 sp. , Affinities: The genus Cyrtoma (= Stigmatopygus] is known from the Cherra Poonji beds of British India (Assam Range), which are evidently of the age of the Arialur group (Senonian) of southern India, where the genus has also been found (see: Medlicott and Blanford, 1879, part i, p. 280, ff., and part 2, p. 688, f., and Stoliczka, 1873, p. 27). It has also been found (Stigmatopygus] in the upper Cretaceous (Senonian) of France (Angouleme). Our species from the Tertiary beds of Patagonia extends considerably the range of this genus in time. It differs at the first glance from all the known species in the much more depressed -test, and in the lack of a ridge between the apex and the anal depression. The discovery of this genus, so far belonging exclusively to the upper Cretaceous, is one of the most important palaeontological results of Mr. Hatcher's explora- tions in Patagonia. 62 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Fam. SPATANGIDA1 Ag. Gen. SCHIZASTER Ag. 2. SCHIZASTER AMEGHINOI v. Ihering. PI. XIII, Fig. !«•». 1897 Schizaster ameghinoi v. Ihering, in: Rev. Mus. Paul., v. 2, p. 338. Test moderately depressed (our specimens have suffered from pressure and are distorted in different directions), highest in the posterior part, back of the apical system, where it is raised to a distinct longitudinal carina between the posterior ambulacra. Outline cordiform, with a deep notch anteriorly. Apex situated back of the middle of the upper surface. Anus in the upper part of the truncated and vertical posterior extremity. The anterior ambulacrum in a deep groove, the margins of which are quite distinct, but rounded, and run almost parallel toward the anterior edge of the test. The sides of the groove are oblique and concave. Pores of the anterior ambulacrum regular, those of each pair separated by tubercles, which form two rows in the groove. Anterior and posterior lateral ambulacra distinctly sunken ; the two anterior curved, first running close to the anterior ambulacrum, then diverging. Posterior lateral ambu- lacra very short, hardly one-third as long as the anterior lateral ones. Peripetalous fasciole forming a sharp, almost rectangular, re-entering angle in the posterior lateral interambulacral space, and forming also a re-enter- ing angle in the anterior lateral interambulacral space. Length, 56 mm; width, 51 mm; height, 29 mm; but growing much larger as indicated by fragments. Remarks: The description given by v. Ihering is entirely insufficient to recognize this species ; but since no other species of the genus is known from Patagonia, it is perfectly safe to assume that we have to deal with this species. Record of specimens : Mouth of Santa Cruz River, 2 sp. ; Paso del Rio Santa Cruz, 3 sp.; San Julian, Oven Point, fragments of 2 specimens, imbedded in matrix. Distribution: Gulf of San Jorge (v. Ih.). Affinities: Micraster atacamensis and valdimanus of Philippi (1887, p. 231, pi. 52, f. 2, 3), the first from the Pliocene Coquimbo beds of Chile, the second from the Navidad beds, belong apparently to Schizaster. S. ORTMANN : TERTIARY INVERTEBRATES. 63 atacamensis is very much higher than our species, and the vertex is pro- duced over the anus. S. valdivianus comes nearer to our species, but the anterior ambulacrum is said to be in a shallow groove, it is more narrowed anteriorly, and the outline of the test is broader. It is hard to say whether there is any closer relation to other fossil species. VERMES. CHAETOPODA. TUBICOLA1 (SEDENTARIA). Gen. SERPULA L. 7. SERPULA PATAGONICA Ortmann. PI. XIII, Fig. 2. 1900 Serpula patagonica Ortmann, in: Amer. Journ. Sci., v. 10, p. 369. Tubes solid, calcareous, cylindrical, irregularly contorted and vermic- ulate, growing upon shells, stones, etc. Outer surface transversely rugose. Diameter : up to 3 mm. Remarks : I did not try to compare this species with any of the known forms, since the characteristic features of these tubes are so very few, that it is impossible to say whether it is a good species or not. It agrees well with the short diagnosis given by Philippi (1887, p. 217) for S. colchag- uensis from Navidad, but without material for comparison it is hard to tell whether it is identical or not. It has been mentioned here only for the sake of completeness. Record of specimens : San Julian, Oven Point; 5 large colonies, chiefly on Pecten geminatus. San Julian, Darwin Station ; i colony, on a stone. Gen. TEREBELLA Cuv. 8. TEREBELLA MAGNA Ortmann. ,Pl. XI, Fig. 5-.'. 1900 Terebella magna Ortmann, in: Amer. Journ. Sci., v. 10, p. 370. Large, cylindrical tubes, isolated or growing in groups of two or three, straight or slightly and irregularly curved. Walls composed of large and 64 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. irregular fragments of shells, Balanids, etc. Inner surface of tubes smooth, outer surface very rough. Length of largest fragment, 1 45 mm, diameter of inner tube (without wall), 12-13 mm- Remarks: We may safely assume that these large and curious tubes have been built by a worm, but we do not have the slightest indication as to its systematic position. I place our species with the genus Terebella, because this is the only fossil one known that builds its tubes by gluing together fragments of shells, sand, etc. (see Zittel, 1880, p. 564). The chief characteristics of these tubes are their large size and the large size of the shell-fragments used for their make-up. They will be easily recog- nized from the description and figure given here. Record of specimens : San Julian, Oven Point; 14 fragments. BRYOZOA. CHILOSTOMATA. Fam. CELLARIID^E Hcks. Gen. CELLARIA Lamx. 9. CELLARIA FISTULOSA (Linnaeus). PI. XI, Fig. 6°'6. 1964 Salicornaria marginata Stoliczka, in: Novara Exp. Geol., v. i, p. 150, pi. 20, f. 11-13. 1880 Cellaria fisttilosa Hincks, Hist. Brit. mar. Polyz., p. 106, pi. 13, f. 1-4 (et synonyma). 1900 Cell. fist. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Zoarium dichotomously branched, articulate, internodes of moderate length, slender, subcylindrical. Zocecia lozenge-shaped, a little longer than broad, contiguous in the same longitudinal row. Orifice arched above, lower lip curved inward, subcentral (situated about in the middle of the zocecium). Ovarian opening (special pore) subcircular, situated in the upper part of the zocecium, in the upper angle of the rhombus. Remarks: I follow Hincks in the identification of this species, although there is some doubt whether the fossil form described by Stoliczka from New Zealand is really identical with this cosmopolitan recent species. ORTMANN I TERTIARY INVERTEBRATES. 65 Our Patagonian specimens agree very well with the New Zealandian fos- sil, but they differ — as well as the latter — from the recent C. fistulosa in the absence of avicularia on the top of the cells. The ankylosis of the joints found sometimes in the recent form, and described by Stoliczka in the fossil form, is also exhibited in a few fragments of the Patagonian fossil. Our material is comparatively poor, consisting of a mass of fragments found in a small piece of rock. The structure of the surface is much ob- scured, only a few fragments showing the form of the cells with sufficient clearness. The form of the ovarian opening (which has sometimes a small tooth, according to Stoliczka) cannot be made out satisfactorily. Record of specimens : Shell Gap, lower horizon; numerous fragments. Distribution: Living, almost cosmopolitan (see Hincks). Fossil: Oligo- cene, Miocene, and Pliocene of Europe (see Stoliczka and Hincks) ; Mio- cene of the Orakei Bay, New Zealand (Stoliczka and Hutton, 1885 a, p. 209). Affinities: If this is really the living species C. fistulosa, its range in time is from the Oligocene to the Recent times. But, as has been said above, our Patagonian form resembles more the New Zealandian Miocene form called by Stoliczka Salicornaria marginata Miinster. Gen. MELICERITA M.-E. 10. MELICERITA TRIFORIS Ortmann. PI. XIII, Fig. 30-6. 1900 Melicerita triforis Ortmann, in: Amer. Journ. Sci., v. 10, p. 370. Zoarium foliaceous, lobate. Zocecia hexagonal, with a raised border, disposed quincuncially on both surfaces of the zoarium. Each zocecium about as long as broad. Orifice transverse, crescentic, large, situated about in the middle of the cell. Ovicells immersed, inconspicuous, indi- cated only by an opening (special pore) in the summit of the cell. Besides, there are two openings in each cell, near the summit on each side of the mouth, which probably represent avicularia. No other avicularia dispersed among the other cells. Remarks: The zocecia of Melicerita have the characteristic shape of the family Cellariidce (genus Cellar ia Lamx., 1 8 1 2 = Salicornaria Cuv., 66 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. 1817), but the genus differs from Cellaria in the foliaceous and compressed zoarium, carrying zocecia on both sides (see Busk, 1859, p. 69, and 1884, P- 95)- Record of specimens : Upper Rio Chalia; 3 fragments. Affinities: This species has some resemblance, in the form of the zoarium and the general shape of the cells, to the type-species of the genus, M. charlesivorthi M.-E. (Busk, 1859, p. 70, pi. 10, f. 4) from the English Crag (Pliocene), but it differs in the absence of avicularian cells among the other cells, in the absence of the raised lines on each side of the mouth, and in the presence of the accessory (avicularian ?) openings on each side of the mouth, and, further, in much larger mouth openings. Of the two recent species described by Busk (1884), M. dubia is quite dif- ferent, and probably does not belong to this genus at all. M. atlantica, however (1. c., pi. 14, f. i), from off Monte Video, 600 fath., is closely allied in the form of the zoarium and position of the mouth, but here also the lateral pores are wanting, and the zoarium is narrower. Fossil species of the genus have been found — aside from the English Crag — in New Zealand and Australia. M. angustiloba Busk (see Cellaria ang. Waters, 1882, p. 260, pi. 9, f. 28-30, and Stoliczka, 1864, p. 155, pi. 20, f. 15-18) is found in the Miocene of Mt. Gambier, S. Australia, and of Victoria (Busk, Waters), and in the Miocene of New Zealand (Stol.), Pareora system of Hutton (1885 a, p. 209). But in this species the zoarium is much narrower, the cells are longer, the mouth is situated in the anterior part of the cell, and the lateral pores are wanting. Busk (1859, p. 70) says that besides the type-species, he was able to find only a single other one, that he refers to this genus : Rschara acaste of d'Orbigny (1852, pi. 662, f. 7-9); but these figures represent Esch. achates. And, when he refers this species (achates] to Melicerita, also Esch. acts, acmon, and perhaps actcea would also belong to it. All these Upper Cretaceous species named differ from that Tertiary species under discussion in the much narrower branches of the zoarium and the absence of all traces of "special pores," and I doubt seriously that they belong to Melicerita. The established range in time of Melicerita would thus embrace — aside from the Patagonian beds — the Miocene of Australia and New Zealand, the Pliocene of England, and the recent South Atlantic. ORTMANN : TERTIARY INVERTEBRATES. 67 Fam. ESCHARIDA1 Johnst. Gen. ASPIDOSTOMA Hcks. 11. ASPIDOSTOMA GIGANTEUM (Busk). PI. XIII, Fig. 4. 1854 Eschara gigantea Busk, Cat. mar. Polyz., Brit. Mus., v. 2, p. 91, pi. 119, f. 3. 1 88 1 Aspidostoma crassum Hincks, in: Ann. Mag. Nat. Hist, ser. 5, v. 7, p. 1 60, pi. 10, f. 6. 1884 Aspid. giganteum Busk, Challenger Polyz., i, p. 161, pi. 33, f. 3. 1891 Eschara (Aspidostoma) gig. Jullien, in: Miss. Cap Horn, v. 6, p. 77, pi. 6, f. 5-6. 1900 Asp. gig. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Zoarium erect, compressed, bilaminate, contorted, divided and coales- cent frequently. Zooecia arranged quincuncially, thick-walled, broadly pyriform or hexagonal, divided by deep sutures, tumid in front, but de- pressed in the center. Mouth at the summit of the depressed area, upper lip arched, with an elevated hood rising into two prominent processes. Lower lip with a broad plate covering the mouth ; margin of this plate thickened, squarely truncated in front. Ocecia rounded or oval (Hincks says, elongated), depressed in the older parts of the colony, more promi- nent in the younger parts. Remarks: There is no doubt that this fossil form corresponds to the living species. The most characteristic features are the hood-like (pent- house-like, Busk), bifid projection of the upper lip of the mouth, the broad plate of the lower lip, and the central depression of the cells. All these characters are well exhibited in our specimens. The figures of Hincks and Busk do not bring out these features very distinctly ; they are, how- ever, better represented in Jullien's figures. Record of specimens : Mouth of Santa Cruz River; many fragments. San Julian, Oven Point ; 2 basal parts of colonies. Distribution : Aspidostoma giganteum has been found so far only living in southern Patagonia: Straits of Magellan, between Patagonia and the Falkland Islands, and at Tristan da Cunha (110—150 fath.). 68 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. CYCLOSTOMATA. Fam. LICHENOPORIDsE Sm. Gen. RETICULIPORA d'Orb. em. Wat. 12. RETICULIPORA PATAGONICA Ortmann. Pi. XII, Fig. 2°-°. 1900 Reticulipora patagonica Ortmann, in: Amer. Journ. Sci., v. 10, p. 37°. Zoarium reticulate, fenestrae of the reticulations 2-4 mm long, and 1-2 mm broad, irregular. Branches much compressed in section, about 2 mm deep. Broad lateral surfaces of the branches with slightly exserted, tubular zocecia, which are rather crowded and form irregular transverse rows. Besides the zocecial openings there are smaller, non-tubular ones at the sides of the branches. On the front of the branches a median lamina rises as a distinct narrow median ridge. On the back part, the branches are rounded, and show very small openings. Remarks: In the possession of intermediate pores this species does not correspond to the original diagnosis of the genus given by d'Orbigny (1859, p. 903), but it agrees with Waters' conception of Reticulipora. For the same reason the present species and the genus Reticulipora of Waters cannot be united with Idmonea, as Zittel does (1880, p. 599). I leave our species in the genus Reticulipora, since it comes extremely near to Ret. transennata Waters. Record of specimens: Mouth of Santa Cruz River; fragments of about 10 colonies. Affinities: This species is very closely allied to Reticulipora transennata Waters (1884, p. 689, pi. 30, f. 2, 3, 6, 7), from Aldinga, South Australia, which locality is regarded as Eocene. Indeed, it resembles this one so much that I entertain some doubt as to the specific difference of both. The only differences I am able to point out are : the branches of the zoarium seem to be stronger in our species (compare fig. 3 of Waters), and the zocecial openings appear to be more crowded (see fig. 7, 1. c.). ORTMANN : TERTIARY INVERTEBRATES. 69 Gen. TENNYSONIA Bsk. 13. TENNYSONIA SUBCYLINDRICA Ortmann. PI. XIII, Fig. s«<6. 1900 Tennysonia stibcylindrica Ortmann, in: Amer. Journ. Sci., v. 10, p. 370. Zoarium stipitate, irregularly branched, branches coalescent and lobate, subcylindrical. Orifices of cells slightly raised above the surface, arranged in straight, uniserial lines, placed only on one side of the branches, and beginning at an imaginary median line on this side. Interspaces between cells and back side of branches with pores (cancelli). Remarks: This fossil resembles so much the living and only known spe- cies of the genus, T. stellata Busk (1875, p. 34, pi. 31, f. 6) from the Cape of Good Hope, that it is possibly identical with it. The only differences are : the branches are a little thinner in the fossil form, and subcylindrical, with hardly any indication of a triangular cross section, so that there is no trace of a median ridge on the front part of the branches, and further, the orifices of the cells are slightly raised above the surface, while they are even with it in T. stellata. Record of specimens : Mouth of Santa Cruz River ; i colony. Gen. HETEROPORA Blv. 14. HETEROPORA PELLICULATA Waters. PI. XIII, Fig. 6. 1879 Heteropora pelliculata Waters, in : Journ. Roy. Micr. Soc., v. 2, p. 390. pl- 15- 1879 H. neozelanica Busk, in: Journ. Linn. Soc., v. 14, p. 725, pl. 15, f. 1-4. 1880 H. n., Nicholson, in: Ann. Mag. Nat. Hist., ser. 5, v. 6, p. 333, textf. i A, B, C. 1900 H. pell. Ortmann, in : Amer. Journ. Sci., v. 10, p. 378. Zoarium erect, arising from an incrusting base, with short, subcylindri- cal, diverging, dichotomous branches, terminating in blunt, rounded ex- yo PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. tremities, and sometimes coalescing. Surface with two kinds of orifices, though scarcely distinguishable in size. The larger ones are subcircular, the others (cancelli), disposed more or less regularly around these, are more or less angular. Diameter of branches 5-10 mm, rarely more. Remarks: Heteropora neozelanica is considered identical with H. pel- liculata, by Waters (see: Nicholson, 1880, p. 329); according to Busk it differs from the latter, (i) in the shorter branches, which are never con- nected with each other, (2) in the absence of the calcareous pellicle or epitheca. Our specimens are, as regards the first character, in some de- gree intermediate : the branches, although comparatively shorter and stouter, coalesce frequently. The "pellicle" is not present in them, but this may be due to fossilization. The surface structure is in some places quite well preserved ; it corresponds to Busk's fig. 4, and still better to Nicholson's fig. i B, and differs in this respect from Waters' figure (copied by Nicholson in fig. i D). Record of specimens : San Julian, Oven Point ; 2 colonies. San Julian, Darwin Station ; i colony. Arroyo Gio ; 3 fragments. Distribution: Living in New Zealand and Japan (Busk, Wat.); fossil at Napier, New Zealand (fide Waters, 1884, p. 696), which beds belong to the Ahuriri series of the Pareora System, Miocene (see Hutton, 1885 a, p. 194, 209). BRACHIOPODA. Fam. RHYNCHONELLID^E d'Orb. Gen. RHYNCHONELLA Fischer. 15. RHYNCHONELLA PLICIGERA v. Ihering. PI. XII, Fig. 3-. 1897 Rhynchonella plicigera v. Ihering, in: Rev. Mus. Paul., v. 2, p. 270, textf. 7. Shell variable in shape, mostly wider than long, irregularly tetrahedral or more or less triangular. Beak more or less acute and slightly re- curved; foramen moderately large, the lower part of it formed by the deltidial plates. Beak-ridges well defined. Smaller valve with a more ORTMANN : TERTIARY INVERTEBRATES. 7 1 or less elevated fold, corresponding to a mesial sinus in the larger valve. Surface ornamented by sharp plaits, about 25 in number, 4 to 7 of which are in the fold and sinus ; these plaits become indistinct and disappear in the upper part of the shell. Measurements: Length 15, 16, 18, 20, 20, 22, 24, 21, 21 mm. Width 14, 20, 1 8, 21, 23, 20, 24, 24, 28 mm. Remarks: This species comes very near to R. nigricans (Sow.) (see: Davidson, 1852, p. 81, pi. 14, f. 30, 31 ; Suess, 1864, p. 60, pi. 14, f. 5; Hutton, 1873, p. 37; Davidson, 1887, p. 169, pi. 24, f. 16-19). The only difference of our fossil Patagonian form from the New Zealandian is the tendency of the plaits to disappear toward the umbo of both valves. V. Ihering compares this species with R. nigricans, and says that it differs in the following particulars: (i) In the more triangular and less transverse shape ; (2) in the straighter and more pointed beak; (3) in the smaller foramen. That the first character has no value is shown by our material, in which the outer form varies considerably, as is seen by a glance at our figures and the measurements given above ; and the same is true of the second character. As to the foramen, the figure of v. Iher- ing does not show a smaller foramen than those of Davidson and Suess, and (although our specimens show, as a rule, a foramen a little smaller than in the figures quoted) this difference is very slight. In a letter, v. Ihering again calls my attention to the difference in the foramen, and compares that of R. plicigera to that of R. psittacea. In the latter spe- cies (judging from specimens collected by myself in Inglefield Gulf, North Greenland) the foramen is narrow, elongated, and the deltidial pieces on each side of it are elongated-triangular, while in R. nigricans (according to Davidson) the foramen is more rounded and the deltidial pieces are broadly and almost equilaterally triangular. Examining our fossil form in this respect, it agrees with R. nigricans, and not with R, psittacea. But there seem to be variations even in R. nigricans, as is seen in David- son's figures, and confirmed by Suess's figures, which have a distinctly elongated foramen. Record of specimens : Las Salinas, i sp. ; Shell Gap, upper hori- zon, i sp. ; east end of Lake Pueyrredon, i sp. ; high bluffs, S. W. of Lake Pueyrredon, 2 sp. ; Lake Pueyrredon (Rio Tarde section), base, 35 sp. 72 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Distribution: Gulf of San Jorge, Patagonian beds (v. Ih.). Specimens sent by v. Ihering to the U. S. Museum and examined by the writer were labelled : Santa Cruz. Affinities: The most closely allied species, R. nigricans, has been found living at New Zealand, and fossil in the Oamaru, Pareora and Wanganui beds of New Zealand, and thus it ranges from the Oligocene upward to Recent times. 16. RHYNCHONELLA SQUAMOSA Hutton. , PI. XII, Fig. 4«-». 1873 Rhynclwnella squamosa Hutton, Cat. Tert. Moll; Echin., New Zea- land, p. 37. 1878 R. ccelata Tenison-Woods, Journ. Proc. Roy. Soc. N. S. Wales, v. ii, p. 77. 1880 R. squamosa Tate, in: Trans. Proc. Roy. Soc. S. Australia, v. 3, p. 1 66, pi. 9, f. 9. 1880 R. nigricans var. pixydata Davidson, Challenger Brach., p. 59, pi. 4, f. 14. 1887 R. nigricans var. pixydata Davidson, in: Trans. Linn. Soc., ser. 2, v. 4, p. 170, pi. 24, f. 10. 1896 R. squamosa Pritchard, in: Proc. Roy. Soc. Victoria, v. 8, p. 143. 1900 R. sqtiamosa Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Shell more or less transversely circular ; beak acute and incurved. For- amen small. Dorsal (small) valve convex, mesial fold scarcely distin- guishable. Ventral valve flatter, with a broad, well-defined mesial sinus. Surface of both valves with about 40-50 radiating ribs, 10-15 of them in the sinus ; closely intersected by squamose, concentric lines of growth, giving an imbricated appearance to the surface. Length 24, 25, 25 mm. Width 26, 28, 26 mm. Remarks: Davidson confirms the identity of R. cczlata with his R. pixy- data, and — as Tate points out — R. ccelata is distinguished from R. nigri- cans by the same characters .that distinguish R. squamosa, being accord- ingly the same as the latter. Our individuals agree well with the figures of R. squamosa, as well as of R. pixydata, especially in the squamose surface markings. The only ORTMANN : TERTIARY INVERTEBRATES. 73 difference is the larger size and the more circular outline, which is not so distinctly transversely-oval. Records of specimens : High bluffs S. W. of Lake Pueyrredon, i sp. ; Lake Pueyrredon, base, 10 sp. Distribution: Miocene of South Australia (Tate) and Tasmania (Pritch.) ; Oligocene (Oamaru system) of New Zealand (Hutt.); and recent, Ker- guelen Islands, 150 fath. (Dav.). Fam. TEREBRATULIDsE King. Gen. MAGELLANIA Chemn. 17. MAGELLANIA LENTICULARIS (Deshayes). PI. XII, Fig. 5M. 1864 Waldheimia lentictilaris Suess, in: Novara Exp. Geol., v. i, p. 56, pi. 10, f. 3, 4. 1873 IV. 1. Hutton, Cat. Tert. Moll. Ech., New Zealand, p. 35. 1886 W. I. Davidson, in : Trans. Linn. Soc., ser. 2, v. 4, p. 52, pi. 9, f. 2-13. 1897 Magellania globosa v. Ihering, in: Rev. Mus. Paul, v. 2, p. 268. 1900 Magellania lenticularis Ortmann, in : Amer. Journ. Sci., v. 10, p. 379. Shell regularly oval or subcircular in outline, lenticuliform or more or less globose. Beak prominent, subacute, incurved ; beak ridges well defined, forming a slightly excavated area. Foramen small. Valves uni- formly convex, without a distinct sinus (a slight indication of a sinus is said to be present in the recent form : in a few of our specimens there is just a suggestion of it, but in most of them there is no trace). Length, 36, 35, 32, 29, 25 mm. Width, 31, 32, 31, 29, 23 mm. Remarks: This species, which is found still living in New Zealand, is characterized chiefly by the regular outline and the very small foramen. Terebratitla fontainiana d'Orb. from the coast of Chili seems to be its American representative ; it is certainly not a synonym of M. venosa, as Davidson (1886, p. 50, 51, pi. 8, f. 6) believes. T. fontainiana differs from T. lenticularis in the more elongated form. I have seen four specimens of v. Ihering's M. globosa, sent by him to the U. S. Nat. Mus., and they agree completely with our species. M. 74 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. globosa is made by Davidson a synonym of M. venosa (Sol.). (Recent, Falkland Islands and Tierra del Fuego.) Our specimens differ from M. I'ciiosa in the small foramen, which is one of the characters in which M. lenticnlaris is said to differ from M. venosa. The other characters given by Davidson are : beak more incurved, dorsal valve uniformly convex, and size smaller. They are, however, hardly of any value. And further, he mentions a difference in the deltidium, but the latter is described in M. venosa in almost the identical terms. Therefore, judging from the foramen, which is the only reliable character, the Patagonian fossil belongs to M. lenticnlaris, and it agrees surprisingly well with the account of that species given by Suess. Record of specimens : High bluffs, S. W. of Lake Pueyrredon, 9 sp.; Lake Pueyrredon, base ; 34 sp. Distribution: Recent, New Zealand (Dav.); fossil: Gulf of San Jorge, Patagonian formation (v. Ih.); Oligocene (Oamaru), Miocene (Pareora), and Pleistocene of New Zealand (Suess, Hutt); Oligocene and Miocene of Chatham Islands (Hutt.). Gen. TEREBRATELLA d'Orb. 1 8. TEREBRATELLA DORSATA (Gmelin). PI. XIII, Fig. 7<-«. 1864 Terebratella dorsata Suess, in: Novara Exp. Geol., v. i, p. 57, pi. 14, f- 6. 1873 T. d. Hutton, Cat. Tert. Moll. Ech., New Zealand, p. 36. 1880 T. d. Davidson, Challenger Brach., p. 44, pi. 4, f. 4. 1887 T. d. Davidson, in: Trans. Linn. Soc., ser. 2, v. 4, p. 75, pi. 14, f. 9-19. 1900 T. d. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Shell very variable jn shape, but mostly transversely-oval, wider than long. Valves moderately convex, ribbed or (rarely) smooth. Dorsal valve with a median sinus, ventral valve with a corresponding fold. Beak produced, slightly incurved and truncated by a rather large, circular for- amen. Beak ridges sharply defined. Length, 44, 43, 38, 31, 31, 28 mm. Width, 36, 45, 40, 26, 29, 22 mm. ORTMANN : TERTIARY INVERTEBRATES. 75 Remarks: The radiating ribs are very variable; in some cases they ex- tend almost over the whole shell, in others (and this is the most common form) they begin at about the middle of the shell and run to the margins, and again in other cases they are visible only near the margins. In very few specimens (about half a dozen out of 81 individuals) the ribs are lack- ing altogether. The external form of the shell varies also. It is generally wider than long, often distorted (as figured by Suess), and sometimes elongated, longer than wide. In the latter case it approaches closely the following species ( T. patagonica], but differs in the presence of ribs. This is seen in a few individuals from Lake Pueyrredon and in the specimen from the mouth of the Santa Cruz River (L. 26, W . 19). This same form has been sent by v. Ihering to the U. S. National Museum, labelled : Gulf of San Jorge. Since T. patagonica assumes sometimes a broader form, and since the ribs of T. dorsata sometimes disappear completely, it is evident that both species may pass into each other, and, indeed, we possess specimens in which it is very hard to say, to which one they belong. In my opin- ion, both species are intimately related to each other. There is hardly any difference between our fossil material and the re- cent form now living on the coast of Patagonia, of which Mr. Hatcher has collected numerous individuals. Among the fossil forms, however, there are much larger shells. The fossil New Zealand form, figured by Suess, has the fold and sinus more sharply defined than any of our specimens, but since there is much variation among them, as well as among the recent ones (the sinus and fold being in some cases quite indistinct), there is no reason for separating the New Zealandian shell from T. dorsata. Record of specimens: Mouth of Santa Cruz River, i sp. (Van) ; Shell Gap, upper horizon, 4 sp. ; East end of Lake Pueyrredon, i sp.; Lake Pueyrredon, base, 81 sp. ; Lake Pueyrredon, 600' above base, 4 sp. Distribution: Gulf of San Jorge (2 specimens sent by v. Ihering to the U. S. Nat. Mus.); Miocene (Pareora) of New Zealand (Suess, Hutt); Recent, S. America (Patagonia, Straits of Magellan, Chili) and Kerguelen Islands (Dav.). 19. TEREBRATELLA PATAGONICA (Sowerby). PI. XIII, Fig. 8"'" and PI. XIV, Fig. i"'4. 1846 Terebmtula patagonica Sowerby, in: Darwin, Geol. Obs. S. Amen, p. 252, pi. 2, f. 26, 27. 76 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. 1873 Wahlhcimia p. Hutton, Cat. Tert. Moll. Ech. New Zealand, p. 36. 1885 W. p. Hutton, in: Quart. Journ. Geol. Soc., v. 41, p. 553. 1887 Tercbratula p. Philippi, Tert. Quart. Verst. Chil., p. 217, pi. 49, f. 2 (after Sovverby). 1897 Magcllania p. v. Ihering, in: Rev. Mus. Paul, v. 2, p. 267. 1899 Tcrcbratella p. Lahille, in: Rev. Mus. La Plata, v. 9, p. 289, ff., pi. i, 2. Shell oval, more or less elongate, rarely almost as broad as long. Both valves nearly equally convex. Beak incurved, with a large foramen ; beak ridges blunt. No sinus on the dorsal (smaller) valve. Surface smooth, without ribs. Measurements: Length, 59, 40, 30, 25, 34, 31, 27 mm. Width, 51, 32, 28, 19, 25, 21, 25 mm. Remarks: This species is a Tcrcbratella as has been shown by Lahille, who figures the complete brachial apparatus (pi. i, f. 53, 54, 55). Al- though I do not possess any specimens showing this apparatus complete, the median septum of many of my specimens shows distinctly a cruciform appearance, /. ^ 29)> is quite, insufficient. Record of specimens : Mouth of Santa Cruz River ; 4 double, 7 right, 3 left valves. (Some of them labelled 200' and 250' above high tide.) Mt. of Observation, lower horizon ; 2 double, i right, 2 left valves. Distribution : Santa Cruz (Sow., v. Ih.); La Cueva and Jack Harvey (v. Ih.); Port Desire (Sow.). Patagonia formation (v. Ih.), Oamaru and Pareora systems (Oligocene and Miocene) of New Zealand (Hutton). Affinities: C. cliilensis Phil. (1887, p. 189, pi. 40, f. 2), from the Navidad beds of Chili comes extremely near to this species, the only difference is the less high and more elongated form. From the Oligocene and Miocene beds of New Zealand Hutton has described three more species, which are hardly distinguishable. Sowerby compared this species with C. dectissata Sow. (= crassatina Lmck.), which is found in Eocene beds of Europe, and a similar form is found in the Eocene of the eastern United States : C. gigantca Conr. (see : Clark, 1896, pi. 30-33). Since species of this genus have not been found in the northern hemisphere in younger Tertiary beds, we are confronted here with a pronounced relation to Eocene : but the value of this case is 9O PATAGONIAN EXPEDITIONS : PALEONTOLOGY. counterbalanced by the fact that the identical species or very closely allied forms are found in beds that are considered Oligocene, Miocene and Plio- cene (Hutton, 1886, p. 365), in Chili and New Zealand, and further by the fact that the genus still exists in the recent seas. 28. CUCULL^EA (CUCULLARIA) DARWINI (Philippi). PI. XXV, Fig. 5"- ». 1887 Area darw. Philippi, Tert. Quart. Verstein. Chiles, p. 188, pi. 36, f. 3. 1897 Cttcitllnria tridentata v. Ihering, in: Rev. Mus. Paul., v. 2, p. 237, pi. 4, f. 22, pi. 5, f. 28. 1900 Cttcullcea danvini Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Shell elongate-oval, subtrapeziform, with fine radial ribs, which are flat and separated by fine but sharp sulci. Toward the lower margin the ribs appear geminate or double, through the development of a slight sulcus in the middle of each rib. The ribs are crossed by distant, distinct lines of growth. Apices swollen, closely approaching each other. Anterior end of shell short, rounded ; posterior broader, with rounded angles. Area very small. Teeth of hinge (according to v. Ihering) : 2-3 horizontal anterior ones, and 3 posterior. Small vertical teeth below the apex. Our largest individual measures : length, 29 mm ; height, 1 7 mm ; but the species grows much larger, since v. Ihering gives : length, 42; height, 24. Remarks: According to v. Ihering this species belongs to the subgenus Cucullaria. My specimens agree completely with his figure in outline, but they are smaller and do not show the hinge. There cannot be the slightest doubt that this species is the same as Philippi's Area darwini. Philippi's reconstruction of the anterior part of the shell is not quite correct (it being too high), and this is the only dif- ference from v. Ihering's figures. The fact that Philippi's species is from Santa Cruz is sufficient to establish this identity. Record of specimens : Mouth of Santa Cruz River, i double, i right valve ; Las Salinas, 2 right valves. Distribution: Santa Cruz (Phil.); Jegua quemada, Suprapatagonian beds (v. Ih.)._ Affinities: The type-species of the subgenus or genus Ctictillaria is C. heterodonta Desh. from the Eocene of the Paris basin (see: Deshayes, ORTMANN : TERTIARY INVERTEBRATES. 91 1860, p. 906, pi. 67, f. 22-25); it differs considerably from our species in size, shape and sculpture, the latter consisting of fine, sharp and nodulose ribs. C. aldrichi Dall (1898, p. 630, pi. 32, f. 19), from the Eocene of Alabama approaches our species more closely in size and shape and also in sculpture, which consists, in C. aldrichi, of fine, flattish, equal, but not geminate ribs. In the sculpture of the shell, C. tceniata Dall (ibid., p. 631, pi. 25, f. i), from the Pliocene of Florida and Carolina, is still more closely allied to our form, showing geminate, and even quadripartite ribs; the shape, however, of C. tceniata is different. Thus, it seems that C. dancim' is intermediate in sculpture between the Eocene C. aldrichi, and the Pliocene C. tceniata, approaching in form more the former. Fam. LIMOPSID^E Dall. Gen. LIMOPSIS Sassi. 29. LIMOPSIS INSOLITA (Sowerby). PI. XXV, Fig. 6. 1846 Trigonoccelia ins. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 252, pi. 2, f. 20, 21. 1864 Limopsis ins. Zittel, in: Novara Exp., p. 48, pi. 13, f. i. 1873 L. i. Hutton, Cat. Tert. Moll. Ech., New Zealand, p. 28. 1886 L. i. Tate, in: Tr. R. Soc., South Australia, vol. 8, p. 134. 1887 L. araucana Philippi, Tert. Quart. Verst. Chiles, p. 191, pi. 46, f. 4 (juv.). 1897 ^- ins- v- Ihering, in: Rev. Mus. Paul., vol. 2, p. 234. 1899 L. i. v. Ihering, in: N. Jahrb. Min. Geol. Pal., vol. 2, p. 14. Shell suboval, very oblique, thick, convex. External surface, in well preserved specimens, with very fine radiating striae (fine grooves separated by flat intervals), and with distinct concentric lines of growth. Umbones not much prominent. Area triangular, high, with a triangular depression in the middle below the apex, bordered by sharp margins. Hinge teeth forming a curved line, anterior and posterior ones larger, median ones smaller. Height, 26 mm ; length, 25 mm. Remarks: The external surface of the shell is finely striated, a char- acter in which L. araucana is said to differ. Sowerby and Zittel, ho\v- 92 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. ever, describe the surface of L. insolita as smooth, and, indeed, in most of the specimens it appears so. But this character is produced by fossili- sation, the fine striae being distinctly visible only in very well preserved individuals. This character has been ascertained already by v. Ihering. We possess young shells that agree in every respect with Philippi's L. araucana, they being less oblique than the older ones. V. Ihering (1899, p. 40, footnote) doubts the identity of the New Zealandian and Patagonian form. Zittel's figure of L. iusolita represents a very large specimen, which is very oblique, and which is smooth. That the latter character does not agree with the Patagonian fossil has been mentioned, but since the latter was described originally and erroneously as smooth, it is very probable that also in New Zealandian specimens, in well preserved individuals, a striation may be present. The external form does not warrant a specific separation, since among our Patagonian material the obliquity of the shell varies considerably in individuals of the same size, and since also — as has been said above — young individuals as a rule are less oblique. Accordingly, the very oblique shape of the figure given by Zittel may be due to age. In the configuration of the area I do not find any differences : although it is hard to understand Zittel's description in this respect ("area — traversed by a narrow, slightly depressed, very indistinct, triangular groove"), his figure corresponds to what we see in the Patagonian fossil : below the apex, there is, on the area, a broadly-triangular depression on a slightly lower level than the lateral parts of the area, and separated from the latter by a sharp, angular line. This triangular depression is slightly concave in the middle. Record of specimens : Mouth of Santa Cruz River, 175 isolated speci- mens, and many more imbedded in matrix; Las Salinas, i sp. Distribution: Santa Cruz (Sow.); La Cueva and Jegua quemada (v. Ih.). According to v. Ihering in both the Patagonian and Suprapata- gonian beds. Navidad beds of Llancahue, Chili (Phil.). New Zealand (Zittel), Miocene Pareora beds (Hutt). South Australia, in so called "Eocene" (? Miocene) beds (Tate). Affinities: By the very strongly developed obliquity, by the very broad triangular groove of the area, by the lack of crenulations of the inner margins, this shell represents a very peculiar type of the genus, that cannot be brought into closer relation to any of the known species. The ORTMANN : TERTIARY INVERTEBRATES. 93 more striking is the fact that this identical type is represented in the Pareora beds of New Zealand and in Australia. Fam. ARCID^E Dall. Gen. ARCA Lamck. 30. ARCA PATAGONICA v. Ihering. PI. XXV, Fig. 3«'4. 1897 -d- Pat- v- Ihering, in: Rev. Mus. Paul., Vol. 2, p. 235, pi. 4, f. 23, pi. 5, f. 30. Shell elongate, anterior part short, rounded, posterior elongated, obliquely truncated. Ventral margin almost straight and parallel to the upper margin (hinge-line). A blunt, but distinct carina runs down from the apex to the posterior ventral angle. Apices remote from each other, area broad, concave, with a large sulciferous rhombus. Surface of shell with radiating ribs, anterior ribs stronger, median ribs finer, and on the posterior part of the shell, above the carina, again 3-5 (according to age) stronger ribs. All ribs noduloso-imbricated by concentric lines crossing them. In' older parts of the shell (near the apex) the ribs are much crowded, but they become more distant on approaching the ventral margin, and finer ribs develop in the intervals. In old individuals, near the ventral margin, from 2 to 4 finer ribs are found between the stronger ones. ' Our largest individual measures : Length, 28 mm ; height, 20 mm ; diameter, 10 (x 2) mm. Records of specimens : Mouth of Santa Cruz River, ca. 30 sp. ; Mt. of Observation, upper horizon, i sp.; Arroyo Gio, i cast. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: A. oxytropis Philippi (1887, p. 188, pi. 37, f. 6) from the Navidad beds of Lebu, Chili — although very incompletely known — seems to be closely allied to this species. In A. oxytropis the following char- acters agreeing with A. patagonica are known : ( i ) the elongated and narrow form, (2) the oblique truncation of the posterior end, (3) the ridge running across the valve, (4) the existence of a few (3) ribs above this ridge. 94 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. V. Ihering compares this species with A. imbricata, which is, according to Dall (1889, p. 40), a living species of the West Indies. In my opinion the most closely allied forms are: A. tetragona Poli (= navicularis Desh.) from the English Crag (see Wood, 1856, p. 76, pi. 10, f. i), and said to be also Miocene and Recent. And further, a closely allied species is A. noce L. (Miocene to Recent, see Hoernes, 1870, p. 324, pi. 62, f. 4). Specimens of the same size of A. tetragona from the Pliocene of Mt. Mario, Rome, differ only in the more anterior position of the apex, finer ribs, and sharper carina. Although the type of A. noce is represented also in Eocene deposits by different species, none of these approach our species so closely as the two forms named (A. tetragona and noce}, so that we have here a distinctly Neogene relation. Similar forms, for instance A. occidentalis Phil, and A. paratina Dall (Philippi, 1851, p. 15, pi. 4, f. 4, and Dall, 1898, p. 621, pi. 33, f. 14), closely allied to the European A. noce, are found in Oligocene, Miocene, Pliocene, and Recent beds in Florida and the West Indies. Also A. pseudonavicularis Tate (1886, p. 139, pi. 1 1, f. 8) from so-called "Eocene" beds of South Australia and Tasmania belongs into this group. Gen. GLYCIMERIS Da Costa. (= Pectunculus Lamck.) 31. GLYCIMERIS IBARI (Philippi). PI. XXVI, Fig. i"-*. 1887 Pectunculus ib. Philippi, Tert. Quart. Verst. Chiles, p. 190, pi. 40, f. 3. 1887 Pectunculus magellaniciis Philippi, ibid., p. 190, pi. 41, f. i. ? 1887 P. araucanus Philippi, ibid., p. 191, pi. 36, f. 2. 1897 P- Pulvinatus cuevensis v. Ihering, in: Rev. Mus. Paul., v. 2, p. 238, pi. 7, f. 46, pi. 8, f. 50. 1899 P. pulv. cuev. v. Ihering, in: N. Jahrb. Min. Geol. Pal., v. 2, p. 14. 1900 P. ibari Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Shell large and thick, suborbicular or more or less oblique. Surface with radial striae, the striae are fine, but sharp furrows, separated by rather ORTMANN : TERTIARY INVERTEBRATES. 95 broad, flat intervals; they are crossed by fine concentric lines of growth. Area triangular, increasing in width with age, with rhombiform sulci up to 6 or 7. Hinge teeth up to 1 1 on each side ; the median ones, in old individuals, covered by the dilated area, so that only 4 to 5 on each side are visible. Only the median teeth slightly geniculate, the lateral ones oblique to horizontal, but straight. Remarks: As v. Ihering has already pointed out, the external form of this species is extremely variable. Sometimes it is perfectly circular, but often more or less oblique, or even transversely elongate ; the latter ex- treme is represented by Philippi's P. magellanicus. We possess from Punta Arenas all intermediate conditions in shape. Also the thickness of the shell is variable. (As to variations in shape and thickness in this genus, compare Wood, 1856, p. 68, under P. g/ycimeris.} Philippi's P. ibari represents an extremely thick and high individual, while his P. magellaiiicns is extremely transverse. Both are large, and possess accordingly a very broad area, which leaves only the 4-5 lateral hinge teeth exposed. The same number of teeth is present in our largest individuals (see below, No. n and 12). We have a smaller indi- vidual (No. 3) that corresponds in shape exactly to P. magellanicus, but has a larger number of hinge teeth. V. Ihering figures a very large indi- vidual of subcircular outline, in which the median hinge teeth are not completely covered. I give here the measurements of some of our specimens, compared with those of Philippi and v. Ihering. No. Height. length. Diameter. Anterior hinge teeth. Sulci. Shape. I 48 50 18 II I almost circular (left). 2 50 54 20 9 ? oblique (left). 3 66 80 27 9 ? very oblique, transverse (left). 4 70 ca. 70 25 8 ? almost circular (left). 5 70 72 30 12 ? subcircular (left). 6 72 7i 26 10 3 almost circular (right). 7 74 77 32 12 ? moderately oblique (left). 8 76 77 31 8 ? almost circular (right). 9 82 92 33 8 4 oblique (left). 10 87 ca.95 35 6 ? moderately oblique (right). 1 1 90 85 33 5 5 moderately oblique, high (left). 12 92 100 38 5 6 oblique (right). 13 90 116 37 5 7 very oblique, transverse. H 98 92 ca. 40 5 ? moderately oblique, high. 15 99 1 06 38 fi Q ca. o 4 or 5 subcircular. 96 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Of these individuals Nos. 1-12 are in our collection from Punta Arenas; No. 13 is Philippi's P. magellamcus, and No. 14 is Philippi's P. ibari. No. 15 gives the measurements of v. Ihering's P. piihnnatus cnevensis. The following are the corresponding measurements of our smaller indi- viduals from Santa Cruz: 16 17 18 28 25 19 29 25 19 9 8 6 10 9 very slightly oblique (left), circular (right), circular (right). Finally, we have here the measurements of an individual from Lake Pueyrredon, base of Tertiary : 19 | 93 90 | ca. 34 | ? ? | slightly oblique, high (left). Individuals from other localities are too poorly preserved to be meas- ured. It may be, that P. colcliagiictisis Phil. (1887, p. 191, pi. 37, f. 8) from the Navidad beds of Chili is identical with our species ; it is of medium size, and the measurements fit well into the series, but Philippi says that the surface shows only lines of growth, and hardly any radial striae ; and further, the sulci of the area seem to be — according to the figure — more numerous and more crowded. P. arancanus Phil, is a very young form ; Philippi says that it differs from other species in the rectilinear hinge line. The same character is present in a few of our smallest and medium sized specimens from Santa Cruz, but it disappears with age. Although these young specimens from Santa Cruz resemble completely Philippi's figure of P. araucanits, it does not seem quite safe to put this species in the synonymy of P. ibari, with- out having compared authentic specimens of the Navidad form. The failure on the part of v. Ihering to recognize the identity of his species with P. inagcllanicns and ibari of Philippi, is apparently due to the fact that Philippi has figured two extremes of this form ; an excep- tionally transverse one, and an exceptionally high one; but we possess from the type-locality (Punta Arenas) not only these two extremes, but also the intermediate forms. The latter prevail, are more circular in out- line, and agree well with v, Ihering's species (so especially our Nos. 4, 5, and 8). The large specimens in our collection from Rio Chalia are all casts, but agree well — as far as can be ascertained — with the specimens from Punta ORTMANN I TERTIARY INVERTEBRATES. 97 Arenas. Large specimens from Arroyo Gio and Lake Pueyrredon pos- sess at least part of the valves and are completely identical with those from Punta Arenas. Record of specimens: Mouth of Santa Cruz River, 14 valves (young and medium) ; Upper Rio Chalia, 8 casts (large) ; 30 miles north of upper Rio Chalia, 4 casts; Arroyo Gio, 2 valves (large and small) ; Lake Pueyr- redon, base of Tertiary, 2 sp. (large and small) ; Lake Pueyrredon, 600' above base, 11 casts (small and medium); Punta Arenas, horizon V, 12 valves (medium and large). Distribution: Punta Arenas (Phil.); Patagonian beds of Santa Cruz (v. Ih.), and Suprapatagonian beds of Jegua quemada and La Cueva (v. Ih.) ; Perhaps in the Navidad beds (Lebu) of Chili (P. arattcanus Phil.). Affinities: V. Ihering considers this form a variety of P. pulvinatus Lmck., and compares it with the form described by Hcernes as P. pilosus L. from the Miocene of the Vienna basin. This is quite right, especially as far as it relates to the sulci of the area. The larger Eocene species of the genus (P. obovatus Lmck., pulvinatus Lmck., polymorphus Desh.) dif- fer at a glance in the more crowded, and accordingly more numerous sulci, while in the Miocene, Pliocene and Recent form, that has been called by the Linnean name P. pilosus, the number of sulci, in individuals of corresponding size, agrees closely with that of P. ibari. In P. pilosus, however (see Hcernes,. 1870, p. 316, pi. 40, 41), these sulci are very in- distinct, while in P. ibari they are well developed and sharp, and further, P. piloszis differs from our species in the teeth of the hinge being more distinctly geniculate, and in the radial striae of the surface being less dis- tinct and more obscured toward the lower margin by the lines of growth. Thus P. ibari has a distinctly Neogene appearance. Fain. PERNID^E Zitt. Gen. PERNA Brug. 32. PERNA QUADRISULCATA v. Ihering. PI. XXIV, Fig. 2"'". 1897 P- y*1- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 231, pi. 9, f. 54. Shell thick, upper margin straight, anteriorly sharply angulated, pos- teriorly alate, angularly produced and sinuate. Hinge broad, with four distant, very broad grooves. 98 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Remarks: Our fragments from Santa Cruz do not show the external form of the shell. According to v. Ihering it is subquadrate, a little broader than high. The specimens from Lake Pueyrredon are chiefly casts ; one of them shows clearly the impressions of the lower margin of three of the cardinal grooves (see pi. XXIV., fig. 2*), agreeing with the specimens from Santa Cruz in being very broad and distant, which is the most prominent character of this species. Only part of the shell remains here. The casts indicate apparently a higher shell, but taking into con- sideration the great thickness of the shell at the anterior end, and the missing part at the posterior, which has been lost by fossilization, the outline of the Pueyrredon specimens would come near to that given by v. Ihering. I do not hesitate to refer them to v. Ihering's species, since the characteristic feature of P. quadrisulcata is exhibited in one of them. Record of specimens : Mouth of Santa Cruz River, 9 fragments; Lake Pueyrredon, base of Tertiary, 2 sp.; Lake Pueyrredon, 600' above base, 4 sp. (2 of them young). Distribution: La Cueva, upper part of Patagonian formation (v. Ih.). Affinities : No known species of Perna can be compared with this one ; but there is some resemblance to Crenatula aviculifonnis Philippi (Navidad beds of Chili), but the latter is too incompletely known to permit any opinion in this respect. Fam. OSTREWsE Lamck. Gen. OSTREA L. 33. OSTREA TORRESI Philippi. PI. XIV, Fig. 3"'". 1887 O. torr. Philippi, Tert. Quart. Verst. Chiles, p. 215, pi. 48, f. 8. 1899 O. torr. Ortmann, in: Amer. Journ. Sci., v. 8, p. 427. Shell in outline broader or narrower oval, rarely suborbicular ; mod- erately thick. Lower valve deeply excavated, upper valve flat. Area moderately large, triangular. Muscular impression about in the middle of the inner surface, only slightly lateral. Both valves with numerous, strong radial plaits, but on the upper valve these are less distinct and sometimes missing. Measurements: Length, 104, 117, 125, 131. An upper valve : 129 mm. Width, 74, 1 06, 74, 95. 102 mm. ORTMANN : TERTIARY INVERTEBRATES. 99 Remarks : Characterized by the well developed radial plaits, and by the deeply and regularly excavated lower valve, which has a spoon-like shape. All our individuals are much worn, but in spite of this fact the radial plaits are distinct. This character, as well as the comparatively thinner shell, are the only differences from the following species, of which it is no doubt the representative in the Magellanian beds. I am not quite satis- fied that it is really a distinct species. Record of specimens : Punta Arenas, Magellanian beds, upper horizon (III), 21 lower, 14 upper valves (5 upper valves from the lower horizon (II) of the Magellanian beds may belong to this species, but since they do not show any plaits, and lower valves are absent, I am not prepared to say that they really belong here). Distribution: Straits of Magellan (Phil.). Our specimens are appar- ently from the type-locality of this species, since most of the fossils described by Philippi from the Straits of Magellan (1887, p. 251) are from Punta Arenas. Affinities: A similar species is O. bellovacina Lmck. (see Wood, 1861, p. 17, pi. 3, f. i, pi. 7, f. 3) from the Lower Eocene of France and En- gland, but in the latter the radial plaits are stronger, and the outline is broader. 34- OSTREA INGENS Zittel. PI. XV, XVI, XVII, XVIII, and XIX, Fig. i". 1864 O. ingens Zittel, in: Novara Exped., p. 54, pi. 13, f. 3. 1864 O. nelsoniana Zittel, ibid., p. 55, pi. 11, f. 7. 1,873 O. nelson. & ing. Hutton, Cat. Tert. Moll. Echin., N. Zealand, p. 34. 1887 O. patagonica Philippi (non d'Orbigny), Tert. Quart. Verst. Chiles, p. 213 (pro parte, not fig. 2 on pi. 48). 1887 O. fermrisi Philippi (non d'Orbigny), ibid., p. 214 (not fig. 5 on pi. 48). 1887 O. botirgeoisi Philippi (non Remond), ibid., p. 215, pi. 48, f. 3. 1896 O. beneckei Moericke, in: N. Jahrb. Min. Geol. Pal. Beil., Bd. 10, p. 574, pi. 13, f. i. 1897 O.ferraresi Pilsbry (non d'Orbigny), in: Pr. Acad. Philad., p. 330. 1897 (Nov.) O. hatcheri Ortmann, in: Amer. Journ. Sci., v. 4, p. 355, pi. ii, f . i . 1897 O. philippii Ortmann, ibid., p. 356, pi. 11, f. 2. IOO PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. 1897 (Dec.) O. percrassa v. Ihering (non Conrad), in: Rev. Mus. Paul., v. 2, p. 221, pi. 9, f. 53, textfig. i. 1897 O. patagonica v. Ihering (non d'Orbigny), ibid., p. 222, pi. 9, f. 2 (O. orbignyi in tab.) (non O. patagonica, ibid., p. 326). 1899 O. hatcheriv. Ihering, in: N. Jahrb. Min. Geol. Pal., v. 2, p. 8. 1900 O. ingens Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Shell very variable in shape, circular, ovate, triangular, or elongated, becoming very large and thick with age. Lower (left) valve more or less convex, concave on inner side. Outer surface with concentric lines of growth, forming lamellae, which are sometimes very far apart from each other, giving thus a graduated appearance, but in other cases the lamellae are much crowded. Radial plaits in many cases completely absent, in others slightly and irregularly developed, but always less than in O. tor- resi. Upper (right) valve flat or slightly concave on the inner side, ex- ternally with lines of growth, forming lamellae, very rarely with traces of radial plaits. Beak of lower valve longer or shorter, often very much elongated, sometimes incurved or distorted. Area triangular, generally longer than broad, often very long, rarely broader than long. Ligament- groove deeper or shallower, broader or narrower. Beak and area of upper valve shorter than in the lower valve. Muscular impression large, situated generally a little below the middle of the shell and a little pos- teriorly (laterally). Margins of upper valve, close to the area, sometimes for a short distance with small crenulations (wrinkles or nodes) corre- sponding to small grooves in the other valve, but in most cases no traces of these crenulations are present. Largest specimen: Length, 255 mm; width, 162 mm. Remarks: The chief characters of this species are : 1. The large size, and extremely thick shell. 2. The situation of the muscular scar. 3. The smooth margin of the inner side of the valves. 4. The slight development of the radial folds. But even these characters are variable in a certain degree. Greater variations are shown in : (i) the outline, (2) the shape of area and beak, (3) the convexity of the valves, (4) size, and development of the surface-sculp- ture, especially of the lines of growth. Since there is an almost unparalleled confusion as to the identification and synonymy of the large oyster of the Patagonian beds, it seems neces- ORTMANN : TERTIARY INVERTEBRATES. IOI sary to support the position taken by the writer by a detailed account of the history of this form, and the results furnished by the examination of our material. It is exceedingly difficult to clear up the synonymy of the large oysters of Patagonia. The oyster described by d'Orbigny as O. patagonica has been frequently and almost generally confused with the present species, but — as will be demonstrated below — the original O. patagonica is not found at all in the Patagonian formation, that is to say, below the Santa-Cruz beds contain- ing mammalian remains. It is to be borne in mind, that O. patagonica is not recorded by d'Or- bigny himself from any locality south of San Julian ; and since our col- lections show, that the true O. patagonica is really present at San Julian, but in a higher horizon, it is very probable, that d'Orbigny did not pos- sess the species from the true Patagonian formation at all. The latter was first mentioned by Darwin (1846, p. 1 1 1 and passim), but without being distinguished from O. patagonica. Philippi (1887) possessed typical specimens of the Patagonian oyster from Punta Arenas, and called them by the name of O. botirgeoisi Remond, which was a mistake, as O. boiirgeoisi is from the Californian Miocene. At the same time he confounds, the true O. patagonica with the species found at Santa Cruz, and introduces two more species: O. ferrarisi d'Orb. and remondi from the Pliocene Coquimbo-beds of Chili. What he calls O. ferrarisi is not the O. ferrarisi of d'Orbigny (= patagonica}, since he dis- tinctly states, that crenulations of the margins are not present in his speci- mens. These crenulations are the only character by which O. patagonica can be distinguished from O. ingens in every case, and accordingly, Philippi's O. ferrarisi must belong to O. ingens. O. remondi, on the other hand, seems to belong to the true O. patagonica (see below). Mcericke, in 1896, mentions from the Pliocene Coquimbo-beds: O. remondi and transitoria Hup., both apparently the same species as Philippi's of the same names, and adds a new species : O. beneckei. The latter is nothing but a very old and large, typical individual of O. iiigen*. Its chief characters, the extreme thickness of the shell, the incurved beak, and the elongated and large area are exhibited in many of our larger in- dividuals. Incidentally Moericke (p. 575) mentions O. patagonica from Santa Cruz, and it seems, he understands by this name the more circular form found at this locality. 102 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. The present writer, in 1897, recognized first the difference of this oyster of the Patagonian formation from the true O. patagonica of d'Orbigny, but he made the mistake of regarding the more elongated form from the upper Rio Chalia and the more circular form of Santa Cruz as different species, and described them under the names of O. pJiilippi (substituting this name for the preoccupied bourge&isi of Philippi) and O. hatcheri re- spectively. In the same year, a few weeks later, v. Ihering called the more rounded form of Santa Cruz by the name of O. percrassa (preoccu- pied by Conrad for a Miocene species of N. America, see : Whitfield, 1894, p. 29, pi. 3, f. 1-4), giving as its chief characters the widely remote lamellae formed by the lines of growth, which gives a terraced or gradu- ated appearance to the external surface. This same character is found in the type-specimen of my O. hatcheri; but I do not regard it as of specific value. Some individuals show it very distinctly developed, but in others we have all transitional conditions to the crowded lines of growth of what I have called O. philippi, and, indeed, both characters — the distant and the crowded lamellae — may be present in one and the same individual at different stages of age (see pi. XV). Moreover, this percrassa-stage— which is apparently due to very vigorous growth of the shell — is not re- stricted to the locality of Santa Cruz or to the particular horizon of the Patagonian beds represented there. Very recently (1899, p. 8), v. Ihering has again discussed the oyster of Santa Cruz, adopting the specific name of hate hen. Here he does not consider the graduated appearance of the shell as of specific value, but still he maintains its specific difference from "patagonica" = philippii Ortm. The chief characters given are : 1. Very thick shell and rounded outline. 2. Short and broad area. 3. Ligamental fossa of upper valve concave or flat, not prominent on the lower margin of the area, or only slightly so. 4. Muscular impression far distant from the margin. Characters (i) and (2) have been discussed above, and are of no specific value: we have at the type-locality elongate shells as well as rounded ones, and individuals with an elongated area. These characters change sometimes in the same individual with age (see pi. XV). In fact, the typical O. hatcheri (or percrassa] is not the only form found at Santa Cruz, and it is not even the prevailing form (see : No. i under record of ORTMANN I TERTIARY INVERTEBRATES. 103 specimens). As to the situation of the muscular scar (character 4), it may be remarked, that the more central position is found in more rounded individuals, and is without doubt directly dependent on the outline of the shell. As to the third character introduced by v. Ihering, taken from the liga- mental groove of the upper valve, I fail to see any constancy in it. Indeed, there is considerable variation, it being sometimes concave, sometimes flat, sometimes slightly convex, and the lower margin is more or less prominent. The convex character, however, of the surface of the groove, is exhibited in comparatively few individuals, and the prominence of the lower margin is not always connected with a convexity of the surface. My type-specimen of O. pliilippii does not differ at all in this respect from O. hatcheri, and my specimens of the true O. patagonica from Entrerios (sent by v.- Ihering to the Princeton Museum) do not show a remarkable con- vexity, and also among my specimens of the true O. patagonica from S. Julian the fossa is sometimes flat and sometimes slightly convex. On the other hand, I have specimens of the typical O. hatcheri with the lower margin of the ligamental fossa very prominent, although its surface is flat, and many individuals from Cape Fairweather (see pi. XVIII) show a dis- tinctly concave fossa. I possess even a few individuals with circular out- line, which have a slightly convex and strongly prominent fossa: on the whole, this latter development is rare among our material, and shows all possible transitions to a concave fossa. Therefore, it is evident, that this character is of no use for distinguishing O. hatcheri and philippii, and also cannot be used for the distinction of the Patagonian oyster from O. patagonica. The form with the crowded lamellae from the Patagonian beds (v. Iher- ing following Ameghino, alleges that it is restricted to the so-called Supra- patagonian beds), has also been confounded by v. Ihering with the true O. patagonica, although he confirms the presence of the characteristic cren- ulations of the upper valve in specimens from Parana (Entrerios). In 1899 (p. 10), he emphasizes this character, but he says that crenulations are not always missing in the oyster of the Patagonian beds and not always present in the Entrerios oyster. But — according to my expe- rience— the first is the case in a very limited degree : the crenulations in Patagonian shells — if present at all — are found only for a very short dis- tance near the area (see pi. XVIII, fig. i6), while, on the other hand, the 104 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. lack of crenulations in the true O. patagonica is always due to an incom- plete state of preservation, the crenulations being worn off or having dis- appeared by the breaking off of the margin (see below under O. patagonica]. 0. pyrofhcrioi'itni of v. Ihering (1897, P- 3T5> textfig. 21, numbered 20 by mistake) seems to be a different species. It has a short, triangular outline, with crenulations of the upper valve. The figure is very poor, but since v. Ihering has kindly sent a specimen of this species to Prince- ton, I have verified his statement that the triangular form is very striking, and furthermore — as v. Ihering has already pointed out — the muscular impression is very peculiar, being very narrow and transversely elongated. The stratigraphical position of this form is said to be in the Pyrotherium beds, but since at. present nobody knows what these beds really are, and since — as Ameghino himself admits (1899, P- :3) — different horizons have been mixed up under this designation, I cannot venture to express an opinion on this subject. But at any rate, a Cretaceous age of these beds is out of question (see Hatcher, 1900, p. 96). I may, however, call atten- tion to the following facts : ( i ) O. pyrotherionim possesses, according to v. Ihering, crenulations on the margin of the upper valve, and (2) I have figured on pi. XX, fig. i^, an upper valve of O. patagonica, . from San Julian, belonging to the lot described below, which has a muscular scar that corresponds exactly to that of O. pyrotherionim. Thus two of the chief characters of O. pyrotherionim are also found in O. patagonica, and it may be, that O. pyrotheriorum is only a form of O. patagonica, and that the two individuals, upon which this species is based, have been picked up just for this peculiarity in external form. If this should prove to be correct, its stratigraphical position would be in much younger beds (Plio- cene), and this would support Mr. Hatcher's opinion, that a part at least of the Pyrotherium-beds of Ameghino belongs in the Pliocene. To sum up, we have the following results : 1. The large oyster of the Patagonian formation is different from the true Ostrea patagonica of d'Orbigny, and the only, but constant difference, is the presence of crenulations all around the margin of the upper valve in the latter. Such crenulations are sometimes present near the hinge in the Patagonian oyster, but they are very much smaller, and found only for a short distance, never all around the margin. 2. There is only one species of large oyster in the Patagonian forma- tion. From our material, I may pick out individuals corresponding to ORTMANN : TERTIARY INVERTEBRATES. -105 every single one of the figures of the different supposed species. And, in addition, we possess numerous intermediate forms, and individuals combining characters of these so-called species. 3. This large species of oyster continues its existence all through the (Miocene) Patagonian formation, as well as upward into the Pliocene Coquimbo beds of Chili. Now, the same is true in Patagonia. The identical form reappears in the Cape Fairweather beds, which are separated from the Patagonian beds by the whole of the Santa Cruz formation. (At the cafion near Sierra Oveja, No. 11, O. ingens has been found in beds interstratified with the lower Santa Cruz beds.) I cannot find any specific difference between the Patagonian and the Cape Fairweather forms, although there are some slight and almost insensible differences in the general features : the Cape Fairweather form does not grow quite as large, the shell — although still very thick — is not quite so gigantic in mass, the outline is more distinctly and more frequently of a triangular shape ; the crenulations of the margin of the upper valve near the hinge occur more frequently, but still there are many specimens which show no traces of them. An oyster closely resembling in these characters the Cape Fairweather form is found in the uppermost oyster bed at Punta Arenas (No. 22), and in the uppermost beds at Lake Pueyrredon (No. 19). As to the identification of the Patagonian oyster with the New Zea- landian form, I would make the following remarks : O. ingens and nelsoniana of Zittel are apparently identical. The latter is smaller, and accordingly younger, but Zittel mentions the character of the extraordinary thickness of the shell. O. ingens is a large individual, of elongated shape, and with a very long and broad area. Zittel says, that the area is limited on each side by a groove (sulcus), a character that is seen in some of our specimens. I do not find a single character by which it is possible to distinguish our species from the New Zealand- form, and, indeed, I can pick out individuals agreeing closely with the figures given by Zittel of O. ingens as well as of O. nelsoniana. A very closely allied, if not identical, form is O. stiirtiana Tate (1886, p. 97, pi. 6, f. i), from the Miocene River Murray Cliffs, Australia. Ostrea rostrata of Hupe (see Philippi, 1887, p. 213), from Coquimbo, of which no figure is known, seems to agree, according to the descrip- tions, with the form called by Moericke O. beneckei. If that should prove IO6 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. to be correct, the name of O. rostrata of Hupe antedates O. ingens of Zittel, and should be used accordingly. 0. transitoria of Hupe (Philippi, p. 213, pi. 49, f. 9), from Coquimbo, Caldera and Navidad (Pliocene and Miocene of Chili) may also belong here. It is very broad, but agrees in this character with many individuals from Patagonia. O. transitoria of Moericke (1896, p. 576, pi. 12, f. i), however, seems to belong to O. patagonica (see below). Lately, several species of Ostrea have been described by Grzybowski (1899, pp. 629-631) from the lower Miocene and Pliocene of northern Peru (Payta and Tumbez), of which O. latiareata (Miocene) and O. oculata and lunaris possibly also belong here. At any rate, they are closely allied to O. ingens, but the material described and figured is too poor to form an opinion upon. I shall now proceed to give a record of our specimens, adding under each locality the necessary notes, which would serve to support the views set forth above. Record of specimens : 1. Mouth of Santa Cruz River; 12 double, 7 lower valves. One of them is my type of O. hatcheri. 2 more of the double valves agree completely : they are broad, with distant lamellae. In 2 more the lamellae are more crowded near the lower margin ; i other is a little elongated, with crowded lamellae in the posterior third of the shell, other- wise like O. hatcheri. 5 are very large, greatly elongated (one of them figured on pi. XV and XVI), with the beak and area more elongated ; one of them with distinct lateral furrows as in O. ingens from New Zealand ; the lamellae are much crowded in the posterior part of the shell, i further specimen is of medium size and elongated ; the lamellae are crowded, except near the beak ; the area is long. Of the single valves 5 small or medium sized are the typical O. hatcheri ; form circular or broadly oval. 2 others are larger and more elongated, one of them very large, is very elongate, with the lamellae crowded in the posterior half; area long and broad, with distinct lateral furrows. The other is a little smaller than this one, ovate, the area triangular, the lamellae crowded posteriorly. In all specimens from this locality the radial ribs of the surface are only slightly or not at all developed. As will be noticed, the typical O. hatcheri is represented, among these 19 individuals, by only 8. The rest are transitions to O. philippii, and a few may be called O. philippii. ORTMANN I TERTIARY INVERTEBRATES. 107 2. Pescadores, Rio Santa Cruz ; i lower, i small upper valve. 3. Paso del Rio Santa Cruz ; i double valve. These are typical, broad O. hatcheri. 4. Mt. of Observation, lower horizon ; 1 2 double, 3 lower, i upper valve. Four of them are large, rounded, with short and broad area, but the lamellae, although a little distant, are more crowded than in the typical O. hatcheri. Only slight traces of radial ribs are present, i individual of medium size has the lamellae as in O. hatcheri, but its form is elongate, with long, triangular area. 3 of medium size have the lamellae more or less crowded, the form elongate, and the area long. The rest are young individuals of very irregular shape. 5. Mt. of Observation, upper horizon ; 2 double valves. One is of medium size, the lamellae distant, the outline elongated-tri- angular ; area very long and narrow. Slight traces of plaits. This is the first individual that shows traces of crenulations near the area on the upper valve. The other is young, ovate-triangular, with a long area. No char- acters of O. hatcheri. 6. San Julian, Oven Point ; 2 double, i lower valves. Both double valves are the typical O. hatcheri, and in one of them — a very rare case — the upper valve has radial ribs. The single valve is of a more ovate form. All 'three lower valves have distinct radial ribs ; in no other set of specimens are these radial plaits so strongly developed as in this one. 7. Port Madryn, New Bay, Terr, of Chubut; 2 lower valves from between tides, 2 lower and 3 upper valves from ca. 25' above high tide. - One lower valve of the lower horizon is the typical O. hatcheri; the others have the lamellae more crowded ; their outline is ovate or elongate. Area broadly triangular. 8. Shore of Salt Lake, 10 miles north of mouth of Rio Chico ; i double, 4 lower, 5 upper valves. Double valve, and i lower, typical O. hatcheri, with distant lamellae. The rest is more ovate or irregular. 2 isolated upper valves show dis- tinct traces of crenulations, in the larger of them the lamellae are remote. 9. Upper Rio Chalia ; i double, 5 lower, 9 upper valves. The double valve agrees absolutely with O. bourgeoisi Phil. One up- per valve is the typical O. hatcheri, with rounded outline and remote lamellae. Another upper valve is the typical O. philippii (= O. bourgeoisi, IO8 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. but with very long area). The rest are young individuals, of various shapes ; one upper valve shows crenulations on the margin ; a very young lower one (pi. XIX, fig. ic) shows the corresponding grooves. 10. 30 miles north of upper Rio Chalia ; 2 double, 6 lower, 2 upper valves. One of the double valves is the type of my O. philippii; the lamellae are on ^ of the shell remote, as in O. hatcheri. The other double valve, and one of the lower ones agree with this in the elongate form and long area, but the lamellae are crowded. Another lower valve is very broad, almost circular, area short and broad, and it agrees in these respects with O. hatcheri; but the lamellae are very crowded. The type of O. philippii has distinct crenulations near the area of the upper valve. 1 1. Canon near Sierra Oveja, Rio Chico (interstratified with Santa-Cruz beds) ; i double, 2 lower, 2 upper valves. The double valve is the typical O. philippii, but the area is a little shorter (= O. bourgeois*). Outline of the others irregularly oval, lamellae not much crowded, but less distant than in O. hatcheri. Area triangular. In one of the upper valves the lamellae are very distant for half its length. 12. Shell Gap, Rio Chico, lower horizon; i double, i lower, i upper valve. The double valve corresponds to O. hatcheri. The lower valve (pi. XVII) is elongated, area very long, lamellae crowded, only near the beak more distant. The upper valve is broadly ovate, the lamellae more crowded than in the typical O. hatcheri. 13. Mayer Basin, upper lignites; 2 lower valves. Elongate, with long area, corresponding to O. philippii. 14. Arroyo Gio; 2 double, 16 lower, i upper valves. One of the double valves is very large, round ; the other one small, circular ; the latter with remote lamellae, and distinct crenulations in the upper valve near the area ; it is a typical O. hatcheri, while in the larger one the lamellae are more crowded. One of the lower valves is very large, round, with lamellae like O. hatcheri, but the area is elongated. The upper valve is an ovate, typical O. bourgeoisi, with crenulations. The rest (15 lower valves) grow in clusters, of very irregular shape; no characters of O. hatcheri are present ; area mostly triangular, in a few very long; 4 of them show grooves corresponding to crenulations near the area ; they correspond to O. bourgeoisi and philippii. ORTMANN : TERTIARY INVERTEBRATES. 109 15. East end of Lake Pueyrredon; i lower valve. Young, oval. 1 6. Lake Pueyrredon, base of Tertiary; i lower valve. Form of O. hatcheri, but lamellae in posterior half more crowded. A fragment found washed out just below these beds, in the same matrix and of the same color, indicates a more elongated shell, with long area and crowded lamellae. 1 7. Lake Pueyrredon, 600' above base ; 2 double valves. The smaller one is a typical O. hatcheri; the larger one is more ovate, with crowded lamellae. 1 8. Lake Pueyrredon, extreme top of Patagonian beds; 2 double valves. Both broadly oval, like O. hatcheri, and with broad area, but lamellae more crowded. 19. Lake Pueyrredon, marine beds overlying Santa-Cruz beds; 8 double valves. No characters of O. hatcheri ; outline irregularly oval or triangular. Area triangular, shorter or longer. Size medium or small. Ribs indis- tinct or none. Crenulations in smaller valve in 3 specimens. Ap- proaches distinctly the Cape Fairweather form. 20. Punta Arenas, horizon V, lower layer ("below V") ; 2 double, 3 upper valves. Both double valves are the typical O. philippii: long-ovate, area very long, in one of them incurved ; lamellae in the latter quite distant. 2 of the upper valves agree with this form, but the third is subcircular, with a broad area, and distant lamellae, except near the margin, and it would thus correspond to O. hatcheri. 21. Punta Arenas, horizon V proper ; i double, 8 lower, 2 upper valves, Lower valves all much elongated, long-oval, beak much produced, nar- rowed toward the tip, area very long (see pi. XIX, fig. i*); correspond- ing to O. philippii. Lamellae not very distant, crowded near the margins. One upper valve is oval, with distinct crenulations near the area. The area of these specimens corresponds to Philippi's figure of the area of O. bourgeoisi. 22. Punta Arenas, "above horizon V" ; 2 double, 2 lower valves. Outline ovate or subtriangular. Area triangular, not very long. In one of the isolated lower valves the lamellae are very distant, as in O. IIO PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. hatclteri. The more oval specimens are fully identical with Philippi's main figure of O. bourgeoisi. These specimens approach closely the Cape Fair- weather form in the triangular outline, moderate area, and medium size and thickness of the shell. 23. Cape Fairweather ; 23 double, 1 2 lower, 5 upper valves. The characters of this form are given above. Crenulations of the mar- gin, close to the area (see pi. XVIII, fig. i*), are shown in 14 upper valves (out of 28). Thus we possess from these 23 different localities 77 double, 76 isolated lower, and 35 isolated upper valves. Altogether, 153 lower, and 112 upper valves are represented in our collection. Distribution: Darwin mentions this species from southern Patagonia generally, Philippi from Punta Arenas, v. Ihering from Santa Cruz and La Cueva (upper part of Patagonian formation and Suprapatagonian respectively); it has been found in the Pliocene Coquimbo beds of Chile (Moer.), and further in New Zealand, in the Oamaru and Pareora beds (Oligocene and Miocene) (Zitt, Hutt). Affinities: Moericke has already compared this oyster with O. crassis- sima Lamck. from the Miocene of Europe, and there is no doubt a strik- ing resemblance between them in general appearance. Still closer appears the relation to that form called O. gingensis (Schloth.), especially when we compare the account and figures given of it by Hoernes (1870, p. 452, pi. 76-80). Hoernes (p. 455) points out the close resemblance of O. gin- gensis and ingens. This type of oyster (cmssisst'ma-type) — although not exclusively found in the Miocene — is a very characteristic feature of Miocene deposits. In older Tertiary beds this giant type is not found. 35. OSTREA PATAGONICA d'Orbigny. PI. XX, Fig. i**. 1842 O.patagonica d'Orbigny, Voy. Amer. merid., v. 3, p. 33, pi. 7, f. 14-16. 1842 O. ferrarisi d'Orbigny, ibid. 1887 O. patagonica Philippi, Tert. Quart. Verst. Chiles, p. 215 (part), pi. 48, f. 2 (after d'Orbigny). 1887 O. ferrar. Philippi, ibid., pi. 48, f. 5 (after d'Orbigny) (non text, p. 214). ORTMANN : TERTIARY INVERTEBRATES. I I I 1887 O. remondi Philippi, ibid., p. 214, pi. 48, f. 4. 1896 O. reni. Moericke, in: N. Jahrb. Min. Geol. Pal. Beil., Bd. 10, p. 575, pi. 12, f. 2. 1896 O. transitoria Moericke, ibid., p. 576, pi. 12, f. i. 1897 O. patagonica Ortmann, in : Amer. Journ. Sci., v. 4, pi. 1 1, f. 4 (after d'Orbigny). 1897 O. patagonica v. Ihering, in: Rev. Mus. Paul., v. 2, p. 326 (non p. 222, nee pi. 9, f. 52). V. Ihering quotes as synonyms the following species of Philippi : O. burmeisteri, O. bravardi, O. longa, O. agghitinans, O. adsociata (Anales Mus. Nac. Chile, 1893): I cannot verify these, since I have no access to that paper. Differs from O. ingens in the presence of distinct crenulations all around the margin of the upper valve. In addition, we may mention, that there are sometimes folds on the lower valve, which are irregular, and comparatively larger than these of O. ingens (see the figure of O. remondi given by Philippi and our figure i* on plate XX). Remarks: These crenulations or wrinkles are the only distinctive char- acter, but it is a very good one. They are shown in one of the original figures of d'Orbigny, which represents an upper valve. V. Ihering men- tions them, and the two upper valves among the material sent by him to the Princeton Museum show them well developed. Among our material, all of the 33 upper valves show these crenulations, and they pass all around the margin in 26 of them. In the 7 remaining the lower margin is either broken away, or so much worn, that the crenu- lations have been obscured, but they are always seen at least in certain parts of the margin. In the 103 upper valves of O. ingens in our collec- tion in which the inner side is exposed, crenulations are never seen all around the margin, and only in 13 valves from the Patagonian beds, and in 14 from the Cape Fairweather beds they are present for a short dis- tance near the area. We cannot wish any better specific character in an oyster! To the crenulations of the upper valve correspond, in the lower valve, little grooves. But these are in most cases obliterated, or have been destroyed by the mutilation of the margins. So it is difficult, in many cases, to identify isolated lower valves, unless radial ribs are present. I 1 2 PATAGONIAN EXPEDITIONS '. PALAEONTOLOGY. Out of 14 lower valves in our material, about 10 show these radial ribs of the outer surface distinctly, the rest indistinctly or not at all. The 4 lower valves sent by v. Ihering do not show them. Our individuals are small or of medium size. The largest lower valve measures: Length, 135 mm; width, 79 mm; the largest upper valve: Length, 142 mm; width, 91 mm. O. ferrarisi is nothing but the young of this species. O. remondi of Philippi is certainly this species ; the lower valve figured by him shows radial folds of a character that is never found in O. ingens, but frequently in O. patagonica, and Philippi expressly states, that the margin of the upper valve possesses crenulations ("margine interne exquisite denticu- late"). These crenulations are wanting in Philippi's O. ferrarisi, and therefore I consider his specimens of this species to belong to O. ingens, while his figure is a copy of d'Orbigny's O. ferrarisi = patagonica. O. remondi of Moericke is apparently the same as O. remondi Phil. O. tramitoria of Hupe and Philippi is doubtful (see above). But O. transi- toria of Moericke is certainly identical with O. patagonica, since it has crenulations ("deutliche Kerbung am Schalenrande"). O. patagonica seems to be the descendant of O. ingens. The crenula- tions of the margin begin to develop in O. ingens, but are rare there, only in the Cape Fairweather form they are more frequent. But they never extend all around the margin. In the true O. patagonica this character is fully developed, and — as it seems — is always present, unless obscured or destroyed by the process of fossilization. Record of specimens: San Julian, Darwin Station, above Patagonian beds; 14 lower, 33 upper valves. (V. Ihering has sent to Princeton from Parana, Entrerios: i double, 3 lower, i upper valves.) Distribution: Entrerios ; Punta Gorda (mouth of Uruguay) ; Rio Negro ; San Julian (d'Orb.). Santa Rosa (or Punta Raza?, see under discussion of Cape Fairweather beds below), between Santa Cruz and San Julian, Tehuelche Formation (v. Ihering, p. 225) ; Parana, Entrerios (v. Ih.). Pliocene of Chili : Coquimbo (Phil.), Caldera (Moer.). ORTMANN : TERTIARY INVERTEBRATES. 113 Gen. GRYPH^EA Lamck. 36. GRYPH^EA CF. TARDA Hutton. PI. XIV, Fig. 4"-'. 1873 G. tard. Hutton, Catal. Tert Moll. Ech. New Zealand, p. 35. 1886 G. tard. Tate, in: Tr. R. Soc. S. Australia, v. 8, p. 98, pi. 6, f. 2. Lower (left) valve ovate-triangular in outline, tumid, with incurved umbo. Exterior smooth, with concentric lines of growth ; posterior margin produced into a distinct lobe. Height, 59 mm; width, 58 mm (but posterior lobe-like expansion damaged). Remarks: According to Tate, G. tarda is very close to the European and North American upper Cretaceous G. vesicularis (Lamck.), and differs chiefly in the more triangular outline and larger lobation. Especially the first character would apply to our specimen, which agrees well also in the side view with Tate's figure. Our specimen, however, does not possess the upper valve, the inner side of the lower is filled with hard matrix and not exposed, and further, the lobe-like expansion of the posterior margin is much damaged. Thus it is hard to say, whether our Patagonian fossil is really identical with the Australian species or not. The close resem- blance to O. vesicularis is quite striking, and there is hardly any doubt that we have to deal with a species of the genus Gryphcza. Record of specimens : High bluffs, S. W. of Lake Pueyrredon, ca. 1000' below Santacruzian beds, i sp. Distribution: G. tarda has been mentioned first by Hutton from the Chatham Islands, from beds, which belong probably to the Oamaru and Pareora series (Oligocene and Miocene). Tate records this species from South Australia (Aldinga Bay and Bunda Cliffs) from supposed Eocene beds (lowest beds of marine series of older Tertiary). The genus Gryphcea is generally supposed to have disappeared at the close of the Cretaceous period, but we must bear in mind that already Whitfield (1885, p. 224) has recorded G. vesicularis from the Eocene marls of New Jersey, and according to Zittel (1885, p. 20) it continues to Recent times. The stratigraphical position of our specimen is not very well ascertained, but it has been found associated with a number of 114 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Brachiopods, which are characteristic for the lower part of the Lake Pueyrredon (Rio Tarde) section, and belong undoubtedly in the Pata- gonian series : thus the Tertiary age of our fossil, and its association with Patagonian fossils seems to be well established. Fam. PECTINID^. Lamck. Gen. PECTEN Mueller. 37. PECTEN PROXIMUS v. Ihering. PI. XXI, and PI. XXII, Fig. i«-. 1897 P- centralist. Ihering, in: Rev. Mus. Paul., v. 2, p. 229, pi. 8, f. 48, 49 (non P. centralis Sow.). 1897 P. proximus v. Ihering, ibid, in tabula. 1900 P. prox. Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Valves equilateral, unequal, the left one flat, the right one convex. Outline suborbicular. Ears large, subequal, the anterior one a little larger ; byssal sinus wanting. Sculpture of the convex (right) valve : about 6 large, rounded principal folds, the 2-4 median ones distinct, the lateral ones indistinct Each fold with 4-8 strong, radial ribs. Intervals between the folds concave, a little narrower than the folds, near the apex of the shell without ribs, and finely squamulate. Toward the lower margin strong ribs begin to develop in the intervals (from i to 5), and, on the margin, the whole surface of the shell, folds as well as intervals, are covered with strong radial ribs. The finely squamulate sculpture is found on the ribs also, but toward the margin it disappears through obliteration by transverse larger squamae, which develop on the ribs. The ribless intervals are different in extent in different individuals : some- times they begin to show ribs at an earlier age than in other cases. Flat (left) valve of a similar sculpture : 5-7 principal folds, but intervals broader than the folds, ribs more strongly squamate, the squamae begin- ning nearer to the apex. Ears in both valves with radiating ribs, which are less strong than those of the valves, subequal, and squamose. Largest valve : Height, 21 cm; width, 22.5 cm (but not quite complete). Remarks: Young individuals, which have the intervals between the large folds smooth (except for minute squamulae) exhibit quite a different ORTMANN : TERTIARY INVERTEBRATES. 115 aspect from large individuals, where the lower half of the valve is covered completely with strong radial ribs. V. Ihering's figure 48 (right valve, Proximiis] represents an individual in which the intermediate ribs have not begun to develop. This species cannot be the P. centralis of Sowerby, as will be seen below. Record of specimens : Mouth of Santa Cruz River; 5 double, 9 right, 4 left valves. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities : A most closely allied form is P. caracolensis Steinmann (1881, p. 254, pi. 14, f. 10) =P. simpsoni Philippi (1887, p. 210, pi. 46, f. i ) from the Navidad beds of Chili, but in P. caracolensis (according to Philippi's figure) the principal folds are more numerous and narrower, and the ribs are less numerous. P. athleta Zittel (1864, p. 49, pi. 10, f. i) from the Oligocene (Oamaru, Hutton, 1873, p. 32) of New Zealand is also closely allied. It differs in the same characters and seems hardly distinct from P. caracolensis. As to the relation to P. caloosaensis Dall see below (under next species). 38. PECTEN PR^ENUNCIUS v. Ihering. PI. XIX, Fig. 2a'». 1897 P- Pr<%n. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 230. Similar to P. proximus in the unequal valves, and the large folds (5-7). The differences are : Right valve with a slight byssal sinus. No radial ribs, but only fine radial striae. Our specimens are smaller than P. proxiimts : Height, 63 mm; width, 54 mm. Remarks: The characters given above seem to agree with v. Ihering's species, and moreover, through the kindness of Dr. v. Ihering, I possess a lead-pencil sketch of P. prcsnuncius, which removes all doubt as to the identity of our specimens with this species. In the two left valves from San Julian the radial folds are quite strong, especially in the one figured (pi. XIX, fig. 2*). The right valve is small, and the folds are of medium size. Of the specimens of Santa Cruz, one of the right valves agrees completely with that from San Julian ; in the other one the radial folds are very slight (pi. XIX, fig. 2"). The left valve from Santa Cruz shows hardly any traces of folds, and is almost perfectly flat. Il6 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Record of specimens : Mouth of Santa Cruz River ; 2 right, i left valves ; San Julian, Darwin Station ; i right, 2 left valves. Distribution: Gulf of S. Jorge, Patagonian formation (v. Ih.). Affinities: There is a strong resemblance to P. caloosaensis Dall (1898, p. 731, pi. 29, f. 12) from the Pliocene of Florida, but in the latter the striae are stronger, more rib-like, and P. caloosaensis approaches in this respect more the younger individuals of P. proximus. The Australian representative of this species is : Pecten palmipes Tate (1886, p. 105, pi. 5, f. 4, pi. 7, f. 4) ; it has been found in so called (?) Eocene beds of Edithburgh, Yorke Peninsula, and of Aldinga Bay (South Australia). 39. PECTEN CF. CENTRALIS Sowerby. PI. XXIII, Fig. i"-6. 1846 P. centr. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 253, pi. 2, f. 31. Sowerby figures a left valve. It is characterized by 5 (text, the figure shows 6) radial ribs, which are thin and sharp, and separated by broad, concave intervals. This character is exhibited in our left valve, but the intervals are finely squamulose, without striae, while Sowerby describes "numerous rough radial lines." Sowerby did not possess right valves. Our right valve, which was found in connection with the left, agrees completely with that of the fore- going species (P. prcznuncius}. Anterior ear without sinus. Both valves are much broken, very delicate, and have been put in a bed of plaster, some fragments of the left valve, however, are not in the proper place ; I give the figures, as is the present condition of the shell. Measurements of left valve : Height, ca. 63 mm ; width, ? There remains some doubt whether this is really Sowerby's species ; but since it comes from one of his type-localities, it may be that it is this species. Record of specimens: Port Desire, N. E. side, i right and i left valve, belonging together. Distribtttion : San Julian and Port Desire, one fragment from the first locality, two of the latter (Sow.). Sowerby and Darwin (1846, p. 113) mention this species also from Santa Cruz. ORTMANN : TERTIARY INVERTEBRATES. I 1 7 40. PECTEN GEMINATUS Sowerby. PI. XXIII, Fig. 2°". 1846 P. gem. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 252, pi. 2, f. 24. 1897 P- quemadensis v. Ihering, in: Rev. Mus. Paul., v. 2, p. 228, pi. 6, f. 38. 1899 P- fissicostalis v. Ihering, in: N. Jahrb. Min. Geol. Pal., v. 2, p. ,n pi. i, f. i. 1900 P. geminatus Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Both valves almost equally convex, the right one a little less so. Outline suborbicular in old individuals, subtriangular in younger ones. Anterior ears considerably larger than the posterior, that of the right valve with a deep byssal sinus. Outer surface of right valve with from 20—30 strong ribs, arranged somewhat irregularly in pairs. All the principal ribs dis- tinct down to the lower margin, but on the lateral parts of the shell they become less distinguishable from the smaller (secondary) ribs. Intervals between the principal ribs a little broader than the ribs, occupied by from i to 5 secondary ribs or striae. All ribs covered with squamae. In very old individuals the number of the intermediate ribs increases to 7, and those adjoining the principal ribs grow larger, giving a fasciculate appear- ance to the principal ribs. The median secondary rib in each interval is usually a little stronger than the rest. In the left valve the character of the ornamentation is practically the same, but the geminate character of the principal ribs is altogether lacking. Measurements: Height, 105, 85, 35 mm. Width, 107, 80, 28 mm. Remarks: The character of the geminate ribs is present only in the right valve, and distinct only on the median part of it ; near the lateral margins principal and secondary ribs are hardly distinguishable. Young individuals, of about the size of Sowerby's figure, ca. 30 mm, show only the principal ribs and a few single striae in the intervals, so that the total number of ribs is only about 20 to 24. In larger individuals, of the size of v. Ihering's figure of P. quemadensis, ca. 35-40 mm, the intermediate ribs become more numerous, and especially the ribs near the lateral mar- gins increase in number, so that we may count 24-30 ribs. In still larger Il8 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. individuals the number of the principal ribs does not increase materially with age, but that of the secondary ribs does considerably, so that we may count, in very large ones, 90-100 ribs of different size. Furthermore, this species varies in the development of the larger ribs. In some cases (so chiefly in large individuals from Santa Cruz) the prin- cipal ribs are very distinct and much larger than the secondary. A few individuals from Oven Point and Darwin Station show the same char- acter, but in the larger number from Oven Point this difference is not so strongly pronounced, although the principal ribs are still well marked. A third variety is found, to which belongs the larger number from Darwin Station ; here the fasciculate appearance of the ribs, shown in old indi- viduals of the typical form, begins at an earlier stage: the intermediate ribs closest to the principal ones become stronger, while the principal rib itself is not so strongly contrasted to them in size, and we have in shells of medium size, already an appearance of the principal ribs being com- posed of from two to four smaller ones. In these specimens the ribs appear to be more numerous, a little finer on the average, and more crowded, and they represent completely the form described by v. Ihering as P. quemadensis. There is no sharp line to be drawn between these different forms, and the form quemadensis, although not found among the larger specimens from Santa Cruz, is exhibited in a few younger individuals from this locality, and in a few larger ones from Oven Point. As has been stated, it is the prevailing form at Darwin Station. There are all possible tran- sitions between the different forms in the development of ribs. The outline of the shell changes with age. Young shells appear more elongate, subtriangular, while larger shells are broader and more rounded (see measurements). V. Ihering's P. fissicostalis is nothing but the cast of a larger individual of this species. We have received from v. Ihering one cast under this name, from Santa Cruz : it agrees completely with casts of P. geminatus represented in our collection, and still connected with the shell. Record of specimens: Mouth of Santa Cruz River, 7 right, 11 left valves; San Julian, Oven Point, 3 double, 17 right, 17 left valves ; San Julian, Darwin Station, 6 double, 12 right, 7 left valves; Canon near Sierra Oveja, i left valve ; Shell Gap, lower horizon, 3 single shells (imbedded in matrix) ; Arroyo Gio, i right valve ; East end of Lake ORTMANN : TERTIARY INVERTEBRATES. 119 Pueyrredon, 2 casts ; Lake Pueyrredon, base of Tertiary, 4 casts ; Lake Pueyrredon, 600' above base, 4 casts; Lake Pueyrredon, marine beds overlying Santa Cruz beds, 9 casts. Distribution: San Julian (Sow.) ; Santa Cruz, LaCueva, Jegua quemada (v. Ih.). According to v. Ihering, P. fissicostalis is from the Patagonian, P. quemadensis from the Suprapatagonian beds. Affinities: A very closely allied form is P. coquimbensis Moericke (1896, p. 577, pi. 13, f. 7-10) from the Pliocene Coquimbo beds of Chili, but in P. coquimbensis both valves are said to be almost flat, and the intermediate ribs are less numerous ; primary ribs 26-28. The following species (P. actinodes] is also closely related, see below. 41. PECTEN ACTINODES Sowerby. PL XXIV, Fig. i°'6. 1846 P. act. Sowerby, in: Darwin, Geol. Observ. S. Amen, p. 253, pi. 3, f- 33- 1897 P- a°t- Pilsbry, in: Pr. Acad. Philad., p. 330. 1897 P- act- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 227. Shell suborbicular. Right valve very slightly convex, almost flat ; left valve convex. Anterior ear much larger, that of the right valve with a deep byssal sinus. Sculpture of both valves consisting of 30-40 principal ribs, which are only slightly elevated, and not very different from the in- termediate ribs, of which 3-7 exist in each interval. Sometimes the in- termediate ribs closest to the principal ribs are a little stronger, giving a sub- fasciculate appearance to the principal ribs. All ribs covered with squamae. Height, 112 mm; width, 104 mm. Remarks: This species differs from P. gemmatns in the complete ab- sence of geminate ribs on both valves, in the larger number and smaller size of the principal ribs, and the flattened right valve. It will be re- marked, that the form quemadensis of P. geminatus makes in some degree a transition toward this species. Record of specimens : Cape Fairweather ; 1 5 right, 1 1 left valves, and a number of fragments. Distribution: San Josef (Sow.); Bay de la Pava, north of Desire, and Punta Rosa (or Raza ?, see under discussion of Cape Fairweather beds below), between Santa Cruz and San Julian, Tehuelche formation (v. Ih.). I2O PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Affinities: There is no doubt that this species is the descendant of P. geminatus. Closely allied is: P. tenuicostatus Hup. (Philippi, 1887, p. 210, pi. 47, f. i, and Moericke, 1896, p. 580, pi. 12, f. 13-16) from the Navidad beds of Tubul, Chili, but in P. tenuicostatus the byssal sinus is less developed, and the ears are not so unequal. P. vidali Phil. (1887, p. 212, pi. 47, f. 5) from the Pliocene of Coquimbo has stronger, more distinctly fasciculate, and less numerous ribs ; the right valve is not so flat. Fam. MYTILID^. Flem. Gen. MYTILUS L. 42. MYTILUS CF. CHORUS Molina. PI. XXV, Fig. I0-6. 1843 M. ch. d'Orbigny, Voy. Amer. men, v. 5, p. 647. 1858 M. ungulatus Reeve, Conch, icon., v. 10, pi. 2, f. 4. 1887 M. chorus Philippi, Tart. Quart. Verst. Chiles, p. 202. 1889 M. ch. Clessin, in: Martini & Chemnitz, Syst. Conch. Cab., v. 8, p. 65, pi. 5, f. i, pi. 9, f. i, 2. 1897 M. cf- ch- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 232, pi. 9, f. 55. Shell elongated-oval, thick, smooth, except for the growth lines ; apex acute ; dorsal margin slightly curved, about half as long as the total length of the shell, forming a very obtuse angle with the posterior mar- gin. Ventral margin almost straight. Meastirements (of Cape Fairweather specimen) ; Length, 1 1 1 mm ; height, 63 mm. Remarks: Our specimens from Rio Chalia are very poor; they are in- complete casts, with a few fragments of the shell remaining. As far as can be made out, they agree in shape well with M. chorus, and since this species is also mentioned by v. Ihering from the Patagonian beds, it seems quite probable, that we have to deal with this species. One of the casts from Cape Fairweather shows well the external form, and is indis- tinguishable from the rest. Record of specimens: Upper Rio Chalia, remains of 6 valves; Cape Fairweather, 2 casts. Distribution: Living : Chili. Fossil ': Santa Cruz, Patagonian beds (v. Ih.); Quaternary of Chili, and doubtfully in the Navidad beds (Phil.). ORTMANN I TERTIARY INVERTEBRATES. 121 43. MYTILUS MAGELLANICUS Chemnitz. PI. XXIV, Fig. 3. 1785 M. mag. Chemnitz, N. syst. Conch. Cab., v. 8, p. 162, pi. 83, f. 742, 743- 1843 M. mag. d'Orbigny, Voy. Amer. mer., v. 5, p. 647. 1858 M. mag. Reeve, Conch, icon., v. 10, pi. 6, f. 22. 1873 M. mag. Hutton, Cat. Tert. Moll. Ech., N. Zealand, p. 25. 1886 M. mag. Hutton, in: Trans. N. Zealand Inst, v. 18, p. 365. 1900 M. mag. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Shell elongated-triangular, apex subacute ; dorsal margin slightly con- vex, forming an obtuse and indistinct angle with the posterior margin ; ventral margin straight or slightly concave, forming a rounded angle with the posterior margin. Surface of shell sculptured by radiating, wrinkled, dichotomous ribs. Length, 50 mm; height, 23 mm. Remarks: I have compared our specimens with recent specimens of M. magellanicus collected by Mr. Hatcher at various localities on the Patagonian coast, and find that they agree completely. Record of specimens: San Julian, Oven Point; internal and external casts of about 10 specimens. Distribution: Living: coast of Patagonia and Straits of Magellan. Fossil: Philippi (1887, p. 249) mentions M. magellanicus from Lota, Chili, "probably quarternary," but does not give it in the text, pp. 200-202. Hutton records it from the Miocene (Pareora), Pliocene (Wanganui) and Pleistocene beds of New Zealand. Gen. MODIOLA Lamck. 44. MODIOLA AMEGHINOI v. Ihering. PI. XXV, Fig. 2. 1897 M. am. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 233, pi. 6, f. 43. Shell oblong, elongate, subtrapeziform, smooth. Cardinal part of dorsal margin very slightly convex, posterior part slightly concave ; these two parts meeting at an obtuse angle ; posterior part passing in a regular curve 122 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. into the posterior margin. Ventral margin slightly concave. Apex situ- ated a very short distance from the anterior end. Remarks: Our specimens differ from the original description in the more ventricose valves and apex, but the general form agrees well with v. Ihering's species, so that I do not think they are different. Record of specimens : Mouth of Santa Cruz River, i right valve ; Mt. of Observation, upper horizon, i left valve. Distribution : Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: Our specimens come very near to M. coquimbana of Philippi (1887, p. 203), especially to the variety figured in Philippi's pi. 44, f. 7, from the Pliocene of Coquimbo, Chili. In M. coqiiimbana, however, the apex is more anterior. 45. MODIOLA ANDINA Ortmann. PL XXIV, Fig. 4. 1900 M. and. Ortmann, in: Amer. Journ. Sci., v. 10, p. 370. Shell small, elongate, about 2^ times as long as high. Apex near anterior end. Both valves convex, with a blunt ridge running down from the apex to the posterior and inferior end. This ridge is curved, the con- cave side of the curve directed toward the lower margin. Upper margin almost straight in its anterior (cardinal) part, forming a blunt angle with the posterior part, which is almost straight, and passes by a regular curve into the rounded posterior margin of the shell. Ventral margin distinctly concave, and forming with the posterior margin a right, but blunt angle. Surface of shell finely radially striated in the upper part, i. e., above the oblique ridge crossing the valve ; the striae most distinct near the poste- rior half of upper and near posterior margin. Lower part of shell, below the ridge, smooth, only with few lines of growth. Anterior end of shell, below and in front of the apex, with a few (5-7) fine striae, which are often very indistinct. Measurements: Length, 24, 23 mm. Height, 9, 10.5 mm. Remarks: No Modiolce are known from South American deposits that might be compared with this one. In M. rugulosa and keviuscula of Philippi (both from Lebu, Chili), radial striae are present, but they are not distributed in the particular manner as in this species, and, furthermore, the outline of the shell is quite different. ORTMANN I TERTIARY INVERTEBRATES. 123 Record of specimens: Lake Pueyrredon, base of Tertiary, i sp.; Lake Pueyrredon, 400' above base, ca. 35 sp.; Lake Pueyrredon, 600' above base, ca. 8 sp. Fam. CRASSATELLITID^E Ball. Gen. CRASSATELLITES Krueger. 46. CRASSATELLITES LYELLI (Sowerby). PI. XXVI, Fig. 9". ». 1846 Crassatella I. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 249, pi. 2, f. 10. 1897 C. 1. v. Ihering, in: Rev. Mus. Paul, v. 2, p. 246. Shell obovate, comparatively thin and flat, rounded anteriorly, angu- lated posteriorly ; posterior dorsal margin oblique, straight or slightly con- cave near the apex ; anterior dorsal margin straight near the apex, hardly concave. Surface with broad concentric grooves, which are separated by blunt ridges. Length, 44 mm; height, 33.5 mm; diameter, 5.5 (x2). Remarks: V. Ihering says that the ventral margin is crenulated. I do not see any crenulations in our specimens. Record of specimens: Mouth of Santa Cruz River, i double, i right, 3 left valves. Distribution: Santa Cruz (Sow.), Patagonian formation (v. Ih.). 47. CRASSATELLITES KOKENI (v. Ihering). PI. XXVI, Fig. io"'*. 1899 Crassatella k. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 17, pi. 2, f. 2. Shell ovate, subtriangular, comparatively thicker than in C. lyelli, and more convex ; rounded anteriorly, angulated posteriorly, but not so much produced as in C. lyelli. Posterior dorsal margin oblique, slightly con- vex near apex; anterior dorsal margin distinctly concave below apex. Surface markings nearly as in C. lyelli, but the ridges separating the con- centric grooves are a little sharper. Ventral margin sharply crenulate. Measurements : Length, 26mm; height, 22 mm; diameter, 6 (x 2) mm. 124 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Remarks: Differs from C. lyelli in the subtriangular, shorter outline, the concave lunular margin, and the crenulations of the lower margin. As to the latter point, compare remarks above. Record of specimens : Mouth of Santa Cruz River, 4 right, 4 left valves. Distribution: Santa Cruz, Patagonian formation (v. Ih.). Affinities: C. plicatilis Deshayes (1860, p. 745, pi. 18, f. 6, 27) from the Middle Eocene of Paris comes near this species, but C. kokeni is narrower posteriorly, the surface markings are much more distant from each other, and the anterior dorsal margin is concave. Still more closely allied is C. sulcata (Sol.) (see Wood, 1871, p. 170, pi. 23, f. n) from the Upper Eocene of England and France, which agrees especially well in sculpture, while the outline is more like that of C. plicatilis, although more produced posteriorly. 48. CRASSATELLITES QUARTUS (Ortmann). PL XXVII, Fig. i. 1900 Crassatella quarta Ortmann, in: Amer. Journ. Sci., v. 10, p. 371. Shell ovato-elongate, comparatively thin, less convex than in the pre- ceding species, but more so than in C. lyelli. Apex only slightly promi- nent. Anterior end of shell rounded, posterior hardly angulated and hardly narrowed. Posterior dorsal margin straight near apex, anterior straight, only with a slight suggestion of concavity close to the apex. Surface ornaments as in C. lyelli, but the ridges more crowded, and a little less strongly developed. Ventral margin without crenulations. Measurements of Santa Cruz specimen: Length, 25 + x ; height, 15 mm; of Lake Pueyrredon specimen: Length, 17 mm; height, 10 mm. Remarks: The single isolated valve from Santa Cruz is broken pos- teriorly, so that the complete length cannot be given. According to the lines of growth, however, we have here the following relations : Length, 1 6 mm; height, 9 mm, which agrees well with the Lake Pueyrredon specimen, and would bring up the total length of the shell to about 30 mm (not quite double its height). Record of specimens: Mouth of Santa Cruz River, i left valve; Lake Pueyrredon, 600' above base, i cast of left valve. ORTMANN : TERTIARY INVERTEBRATES. 1 25 49. CRASSATELLITES LONGIOR (v. Ihering). PI. XXVII, Fig. 2. 1897 Crassatella I. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 247, pi. 5, f. 34, pi. 6, f. 37. Shell elongate-triangular, slightly convex, thick. Apex at about y$ of the length. Anterior end rounded, posterior produced and narrowed. Posterior dorsal margin oblique, forming a blunt angle with the pos- terior margin. Ventral margin curved anteriorly, almost straight pos- teriorly, forming a distinct acute angle with the posterior margin. An indistinct angulation runs down from the apex to the posterior angle. Surface marked with shallow concentric furrows. According to v. Iher- ing, the ventral margin is crenulated. Length, 44 mm; height, 27 mm; diameter, 8 (X2). Attains, accord- ing to v. Ihering, almost double that size. Record of specimens : Lake Pueyrredon, base of Tertiary ; i right valve. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities : This species resembles C. melina Conr. (see Whitfield, 1894, p. 60, pi. 8, f. 11-13) from the Miocene of New Jersey, but it is more elongated and less convex, and the ventral margin is straight in the pos- terior part. A more remote resemblance exists with C. ponderosa Phil, from the Navidad beds of Chili, the latter being larger, thicker, and less elongate, with the ventral margin arcuate throughout. Fam. CARDITID^. Gill. Gen. CARDITA Brug. 50. CARDITA ELEGANTOIDES Ortmann. PI. XXVI, Fig. 5«-c. 1899 C. el. Ortmann, in: Amer. Journ. Sci., v. 8, p. 428. Shell subcircular, slightly oblique, with 18-19 radial ribs, which are convex and obtuse, about as broad as the intervening furrows, and nodu- lose. Lunula small, oblong. 126 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Measurements (of type specimen): Length, 16 mm; height, 14 mm; of specimens from Santa Cruz : Length, 18.5, 15. 5 mm; height, 17, 14.5 mm. Remarks: This species was first described by the present writer in 1899 from a few poorly preserved individuals from the Magellanian beds of Punta Arenas. Later, among the last set of fossils brought home by Mr. Hatcher, I found a larger number of better preserved valves of a small Cardita from the Patagonian beds of Santa Cruz, which agree completely with the Punta Arenas fossil. The chief characters of this species are : the shape and the size of the shell, and the number and shape of the ribs. Young individuals of C. incequalis, of the same size as this species, may be distinguished at a glance by the larger number of ribs. C. volckrnanni, which is closely allied, differs in the smaller number of ribs. Record of specimens: Punta Arenas, horizon III (upper Magellanian), 3 isolated right valves, 2 valves in matrix ; Mouth of Santa Cruz River, 23 right, 17 left valves; Mt. of Observation, upper horizon, i double, 4 right, i left valves. Affinities: I compared this species with C. elegans Lmck. of the Euro- pean Eocene. But since there are so many similar species known from Tertiary deposits, it is impossible to say that just this one is the most closely allied form. This is so far the only species that is common to the Magellanian and Patagonian beds. 51. CARDITA VOLCKMANNI Philippi. PI. XXVI, Fig. 6. 1887 C. v. Philippi, Tert. and Quart. Verstein. Chiles, p. 173, pi. 37, f. 4. 1900 C. v. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. The only difference from the preceding species is the number of the radial ribs: there are only 15 of them. The ornamentation of the ribs is not preserved in our specimens. Record of specimens : Lake Pueyrredon, 600' above base of Tertiary, 6 casts. Distribution: Navidad beds of Tubul, Chili (Phil.). ORTMANN : TERTIARY INVERTEBRATES. 127 52. CARDITA INL-EQUALIS Philippi. PI. XXVI, Fig. 7-*. 1887 C. in. Philippi, Tert. and Quart. Verst. Chiles, p. 173, pi. 37, f. 5. 1897 C- i*1- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 245. 1897 C. patagonica v. Ihering, ibid., p. 244. 1899 C. pat. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 16. 1900 C. incequ. Ortmann, inj Amer. Journ. Sci., v. 10, p. 380. Shell ovato-subquadrate, oblique. Surface with 21-26 radiating ribs, which are sharply angular and high, crossed by lines of growth, and ren- dered sharply nodulose by them. In old individuals the ribs become more rounded, and the nodules disappear toward the ventral margin. Length, 43 mm ; height, 40 mm. Remarks: In outline this shell is somewhat variable: the apex is more or less inclined. Philippi's specimens had all lost the outer layer of the shell. In our specimens the latter is in most cases at least partly pre- served, and the shell has, as regards the sculpture, quite a different appear- ance. The radial ribs are sharp, angulated, and on their upper edge is a series of sharp nodules. The valleys between the ribs are deep, but their bottom is flat. If the outer layer is gone, the ribs appear as flat, broad, and smooth elevations. Toward the ventral margin, in some of the old individuals, the ribs become comparatively narrower with broader intervals, while the general character of the ribs remains the same ; in others, however, the ribs become more rounded and broader, and lose their nodules. Although individuals of medium and large size may be easily recog- nized, young ones (smaller than 15 mm) do not always exhibit distinctly the characteristic outline, and especially the apex is not so much inclined. It is sometimes difficult to distinguish these from C. elegantoides, but, as a rule, in C. elegantoides the apex is more upright, and the posterior hinge tooth of the left valve is shorter than in the present species. Furthermore, the number of ribs is different. I have 12 shells from Santa Cruz — all small — which have the ribs of C. incequalis (over 20), but the shape of the shell is that of C. elegantoides. In C. incequalis the posterior hinge tooth in the left valve is about four times as long as the anterior, but there are considerable variations in the thickness of the hinge teeth. 128 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. V. Ihering has sent to us one right and one left valve of this species under the name of C. patagonica ; they are from Santa Cruz, and he called this species apparently by that name (patagonica) in 1899. But the ex- ternal shape of the true C. patagonica of Sowerby is entirely different, and we cannot regard it as this species, unless we assume that Sowerby's figure is all wrong. There is, however, hardly anything that could warrant this assumption, although there are discrepancies between Sowerby's diagnosis and his figure. Record of specimens: Mouth of Santa Cruz River, 2 double, 20 right, 29 left valves ; Las Salinas, 3 right, 4 left valves ; Mt. of Observation, upper horizon, 2 double, 12 right, 2 left valves. Distribution: Santa Cruz (Phil., v. Ih.); Patagonian formation (v. Ih.). Suprapatagonian beds of Jegua quemada and La Cueva (v. Ih.). 53. CARDITA PATAGONICA Sowerby. PI. XXVI, Fig. 8M. 1846 C.pat. Sowerby, in : Darwin, Geol. Observ. S. Amer., p. 251, pi. 2, f. 17. 1897 C. pat. var. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 245. 1899 C. psetidopatagonica v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 16. 1900 C. pat. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell ovate, subtriangular, hardly oblique, as high as long, or higher; apex very slightly inclined. Surface with 23-25 radial ribs, which are rounded, not angular, and slightly squamuloso-nodose near the apex, but toward the ventral margin they are crossed only by lines of growth. Measurements: Length, 20, 19 mm; height, 20, 20 mm. Remarks: In the external form our specimens agree with Sowerby's figure, especially the almost upright apex is very striking. Sowerby, in the diagnosis, calls the ribs narrow, angular and squamoso-serrate, char- acters which are not supported by the figure, where the ribs appear broad, rounded, and slightly nodulose. In all essential respects the figure cor- responds with our specimens, with only the exception that it is almost double the size (length, 35 ; height, 37). Since a large form, agreeing with this figure has never been found by other collectors at Santa Cruz, it seems possible that Sowerby's figure was drawn on an enlarged scale. V. Ihering, in 1897, mentions this smaller form from the so called Santacruzian (= Suprapatagonian) beds, and calls it in 1899 C. pseitdo- ORTMANN : TERTIARY INVERTEBRATES. 1 29 patagonica, and this is certainly identical with our species. Besides, v. Ihering records from the Patagonian formation a C. patagonica, but it seems to me that this identification is not correct. He does not give any detailed description of this supposed C. patagonica, but (as has been said above) he has sent to the Princeton Museum 2 specimens of C. inccqualis under that name, so that there is no doubt that his C. patagonica is really C. inccqnalis. If my presumption is correct that Sowerby's figure is enlarged, then it is beyond doubt that this small form represented in our collections is the typical C. patagonica. Hutton (1886, p. 364) identifies his Venericardia intermedia (1873, p. 24) with "C patagonica" but I am unable to say whether this species corresponds to our C. patagonica. Record of specimens : Mouth of Santa Cruz River; 9 right, 4 left valves. Distribution: Santa Cruz (Sow.), ibid., Patagonian formation (v. Ih.); Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: Sowerby compares this species with the European Eocene C. acuticostata Lmck. (Wood, 1861, p. 142, pi. 22, f. 5), but there is hardly any close relation with it. C. caiimotiensis Deshayes (1860, p. 774, pi. 61, f. 6-8) from the Eocene of France has a similar outline, but is much smaller and has more numer- ous and finer ribs. C. gram-data Say (Whitfield, 1894, p. 56, pi. 9, f. 1-4) from the Miocene of New Jersey agrees in the slightly oblique outline and the number and character of ribs, but it is more circular and the apex more incurved. The most closely allied form seems to be : C. dunkeri Phil. (1846, p. 50, pi. 7, f. 7) from the Lower Oligocene of Germany. It agrees well in sculpture and outline, but the latter is more circular, with hardly an indication of triangular shape, and the apex is slightly more inclined. In this species also the hinge teeth of the right valve closely correspond to C. patagonica. Earn. LUCINID^ Flem. Gen. LUCINA Brug. 54. LUCINA NEGLECTA spec. nov. PI. XXVII, Fig. 3. Shell suborbicular, lentiform ; posterior dorsal margin slightly convex, forming an obtuse angle with the posterior margin. Anterior dorsal 130 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. margin slightly concave, forming an indistinct, rounded angle with the anterior margin. Anterior, ventral, and posterior .margins forming part of a regular circle. Beak small. Surface with concentric, elevated lines, which are crowded, especially near the apex. Hinge teeth not visible. Length, ca. 22 mm; height, 21 mm; diameter, 4 mm (x 2). Remarks: In my preliminary report on the Magellanian beds of Punta Arenas, I confounded these shells with Dosinia magellanica (see below), a mistake that was due to the fact, that they are imbedded in the same piece of rock with the latter species, and had not been worked out of the matrix sufficiently. This species resembles much the following (L. promaucanci], but is distinguished by the concentric lines, which, in L. neglecta, are more crowded, especially toward the apex, and which are more irregular toward the ventral margin. L. neglecta seems also to be a little less convex, and the lunula — as far as can be seen — is less concave and less distinct. Record of specimens: Punta Arenas, horizon II (lower Magellanian), i double, i right, 2 left valves. 55- LUCINA PROMAUCANA Philippi. PI. XXVII, Fig. 4"'6. 1887 L. prom. Philippi, Tert. & Quart. Verst. Chiles, p. 181, pi. 24, f. 6. 1897 •£• Prom. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 47, pi. 5, f. 32. Shell suborbicular, lentiform, posterior dorsal margin almost straight, forming an obtuse angle with the posterior margin ; anterior dorsal margin slightly concave, forming an indistinct, rounded angle with the anterior margin. Anterior, ventral, and posterior margins forming part of a regular circle. Beak small. Surface with concentric, rather widely distant, elevated lines. Hinge (of right valve) with 2 cardinal teeth, the posterior slightly divided ; anterior and posterior lateral tooth small, but distinct. Our largest specimen (Punta Arenas) measures: Length, 25 mm; height, 22 mm ; diameter, 5 (x 2), another one (Paso del Rio Santa Cruz) : Length, 21 mm; height, 19 mm; diameter, 5 mm (x 2). According to Philippi and v. Ihering it attains the length of 31 mm by a height of ca. 28 mm. ORTMANN : TERTIARY INVERTEBRATES. 131 Record of specimens: Mouth of Santa Cruz River, 2 double valves; Paso del Rio Santa Cruz, i right valve; Upper Rio Chalia, 2 casts; Punta Arenas, horizon V (Patagonian beds), 2 right valves. Distribution: Jegua quemada, Jack Harvey, La Cueva, Suprapatago- nian beds; perhaps also from Santa Cruz, Patagonian beds (v. Ih.); Santa Cruz (Phil.). Navidad beds of Chili: Navidad, Matanzas, Lebu (Phil.). Affinities: Philippi compares this species with L. radula Lmck. = bore- alis (L.) (see : Hoernes, 1870, p. 229, pi. 33, f. 4), to which it is very closely allied indeed. L. borealis is Miocene to Recent in Europe, and also Miocene to Recent in California (see: Gabb, 1869, p. 100). In L. pro- inancana the lateral hinge teeth are more distinct than in the European species ; I have seen, however, in the Princeton collections, individuals of L. borealis from the Pliocene of Italy, which also have distinct lateral teeth. L. pmecedens v. Koenen (1868, p. 246, pi. 28, f. 8) from the Mid- dle and Upper Oligocene of Germany is hardly distinguishable from L. borealis. 56. LUCINA ORTMANNI v. Iliering. PI. XXVII, Fig. 5. 1899 L. o. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 18, pi. 2, f. 3. Shell suborbicular, convex; posterior dorsal margin straight, forming an obtuse angle with the posterior margin. Anterior dorsal margin straight, forming an obtuse angle with the anterior margin. No lunula. Beak small. Anterior, ventral, and posterior margins forming part of a circle. Two very blunt ridges running from the apex toward the middle and lower part of the posterior margin (according to v. Ihering) : these ridges are not shown in his figure. In our specimen an indistinct depres- sion runs toward the lower posterior margin, which would correspond to the space between the ridges mentioned by v. Ihering. Surface almost smooth, very finely concentrically striated, and with concentric growth- lines, but no elevated lines are present as in the two preceding species. Length, 22 mm; height, 20 mm; diameter, 6 ( x 2) mm. Record of specimens : Mouth of Santa Cruz River; i right valve. Distribution: Santa Cruz, Patagonian beds (v. Ih.). Affinities: v. Ihering compares this species with L. globosa Ad. We know several Tertiary species, which resemble this one in form and sculp- 132 PATAGONIAN EXPEDITIONS 1 PALEONTOLOGY. ture, but until we know the interior of the shell and the hinge, it is im- possible to say whether there are closer relations to any of them. Fam. CARDIID^. Fisch. Gen. CARDIUM L. 57. CARDIUM PHILIPPII v. Ihering. PI. XXVII, Fig. 6. 1887 C. multiradiatum Philippi, Tert. & Quart. Verst. Chiles, p. 178 (pro parte, non C. miiltiradiatnm Sowerby). 1897 C- philippiiv. Ihering, in: Rev. Mus. Paul., v. 2, p. 49, pi. 6, f. 40. 1899 C. ph. var. paiiciradiata v. Ihering, in: N. Jahrb. Miner., etc., v. 2, P- IS- Shell large, subglobular, with about 45 radiating ribs, of which i or 2 near the anterior end are large and geminate. The other ribs are high and smooth, those of the posterior end finer and tuberculate. Remarks: Our specimens are all badly preserved, but they agree, espe- cially those from Santa Cruz, with v. Ihering's species, which is, according to v. Ihering, different from C. multiradiatum of Sowerby. They seem to correspond to the variety paiiciradiata. Our best individual shows only one geminate rib anteriorly. Two casts from Lake Pueyrredon, 600' above base, show distinctly 2 larger geminate ribs. One of them is very small (about as large as the following species), but it is distinguished at once by the smaller number of ribs. The casts from Arroyo Gio are doubtful : only traces of the ribs are seen, which seem to correspond to this species. Record of specimens : Mouth of Santa Cruz River, 2 right valves, and several fragments ; Arroyo Gio, 2 casts ; east end of Lake Pueyrredon, i cast; Lake Pueyrredon, base of Tertiary, i cast; Lake Pueyrredon, 600' above base, 6 casts. Distribution : Suprapatagonian beds of Jegua quemada, and Patagonian beds of Santa Cruz (v. Ih.); Santa Cruz (Phil.). ORTMANN : TERTIARY INVERTEBRATES. 133 58. CARDIUM PUELCHUM Sowerby. PI. XXVII, Fig. 7. 1846 C. p. Sowerby, in : Darwin, Geol. Observ. S. Amer., p. 251, pi. 2, f. 15. 1899 C. p. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 15. Shell subglobular, posterior end with an indistinct angulation running down from the apex toward the posterior margin. Surface of shell with very numerous (60-80) radiating ribs, which are about as broad as the intervals. On the posterior part of the shell these ribs are a little higher and sharper. In all our specimens (as in Sowerby's) the outer layer of the shell is gone : only in that from Las Salinas remains of it are still present; here the ribs are flat, and appear to be separated by narrow, impressed lines. Length, 30 mm ; height, 30 mm ; diameter (double shell), 20 mm. Record of specimens : Mouth of Santa Cruz River, 2 double, 7 single valves ; Las Salinas, i double valve ; Mt. of Observation, upper horizon, i double valve ; Canon near Sierra Oveja, 2 casts: Distribution: Santa Cruz (Sow.), ibid., Patagonian beds (v. Ih.) Affinities: C. coinatnliitn Bronn (see: Speyer, 1864, p. 301, pi. 41, f. 10, and v. Koenen, 1868, p. 244, pi. 29, f. 1,2) from the Middle and Upper Oligocene of Germany, and the Miocene of the Azores, seems to be closely allied. It has the same general form, but is smaller, and the radiating ribs are less strongly developed. The same type of Cardium is continued from the Miocene to the Recent time in Europe by C. fragile Brocchi (see: Hoernes, 1870, p. 173, pi. 30, f. 6). An Eocene representative of these species is C. difficile Deshayes (1860, p. 572, pi. -55, f. 6, 7), but it is distinguished by the distinctly broader form and more distinct posterior angulation. C. puelchum is clearly more closely allied to the Oligocene and Neogene species mentioned. 59. CARDIUM PISUM Philippi. PI. XXVII, Fig. 8. 1887 C. p. Philippi, Tert. & Quart. Verst. Chiles, p. 179, pi. 9, f. 9. Shell small, ovate, subglobular, scarcely oblique. Surface with 25-30 radiating ribs, which are rounded and crossed by concentric lines of growth. 134 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Length, 9.5 mm; height, 10.5 mm; diameter, 4.5 (x 2) mm. The specimen from Las Salinas : height, 1 1 mm. Remarks: v. Ihering (1897, p. 251) hints that this may be only a variety of C. puelclinm : but the very much smaller number of ribs does not sup- port this view. Philippi says that the radiating ribs are indistinct near the anterior and posterior margins: this is true in our specimen from Santa Cruz, but this feature is due to the exfoliation of the upper layer of the shell. In the specimens from Lake Pueyrredon the shell is partly preserved, and the ribs are distinct also near the anterior and posterior margins, although a little finer and less high than in the middle. Record of specimens: Mouth of Santa Cruz River, i sp. ; Las Salinas, i sp. ; Arroyo Gio, i cast; Lake Pueyrredon, base of tertiary, 11 sp. Distribution : Santa Cruz (Phil.). 'Affinities: A closely allied species is C. sphccridiuui Phil, from Lebu (Navidad beds), but in the latter species the ribs are finer and more numerous. Gen. AMATHUSIA Phil. 60. AMATHUSIA ANGULATA Philippi. PI. XXVII, Fig. 9"'». 1887 A. ang. Philippi, Tert. & Quart. Verst. Chiles, p. 135, pi. 23, f. i, pi. 25, f. i. 1897 ^- anS- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 257, textf. 2. Shell large, smooth, subcordate, oblique, with irregular concentric lines of growth. Apex at */$ of the length of the shell. Anterior dorsal margin straight, posterior first straight and horizontal, then oblique, forming with the posterior margin a rostrum. Length, ca. 190 mm; height, ca. 150 mm. Remarks: There cannot be any doubt that the proper position of the genus Amathusia is near Cardium, and in the family CardiicUz. Philippi points out the resemblance of the hinge to that of Cardimu, but relying on the external form of the shell he prefers to place it with JScnns. V. Ihering (1899, p. 38) believes that AmatHusia is related to Glycinieris (Panopcea], but I cannot see on what grounds. Indeed, there are no characters at all, which would warrant the position of this genus with ORTMANN : TERTIARY INVERTEBRATES. 135 either the Veneridce or the Saxicavidce (Glycimeridce}. On the other hand, the hinge, with the exception of the anterior part, agrees so closely with that of the Cardiidce, that, comparing large species of Cardium, for in- stance C. discrcpans Bast., C. laqueatum Conr., C. sulcatum Lmck., one is at once struck by the close resemblance. Indeed, the hinge is identical, but for the complete lack of the anterior lateral tooth in Amathusia. The hinge has two cardinal teeth in each valve, and one posterior lateral tooth ; the lack of the anterior lateral tooth cannot be regarded as a serious reason for separating this shell from the Cardiidce, since in this family the lateral teeth are obsolete in other genera. The ligamental plates (nymphae) are very high in Amathusia, and sepa- rated from the umbones by a very deep furrow, a condition that is often seen in species of Cardium (for instance C. discrepans, see : Hoernes, 1870, pi. 24, f. i, 2), where it is developed almost in the same degree as in Amathusia. The pallial impression in Amathusia possesses an almost rectangular upward curve posteriorly, which can hardly be called a sinus. The same character, and even a distinct sinus is found in some Cardiidce, so that this character also does not argue against the position with the Cardiidce. The most striking characters that distinguish AmatJinsia from Cardium are : ( i ) the lack of the anterior lateral tooth, of which no trace is pres- ent; (2) the complete absence of radial sculpture of the shell, and the lack of crenulations of the lower margin. Record of specimens: Mouth of Santa Cruz River (just above high tide) ; 2 double, 2 left valves. Distribution: Navidad, Chili (Phil.); Jegua quemada, Suprapatagonian beds (v. Ih.) (p. 257; on p. 258, v. Ihering says that his specimens are from Santa Cruz). Fam. VENERIDA1 Leach. Gen. VENUS L. 61. VENUS DIFFICILLIS Ortmann. PI. XXVIII, Fig. i"-". 1899 V. d. Ortmann, in: Amer. Journ. Sci., v. 8, p. 428. Shell thick, oblique, inflated, posteriorly a little narrowed; apex situ- ated in advance of or at l/4 of the length. Area long, occupying almost 136 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. the whole of the posterior dorsal margin of the shell. Lunula oval, flat. Surface with close and regular concentric furrows, and with some concen- tric lines of growth, the latter more crowded near the lower margin and irregular. Concentric furrows sharp, not interrupted, y2 to i mm distant from each other. Margin of shell not crenulated within. Length, 75 mm; height, 64 mm; diameter, 18 ( x 2) mm; apex at 16 mm from anterior end. Remarks : Surface more or less well preserved in specimens from the lower horizon, but in those from the upper horizon obscured by adhering coarse matrix, although still recognizable. Record of specimens : Punta Arenas, horizon II (lower Magellanian) ; 2 double, 2 left valves. Punta Arenas, horizon III (upper Magellanian) ; 2 right, i left valves, and several fragments. Affinities: I have compared, in my preliminary report, this species with V. subsulcata Phil. (1887, p. 115, pi. 17, f. 7) from the Cretaceous beds of Chile. And indeed, this seems to be the most closely allied form. 62. VENUS ARENOSA Ortmann. PI. XXVIII, Fig. 2"'". 1899 V. a. Ortmann, in: Amer. Journ. Sci., v. 8, p. 428. Shell transversely elliptical, moderately swollen. Posterior end hardly narrower than the anterior. Apex situated at about y$ of the length. Area indistinct, shorter than the posterior part of the dorsal margin. Nym- phae X~K as l°ng as ^e area. Lunula indistinct. Exterior surface with strong concentric lines of growth, which have between them finer concentric striae. Hinge that of a true Venus. Length, 60 mm; height, 44 mm; diameter, 15 (x 2) mm. Remarks: The surface markings are obliterated on account of the closely adhering matrix. Record of specimens : Punta Arenas, horizon III (upper Magellanian), 3 right valves. Affinities: This species possesses a very characteristic, elongated out- line, and resembles in this character — as I have pointed out in my pre- liminary report — V. landbecki Phil. (1887, P- IJ6> pi- 20, f. 8) from the Cretaceous of Chili. V. landbecki, however, differs in the position of the apex, more inflated valves, and more distinct area. ORTMANN : TERTIARY INVERTEBRATES. 137 63. VENUS CHILOENSIS Philippi. PI. XXVII, Fig. 10. 1887 V. cJi. Philippi, Tert. and Quart. Verst. Chiles, p. 121, pi. 15, f. 6. 1900 V. ch. Ortmann, in: Amer. Journ. Sci., v. 10, p. 378. Shell ovato-elliptical, moderately swollen, both ends well rounded ; apex at about one-fifth of the length. Area indistinct, nymphae immersed. Lunula ovate, flat. Surface with widely distant, elevated lines, and sharp radiating striations. Measurements: Length, 44 mm; height, 37 mm; diameter, 12 (x 2) mm ; according to Philippi : Length, 62 mm ; height, 52 mm ; diameter (double), 32 mm. Remarks: We possess only a single right valve, which is much smaller than Philippi's figure, but agrees in all essential respects very well with it. Since Philippi mentions this species from the Straits of Magellan, we may safely assume that our individual comes from one of the type-local- ities, as most of the fossils recorded by Philippi from "Magellanes" are from Punta Arenas. Record of specimens : Punta Arenas, horizon V (Patagonian), i right valve. Distrfoition : Ancud, Chile, and Straits of Magellan (Phil.). Affinities: This species comes near V. meridionalis in sculpture, but differs in size and outline. Other relations see under V. meridionalis. 64. VENUS MERIDIONALIS Sowerby. PI. XXVII, Fig. na'6. 1846 V. mer. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 250, pi. 2, f. 13. 1887 K mer. Philippi, Tert. and Quart. Verst. Chiles, p. 120, pi. 14, f. 8. 1897 V. mer. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 251. 1899 V. mer. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 19. Shell ovate, convex, both ends rounded. Apex between ^ and % of the length. Area indistinct, nymphae immersed. Lunula well marked, broadly lanceolate, a little convex and prominent. Surface with rather 138 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. widely distant, elevated and sharp concentric lines, which are, toward the ventral margin, more crowded ; besides, there are distinct radiating striae. Margin of shell very finely crenulated. Length, 33 mm; height, 27 mm; another individual: Length, 31 mm; height, 24 mm; diameter (double), 15 mm. Remarks: This species differs from V. chiloensis chiefly in the outline; it is more elongated (rel. H.: L. = i : more than 1.2, while in V. chiloensis it is = i : less than 1.2). Further, the position of the apex is different, and the lunula is convex and distinctly elevated in the middle in V, meridionalis. Young individuals, however, have a more circular outline. Record of specimens : Mouth of Santa Cruz River, 14 double, 57 iso- lated valves ; Las Salinas, 3 isolated valves ; Mt. of Observation, upper horizon, 2 isolated valves ; Shell Gap, Rio Chico, upper horizon, 2 casts ; Lake Pueyrredon, 600' above base, 17 casts. Distribution: Patagonia'n beds of Santa Cruz (Sow., v. Ih.); Supra- patagonian beds of Jegua quemada (v. Ih.); Navidad beds of Chili : Navi- dad (Sow., Phil.), Ranquil near Ancud (Phil.). Affinities: A species similar in outline and sculpture is Mercenaria cancellata Gabb (see: Whitfield, 1884, p. 68, pi. 12, f. 2, 3), from the Miocene of New Jersey. It is intermediate in outline between V. merid- ionalis and chiloensis. In V. clathrata Duj. (see Hoernes, 1870, p. 125, pi. 13, f. 3), from the Miocene of Europe the same type of surface orna- mentation is seen, but much more strongly developed. Moreover, V. cla- thrata is much higher and more rounded than even V. chiloensis. This type of ornamentation in the genus JSenus (cancellated surface) is characteristic for species from Miocene to Recent deposits. It is repre- sented in Australia and Tasmania by V. multitceniata Tate (= mnltilamel- lata Tate, 1887, p. 154, pi. 15, f. 6), and V. hormophora Tate (ibid., p. 155, pi. 15, f. i), said to be Eocene, but being probably Miocene. According to Hutton (1886, p. 362), Cliione uellicata Hutt. (1873, p. 21), is identical with V. meridionalis, and thus this species would also belong to the Pareora and Wanganui beds of New Zealand. ORTMANN : TERTIARY INVERTEBRATES. 139 65. VENUS VOLCKMANNI Philippi. PI. XXVIII, Fig. 3«-». 1887 V. v. Philippi, Tert. & Quart. Verst. Chiles, p. 121, pi. 14, f. 9. 1897 ^ v- var- argcntina v. Ihering, in : Rev. Mus. Paul., v. 2, p. 252, pi. 7, f. 45- 1899 V. v. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 20. Shell suborbicular, very convex; posterior dorsal margin slightly con- vex, forming an obtuse angle with the posterior margin, which forms, with the ventral margin, part of an almost regular circle. Apex between l/z and % of the length. Area indistinct, nymphae immersed. Lunula cor- date, elevated in the middle. Surface with rather widely distant, elevated lines, and distinct radiating striae. Measurements: Length, ca. 73 mm ; height, 70 mm ; diameter, 27 (x 2). Another individual: length, 55 mm; height, 51 mm; diameter (double), 35 mm. In the latter, the apex is 15 mm from the anterior end. Remarks: Philippi gives the position of the apex as 2^ = ca. jof the length, v. Ihering as ^ = ca. % . This difference is apparently in part due to the different position given to the shell, as the regular outline of the shell renders it difficult to give a uniform position to different specimens. The form from Santa Cruz is hardly distinguishable as a variety from that from Chili (in 1899 v. Ihering mentions typical individuals from Santa Cruz). The convexity of the dorsal margin is variable, and the anterior end (below the lunula) is more or less produced in different individuals : There are specimens from Santa Cruz completely agreeing with Philippi's figure in this respect. The cast from Shell Gap does not show any surface markings, but agrees in form. The casts from Lake Pueyrredon, however, show distinct remains of the cancellations. The cast from the mouth of Santa Cruz River shows the impression of the cardinal teeth : they are those of a true JSenus. Record of specimens : Mouth of Santa Cruz River, i double cast (near high water mark), i double valve, 2 isolated valves (250' above high tide). Shell Gap, Rio Chico, upper horizon ; i cast. Lake Pueyrredon, base of Tertiary ; 3 casts. Lake Pueyrredon, 600' above base ; 2 casts. Distribution: Patagonian beds of Santa Cruz (v. Ih.) ; Navidad beds of Chili: Navidad, Tubul, Millanejo and Lebu (Phil.). I4O PATAGONIAN EXPEDITIONS '. PAL/EONTOLOGY. Affinities: In sculpture, this species is closely related to the two fore- going. It is represented in the recent seas of Chili and Patagonia by V. antiqua Kg. (see: v. Ihering, 1897, p. 253). 66. VENUS DARWINI Philippi. PL XXVIII, Fig. 4. 1887 V. d. Philippi, Tert. & Quart. Verst. Chiles, p. 122, pi. 17, f. 2. 1899 V. d. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 19. Shell ovato-orbicular, subquadrate, convex. Posterior dorsal margin convex, forming an indistinct angle with the posterior margin. Anterior extremity distinctly narrower than posterior. Apex at about one-fifth of the length. Area indistinct, nymphae immersed. Lunula lanceolate, depressed in the middle. Surface with regular, rather widely distant, elevated concentric lines, and very slight indications of radiating stria::. Inner margins crenulate. Length, 73 mm; height, 65 mm; diameter, 17 (x 2) mm. Apex at 15 mm from anterior end. Remarks: Radiating striae are not shown in Philippi's figure, and, indeed, there are hardly any traces of them in our specimens. This species corresponds in size to V. volcknianni, but is more elon- gated, and the concentric lines are a little more widely distant in I/, volck- manni. There is also a slight resemblance to V. difficilis (see above), but in the latter the posterior end of the shell is narrower, and the surface ornaments are quite different. Record of specimens: Mouth of Santa Cruz River, 2 right, 4 left valves (in matrix). Distribution: Patagonian formation of Santa Cruz (Phil., v. Ih.). Affinities: Closely allied in form and sculpture is V. burdigalensis Mayer (see: Hoernes, 1870, p. 129, pi. 15, f. i) from the Miocene of Europe, but the latter differs in the more closely set, and more distinctly lamellar concentric lines, and further, in V. burdigalensis, the hinge makes a transition to the genus Meretrix, while V. darwini seems to be a true Venus. Philippi and v. Ihering do not describe the hinge ; in our specimens only part of it is seen, and seems to possess, in the left valve, only three teeth : at any rate, I do not see any trace of a fourth (lunular) tooth ; this part of the hinge, however, is incompletely exposed. ORTMANN : TERTIARY INVERTEBRATES. 141 67. VENUS NAVIDADIS Philippi. PI. xxvn, Fig. i2«:». 1887 V. nav. Philippi, Tert. & Quart. Verst. Chiles, p. 126, pi. 14, f. 4. 1897 V. striatolatnellata v. Ihering, in: Rev. Mus. Paul., v. 2, p. 253, pi. 7. f- 44- 1900 V. nav. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell ovate, slightly convex, anterior and posterior ends rounded. Apex at about l/± of the length. Area indistinct, nymphae immersed. Lunula lanceolate, concave. Surface with concentric lamellae, between the lamellae fine concentric striae. Margin on inner side not crenulate. Length, 63, 39 mm; height, 50, 31 mm; % diameter, 10, 7 mm. Remarks: As v. Ihering has already mentioned, the external form is a little variable. His type specimen measures: L. 79, H. 68 (ratio H. : L. = i : 1.16), while another one is only 64 high by 79 long (ratio = i : 1.23). Philippi gives: L. 47, H. 38 (ratio = i : 1.23). The ratio of our specimens given above is = i : 1.25, and i : 1.26; they are, accordingly, a little more elongated than even v. Iliering's second individual, while that figured by v. Ihering appears exceptionally high. In all other respects our specimens agree well with v. Ihering's description of V. striato-lamcl- lata, and I cannot discover any difference between this species and V. navidadis. V. Ihering believes them to be closely allied, but points to a difference in the lunula, which he takes from the figure of V. navidadis. In the discription of both forms, however, Philippi and v. Ihering use the identical words: "lunula profundata," so that it is impossible for me to see the difference. In the description Philippi gives the position of the apex at *4 of the length ; but his figure shows it distinctly at l/± . The casts from Upper Rio Chalia possess the outline of this species, but no traces of the shell are preserved. Record of specimens : Mouth of Santa Cruz River ; 2 left, 5 right valves. Upper Rio Chalia, 4 casts. Distribution: Suprapatagonian beds of Jegua quemada (v. Ih.) ; Navi- dad (Phil.). Affinities : This species comes near V. arenosa described above, but K arenosa is still more elongated (ratio = i : 1.36 to i : 1.5), and further, the 142 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. apex is, in V. navidadis, distinctly more anterior, and the sculpture seems to be different. According to v. Ihering, this species is closely allied to the living species V. exalbida Ch. from S. America. Gen. MERETRIX Lam. 68. MERETRIX (?) PSEUDOCRASSA (Ortmann). PI. XXIX, Fig. i">6. 1899 Cytherea ps. Ortmann, in: Amer. Journ. Sci., v. 8, p. 429. Shell very thick and solid, very convex. Outline almost circular, pos- terior end rounded. Apex at two-sevenths of the length. Lunula and area indistinct, nymphs deeply immersed. Exterior surface concentrically striated, but the adhering matrix obscures the details of sculpture. Ven- tral margin of shell not crenulated. Hinge with two strongly developed teeth, and a smaller anterior one. Posterior tooth distinctly divided by a groove, the middle one also divided on upper side. Length, 62 mm; height, 60 mm; diameter, 25 (x 2) mm. Remarks: I am unable to decide whether this species belongs really to Meretrix or not. The division of the posterior hinge teeth is in favor of this view, but I cannot make out whether there was a fourth (lunular) tooth ; we have only the right valve, and that part of it, where we should look for the groove that receives this tooth, is broken out. Perhaps it would be better to leave this species with Venus. Record of specimens : Punta Arenas, horizon III (upper Magellanian); i right valve. Affinities: I have compared this species with the Pliocene V. crassa Phil., and the Cretaceous K alta Phil., both from Chili. 69. MERETRIX IHERINGI Cossmann. PI. XXVIII, Fig. 5"'6. 1897 Cytherea splendida v. Ihering, in: Rev. Mus. Paul., v. 2, p. 255, pi. 6, f. 42 (non C. splendida Merian). ORTMANN : TERTIARY INVERTEBRATES. 143 1898 Meretrix iheringi Cossmann, in: Rev. crit. Paleozool., v. 2, No. 3, p. 109. 1899 Cytherea ih. Ameghino, in: Segundo Censo Arg. Supl., p. 4. Shell large, swollen, with concentric sulci. Apex slightly prominent, at y^ of the length of shell. Anterior dorsal margin straight, posterior slightly arcuate. Anterior end of shell shorter, rounded ; posterior sub- rostrate. Lunula circumscribed by a slightly impressed line. Three car- dinal teeth ; the anterior and middle one diverging from the apex. Lateral (lunular) tooth of left valve strong, lamellifonn, and parallel to the lunula. Remarks: We possess only an incomplete left valve, but it agrees— according to the lines of growth — with this species in form. The hinge is well preserved, and corresponds well to the description given by v. Ihering: especially the large lunular tooth is very striking. Record of specimens : Punta Arenas, horizon V (Patagonian); i left valve. Distribution: Jegua quemada and La Cueva, Suprapatagonian beds (v. Ih.). 70. MERETRIX ROSTRATA (Koch). PI. XXVIII, Fig. 6. 1845 Cytherea rostr. Koch, in: Philippi, Abbild. neu. Conchyl., v. i, p. 150, pi. i, f. 3. Shell cordate-ovate, oblique, swollen ; surface with concentric lines of growth. Apex much produced and strongly incurved. Anterior end of shell short, rounded ; posterior longer, rounded and a little produced. Lunula very large, cordate, almost flat, circumscribed by a distinct im- pressed line. Pallial sinus triangular. Height, 34, 39 mm; length, 38, ca. 42 mm; >£ diameter (of first speci- men), ca. 14 mm. Remarks: Our material consists chiefly of casts, but some show large portions of the shell preserved. The agreement with the recent form is complete. The outline of the shell is a little variable, the apex being more or less produced. Record of specimens: Cape Fairweather; remains of 14 valves. Distribution: So far known only Recent from Brazil (Koch); I have seen, in the collections of the Academy of Philadelphia, specimens from Maldonado Bay, Uruguay. 144 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Gen. DOSINIA Scop. 71. DOSINIA MAGELLANICA spec. nov. PI. XXVII, Fig. 13. 1899 D. complanata Ortmann, in: Am. Journ. Sci., v. 8, p. 429 (non D. complauata Phil . ) . Shell orbicular, compressed. Posterior dorsal margin forming, with the posterior margin, a regular, circular curve. Lunula ovate, concave, but elevated in the middle. Surface with regular concentric impressed lines, which are not very crowded (y* to ^ mm distant from each other), dis- tinct all over the shell ; intervals between these lines perfectly flat, but near the anterior and posterior margins a little elevated. Length, 28 mm; height, 26.5 mm; diameter, about 4 (X2) mm. Remarks: In my first publication I made a mistake: I believed some fragments of Lucina neglecta (see above, p. 130) imbedded in the same piece of rock to belong to this species. But now, after succeeding in working them out of the matrix more satisfactorily, I see that we really have to deal with two different forms, and that the characters of the sur- face markings and of the curve of the posterior and dorsal margins, on which I founded the identification with Philippi's D. complanata, are taken from individuals of the new Lucina. D. magellanica differs from D. complanata in the shell's being a little more convex, in the posterior dorsal margin's forming no angle with the posterior margin, and in the surface ornaments, which consist of impressed lines, not of elevated striae. In the following species (D. meridionalis), the curve between posterior dorsal and posterior margins is not regularly circular, but there is a sug- gestion of a blunt angle ; the lunula is not elevated in the middle, the concentric ornaments are more irregular, and the intervals between the impressed lines are more or less convex ; the shell itself is much larger. Record of specimens : Punta Arenas, horizon II (lower Magellanian), 2 right, i left valves. Affinities: The most closely allied form seems to be: D. semilcevis (Artemis s., Philippi, 1887, p. 113, pi. 13, f. 22) from Navidad, especially as regards the concentric lines, which are said to be "remote" from each ORTMANN : TERTIARY INVERTEBRATES. 145 other. But in D. semilcems they appear still more widely distant than in D. magellanica, and the outline of the shell is different. 72. DOSINIA MERIDIONALIS V. Ihering. PI. XXIX, Fig. 2"-c. 1897 D. m. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 256, pi. 6, f. 41. Shell circular, compressed ; posterior dorsal margin forming, with the posterior margin, a strong curve, giving the appearance of a blunt and indistinct angle. Lunula concave, oblongo-cordate, its margin obtuse and indistinct. Surface with concentric impressed lines, which are somewhat irregular (more or less deep, and more or less widely distant), on the whole quite crowded ; the intervals between these lines are distinctly con- vex, but not much so. Measurements : Length, 55 mm; height, 51 mm; diameter (double), 23 mm. V. Ihering gives the following measurements: Length, 83 mm; height, 74 mm; diameter, 21 (x 2) mm, but his figure is much smaller. Remarks : V. Ihering says that the concentric sculpture tends to become obsolete in the middle of the shell : in his figure it is quite distinct every- where, and is also in our specimens well developed in the middle. Toward the margins the sculpture becomes more strongly pronounced. Young individuals, about as large as D. magellanica, differ at once from the latter in the concentric lines' being more crowded, and the intervals' being distinctly convex. . The casts from upper Rio Chalia do not show any details of sculpture, but agree — at least two of them — completely in outline. The internal cast from Cape Fairweather is very poor, but the cast of the surface sculpture, as well as the size and form of the pallial sinus agree with this species. Record of specimens: Mouth of Santa Cruz River, 2 double, 7 right, 5 left valves ; Upper Rio Chalia, 6 casts ; Arroyo Gio, i double valve ; Cape Fairweather, i inner cast of right valve, and part of external cast of right valve. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). According to v. Ihering (1897, P- 33 0 this species is also found in the "Paranense" formation of Parana, Entrerios. As only casts have been 146 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. found there, v. Ihering says himself (1899, p. 43) that this identification is doubtful. But the presence of this species in the Cape Fairweather beds affords some support to v. Ihering's original opinion. Affinities: As v. Ihering points out, this species is very closely allied to Artemis foncferosn Gray (see: Philippi, 1887, p. 113, pi. 14, f. 5, and Moerickc, 1896, p. 585), which is found in the Pliocene Coquimbo beds of Chili, and living on the western coast of Mexico. The latter form, however, is larger, and the concentric sculpture is less marked in the middle of the shell. The hinge agrees closely in both species, and even the lunula, although there is a slight difference as v. Ihering points out, is almost the same, and differs from other species of Dosiuia. D. ponderosa is represented in the Miocene deposits of California by D. matlicwsoni Gabb (1869, p. 57, pi. 15, f. 16), which differs from D. meriditmalis in the more swollen form, but agrees well in size. A very close resemblance exists also to D. acctnbnlnin (Conr.). Al- though, in comparing the figure given by Whitfield (1894, pi. 13, f. 2), and copied from Conrad, this resemblance is not so very striking as regards the sculpture, I have compared specimens from Virginia, in which the sculpture is essentially identical. The only difference is the large size, and the less excavated lunular margin of D. acetabnlum. D. acetabiiliuu is from the Miocene of the Atlantic coast of N. America. D. denselineata Pritchard (1896, p. 135, pi. 4, f. 5-7) seems to be closely allied to D. meridionalis, especially in sculpture, but the outline is dif- ferent : the posterior dorsal margin appears longer, and forms a more dis- tinct angle posteriorly. It is from Table Cape, Tasmania and Spring- Creek, Victoria. Thus the presence of this comparatively large and typical Dosinia in the Patagonian beds points clearly to a Neogene age, and this view is still- more supported by the fact, that D. meridionalis is apparently closely related to, and perhaps the ancestral form of a species that is still found living on the western coast of America. Probably all the forms mentioned above are connected genetically. ORTMANN I TERTIARY INVERTEBRATES. 147 73. DOSINIA L/EVIUSCULA (Philippi). PI. XXVIII, Fig. 7. 1887 Artemis lcev. Philippi, Tert. £ Quart. Verst. Chiles, p. 115, pi. 19, f. i. 1899 Dosinia Iccv. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 20. Shell circular, not much compressed, but comparatively swollen. Pos- terior dorsal margin convex, hardly forming an angle with the posterior margin. Lunula ovate, slightly concave, indistinct. Surface almost smooth, only with very fine and crowded concentric lines. Length, 22 mm; height, 20 mm; diameter, 5 (x 2) mm. Remarks: Although much smaller than Philippi's and v. Ihering's spec- imens, and more circular in outline, I believe our individuals to belong here, since the characters of the surface agree completely. Philippi gives : L. 50, H. 43, D. 25 mm, and v. Ihering: L. 75, and perhaps up to 85 mm, while our largest is only: L. 24, H. 21.5 mm. Our specimens retain the original shell in many places, and show only very fine, concentric, impressed lines, which are quite different from the distinct impressed lines with more or less prominent intervals in D. mcri- dionalis. And further, the character mentioned by Philippi, that the shell is comparatively more swollen than in other species, is also shown in our individuals: a young one, L. 17.5, H. 17, has a diameter of both valves of 9 mm. I think the more circular form of our specimens is due to their young age. Record of specimens : Mouth of Santa Cruz River, ca. 30 internal and external casts in hard matrix, with remains of the shell adhering. Distribution: Patagonian beds of Santa Cruz (Phil., v. Ih.). Fam. -TELLINIDsE Desh. Gen. TELLINA L. 74. TELLINA TEHUELCHA v. Ihering. PI. XXIX, Fig. 3. 1899 T. t. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 21, pi. 2, f. 4. Shell oblong-oval, subequilateral, inequivalve ; left valve convex, right one flat (according to v. Ihering). Apex at about the middle of the 148 PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. length. Anterior end rounded, posterior indistinctly biangulate. Pallial sinus linguiform, reaching hardly to the middle of the shell. Length, 28.5 mm; height, 20 mm; another: length, 27111111; height, i8mm. Remarks: Our casts of the left valve agree completely with v. Ihering's figure, and one of them shows also the pallial sinus. The external cast of a right valve, however, is convex, and perhaps it does not belong here. In the largest individual, near the ventral margin, faint indications of radiating striae are seen. Of this species, so far only the cast is known. Record of specimens : Shell Gap, Rio Chico, upper horizon, 2 internal casts of left valve, i external cast of right valve. Distribution: Santa Cruz, Patagonian formation (v. Ih.). 75. TELLINA JEGUAENSIS v. Ihering. PI. XXVIII, Fig. 8. 1897 T. j. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 260, pi. 5, f. 33. Shell thin, subpyriform, smooth. Anterior end longer, rounded, pos- terior narrowed and acuminate, near the posterior dorsal margin a radi- ating angulation. Length, 20 mm; height, 14 mm (according to v. Ihering: length, 25 mm; height, 16 mm; diameter, 4 mm). Remarks: Our individuals from Santa Cruz are smaller, but otherwise agree well with v. Ihering's figure. The specimens from Arroyo Gio are a little higher comparatively, and the anterior end is less elongated. Record of specimens : Mouth of Santa Cruz River, 2 valves ; Arroyo Gio, 5 casts. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: T. promancana Phil. (1887, p. 141, pi. 26, f. 9) from Navi- dad comes near this species, but it is equilateral (anterior end not longer than posterior), and higher (L., 26; H., 19; rel. = 1.26: i). Our indi- viduals from Arroyo Gio are intermediate in form between T. promancana and jegnacnsis: rel. = 1.42:1, while in the typical form, according to v. Ihering, the rel. is = 1.56: i. The anterior end in the Arroyo Gio speci- mens is less elongated than in the typical form, thus approaching 7\ promaucana in this character also. T. capillifera Conr. (see: Whitfield, 1894, p. 76, pi. 14, f. 8-10) from the Miocene of New Jersey is hardly distinguishable from this species. ORTMANN I TERTIARY INVERTEBRATES. 149 Fam. PSAMMOBIIDsE Ball. Gen. PSAMMOBIA Lam. 76. PSAMMOBIA PATAGONICA Philippi. PI. XXIX, Fig. 4. 1887 P. pat. Philippi, Tert. and Quart. Verst. Chiles, p. 143, pi. 26, f. 17.' 1899 P. pat. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 21. Shell elongated-elliptical, compressed, smooth, almost equilateral, the posterior end a little shorter, both ends evenly rounded. Anterior and posterior dorsal margins straight, forming an obtuse angle. Ventral margin very slightly curved. Length, 28 mm; height, 16 mm; diameter, 3 (x 2) mm. Record of specimens : Mouth of Santa Cruz River, 10 valves; Las Salinas, 6 valves ; 30 miles north of upper Rio Chalia, i double cast ; Arroyo Gio, 3 casts. Distribution: Patagonian beds of Santa Cruz (Phil., v. Ih.). Affinities : This species has a distinct Eocene appearance, and resembles much the forms described by Deshayes (1860, p. 370, ff.) under the names of P. nitida (pi. 24, f. i, 2), and P. tenera (pi. 24, f. 6-8), especially the first, with which it also agrees in size. A very closely allied species is P. hamiltonensis Tate (1887, p. 167, pi. 1 6, f. 13) from Muddy Creek, Victoria, and Table Cape, Tasmania, from so-called "Eocene," but probably Miocene beds. Fam. MACTRID^. Gray. Gen. MACTRA L. 77. MACTRA (?) DARWINI Sowerby. PI. XXIX, Fig. 8. 1846 M. d. Sowerby, in : Darwin Geol. Observ. S. Amer., p. 249, pi. 2, f. 9. Shell triangularly subovate, almost equilateral, compressed, but con- vex toward the apex. Surface smooth, with concentric lines of growth. 'There arc two figures numbered 17 on Philippi's plate, but only the smaller one represents this species, the larger one is Tcllina subfalcata. 150 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Anterior and posterior ends rounded, posterior very little more produced than anterior. Length, 48 mm ; height, 36 mm ; diameter, 9 ( x 2 ) mm. Remarks: Sowerby calls the posterior end "subquadrate," but his figure does not show this character, nor is it seen in our specimens. The generic position is doubtful : I do not see the hinge in any of our individuals. The external form agrees completely with Sowerby's figure, but most of the specimens are larger. Record of specimens : Mouth of Santa Cruz River, 19 double, 9 single valves ; Mt. of Observation, upper horizon, 4 fragments ; Shell Gap, Rio Chico, upper horizon, i double cast. Distribution: Santa Cruz (Sow.). 78. MACTRA GARRETTI spec. nov. PI. XXIX, Fig. 9"^. Shell thin, triangularly-subovate, compressed, a little inequilateral, smooth, with concentric lines of growth. Anterior end a little longer, rounded, posterior subtruncate, with an indistinct keel running down from the apex to the posterior ventral angle. Length, 19 mm; height, 15 mm; diameter (double), 7 mm. Remarks: At first I believed that this is M. indistincta of v. Ihering; but having sent some specimens to the author, he informs me that it is not his M. indistincta, but probably new, and so I describe it as new, con- necting with it the name of Mr. J. W. Garrett. This species is a true Mactra (see: Ball., 1898, p. 874), as shown by the hinge-teeth (pi. XXIX, Fig. 9 c-d). Record of specimens: Mt. of Observation, upper horizon, ca. 40 well- preserved shells, and many fragments ; Lake Pueyrredon, 600' above base, i cast of right valve. " Affinities: Among the numerous species of Mactra described by Phil- ippi from Chili, there is one that might be compared with M. garretti: M. trnncatnla (Phil., 1887, p. 154, pi. 27, f. 15) from Navidad. It agrees in the truncation of the posterior end, which is shorter than the anterior: but it is smaller, the anterior end is distinctly narrower than the posterior, and longer comparatively than in M. garretti. ORTMANN I TERTIARY INVERTEBRATES. 151 There are some species in the Eocene beds of France, which resemble this one in outline, although they differ in other respects: M. levesqnci Desh., M. lamberti Desh., and M. contortiila Desh. (Deshayes, 1860, p. 289, ff. pi. 1 8). Another closely allied form is M. trinacria Semp. (Speycr, 1866, p. 34, pi. 3, f. 4) from the Oligocene of northern Germany, but it is smaller and a little higher. • Gen. LUTRARIA Lam. 79. LUTRARIA (?) UNDATOIDES Ortmann. PI. XXX, Fig. 3. 1899 L. u. Ortmann, in: Amer. Journ. Sci., v. 8, p. 429. Shell almost elliptic, 1 1/2 times as long as high. Surface with strong, and somewhat irregular, undulated concentric folds. Dorsal margin almost straight, ventral margin slightly arcuate. Apex at y$ of the length, prominent, rather sharp, incurved. Anterior and posterior ends evenly rounded, and of the same height. Length, 32 mm ; height, 2 1 mm ; apex at 1 1 mm from anterior end. Remarks: Only the remote resemblance to L. tmdata Phil. (1887, p. 164, pi. 33, f. 8-1 1 ) induces me to place this species with the genus Lntrarin. Record of specimens : Punta Arenas, horizon II (lower Magellanian), i double valve (cast). Fam. CORBULID^E Flem. Gen. CORBULA Lam. 80. CORBULA HATCHERI Ortmann. PI. XXX, Fig. 4"-'. 1900 C. h. Ortmann, in: Amer. Journ. Sci., v. 10, p. 371. Shell small, solid and thick, subovato-triangular. Right valve very little larger than left, both moderately convex. Anterior end rounded, posterior produced, subtruncate, an angular ridge running down from the apex to the posterior angle. Ventral margin arcuate, posteriorly a little concave. Lower margin of right valve reflected toward the left valve. Surface with concentric ribs, which are rounded and rather crowded. 152 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Length, 11 mm; height 7.5 mm; diameter (of right valve), 2.5 mm. Cast from San Julian : length, 13 mm; height, 8 mm; diameter (double), 5 mm. Remarks: So far no Corbula has been known from Patagonia. In C. birostris Phil., from Lota, Chili (Navidad beds), the genus is very doubtful. Record of specimens: Mouth of Santa Cruz River, 8 double, 6 right, 8 left valves ; Las Salinas, i left valve ; Mt. of Observation, upper horizon, i left valve ; San Julian, Darwin Station, 2 double casts. Affinities: The most closely allied form, in my opinion, is C. subcequi- valvis Sandb. (see: Boettger, 1870, p. 41, pi. 9 (8b), f. 16), from the Oligo- cene of Germany and Switzerland. In C. hatcheri the posterior truncation is less distinct, the lower margin is more arcuate, and the ribs of the sur- face seem to be stronger (Boettger calls them "thin" in C. subccquivalms}. There are also some Miocene species of Europe and North America, which might be compared with C. hatcheri, but they do not approach it so closely as C. subcequivalvis. Fam. SAXICAVID^E. Gr. Gen. PANOPEA Men. 81. PANOPEA IBARI Philippi. PI. XXIX, Fig. 5. 1887 Panopcea ibari Philippi, Tert. & Quart. Verst. Chiles, p. 167, pi. 35, f. 4. 1899 Glycimeris ib. Ortmann, in : Amer. Journ. Sci., v. 8, p. 429 (non Gly- cimeris ibari (Phil.), see above, p. 94). Shell elongate-oval, moderately convex, with concentric undulations. Apex at two-fifths of the length, prominent. Both ends rounded, poste- rior distinctly narrower than anterior. Ventral margin arcuate. Length, 84 mm; height, 45 mm; diameter, 15 mm; apex at 35 mm from anterior end. Rel. H. : L. = i : 1.8; in another individual : = i : 2.0. Record of specimens : Punta Arenas, lower horizon, II (lower Magella- nian), i double, 2 right, 2 left valves. Distribution: Magellanes and Skyring Water (Phil.). ORTMANN ! TERTIARY INVERTEBRATES. 1 53 82. PANOPEA SUBSYMMETRICA (Ortmann). PI. XXIX, Fig. 6. 1899 Glycimeris subs. Ortmann, in: Amer. Journ. Sci., v. 8, p. 429. Very near the preceding species (P. ibari), but more swollen, not so much elongated, and posterior end not narrowed. Length, 69 mm; height, 45 mm; diameter, 18 mm. Apex at 28 from anterior end. Rel. H. : L. = i : 1.6. Remarks: Differs from the following species (P. regularis) in the posi- tion of the apex (nearer to the middle), and the stronger convexity, especially of the posterior part of the shell. Record of specimens : Punta Arenas, horizon III (upper Magellanian), i right valve. 83. PANOPEA REGULARIS (Ortmann). PI. XXX, Fig. }"'". 1900 Glycimeris reg. Ortmann, in: Amer. Journ. Sci., v. 10, p. 371. Shell elongate, convex, with concentric lines of growth and undulations. Apex at one-third of the length, incurved. Anterior end rounded, pos- terior subtruncate, not narrower than the anterior. Ventral margin straight in the middle. Length, 101 mm (incomplete); height, 61 mm; diameter, ca. 30 mm; apex at 40 mm (Santa Cruz). Length, 78 mm ; height, 45 mm ; apex at 25 mm (San Julian). 'Length, 105 mm (incomplete); height, 75 mm; apex at 31 mm (Lake Pueyrredon). Remarks: This species differs from P. ibari: (i) in the anterior and posterior ends of the shell being equally high; (2) in the straight lower margin ; (3) in the situation of the apex. It differs from P. sub symmetric a : in the more anterior apex and less convex shell ; from P. quemadensis in the more anterior apex and higher posterior end. P. nucleus v. Ih. (1899, P- 23> pl- r> f- 7). from Santa Cruz, agrees in the position of the apex, but it is considerably narrower posteriorly, and I do not see any traces of the grooves or depressions described by v. Ihering as run- ning down from the apex (these grooves are not visible in v. Ihering's figure). 1 54 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Record of specimens : Mouth of Santa Cruz River, i double valve : San Julian, Darwin Station, i double valve; Lake Pueyrredon, base of Ter- tiary, i double valve ; Lake Pueyrredon, 600' above base, i double valve (jun.). 84. PANOPEA QUEMADENSIS (v. Ihering). PI. XXX, Fig. 2. 1897 Glycimeris qu. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 264, textf. 4. Differs from P. regnlaris in the apex, which is situated only a little in advance of the middle of the shell, and in the posterior end's being nar- rower than the anterior. Length, no mm; height, 73 mm; diameter, 54 mm; apex at 46 mm (Santa Cruz). Length, 83 mm ; height, 53 mm ; apex at ca. 35 mm (Lake Pueyrredon). Record of specimens: Mouth of Santa Cruz River, 2 double valves; San Julian, Darwin Station, i double cast; Upper Rio Chalia, i cast of left valve; 30 miles north of upper Rio Chalia, 2 double casts; Lake Pueyrredon, base of Tertiary, i double cast; Lake Pueyrredon, 600' above base, 5 double casts. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). * 85. PANOPEA PILSBRYI spec. nov. PL XXIX, Fig. 7. Shell short and high, very convex, with concentric lines of growth. Apex about in the middle of the length, incurved. Anterior end rounded, posterior subtruncate, hardly narrower than the anterior. Ventral margin straight in the middle. Length, 79 mm; height, 60 mm; diameter, 22 (x 2) mm; apex at 38 mm. from anterior end. Remarks: In the comparatively very short and high outline of the shell, with the apex almost in the middle of the length, and the posterior and anterior ends of the shell of about the same height, this species differs from all others mentioned here. We possess one good cast, which shows the pallial sinus well preserved. The rest are only fragments, but, as far as can be seen, possess the same characteristic features, especially the short and high outline. ORTMANN I TERTIARY INVERTEBRATES. 155 Record of specimens : Cape Fairwcather, i cast and 5 fragments of casts. Note: I am not quite satisfied that my identifications of the species of Panopea are correct. As to P. ibari and qtiemadensis I think I have recognized them correctly, but the question remains, whether the new species are really different. I have characterized them only by the ex- ternal form, since no other parts are visible, only in a few cases the shell itself being preserved. I did not find any form that corresponds to P. nucleus of v. Ihering. Possibly all the Patagonian species described are really nothing but variations of one and the same form. Fam. PHOLADID^E Fisch. Gen: MARTESIA Leach. 86. MARTESIA PATAGONICA (Philippi). PI. XXX, Fig. 5. 1887 Pholas pat. Philippi, Tert. & Quart. Verst. Chiles, p. 171, pi. 42, f. 8. 1897 Martesia pat. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 266. 1899 Mart. pat. v. Ihering, in : N. Jahrb. Miner., etc., v. 2, p. 23, pi. 2, f. 6. Shell subovate, convex, gaping anteriorly, the anterior end forming a blunt right angle ; the gap is closed by two callous plates. From the beak toward the ventral margin runs an oblique groove ; posterior to this groove only concentric lines of growth, anterior to it sharp ribs, which run paral- lel ,to the (gaping) anterior margin of the shell, and bear sharp, upturned spinules, disposed, toward the anterior end, in distinct radiating lines. These spinules become smaller near the oblique furrow, and finally, close to it, are indicated only by undulations. The ribs of the anterior part of the shell form, with the lines of growth of the posterior, almost a right, or slightly obtuse angle. Anterior dorsal margin reflected. No accessory umbonal plates preserved in our specimens. On the cast, the radial groove forms a distinct impression, and the gaping anterior margin, where it joins the callous plate, forms another impression running from the lower end of the first impression obliquely upward toward the anterior margin. Length, 22 mm ; height, 14 mm ; diameter, 7 mm (x 2) ; (Mt. of Ob- servation). 156 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Length, 24 ; height, 15 ; diameter, 14.5 (double) ; (cast from Las Salinas). It seems, however, that this species grew much larger: specimens from Mt. of Observation and Shell Gap, only in fragments, indicate a length of about 60 mm, by about 30 mm in height. V. Ihering says that his figured specimen measures : length, 37 mm ; height, 23 mm; but the figure itself, which is said to be natural size (i/i), is only: length, 23 mm; height, 15. Record of specimens : Mouth of Santa Cruz River, i right valve ; Las Salinas, i double cast; Mt. of Observation, upper horizon, 4 double, i right, i left valve, 7 casts ; Shell Gap, Rio Chico, upper horizon, i double cast; Lake Pueyrredon, 600' above base, i cast of right valve. Distribution: Patagonian beds of Santa Cruz (Phil., v. Ih.). Affinities: Although this genus is not rare in Eocene deposits, none of the Eocene species agrees so well with M. patagonica, as M. peroni Cossm. & Lamb, from the Oligocene of Switzerland (see: Kissling, 1896, p. 45, pi. 4, f. 4, 5), especially as regards the general form and position of the im- pressions on the cast. M. peroni, however, is smaller than our species. On the other hand, our species resembles also the Pliocene and recent type-species of the genus Pholadidea, P. papyracea (Sol.) (see: Wood, 1856, p. 298, pi. 30, f. 10, Woodward, 1854, pi. 23, f. 20, and Philippi, 1851, p. 128, pi. 2, f. 3), and it is possible that it belongs to this genus, which fact would point to a much younger age (Neogene). 87. MARTESIA PUMILA Ortmann. PI. XXX, Fig. 6"-*. 1900 M. p. Ortmann, in: Amer. Journ. Sci., v. 10, p. 371. This form resembles M. patagonica, but it is much smaller, the callous plate of the anterior margin is very small, and the ribs of the anterior part of the valve form a very obtuse angle with the lines of growth of the posterior part. The radiating furrow is narrower, and runs more inclined posteriorly, so as to render the posterior part of the shell smaller in com- parison with the anterior. The ribs of the anterior part are less in number, and the gaping of the anterior margins is very slight : indeed, in the cast (and we possess only casts with slight traces of shell remaining), the anterior end seems to be closed : only the lowermost rib does not run parallel to the lower anterior ORTMANN : TERTIARY INVERTEBRATES. 157 margin, but cuts off a small, crescentric piece, which seems to represent the impression of the callus. The ribs (on the cast) are simple, but near the anterior end of the shell they are crossed by three to four radiating lines, forming small nodes at the points of intersection. Anterior end of shell not rectangular, but rounded. Length, 9 mm ; height, 4 mm ; diameter (double), 4 mm. Largest individual : height, 7 mm, but incomplete in length, which was probably about 1 6 or 18 mm. Remarks: I was first inclined to regard this form as the young stage of M. patagonica. But we possess a single, fragmentary valve, coming from the same piece of rock with the rest, which is a little larger (see measurements above), and corresponds in size to our smallest individuals of M. patagonica. Yet this specimen shows the narrow radiating groove, the obtuse angle between the lines of growth and the concentric ribs, and the simple character of the latter, while the young specimens of M. patagonica (from Mt. of Observation) agree in these respects with larger individuals of that species. Record of specimens: Mouth of Santa Cruz River, ca. 100 specimens. SCAPHOPODA. Fam. DENTALIID^E. Gray. Gen. DENTALIUM L. 88. DENTALUM SULCOSUM Sowerby. PI. XXXI, Fig. !«•». 1846 D. sulc. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 263, pi. 2, f. 2. 1846 D. majus Sowerby, ibid., p. 263, pi. 2, f. 3. 1887 D. sulc. Philippi, Tert. & Quart. Verst. Chiles, p. 106 (partim). 1887 D. maj. Philippi, ibid., p. 106, pi. 12, f. n. 1887 D. gayi Philippi, ibid., p. 107, pi. 12, f. 19 (juv.). 1889 D. patagonicum Rochebrune & Mabille, in: Miss. Sci. Cape Horn, v. 6, p. 98. 158 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. 1899 D. pat. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 24. 1900 D. sulc. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell rounded, in young stage angular and curved, almost straight when old, with longitudinal ribs. Principal ribs, in young individuals, 10 to 14, between which, in older ones, smaller ribs (i to 3 in each interval) are intercalated, bringing up the total number to from 20 to 30; but always about 14 ribs are stronger than the rest. The ribs, which are rather sharp in the young stage, are rounded in older individuals; in young individuals the intervals are broader and flat, in older ones the ribs appear more crowded. In very old specimens the ribs are less dis- tinct, and are sometimes crossed by distinct lines of growth. Remarks: We possess numerous fragments, but no complete indi- vidual. This species grows very large: the largest diameter is 15 mm, while the smallest is i mm (upper end of the smallest fragment is 2.5 mm in diameter, by a length of 18 mm). Between these small ones and the largest we possess all transitions. One of the smallest has only 9 ribs ; another one, as small as this one, has 10. Individuals of from 3 to 5 mm in diameter have 10 to 14 ribs: in all these very young ones the ribs are angular, the intervals flat, and the shell has a polygonal cross section. A fragment, diameter 7 mm, has 15 principal ribs, and 3 very small ones intercalcated. At this diameter, from 6 to 10 mm, the characters of D. sulcosnm Sow. are typically exhib- ited: about 14 ribs, with flat intervals, but sometimes a few intermediate ribs are present. An individual of from 6 to 9.5 mm diameter (on the lower and upper end respectively, length 54 mm), has on the narrow end 1 1 large, and i small rib, on the wider end 12 larger, and 1 1 smaller (and be- sides a few striae), and represents thus, on the wider end, distinctly/), inajits of Sowerby. In still larger fragments, more intermediate ribs are interca- lated, and at a diameter of from 1 1 to 12 mm there are about 14 larger, and 14 to 15 smaller ribs. This intercalation of ribs is not regular : in some of the intervals there are none ; in others 2 to 3. In very large fragments, the ribs often become indistinct, and further, the peculiar exfoliation of the outer layer of the shell tends in many individuals to obscure the ribs, especially the smaller ones, so that only 11-14 principal ribs are visible. Often distinct lines of growth are visible in larger fragments. The curvature of the young shell is a little stronger than that of the rest, which in most cases is almost straight, in some cases perfectly so. ORTMANN : TERTIARY INVERTEBRATES. 159 The taper of the young shell is apparently more considerable than in the old : a fragment 29 mm long is, on the one end, 3 mm in diameter, on the other, 5.5. Another one, 58 mm long, is on one end 7 mm in diameter, on the other 10 mm: this latter fragment thus increases only 3 mm, where we should expect — at the rate of the first one — about 5 mm. In D. giganteum Sowerby (1846, p. 263, pi. 2, f. i), and Philippi (1887, p. 105, pi. 12, f. 9), from Navidad, the sculpture is different, consisting of shallow and narrow grooves, separated by rounded and low intervals. According to the figure, D. sulcosum of Philippi (pi. 12, f. 10), from Navidad is certainly different from D. sulcosum of Sowerby, and it closely resembles D. giganteum, but the furrows are much deeper and more distinct. The specimens, however, mentioned by Philippi from Santa Cruz as belonging to D. sulcosum, belong no doubt to the true D. sulcosum. D. lebuense Philippi (1887, p. 106, pi. 12, f. 18), from Lebu and Llan- cahue resembles a young D. sulcosum, but it is much straighten D. gayi Philippi (Matanzas and Carauma), comes no doubt into the synonymy of D. sulcosum : the figure resembles exactly a young individual of the Santa Cruz species. D. patagonicum of Rochebrune and Mabille, and v. Ihering, is certainly identical with our species : we have received from v. Ihering a fragment of medium size under this name, and this agrees completely with our D. sulcosum. D. mantelli Zittel (1864, p. 45, pi. 13, f. 7), from New Zealand, Victoria arid South Australia (Tate, 1887, b, p. 190), and Tasmania (Pritchard, 1896, p. 126), comes very near to D. sulcosum. The only difference (ac- cording to the figure) is the much stronger curvature of the shell. (Tate, 1887, b, p. 191, says : D. giganteum of which solidum Hutton is a synonym is distinct though conspecific with D. mantelli, a sentence that is not quite clear.) Record of specimens : Mouth of Santa Cruz River, 59 fragments ; San Julian, Darwin Station, 2 casts ; Lake Pueyrredon, base of Tertiary, i broken and compressed specimen. Distribution: Santa Cruz (Phil., Roch. & Mab., v. Ih.), Patagonian formation (v. Ih.) ; Navidad beds of Chili : Navidad, Huafo Island (Sow.), Ancud, Llancahue, Tubul, Matanzas, Carauma (Phil.). l6o PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Affinities: As has been pointed out above, the New Zealandian and Australian species D. mantelli is closely allied to this species. It is found in New Zealand, in the Pareora system of Miocene, according to Hutton. A closely allied species seems to be : D. gabbi Pilsbry and Sharp (1897, p. 470, pi. i o, f. 6, 7, 13, pi. n, f. i, 2), from the Oligocene or Miocene beds of San Domingo, but it is smaller, and the number of ribs near the apex is less. Of the European species, D. kickxi Nyst(see: Speyer, 1870, p. 199, pi. 21, f. 8-1 1 ) from the Middle and Upper Oligocene seems to be the most closely allied species. 89. DENTALIUM OCTOCOSTELLATUM Pilsbry and Sharp. PL XXXI, Fig. 2. 1897 D- octocostatum v. Ihering, in: Rev. Mus. Paul., v. 2, p. 266, pi. 4, f. 16 (non D. octocostatum Fraas, 1867). 1898 D. octocostellatum Pilsbry and Sharp, Man. Conch., v. 17, p. 211. Shell small, slightly curved, with 6 to 8 longitudinal ribs, which are distant, with flat intervals, rendering the cross section polygonal. Diameter up to 3.5 mm. Remarks: This species differs from the foregoing: (i) in the small size ; (2) in the smaller number of ribs ; v. Ihering gives 8—9, but his figure shows — as far as can be made out — an individual that had certainly not more than six ribs. In our specimens, one (the largest) has only 6 distinct ribs, with two very indistinct ones (on the lateral intervals). The other individuals have 7 or 8. D. sulcosum, at the same size, has always at least 9 or 10 ribs. (3) The taper in both species is different. Our largest individual of the present species is 25 mm. long, the diameter at one end is 2.3 mm, at the other 3.5 mm. An individual of D. sulcosum, 24 mm long, increases from 2.3 to a little over 4 mm. The specimens from Arroyo Gio are casts, but they agree in size, curv- ature and taper with this species. Records of specimens: Mt. of Observation, upper horizon, 6 fragments; Arroyo Gio, 2 casts. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). ORTMANN : TERTIARY INVERTEBRATES. l6l GASTROPODA. Fam. PATELLIDsE Carp. Gen. PATELLA L. 90. PATELLA PYGM/EA Ortmann. PI. XXX, Fig. r~c- 1899 P. p. Ortmann, in: Amer. Journ. Sci., v. 8, p. 430. Shell subconical, outline regularly oval. Apex situated in the anterior half. Surface with fine, crowded, a little unequal, radial ribs, crossed by concentric lines of growth, and very finely granulated. Length, 7 mm ; width, 5 mm ; height, 4 mm. Remarks: This may be an Acmcea. Pilsbry (1891, p. 7) says of Acmcea: the shells may generally be distinguished from Patella by the different texture and the marginal border of the inside. The presence or lack of this border cannot be ascertained in our specimens, since the in- terior is filled with matrix. There already exists an Acmcea heroldi forma Pygmcea Dunker (see : Pilsbry, 1891, p. 45). If our species should prove to belong to Acimca, the specific name is ill chosen, but cannot be changed, unless pygnicca Dunker is regarded as a good species. Record of specimens: Punta Arenas, horizon III (upper Magellanian) ; i sp. Fam. FISSURELLIDA1 Riss. Gen. FISSURELLA Brug. 91. FISSURELLA EURYTRETA Cossmann. PI. XXX, Fig. 8. 1899 F. e. Cossmann, in : Journ. Conchyliol., p. 3 (of sep. cop.), pi. 1 1, f. i. Shell depressed, elongated-elliptical ; foramen subcentral, large, ellip- tical. Surface with radiating ribs. Length, 21 mm; width, 12 mm; altitude, 5.5 mm. Remarks : Surface ornaments not seen in our individual, which is a cast. Record of specimens : Upper Rio Chalia ; i cast. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). 1 62 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY Fam. DELPHINULID^. Fisch. Gen. LIOTIA Gray. 92. LIOTIA SCOTTI Ortmann. PI. XXX, Fig. io-«. 1900 L. s. Ortmann, in: Amer. Journ. Sci., v. 10, p. 372. Shell small, rounded, flat above, with a large, open umbilicus below. Spire with 4 rounded whorls, increasing rapidly ; suture deep. Last whorl with six revolving, equidistant keels, the keel nearest the umbilicus the smallest, and disappearing within the umbilicus ; the upper whorls show only the two uppermost keels. The keels are crossed by very fine striae, and a number (15) of strong'radial ribs; at the points of intersection of these ribs and the revolving keels there are small conical tubercles. Last whorl a little deflected toward the mouth, which is circular and thickened. Height, 4 mm ; diameter, 8 mm. Remarks: This species has been named in honor of Professor W. B. Scott, of Princeton University. Record of specimens : Mouth of Santa Cruz River ; i sp. Affinities: This species is very closely allied to the recent L. (Lippistes) acrilla Dall (1889, pi. 32, f. 6, 11) from Florida, but the latter has only five revolving keels, and the upper side of the shell is a little concave. Fam. TROCHIDsE Ad. Gen. LEPTOTHYRA. 93. LEPTOTHYRA PHILIPPI Cossmann. PI. XXX, Fig. 9"'6. 1899 L. p. Cossmann, in: Journ. Conchyliol., p. 4 (of sep. cop.), pi. 10, f. 10, II. Shell small, subgloboso-conical. Spire short. Whorls convex, suture deep. Surface of upper whorls with 6 to 9 revolving ribs, which are ob- tusely granulate. Last whorl about ^ of the height of the shell. Base rounded, imperforate. Mouth contracted, circular, very oblique. ORTMANN : TERTIARY INVERTEBRATES. 163 Height, 8 mm ; diameter, 6.5 mm. Record of specimens : Mouth of Santa Cruz River, 6 sp. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). Affinities: Cossmann compares this species with the French Eocene species L. obtusalis Baud, and L. montensis Br. & Corn., but it differs con- siderably from them. Gen. SOLARIELLA Wood. 94. SOLARIELLA DAUTZENBERGI Cossmann. PI. XXX, Fig. u"-". 1899 S. d. Cossmann, in : Journ. Conchyliol., p. 8 (of sep. cop.), pi. 10, f. 14. Shell small, conical ; spire short, scalariform. Whorls with two spiral angulations, a little concave between them ; upper part, above upper an- gulation, horizontal, lower part, below lower angulation, vertical. Whole surface (aside of these angulations) with fine revolving striae, 3 on upper part, 4-6 on the middle (concave) part, and 3 on the lower. Angulations distinctly granulated (in well preserved specimens), and very fine and in- distinct granulations are sometimes seen on the striae, especially in the middle part of the whorls. Last whorl with an angulation on the periphery. Base plane, with 7-9 revolving striae, and a stronger and beautifully granu- lated rib near the deep umbilicus. Fine revolving striae are also present within the umbilicus, where they are finely and indistinctly granulated. Height, 6 mm ; diameter, 7 mm. Remarks : Cossmann does not mention the granulations of the angula- tions of the upper whorls, and describes, within the umbilicus, only obtuse crenulations, but I think, in his specimens, these fine ornaments were worn off, as is also the case in two of ours. Record of specimens: Mouth of Santa Cruz River, 4 sp.; Lake Pueyr- redon, 600' above base, i cast. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). Affinities: Trochus stoliczkai Zittel (1864, p. 40, pi. 15, f. 7), from the Miocene (Pareora, Hutton, 1873, p. 15), of New Zealand seems to repre- sent this form in New Zealand : but the whorls are more rounded, and not so distinctly angular, and the fine revolving striae are lacking. Cossmann compares this species with the "Eocene Solariellce" from Paris, from which it is said to differ in the lack of granulations : but as we 164 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. have seen above, these granulations are present in the Patagonian fossil. On the other hand, there is a Miocene species, S. turritella Dall (1892, p. 408, pi. 23, f. 2), from Florida, and even a Pliocene species, Margarita maculata Wood (1848, p. 135, pi. 15, f. 3), which much resemble our species, so that it is impossible to say that the Patagonian shell has a distinctly Eocene appearance. Gen. CALLIOSTOMA Swains. 95. CALLIOSTOMA PHILIPPII (Ortmann). PI. XXX, Fig. i2"'». 1899 Trochus ph. Ortmann, in: Amer. Journ. Sci., v. 8, p. 430. Shell low, conical, not umbilicated. Whorls almost flat, only very slightly convex. Last whorl sharply angular on the periphery; above this angular ridge there are 4 revolving ribs. Lower surface slightly convex, with 5 strong, revolving ribs, the most exterior of them separated from the peripheral ridge by a broad groove. Ribs of lower surface with regular, strong granules ; similar granules seem to have been present on the ribs of the upper part of the whorls. Height, 7 mm ; diameter, 1 1 mm. Remarks: There already exists a Trochus pliilippii Koch (see : Pilsbry, 1889, p. 52), but since our species comes in the genus Calliostonia, no change of the specific name is necessary. The upper surface of the whorls is not well preserved in our specimens, so that the characters of the sculpture, especially the granulations, are not plainly recognizable. This species differs from T. fricki Philippi (1887, p. 101, pi. 12, f. 7), from the Navidad beds of Chili in the following details: (i) the whorls are slightly convex (in T. fricki perfectly flat) ; (2) the revolving ribs are less numerous (in T. fricki there are 6 on the upper part, and at least 7 on the lower part) ; (3) the umbilicus is absent. C. pliilippii differs from C. observations in the following particulars : ( i ) the revolving ribs are less numerous (in C. observationis 5-6 in the upper part, 9-10 on the base); (2) the ribs of the base are separated from the periphery by a groove ; (3) there are granulations present. Record of specimens : Punta Arenas, horizon III (upper Magellanian) ; 2 sp. ORTMANN I TERTIARY INVERTEBRATES. 165 Affinities: In the Navidad beds of Chili we have a closely allied species, T. fricki. I cannot say, however, that any species of the northern hemi- sphere shows a marked affinity to this one, although there are many, which might be compared with it. 96. CALLIOSTOMA OBSERVATIONS Ortmann. PI. XXX, Fig. i3a-». 1900 C. o. Ortmann, in: Amer. Journ. Sci., v. 10, p. 372. Shell low, conical, not umbilicated. Whorls flat, last whorl bluntly angular on the periphery. Above this angulation there are 5 distinct revolving ribs ; near the mouth, between the 2d and 3d (counted from the upper part), a sixth rib begins to appear, and, in our largest individual, the peripheral angulation is divided by a fine impressed line into two ribs. On the upper whorls, near the apex, the 2d and 4th ribs disappear, so that only three ribs remain (ist, 3d, and 5th), besides the peripheral angulation, which appears as a 4th rib immediately above the suture. All the ribs, when fully developed, are subequal, flattened, smooth, about as broad as the intervals between them. The base of the shell shows 9—10 revolving ribs of the same character, which are, near the umbilicus, as broad as the intervals, but more crowded toward the periphery. The outermost of them is not separated from the peripheral angulation by a broader interval. Height, 10.5 mm; diameter, 12 mm. Remarks: The lack of granulations distinguishes this species at once from C. fricki and philippii. Furthermore, in C. fricki, the spire is more depressed, but more acute ; the revolving ribs of the base are more crowded, and a small umbilical excavation is present. Record of specimens : Mt. of Observation, upper horizon ; 10 sp. 97. CALLIOSTOMA PERARATUM Cossmann. PI. XXXI, Fig. 3"' ". 1899 C. p. Cossmann, in : Journ. Conchyliol., p. 9 (of sep. cop.), pi. 10, f. 6. Shell conical, about as high as broad, not umbilicated. Whorls flat, last whorl bluntly angulated. Upper whorls with 2 strong and i finer revolving keels ; the lower keel is formed by the peripheral angulation ; 1 66 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. above this is a similar one, which is strong, but not quite as broad as the first one ; near the suture, on the upper part of the whorls, there is a much finer, but distinct keel. No granulations on the keels. The two larger ones broad and rounded, and, on the last whorl, near the mouth of the shell, bifid. All three keels separated by broad, smooth grooves, about as broad as the middle keel. Base of shell almost flat, with 8-10 revolv- ing ribs, which are flat and smooth, more crowded toward the periphery, a little more widely distant toward the umbilicus ; the furrows between them, however, are always narrower than the ribs. Height, 8.5 mm ; diameter, 9 mm. Record of specimens : Mouth of Santa Cruz River; 3 sp. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). Affinities: Cossmann says that, among the fossil species, C. cmdebardi Bast, from the Lower Miocene of Bordeaux is the most closely allied form. 98. CALLIOSTOMA COSSMANNI Ortmann. PI. XXXI, Fig. 4"'". 1900 C. c. Ortmann, in: Amer. Journ. Sci., v. 10, p. 372. Shell conical, higher than broad, not umbilicated. Whorls flat, the last one angulated, with a keel on the periphery, which is wholly exposed on the upper whorls, being situated close to, but above the suture. Upper whorls with 5 revolving keels, the lowermost (the keel of the periph- ery just mentioned) the strongest and almost smooth, with hardly any traces of granulations. The uppermost (first) and the third keels stronger than the 2d and 4th; the ist, 2d, and 3d distinctly granulated, the 4th with finer granulations. Toward the apex of the shell, the 2d and 4th keels disappear, so that only three keels are present, the two upper ones granulated, the lower one (peripheral angulation) smooth. (Perhaps it was also finely granulated, but if so, the granules are worn off.) Base of shell hardly convex, with 6 revolving keels, which are subequal, smooth, narrower than the intervals. One or two of the keels, nearest to the periphery, appear bifid toward the mouth of the shell. Height, 8 mm ; diameter, 6.5 mm. Record of specimens: Mouth of Santa Cruz River, 4 sp. ORTMANN I TERTIARY INVERTEBRATES. 1 67 99. CALLIOSTOMA SANTACRUZENSE Cossmann. PI. XXXI, Fig. 5«-». 1899 C. s. Cossmann, in : Journ. Conchyliol., p. 10 (of sep. cop.), pi. 10, f. 13. Shell conical, higher than broad, not umbilicated. Whorls slightly convex, suture shallow. Last whorl angulated, with a keel on the angu- lation. Above this keel, which is wholly exposed on the upper whorls, there are 2 to 3 other revolving keels, which are remote from each other, the uppermost close to the suture. They are crossed by oblique longi- tudinal ribs, which form, at the intersection with the revolving ribs, small tubercles or conical granulations. These longitudinal ribs are fairly remote from each other in the upper whorls, but more crowded near the mouth of the shell, resembling there lines of growth. Base of shell oblique, a little convex, with 6 spiral ribs, the outermost a little more distant from the peripheral keel than from the rest. These ribs of the base are crossed by lines of growth (or ribs) in the same way as those of the upper part of the whorls, and are also granulated at the points of intersection. Height, 9 mm ; diameter, 7 mm. Remarks: Our specimens differ a little from the original description. Cossmann calls the whorls convex, but they are very slightly so (also in Cossmann's figure1). Further, Cossmann gives 4 spiral ribs on the upper whorls, with a finer one intercalated on the last whorl. In our largest individual, although a little larger than Cossmann's, I can distinguish 4 ribs only near the mouth, with no traces of intermediate ones, and the upper whorls show only three ribs. But since the general character of sculpture : revolving ribs, crossed by longitudinal ones, with granules at the points of intersection, is well exhibited in our specimens, I have no doubt that they belong to Coss- mann's species. Record of specimens : Mouth of Santa Cruz River, 4 sp. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). 1 Cossmann's figures do not help materially in the identification of the species ; they give a good representation of the external form and dimensions, but in most cases the details of sculpture, etc., are obscure. The fault lies with method of reproduction. 1 68 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Affinities: Cossmann compares this species with C. podolicum (Dub.) from the upper Miocene (Sarmatian) of southeast Europe, and C. pseudo- tnrricula Dollf. from the middle Miocene of France. There is no doubt that the first one (C. podolicum, see Hoernes, 1856, p. 447, pi. 45, f. 2) represents a similar type, although the details of sculpture are different. 100. CALLIOSTOMA GARRETTI Ortmann. PI. XXXI, Fig. 6"'". 1900 C. g. Ortmann, in: Amer. Journ. Sci., v. 10, p. 373. Shell conical, as high as broad, not umbilicated. Eight whorls, which are very slightly convex; suture shallow. Last whorl very bluntly angular at the periphery, without a distinct peripheral keel. Surface of whorls, above the periphery, covered with numerous fine revolving threads : there are, on the third whorl, about 7 of them, increasing to about 17 on the last whorl. The number of threads increases by intercalation, the new ones being at first smooth ; with increasing size they equal the older threads, and become, like the latter, finely but distinctly granulated. These granulations, however, are developed only in the upper three quarters of the whorls : the lower 4 or 5 threads remain smooth. Intervals between the threads about as broad as these, and crossed by very fine lines of growth, giving a pitted appearance to them. The revolving threads con- tinue over the periphery to the base of the shell, which is slightly convex ; their number is about 24 on the base, and they are smooth, without granu- lations, resembling in all other respects those of the upper part of the whorls. Height, 17 mm; diameter, 17 mm. Remarks: The specific name is given in honor of Mr. J. W. Garrett. Record of specimens: Mouth of Santa Cruz River, 6 sp. Affinities: Trochus macspormni Philippi (1887, p. 102, pi. 12, f. 6), from the island of Santa Maria, Chili (Navidad beds?) seems to represent this species in Chili, but in T. macsporrani the granulations are wanting, and the periphery is more sharply angulated. C. garretti exhibits distinctly Miocene relations, with a group of species found in Miocene beds of the southern United States, described by Ball (1892, p. 390 ff., pi. 18 and 22). The most closely allied of them seem • to be: C. philanthropies (Conr.) (see: 1. c., p. 390, pi. 18, f. ga) from the ORTMANN I TERTIARY INVERTEBRATES. 169 Miocene of Virginia, which differs, however, in the more strongly angu- lated periphery, flatter whorls and coarser threads, and C. metriiim Dall (p. 394, pi. 22, f. 27) from the older Miocene of Florida, which differs in the flat whorls and the finer sculpture. 101. CALLIOSTOMA IHERINGI Ortmann. PI. XXXI, Fig. 7"'". 1900 C. i. Ortmann, in: Amer. Journ. Sci., v. 10, p. 373. Shell conical, broader than high, scalariform, umbilicated. Six whorls, which are sharply angulated ; one angulation is formed by a sharp revolv- ing keel in the upper part of the whorls, a second one — exposed only on the last whorl — is formed by a keel on the periphery. Suture distinct, forming an obtuse angle ; upper part of whorls (above upper keel) oblique, flat, with 5 to 6 revolving threads, which are slightly granulated ; lower part (below that keel) vertical, slightly concave on the last whorl, with 5 to 7 fine, smooth threads. Base of shell slightly convex, depressed toward the umbilicus, which is moderately large. About 18 revolving threads on the base, which are smooth, more crowded and finer toward the periphery, a little stronger near the umbilicus, where the intervals are about as broad as the threads. Height, 9.5 mm; diameter, 12 mm. Remarks: The presence of an umbilicus brings this species into the subgenus E^ltrochus. Record of specimens : Mouth of Santa Cruz River, i sp. 'Affinities: A species that resembles this one in the presence of two angulations on the last whorl, and belongs also to the subgenus Eutrochiis is: C. cychis Dall (1892, p. 403, pi. 23, f. 20) from the Miocene of North Carolina, but this one is much lower. Another species with the same double angulation and open umbilicus is Trochiis biangnlattis Eichw. (see : Hoernes, 1856, p. 460, pi. 45, f. 5) from the Miocene of Europe, but it is much higher. The latter species is said to be identical with T. ditropis Wood (1848, p. 133, pi. 14, f. 9) from the Pliocene of England. The latter, however, in external form, is more like our Patagonian fossil, being less high than the Miocene form, and it differs from C. iheringi in the upper angulations being more prominent, the number of revolving threads being different, especially on the base, which is said to be finely imbricated. 170 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Gen. GIBBULA Riss. 102. GIBBULA L/EVIS (Sowerby). PI. XXXI, Fig. 8. 1846 TrocJms Icevis Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 256, pi. 3, f. 46, 47 (adult). 1846 T. collaris Sowerby, ibid., p. 256, pi. 3, f. 44, 45 (junior). 1887 T. Icevis Philippi, Tert. & Quart. Verst. Chiles, p. 101, pi. 12, f. 5. 1897 Gibbula coll. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 273. 1899 £• c- v- Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 24. Shell conical, broader than high, umbilicated. Surface smooth, whorls almost flat, only the last one slightly convex. Periphery sharply angu- lated. Upper whorls, close to the suture, with a series of small tubercles. Base very slightly convex, with a deep umbilicus of medium size, and a number of fine revolving striae, which are distinct near the umbilicus, but disappear toward the periphery. Height (not complete), 28 mm; diameter, 51 mm; according to Phil- ippi : height, 38 mm ; diameter, 50 mm. Remarks: Philippi was the first to recognize that T. collaris of Sowerby is only the young stage of T. lcevis, and he retains the specific name of the old stage. V. Ihering again uses the specific name of collaris, preced- ing that of Iczvis in Sowerby's text, but according to the rules of nomen- clature, we are to follow Philippi, who was the first to make a selection between the two names available. In some specimens the small tubercles near the suture disappear later than in others. The lower surface of the shell has spiral striae, which in very young ones are very faint near the periphery, and disappear in old shells altogether, with the exception of 5 to 7 close to the umbilicus (see : Sowerby's figure 47). Record of specimens: Mouth of Santa Cruz River, i2sp.; Las Salinas, i sp.; San Julian, Oven Point, 2 casts; Lake Pueyrredon, base of Ter- tiary, 2 casts ; Lake Pueyrredon, 600' above base, i cast. Distribution: Patagonian beds of Santa Cruz (Sow., v. Ih.); Navidad (Sow., Phil.), Lebu (Phil.). Affinities: A very closely allied species is T. veneficus Philippi (1887, p. 101, pi. 12, f. 8), from Navidad, but the latter has a blunt, but distinct angulation near the suture, on which the tubercles are placed. ORTMANN : TERTIARY INVERTEBRATES. I 7 1 Philippi compares this species (veneficus) with T. magus L. (Pliocene and Recent, Europe), and indeed, this is the only relation with any other known form, but it is very remote. A little closer is the affinity of T. magus with the next species, as we shall see below. 103. GlBBULA DALLI V. Ihering. PI. XXXI, Fig. 9"'». 1897 £• duUi v- Ihering, in: Rev. Mus. Paul., v. 2, p. 272, pi. 3, f. i, pi. 4, f- 13- 1897 G.fracta v. Ihering, ibid., p. 273, pi. 3, f. 2. 1900 G. dalli Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell conical, broader than high, widely umbilicated. Whorls convex, especially in the upper part, near the suture. Last whorl obtusely cari- nated on the periphery. Surface ornamented with revolving striae, and oblique, tubercular, radial ribs near the suture. The latter are sometimes short, and more like tubercles, in other cases they are more elongated, extending almost to the middle of the upper (exposed) part of the whorls, Revolving striae very unequal, 5 to 6 larger ones are found in the region of the tubercles, with intermediate finer ones ; the rest, toward the periph- ery, are fine. Very often, and especially in the young shell, between the tubercles and the peripheral angulation (which shows partly on the upper whorls above the suture), there is another revolving, blunt ridge, resemb- ling the peripheral angulation. Base of shell slightly convex or flat, um- bilicus deeply depressed, wide. Base with a number (7 to 8) of revolving ribs, and numerous fine striae between them ; 3 to 5 of the larger ribs, near the umbilicus, are distinctly granulated by fine lines radiating from the umbilicus. Height, 17 mm; diameter, 36 mm; our largest cast measures: height, about 28 mm ; diameter, 52 mm ; v. Ihering gives : height, 35 mm ; diam- eter, 63 mm. Remarks : I regard G. fracta as the young stage of this species. The only difference — according to the diagnosis ; the figure does not show any differences except that it is smaller — is the presence of two spiral angula- tions in the lower part of the upper whorls. These blunt ridges are present in almost all individuals, and even those that are large and show the typical character of G. dalli on the last whorl, exhibit them on the upper 172 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. whorls. But the development of these ridges is very variable : sometimes, and especially in the young shell, they are very distinct, and such speci- mens represent v. Ihering's G. fracta. In a few individuals the upper ridge is completely absent, even in the young shell, and this part of the shell appears depressed between the row of tubercles and the peripheral angulation : such individuals are mentioned by v. Ihering (p. 274) as variety cuevensis. With increasing age these spiral ridges disappear, and on the last whorl of large shells no trace of the upper one is found, and the lower one (on the periphery) becomes indistinct. At the same time, the tubercles become less pronounced, and the upper half of the whorl appears evenly convex, without depression, representing thus the typical G. dalli. We possess all intermediate stages. There is also considerable variation in the development of the revolving striae, both of the upper side and of the base. In older shells they become more uniform and less distinct, especially the finer ones. The larger striae are often beautifully waved in crossing the tubercles near the suture. Record of specimens : Mouth of Santa Cruz River, 1 6 sp. ; Upper Rio Chalia, 5 casts; Shell Gap, Rio Chico, upper horizon, 12 sp. (mostly casts); Arroyo Gio, i cast ; Lake Pueyrredon, 600' above base, 2 casts. Distribution: Jack Harvey and Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: This species has a better claim than the preceding one to be compared with G. magus (L.), both having radial and spiral sculpture combined ; for the rest, in the details of sculpture, general form of shell, and shape of umbilicus, there are numerous differences. But we may say, if there is any relation of this species, it is with Pliocene and recent forms. 104. GIBBULA DIAMETRALIS Cossmann. PI. XXXI, Fig. lo"-6. 1899 G. d. Cossmann, in : Journ. Conchyliol., p. 5 (of sep. cop.), pi. 10, f. 1-3. Shell conical, broader than high, umbilicated. Whorls convex, suture distinct. Upper surface of whorls with 2 strong, rounded, revolving ribs, and a similar angulation on the periphery, which is partly exposed on the upper whorls. Upper rib with very elegant crenulations or plications, formed by short, oblique ribs radiating from the suture. These crenula- tions are restricted to the upper part of this rib, the lower being occupied ORTMANN : TERTIARY INVERTEBRATES. 173 by 2 to 3 sharp revolving threads. Similar threads or keels (4 to 5) are on the second rib, and a number of very fine, but distinct striae are on the peripheral angulation. Intervals between the ribs deep, with indis- tinct revolving striae. Base of shell with about 9 revolving keels, which are strong and distinctly crenulated near the umbilicus, becoming finer toward the periphery. Umbilicus deep, of medium size. Height, 12 mm; diameter, 15 mm. Remarks: I am not quite satisfied that my identification of this species is correct. Cossmann mentions crenulations on the inferior part of the whorls, while my individuals show them only in the upper part : this dis- crepancy is not cleared up by Cossmann's figure, which shows no crenu- lations at all. In all other respects our individuals agree with Cossmann's figure and description, but it is to be remarked that the figure represents an individual that is apparently considerably worn. In our figured speci- men the surface ornaments are in a beautiful state of preservation, and all of them are more distinct and more pronounced than in Cossmann's figure. In another of our specimens the upper revolving rib does not show any finer keels, but is bipartite, the lower division being a little narrower and showing also traces of crenulations. The third individual is very small, while the other two are considerably larger than the measurements given by Cossmann. Record of specimens : Mouth of Santa Cruz River, 3 sp. Distribution: Jequa quemada, Suprapatagonian beds (Cossm.). Fam. PYRAMIDELLID^ Gr. Gen. ODONTOSTOMIA Jeffr. 105. ODONTOSTOMIA SUTURALIS (v. Ihering). PI. XXXIII, Fig. 7"'6. 1897 Odostomia sut. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 275, textf. 10. 1899 Odontostomia synarthrota Cossmann, in: Journ. Conchyliol., p. 12, (of sep. cop.), pi. n, f. 4. 1900 Odontostomia sut. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell pyramidal, smooth; whorls 7 (9 according to v. Ihering), flat, angulated near the suture, last whorl rounded. Mouth oblong, inner lip with a fold below. .174 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. Height, 9 mm; diameter, 3.5 mm (v. Ihering gives: height, 19; diam- eter, 7; Cossmann: height, 5; diameter, 2 mm). Remarks: One of our specimens is isolated, and it is smaller than v. Ihering's, but larger than Cossmann's. The only difference of Cossmann's species is the small size and small number of whorls (5-6). Our in- dividual is exactly intermediate in these respects between both, and so there is no doubt that Cossmann's species is only the young stage of this species. The fold on the inner lip is well seen in our specimen. The other specimen in our collection is larger, but as it is imbedded in matrix, no exact measurements can be given. In the use of Odontostomia for Odostomia I follow Dall (1892, p. 248). Record of specimens : Mouth of Santa Cruz River, 2 sp. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih., Cossm.). Affinities: According to Cossmann, this species is distinguished by the angulated whorls from the Eocene species of France. I do not know any other species in which this feature is shown, except a variety of O. con- oidea Brocchi (Dall, 1892, p. 250). This variety is mentioned by Wood (1848, p. 86), from the Pliocene of England. He says "the angulated edge of the volution gives a subcanaliculated form of suture to another variety" (fig. 3a on pi. 9). O. conoidea is found from Miocene to Recent times in Europe, the United States, and the West Indies. Gen. TURBONILLA Riss. 1 06. TURBONILLA CUEVENSIS v. Ihering. PL XXXIII, Fig. 8"'". 1897 T. c. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 276, textf. u. 1899 T. iheringi Cossmann, in: Journ. Conchyliol., p. 13 (of sep. cop.), pi. 10, f. 12. 1900 T. c. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell elongate-pyramidal, surface with longitudinal ribs and indistinct spiral striae. Longitudinal ribs about 20 on the last whorl, not extending to the base. Whorls flat, suture distinct and sharply canaliculate. Mouth ovate, columella with an indistinct fold. ORTMANN I TERTIARY INVERTEBRATES. 175 Height, 12 mm; diameter, 3 mm (v. Ihering gives: height, 5; diam- eter, 1.5; Cossmann: height, 7; diameter, 1.5 mm). Remarks: V. Ihering's figure is given as half natural size, but it is— according to the measurements given — about 8 times enlarged. Cossmann's description differs from v. Ihering's only in that he men- tions an indistinct fold on the columella, which may have been overlooked by v. Ihering. (Cossmann calls it "hardly visible.") This fold exists in our individual, but, indeed, it is hardly noticeable. It is situated well up- ward on the columella. Cossmann says it is situated in the lower part, but it is to be remarked that he turns all his shells upside down ; in his figure something like a columellar fold is visible, situated exactly as in our individual, but the figure is too poor to make sure whether this is really this fold. The spiral striae are hardly visible in our specimen. Record of specimens : Mt. of Observation, upper horizon, i sp. Distribution: Suprapatagonian beds of La Cueva (v. Ih.) and Jegua quemada (Cossm.). Affinities: According to Cossmann, this species differs from those of the Eocene of Paris in being much more conical, that is to say, apparently, being less slender. I can, however, hardly support this view, since many of the Eocene species do not differ at all in form from the Patagonian species (see Deshayes, 1864, pi. 20 and 21). It is hard to say to which one of the numerous Tertiary Turbonillce the present species bears the closest resemblance. Earn. SCALARIIDA1 Brod. Gen. SCALARIA Lmck. 107. SCALARIA RUGULOSA Sowerby. PI. XXXI, Fig. ii". 1846 S. rug. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 255, pi. 3, f- 42, 43- 1864 S. lyrata Zittel, Novara Exp. Geol., p. 41, pi. 9, f. 8. 1864 S. browni Zittel, ibid., p. 42, pi. 9, f. 9. 1873 S. browni & lyrata Hutton, Catal. Tert. Moll. Ech. N. Zealand, p. 9. 1885 S. browni & lyrata Hutton, in : Quart. Journ. Geol. Soc., v. 41, p. 550. I 76 PATAGONIA EXPEDITIONS : PALEONTOLOGY. 1887 5. rug. Philippi, Tert. & Quart. Verst. Chiles, p. 83, pi. 9, f. 15. 1897 S- ruS- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 277. Shell large, solid, conical, elongate, not perforated. Whorls convex, with longitudinal ribs and spiral striae. Suture deep. Longitudinal ribs subequal, high, variciform, moderately thick in the young shell, very thick and swollen in the old, the anterior side (directed toward the mouth) is rounded, the posterior excavated, with a sharp edge. The number of ribs varies from 8 to 1 6 in different individuals, and even in one and the same individual the number sometimes varies considerably (for instance from 10 to 1 6 in a large one from San Julian) ; other individuals preserve the number throughout (so a large one from San Julian with 8 ribs on all whorls). In young individuals, where the longitudinal ribs are thinner, they often show, on the upper part, near the suture, a small point or angle. Ribs and intervals crossed by revolving striae, which are more or less distinct (when indistinct, probably worn off). Base of shell with a spiral carina interrupted by the longitudinal ribs. Mouth almost circular, lip reflected. Measurements: Almost complete individual: height, 52 mm, diameter, 18.5 mm; almost complete individual: height, 77 mm, diameter, 25 mm; end broken off: height, 78 mm, diameter, 31 mm. Remarks: The number of revolving striae is very variable. The only difference from the Patagonian shell observed in the New Zealand fossil is the small number (8) of revolving ribs, but I possess from San Julian an individual of medium size that shows exactly this number, and younger ones that possess still less (only 6). In larger individuals this number of revolving striae increases considerably in the Patagonian shell (up to about 20). S. browni is said to possess 16 to 18 longitudinal ribs, a number which is not supported by the figure : the last whorl, in the figure, has on the side directed toward the spectator only 6 ribs, which would bring the total number hardly over 15. Hutton says, that he does not see why 5. lyrata should be different from S. mgulosa, and that S. browni is only a variety of S. lyrata, while Zittel regards S. browni as hardly distinguish- able from S. rugulosa. In my opinion, all these alleged species are forms of one and the same species, which varies in the number of longitudinal ribs, their thickness, and the development of the spiral striae. Indeed, among our material, the young ones completely resemble the figures of S. rugulosa, as well as of S. browni, and some of the larger ones completely resemble S. lyrata, ORTMANN : TERTIARY INVERTEBRATES. 177 while others have the same form, but the rugosities of the ribs caused by the spiral striae are less pronounced. It is impossible to distinguish, among our material, more than one species: if we wanted to do so- according to the number of ribs and the development of the striae — we should be forced to accept about half a dozen species, with as many transitional forms. Record of specimens : Mouth of Santa Cruz River, 5 sp. ; Mt. of Obser- vation, upper horizon, 4 sp.; San Julian, Darwin Station, 21 sp.; Upper Rio Chalia, 2 sp.; 30 miles north of Upper Rio Chalia, i sp.; Canon near Sierra Oveja, Rio Chico, i sp. ; Lake Pueyrredon, base of Tertiary, i sp. Distribution: San Julian (Sow.); Santa Cruz, La Cueva, Jegua quemada, Suprapatagonian, and possibly Patagonian beds (v. Ih.); Chili: Navidad, Matanzas, Lebu (Phil.); New Zealand, Oamaru and Pareora systems, Oligocene and Miocene (Zittel, Hutton). A variety, obsoleta, without spiral striae is mentioned by v. Ihering from Santa Rosa or Punta Raza (see pp. 112 and 119), Tehuelche beds. Affinities: Very closely allied is the Oligocene S. incequistriata v. Koenen (1867, p. 107, pi. 6, f. 14), from northern Germany: the longitudinal ribs are from 14 to 20, and the spiral ribs are more numerous (30 to 40 on the last whorl in individuals, which are only as large as a medium sized 5". rugulosa]. The latter character is the only difference I can discover. On the other hand, the Miocene S. lamellosa (Broc.) (see: Hoernes, 1856, p. 474, pi. 46, f. 7) comes very near (as has been pointed out by Zittel under ^S. lyrata]: it has ribs of the same character, but the number of spiral striae is less (only 6), and the whorls increase (according to the figure) a little more rapidly. In Eocene deposits this type of Scalaria seems to be lacking. Fam. CAPULID^E Cuv. Gen. CRUCIBULUM Schum. 1 08. CRUCIBULUM DUBIUM Ortmann. PI. XXXII, Fig. 3«-». 1900 C. d. Ortmann, in: Amer. Journ. Sci., v. 10, p. 373. Cast subcircular, depressed-conical. Apex subcentral. On one side is seen the impression of the internal cup-shaped lamina, which was attached to the inner wall of the shell. 178 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Diameter, 20 mm ; height, 8 mm. Record of specimens : Arroyo Gio, i cast. Affinities: There are some species living on the western coast of South America ; but without knowledge of the external surface our cast cannot be compared with them. The genus Cntcibidum is found, according to Zittel (1885, p. 215), from the Miocene to recent times in North America and the West Indies. Gen. INFUNDIBULUM Montf. 109. INFUNDIBULUM MERRIAMI (Ortmann). PI. XXXII, Fig. 4"-6. 1887 Trochita costellata Philippi, Tert. & Quart. Verst. Chiles, p. 93, pi. u, f. 4 (non T. costellata Conrad, 1855). 1899 T. merriami Ortmann, in: Amer. Journ. Sci., v. 8, p. 430. Shell thin, depressed-conical ; apex central. Surface with numerous, fine, radial ribs, or rather fine, radial forrows, separated by flat ribs. Height, 12 mm; diameter, 24 mm. (The measurements given by Philippi in text and figure do not agree, although they refer to one and the same individual ; the text gives : height, 9 mm ; diameter, about 25 ; the figure shows: height, 14 mm; diameter, 18 mm.) Remarks: The genus Trochita Schum. 1817, is synonym with Infnn- dibulum of Montfort 1810, but not with Infundibuhim of Klein 1753. But since Klein's genus is Pre-Linnean, Infundibuhtm Montfort is to be used as Tryon (1886, p. 103) does (as subgenus of Calyptrcea]. The chief generic character is found in the distinct spiral diaphragm of the interior, the columellar margin of which does not form a false umbilicus, that is to say : although reflected, the reflected part of the diaphragm is closely appressed, to that no opening remains. This species differs from the following (I. corrugatum) in the much finer radial sculpture. This sculpture is well preserved in one specimen from the lower beds, but hardly at all in that from the upper beds, which is poor. Record of specimens : Punta Arenas, horizon II (lower Magellanian), 2 sp.; Punta Arenas, horizon III (upper Magellanian), i sp. Distribution: Lebu, Chili (Phil.). ORTMANN : TERTIARY INVERTEBRATES. 179 Affinities: A similar species is Trochita filosa Gabb (1869, p. 15, pi. 2, f. 25) from the Miocene of California, but in T. filosa the radiating striae are still finer, and often dichotomous. no. INFUNDIBULUM CORRUGATUM (Reeve). PI. XXXII, Fig. VM. 1859 Trochita corr. Reeve, Conch. Icon., v. 11, pi. 2, f. 9. 1867 Clypeola corr. Gray, in: Pr. Zool. Soc. London, p. 735. 1886 Calyptrcea ( Infundibulum) radians Tryon (pro parte), Man. Conch., v. 8, p. 121, pi. 35, f. 84-88. 1897 Trochita corr. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 279, pi. 4, f. 1 8, pi. 5, f. 26. 1899 Tr. corr. v. Ihering, in : Jahrb. Miner., etc., v. 2, p. 25. Shell subcircular, conical, more or less elevated. Apex more or less central. Surface with distinct radial ribs; the latter rounded, about as broad as the intervals, and crossed by concentric or spiral lines, which are not parallel to the suture. Interior with a spiral diaphragm, which is a little reflected at the columellar side, but does not form a false umbilicus. Height, 15 mm; diameter, 20 mm. Height, 8mm; diameter, 19 mm. Remarks : This species is very variable in external form, higher or more depressed, and in the development of the radiating ribs. The latter are more or less distinct, sometimes quite indistinct. These ribs are some- times visible on the cast, but in most cases they are not, which renders it impossible to distinguish casts of very depressed individuals from the following species. Tryon identifies the living T. cormgata with radians of Lamarck, and possibly he is right : T. radians differs only in the much larger size (see : Reeve, pi. i, f. 3). From San Julian we have a very large cast (pi. XXXII, f. 5'), which would correspond in size to T. radians, except for the higher conical form. This cast appears to be smooth, but I think I can see indistinct traces of radial ribs. Mr. Hatcher has collected a large number of specimens of the recent /. corrugatum at various localities on the coast of Patagonia, and they l8o PATAGONIAN EXPEDITIONS I PALEONTOLOGY. agree completely with the common fossil form found on the mouth of the Santa Cruz River. Record of specimens : Mouth of Santa Cruz River, 25 sp. ; Mt. of Obser- vation, upper horizon, 14 sp. ; San Julian, Oven Point, 2 sp. ; 30 miles north of upper Rio Chalia, 22 sp. ; Shell Gap, Rio Chico, upper horizon, 2 sp. ; Lake Pueyrreden, 600' above base, 1 7 sp. Distribution : Fossil : Santa Cruz and Jegua quemada, Patagonian and Suprapatagonian beds (v. Ih.). Recent: West coast of South America (Chili, Peru, and perhaps Cali- fornia), and East coast of Patagonia (coll. Hatcher: Punta Arenas, Cape Fairweather, Santa Cruz). iioa. INFUNDIBULUM CORRUGATUM VAR. ELATUM var. nov. PI. XXXII, Fig. 5'. We possess three individuals, which are very high, and — although the shell is well preserved — it does not show any radiating ribs, but numer- ous concentric lamellae. They agree in the latter character apparently with the following species, but are very much higher. Among the material from 30 miles north of Rio Chalia, casts are found, which agree in form with this variety, but some of them show traces of radial ribs. To this variety may belong: Trochita parmila of Philippi (1887, p. 93, pi. 11, f. 2), from Navidad. Philippi had only a single small individual, which was apparently poorly preserved. And further, the var. lasvis Gray (1. c.), recent, Falkland Island may belong here, but it is said of to be smooth (without concentric lamellae). We possess this smooth form among the recent material collected by Mr. Hatcher at Cape Fair- weather. Record of specimens : Mouth of Santa Cruz River, 3 sp. in. INFUNDIBULUM CLYPEOLUM (Reeve). PI. XXXII, Fig. 6"-». 1859 Trochita clypeolum Reeve, Conch. Icon., v. n, pi. 3, f. 14. 1867 Clypeola magellanica Gray, in: Pr. Zool. Soc. London, p. 735. 1897 Trochita mag. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 280, pi. 4, f. 17, pi. 5, f. 25. ORTMANN I TERTIARY INVERTEBRATES. l8l 1899 ^ m- v- Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 25. 1900 Infundibulum clypeolum Ortmann, in: Amer. Journ. Sci., vol. 10, p. 380. Like /. corrugatum, but shell much depressed, without radial ribs, and with strong concentric striae. Our largest individual measures: Height, 12 mm; diameter, 30 mm ; v. Ihering gives: Height, 21 mm; diameter, 54 mm. Remarks: I hardly believe that this is a good species. We have seen that in /. corrugatum the radial ribs sometimes disappear, and that the external form is very variable ; indeed, we possess individuals, which are much more depressed than the typical /. clypeolum, but have distinct ribs. But until the relations of the living /. clypeolum and corrugatum are set- tled, I am not prepared to make any change in the accepted nomenclature of these fossil forms. Record of specimens : Mouth of Santa Cruz River, 2 sp. Distribution : Fossil : Santa Cruz and Jegua quemada, Patagonian and Suprapatagonian beds (v. Ih.). Recent: Straits of Magellan. Gen. GALERUS Gray. 112. GALERUS ARAUCANUS (Philippi). PI. XXXII, Fig. ;«-». 1887 Trochita ar. Philippi, Tert. & Quart. Verst. Chiles, p. 92, pi. 1 1, f. i. 1960 Galerus a. Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Cast subcircular, conical, more or less depressed. Spiral diaphragm forming an impression on the cast running down from the apex to the periphery, and occupying not more than about one-fourth of the circum- ference. A second impression, very much shorter, starts also from the apex, and has been formed, apparently, by the reflected part of the dia- phragm. Of the outer surface of the shell only a few traces are seen, and it was apparently smooth. Height, 8 mm; diameter, 19 mm. Remarks: That the second, shorter impression on the cast is formed by the reflected margin of the diaphragm, is positively shown by an indi- vidual from Lake Pueyrredon, in which the apex of the cast is broken 1 82 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. away. This species may be recognized by the two impressed lines on the cast, and these casts are distinguishable at a glance from the casts of Infnndibnlnni corrugatum — which are found associated with them — by the much shorter diaphragm, which does not form a complete revolution, but only part of it. The generic name Galerus Humphreys, 1797, takes precedence over Calyptrcea Lamarck, 1799 (see Tryon, 1886, p. 103), and this genus is distinguished from Infundibulum chiefly in the columellar margin of the diaphragm, which is reflected, and forms a false umbilicus. Record of specimens : Shell Gap, Rio Chico, upper horizon, i cast; Lake Pueyrredon, 600' above base, 14 casts. Distribution: Lebu and Guayacan, Chili (Phil.). 113. GALERUS MAMILLARIS (Broderip). PI. XXXII, Fig. S"'". 1859 Trochita m. Reeve, Conch. Icon., v. 11, p. 3, f. 12. 1886 Calyptrcea m. Tryon, Man. Conch., v. 8, p. 120. 1897 Cal. aff. mam. Pilsbry, in: Pr. Acad. Philad., p. 330. Shell subcircular, conical, elevated or depressed. Apex central. Sur- face with concentric lines of growth, otherwise smooth. Interior with a spiral diaphragm, making i to 2 complete revolutions, the columellar margin is reflected, and forms a distinct false umbilicus. Height, 1 6 mm, diameter, 20 mm; height, 10 mm, diameter, 18 mm; height, 9 mm, diameter, 24 mm. Remarks: There is some variability as to the external form: the shell is more or less high, as shown by the measurements given above. We possess only inner and outer casts, and, superficially, the internal casts very much resemble Infundibulum corrugatmn. But the radial ribs of the latter are completely absent, as is shown by a number of external casts, and a closer examination reveals the fact that the columellar margin of the diaphragm was reflected, forming an umbilicus. This umbilicus is filled with matrix in the cast, and, after breaking away the upper whorls, the cast of the umbilicus is well exhibited in a number of specimens (see plate XXXII, fig. 8*). This species differs from G. araucanus at once in the number of revo- lutions of the diaphragm : while in G. araucanus the diaphragm does not ORTMANN : TERTIARY INVERTEBRATES. 183 form a complete revolution, it completes, in G. mamillaris, the circle at least once (in small individuals), and 'i ^ to 2 times in larger individuals. Record of specimens : Cape Fairweather, 3 1 casts. Distribution : This species is found living on the western coast of South America, from Chili to California (see Tryon, 1. c., p. 120). Gen. SIGAPATELLA Less. 114. SIGAPATELLA AMERICANA Ortmann. PI. XXXII, Fig. 9"'*. 1900 S. a. Ortmann, in: Amer. Journ. Sci., v. 10, p. 373. Shell subcircular or subelliptical, depressed. Apex distinctly excentric. Surface with irregular, concentric, slightly lamellate striae, crossed by very fine radial rugosities. Internal diaphragm spiral, columella excentric, margin of diaphragm slightly concave and slightly reflected at the columella. Measurements of a specimen from Punta Arenas : Height, 3 mm, diam- eter, 17 mm; of a specimen from Santa Cruz: Height, 16 mm, diameter, 49 mm ; of another from Santa Cruz : Height, 5 mm, diameter, 27 mm. Remarks: As to the generic name Sigapatella see Tryon, 1886, p. 104. The radial plications or rugosities are very fine and sometimes indis- tinct. In our smaller individual from Santa Cruz, which is a little worn, there are no traces of them left. Record of specimens : Mouth of Santa Cruz River, 2 sp. ; Punta Arenas, above horizon V (Patagonian), 4 sp., found inside of Ostrea ingens. Affinities: Trochita colchaguensis Philippi (1887, p. 93, pi. 1 1, f. 5) from La Cueva, Colchagua, Chili (age doubtful), may be identical, as well as the recent species S. calyptrcsiformis Lmck. (= tomentosa and mac^llata, Qu. & Gaim.), from Australia and New Zealand (see Tryon, 1886, p. 122, pi. 35, f. 96-99): but these forms do not show the sculpture of our species. A species has been mentioned under the name of Trochita dilatata Sow. = maculata Qu. & Gaim., by Zittel (1864, p. 43, pi. 15, f. 8), from the New Zealand Miocene, but it seems doubtful whether this is a Sigapatella at all. Calyptrcea maculata Qu. & Gaim., is given by Hutton (1873, p. 13) from the Oamaru, Pareora, and Wanganui beds of New Zealand, and thus it would pass from the Oligocene upward to Recent times. No other fossil Sigapatella being known, our species would point most distinctly to Neogene age. 184 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Gen. CREPIDULA Lmck. 115. CREPIDULA GREGARIA Sowerby. PI. XXXII, Fig. io<-«. 1846 C.g. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 254, pi. 3, f. 34- 1864 C. incurva Zittel, in: Novara Exp. Geol., p. 44, pi. 15, f. 9. 1873 Crypta i. Hutton, Catal. Tert. Moll. Echin. New Zealand, p. 14. 1887 Crepidula uncinata Philippi, Tert. & Quart. Verst. Chiles, p. 94, pi. 11, f. 6. 1887 C. gregaria Philippi, ibid., p. 94, pi. 12, f. i (after Sowerby). 1887 Haliotis imperforata Philippi, ibid., p. 102, pi. 12, f. 2. 1897 Crepidula gregaria v. Ihering, in: Rev. Mus. Paul., v. 2, p. 278. Shell oblong, thick, smooth, except for growth-lines, very convex. Apex incurved, marginal, sometimes a little produced ; apical margin of mouth thickened. Diaphragm concave or plane, reaching about to the middle of the shell or slightly beyond, its margin concave. Measurements of largest individual from Punta Arenas : Length, 92 mm, width, 44 mm, height, 34 mm. Remarks: This shell is very variable in shape, broader or narrower. The apex, in smaller individuals, is produced and prominent beyond the line of the curvature of the margin of the shell, and simply incurved ; in very old and large individuals it is subspiral, and raised a little upon the upper surface. C. incurva of Zittel is no doubt this species : the figure agrees perfectly with our complete specimen of small size from Santa Cruz. C. uncinata of Philippi is also a small individual of this species. Haliotis imperforata Phil, comes from Skyring Water: there is no doubt that our giant specimens from Punta Arenas represent this form, and they are nothing but very large C. gregaria. Record of specimens : Mouth of Santa Cruz River, 6 sp. (one of them almost complete) ; Upper Rio Chalia, 31 casts; Arroyo Gio, 3 sp. ; Punta Arenas, horizon V (Patagonian), 6 sp. ; (5 of them very large). Distribution: Santa Cruz (Sow.) ; Jegua quemada and La Cueva, Pata- gonian and Suprapatagonian beds (v. Ih.) ; Navidad beds of Chili : Lebu, Matanzas, Guayacan, La Cueva (Phil.). ORTMANN I TERTIARY INVERTEBRATES. 185 New Zealand : Pareora beds = Miocene (Zitt, Hutt). Affinities: The living C. fornicata L. (see Ball, 1889, pi. 50, f. 23, 24) from the Atlantic coast of North America and the West Indies is very closely allied, and its range in time begins in the Miocene. The Pata- gonian fossil, however, is much larger, generally more elongate, more solid and thicker, and especially the apical margin is more thickened. I cannot distinguish the Patagonian fossil from C. prcerupta Conrad (1849, P- 727» pl- 19< f- 9. Io) from the Miocene of Astoria, Oregon, which, according to Gabb (1869, p. 81) is identical with C. princeps Conrad (1856, p. 326, pl. 6, f. 52) from Subrecent beds of St. Barbara, California (also living). And, further, Gabb identifies this species with Crypta gran- dis Middendorf (1849, P- IOI» pl- IZ» f- 8-10) from Bering Sea. If all these should really prove to be forms of one and the same species, the range would be — in the northern Pacific — also from Miocene to Recent times, and give to the Patagonian beds a distinctly Neogene age. 1 1 6. CREPIDULA DILATATA Lamarck. Pl. XXXII, Fig. ii. 1843 C. d. d'Orbigny, in: Voy. Amer. Merid., v. 5, p. 465, pl. 58, f. 6. 1859 C. d. Reeve, Conch. Icon., v. n, pl. i, f. 3. 1886 C. d, Tryon (pro parte), Man. Conch., v. 8, p. 127, pl. 37, f. 3i» 32- Shell rather thick, broadly ovate or irregularly circular, depressed ; sur- face smooth except for growth-lines. Apex obliquely curved, marginal. Diaphragm slightly concave, hardly reaching to the middle of the shell, its margin sinuate. Measurements (of a cast) : Length, 20 mm; width, 18 mm; height, 6.5 mm. Remarks: This species differs at once from the foregoing in the broader and almost circular outline. All our individuals are casts and compar- atively small, and it seems that the shell was not as thick as that of C. gregaria. Tryon (1. c., p. 127) unites with this species the C. grandis of Middendorf (see above) from the North Pacific, but all the figures pub- lished of the latter are more elongate, so that I believe it comes nearer to C. gregaria. Record of specimens : Cape Fairweather ; 5 casts. 1 86 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY Distribution: This species has not been found previously in the fossil state, but is known living from the Falkland Islands and the Straits of Magellan along the western coast of South America. Fam. NATICID^ Forb. Gen. NATICA Lmck. 117. NATICA CHILOENSIS Philippi. PI. XXXIII, Fig. itt<>>. 1887 N. c. Philippi, Tert. & Quart. Verst. Chiles, p. 89, pi. 10, f. 12. 1899 A7! c. Ortmann, in: Amer. Journ. Sci., v. 8, p. 431. Shell ovate, thick, smooth, except for lines of growth. Spire conical, about y± of the height of the shell. Umbilicus small. Callus very thick, covering most of the umbilicus. Mouth ovate, not dilated. Measurements of a very large specimen : Height, 34 mm, diameter, 25 mm, height of mouth, 21 mm; of a smaller one: height, 24 mm, diameter, 19, height of mouth, 15 mm. Remarks: This species is recognized by the oval outline (which is, however, a little variable), by the thick and solid shell, and the thick cal- lus. The callus leaves only a narrow slit open at the umbilicus. N. gance of Philippi (Cretaceous of Quinquina) is allied, but has a higher spire .(one-third of the height of the shell), while in N. chiloensis it is between one-fourth and one-fifth, rarely more than one-fourth. The external form is a little variable in our species, some individuals being more rounded. The callus of the inner lip is very thick, suddenly nar- rowed near the umbilical region, leaving a narrow, oval or crescentic opening at the umbilicus. The suture is not impressed : in larger indi- viduals, however, where the outer layer of the shell is exfoliated, the suture appears as a deep groove giving a scalariform appearance to the spire, a feature which reminds one of N. chilina and auca d'Orb. from Puerto de Hambre. One individual from horizon III (upper) has the callus in the umbilical region broader, with the slit hardly visible, and approaches thus (as also in the more globular form) N. pachystoma Hupe from the Navidad beds (see Philippi, 1887, pi. 10, f. la), to which also N. barroisi Phil, seems to ORTMANN : TERTIARY INVERTEBRATES. 1 87 belong. The specimen figured by Philippi in fig. ic is probably a differ- ent species from N. pachystoma (N. oyarzuni Phil.). Record of specimens: Punta Arenas, horizon II (lower Magellanian), 25 sp.; Punta Arenas, horizon III (upper Magellanian), 2 sp. Distribution: Chiloe : Cueva de Cucao (Phil.). Affinities: In the ovate form, thickness of shell and callus, and the small umbilicus, this species resembles some Eocene species from the Paris basin, especially N. venusta Deshayes (1866, p. 38, pi. 68, f. 78), but the latter has a much more distinct suture, and the callus is not quite so thick as in our species ; there are other slight differences, but on the whole, N. vemtsta is the only species that I was able to compare with our Punta Arenas fossil. 1 1 8. NATICA OVOIDEA Philippi. PI. XXXIII, Fig. 2. 1887 N. o. Philippi, Tert. & Quart. Verst. Chiles, p. 89, pi. 10, f. 10, a, b (and perhaps fig. 18, as N. solidd]. 1887 N. famula Philippi, ibid., p. 89, pi. 10, f. 13, a, b. 1897 N.f. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 285. 1899 N. f. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 28. 1900 N. ovoidea Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell ovate, moderately thick, smooth. Spire conical, about one-fourth to one-fifth as high as the shell. Umbilicus open, only partly covered by a comparatively thin callus. Mouth ovate, slightly dilated. Height, 27 29 24 20 12 mm. Diameter, 21 24 16.5 16 9 mm. Mouth: 21 23 1 8 1 6 9 mm. Remarks: N. famula is distinguished from N, ovoidea (and its allies), according to Philippi, by its smaller size (height, 15 to 18 mm) and thicker callus. Of our individuals, one is only 1 2 mm high, the rest are larger, and approach N, ovoidea (height 30 to 31 mm, according to Philippi). The callus may be called thick or thin, according to the species selected for comparison, but at any rate, in Philippi's figure of N. famula, it is not thicker than that of N. ovoidea. Our large individuals agree completely with N. ovoidea, and since it is thus shown that this form is also found at Santa Cruz, it seems very likely that N. famula is only the young state of this species. 1 88 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. V. Ihering believes (1897, P- 2%l) that Philippi's figure 18, which is given as N. solida, belongs to N. ovoidea: if so, this species must grow to a much more considerable size. N. ovoidea comes near N. chiloensis, but is distinguished by the thinner shell and much thinner callus, which leaves a larger portion of the umbilicus open ; and further, the mouth is wider. Record of specimens: Mouth of Santa Cruz River, 5 sp.; San Julian, Darwin Station, i cast. Distribution: Santa Cruz, Patagonian beds (Phil., v. Ih.); Navidad beds of Chili : Navidad, Tubul, Llancahue (and perhaps Lebu, of fig. 18 belongs here). (Phil.) 119. NATICA SECUNDA Rochebrune & Mabille. PL XXXIII, Fig. 3".*. 1885 N. secunda Rochebrune & Mabille, in: Bull. Soc. Philom. Paris, sen 7, v. 9, p. 103. 1887 N. obtecta Philippi, Tert. & Quart. Verst. Chiles, p. 88, pi. 10, f. 2, a, b. 1887 N. vidali Philippi, ibid., p. 91, pi. 10, f. 17. 1889 TV. secunda Rochebrune & Mabille, in: Miss. Sci. Cape Horn, v. 6, p. 39. 1897 N. obtecta v. Ihering, in: Rev. Mus. Paul., v. 2, p. 282. 1899 A", o. v. Ihering, in : N. Jahrb. Miner., etc., v. 2, p. 27. 1900 N. secunda Ortmann, in: Amer. Journ. Sci., v. 10, p. 180. Shell semi-ovato-globular, thick, smooth. Spire short, suture incon- spicuous. Umbilicus large, covered in part by a thick callus, which is divided by a groove. Height, 40 mm, diameter, 42 mm ; another individual : Height, 32 mm, diameter, 33 mm. Remarks: As v. Ihering (1899, p. 6) has pointed out, N. secunda of Rochebrune and Mabille is identical with N. obtecta of Philippi ; since the specific name of secunda was already published in 1885 in a prelimi- nary note, this name has precedence over that given by Philippi. N. obtecta of Moericke (1896) is different; v. Ihering calls it (1897, p. 283) N. pachystonta var. moerickei. Record of specimens: Mouth of Santa Cruz River, 16 sp. Distribution: Santa Cruz (Phil., v. Ih.), Jegua quemada and La Cueva (v. Ih.), Patagonian and Suprapatagonian beds (v. Ih.) ; Navided beds of Chili: Navidad, Matanzas, Chiloe (Phil.). ORTMANN : TERTIARY INVERTEBRATES. 189 Affinities: This species, as well as the closely allied N. moerickei v. Ih. (=N. obtecta Moer.) from the Navidad beds, is characterized by the umbilical callus, which is divided by a groove. Moericke has pointed out that the most closely allied European form is the Oligocene N. hantoni- ensis Sow. (see v. Koenen, 1867, p. 148, pi. 12, f. 9), and that other related species are found in Cretaceous and Miocene beds of California. The Californian Miocene species (N. callosa Gabb, 1869, p. 10, pi. 2, f. 17) more resembles Moericke's form, while the European N. hantoniensis is more like the Patagonian N. secunda. Since the California Cretaceous species, N. secta and globosa, are a little more different in external form, the closest relation of our species is with one of Oligocene age {N. han- toniensis]. 120. NATICA DARWINI Hutton. PI. XXXIII, Fig. 4. 1846 N. solida Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 255, pi. 3, f. 40, 41 (non N. solida Blainville). 1864 N. sol. Zittel, in: Novara Exp. Geol., p. 42, pi. 15, f. 6. 1873 N. sol. Hutton, Cat. Tert Moll. Echin. N. Zealand, p. 9. 1885 N. sol. Hutton, in: Quart. Journ. Geol. Soc., v. 41, p. 550. 1886 N. darwini Hutton, in: Tr. N. Zealand Instit., v. 18, p. 334. 1887 N. sol. Philippi, Tert. & Quart. Verst. Chiles., p. 91, pi. 10, f. 10 (nee. fig. 1 8). 1889 N. sol. Rochebrune & Mabille, in : Miss. Sci. Cap Horn, v. 6, p. 29. 1896 N. sol. Moericke, in: N. Jahrb. Miner, etc., Bcil. Bd. 10, p. 558. 1897 N. sol. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 280. 1899 N. darwini v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 29. 1900 N. darw. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell subglobular, thick, smooth. Spire short, suture inconspicuous. Umbilicus large, open, not covered by the callus ; labial callus thick in the upper part, truncated at the umbilicus. Mouth ovate, large. Measurements : Height, 35 mm, diameter, 34 mm (but it grows larger). Remarks: I cannot find any difference between the Patagonian fossil and Zittel's figure of the New Zealandian form, except that in the latter the callus is not truncated at the umbilicus. In weathering, in this species as well as in N. secunda, a comparatively larger amount of shell substance is removed at the suture, so that the latter appears to be situated in a groove or channel. I9O PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Record of specimens : Mouth of Santa Cruz River, 4 sp.; 30 miles north of upper Rio Chalia, i good specimen, 3 doubtful casts. Distribution: Santa Cruz (doubtful casts, according to Sowerby) (Roch. & Mab.); La Cueva and Jegua quemada, Suprapatagonian, but not Pata- gonianbeds (v. Ih.); Navidad beds of Chili; Navidad (Sow., Phil., Moer.), Lebu (Phil.). New Zealand (Zitt), Pareora beds (Miocene, Hutt); Chatham Isl. (Hutt). Affinities: This species comes near those which have been called by the collective name of Natica heros, and are found from Miocene to Re- cent times on the Atlantic coast of North America, and which are divided by Dall (1892, pp. 372, 373) into three species. One of them, N. internet Say, resembles our Patagonian fossil in the thick callus, but the callus is not suddenly truncated at the umbilicus, the umbilicus is narrower, and has a spiral rib inside, and further, the suture is more distinct. (I have compared 4 specimens from the Miocene of St. Mary's River, Maryland, one of which seems to be a true N. heros, the others belonging to N. internet}. In N. perspective!, the suture is more like that of our species, but it has a sharp umbilical rib, while in the true N. heros this rib is want- ing, but the callus is less developed, and the suture deeper. Thus it is hard to say, to which of these forms N. darwini shows the closest affin- ity, but at any rate, it is closely allied to this Neogene group of species. 121. NATICA SUBTENUIS v. Ihering. PI. XXXIII, Fig. 5. 1897 N. s. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 284, textfig. 13. Shell ovato-subglobose, comparatively thin, smooth. Spire short, suture inconspicuous. Umbilicus small. Inner lip with thin callus. No callus on the columellar lip. Mouth large. Height, 40 mm ; diameter, 35 mm, another individual : height, 29 mm, diameter, -24 mm. Remarks: This species is closely allied to the foregoing, but the shell is thinner, the callus is less developed, the umbilicus narrower, the suture a little more depressed, and the external form a little higher. Record of specimens : Mouth of Santa Cruz River, 2 sp. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). ORTMANN I TERTIARY INVERTEBRATES. 191 Affinities: This species approaches more nearly the typical form of N. heros Say (see Ball, 1889, pi. 51, f. n) from Miocene to Recent, but the latter has the suture more depressed. 122. NATICA CONSIMILIS v. Ihering. PI. XXXIII, Fig. 6. 1897 N. c. v. Ihering, in : Rev. Mus. Paul., v. 2, p. 283, textfig. 12. 1899 N. c. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 28. Shell subglobular, rather thick, smooth. Spire obtusely conical, about y± of the height of the shell ; whorls convex, indistinctly angulated near the distinct and deep suture. Umbilicus small, open, without callus. Columellar lip thin, the free part at the umbilicus slightly reflected and thickened. Mouth large. Height, 28 mm ; diameter, 26 mm. Remarks: The distinct and sharply depressed suture distinguishes this species from all other Patagonian species of the genus. In partly exfo- liated individuals the suture is very deep. It seems doubtful whether N. omoia Rochebrune & Mabille (1885, p. 138, and 1889, p. 31), belongs to this species. According to the descrip- tion, the union seems hardly warranted ; but since no figure of N. omoia is given, this question is to remain open. There seems to be a slip of the pen in v. Ihering's description of this species in 1899. After describing N. consimilis, he says: It seems prob- able that also N. consimilis belongs here as a synonym. N. subtenuis cannot be the species intended, since it is quite different, and thus this sentence remains unintelligible. Record of specimens: Mouth of Santa Cruz River, 16 sp. Distribution: Santa Cruz and La Cueva, Patagonian and Suprapata- gonian beds (v. Ih.). Affinities: There is quite a number of species in Eocene and Miocene deposits of the northern hemisphere, which resemble this one in external form, but I cannot point any particular one, that agrees with this one more closely than others. Note: The Australian species of Natica (Tate, 1893, p. 318, ff.) require closer inspection : there are many forms similar to the Patagonian. 1 92 PATAGONIAN EXPEDITIONS-: PALAEONTOLOGY. Fam. TURRITELLID^E Gray. Gen. TURRITELLA Lmck. 123. TURRITELLA EXIGUA Ortmann. PI. XXXI, Fig. 12"' \ 1899 T. e. Ortmann, in: Amer. Journ. Sci., v. 8, p. 430. Shell small, with 10 whorls, about 4 times as high as broad at the base. Suture deep, whorls convex, with 5 to 7 spiral ribs, which are rather crowded, and often alternately stronger and weaker. The stronger ribs sometimes appear to be slightly granulated. Height, 15 mm; diameter, about 4 mm. Remarks: In the small size of the shell this species differs from all other Patagonian Turritellce, and agrees with the two dwarf forms described by Philippi from the Navidad beds of Chili : T. trilirata and parvula. But the latter can easily be distinguished by their flat whorls and smaller number of ribs. Record of specimens : Punta Arenas, horizon II (lower Magellanian), over 100 sp. Affinities: As regards the sculpture and the convex whorls, this species finds many analogous forms in Eocene, Miocene and Pliocene beds, as the type of which we may take the Oligocene and Miocene T. turris Bast, of Europe. But taking into consideration the small size of the present form, we find that the species younger than Eocene are very much larger, and only in Eocene beds do we find a few that might be compared with our species in this respect also. One of the most closely allied forms seems to be : T. granulosa Deshayes, from the Eocene of France, but this one is still considerably larger than T. exigua and the whorls increase more rapidly. 124. TURRITELLA AMBULACRUM Sowerby. PI. XXXI, Fig. i3°>6. 1846 T. ambulacrum Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 257, pi. 3, f. 49. 1846 T. suturalis Sowerby, ibid., p. 257, pi. 3, f. 50. ORTMANN I TERTIARY INVERTEBRATES. 193 1873 T. ambulacrum Hutton, Cat. Tert. Moll. Ech. New Zealand, p. 12. 1887 T. sowerbyana ( = suturalis Sow.) Philippi, Tert. & Quart. Verst. Chiles, p. 76, pi. 9, f. 2 (after Sowerby). 1887 T. ambulacrum Philippi, ibid., p. 76, pi. 9, f. la (after Sowerby). 1889 T. ambulacrum and suturalis Rochebrune & Mabille, Miss. Cape Horn, v. 6, p. 43. 1896 T. affinis Moericke, in: N. Jahrb. Miner., etc., Beil Bd. 10, p. 555, pi. 11, f. 3 (nee T. affinis Hupe, Philippi, Gray, see: Dall, 1892, p. 308). 1897 ^ ambulacrum v. Ihering, in: Rev. Mus. Paul., v. 2, p. 286. 1897 ^ argentina v. Ihering, ibid., p. 286. 1897 T. steinmanni v. Ihering, ibid., p. 289. 1899 T. ambiilacrum and argentina v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 25. Shell elongated, forming an angle of about 16 to 21°. Suture in a deep furrow, whorls flat or slightly concave, with 3 principal revolving ribs, the upper and lower one the strongest, the middle one a little weaker. The upper rib forms a distinct angulation. Between the princi- pal ribs, 3 to 5 finer striae, and below the lower principal rib, i to 3 fine striae. Principal ribs, especially the uppermost, sometimes indistinctly crenulated by the lines of growth, but in no case with granulations. Height, 56'mm; diameter, 15 mm. Remarks: This species is on the one hand very variable, and on the other it assumes a different aspect according to the state of preservation. In young specimens the suture is not so deep as in older ones, and if the spiral striae are well preserved, they represent v. Ihering's T. argentina. Large individuals show a very deep suture, and, as a rule, the surface ornamentation is destroyed to a great degree, so that the finer striae are in most cases completely obliterated, and only the three principal ribs remain. In many cases, only the upper and lower principal ribs are pre- served, and such individuals correspond to Sowerby's type of T. ambula- crum. Individuals, in which the difference between the stronger and finer ribs is less pronounced, form v. Ihering's T. steinmanni. In some cases the lower principal rib is indistinct, and such individuals are Sowerby's T. suturalis (sowerbyana of Philippi). T. affinis of Moericke is a typical young T. ambulacnim ( = argentina v. Ih.). One must bear in mind that Moericke's figure is 4 times natural size. 194 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. I do not think that T. argentina is a distinct variety, but it is only the younger, and better preserved state of the larger, and partly worn T. ambulacrum. All our individuals of the typical T. ambulacrum show in the upper part of the shell the sutural furrow less deep. The external sculpture is essentially the same in both, only in old shells it is much worn, and shows the striae less plainly, or not at all. V. Ihering, in 1899, hinted at the identity of T. ambulacrum, argentina and steinmanni. Record of specimens : Mouth of Santa Cruz River, about 250 speci- mens, and many fragments ; Paso del Rio Santa Cruz, 3 sp. ; Las Sa- linas, 10 sp., and 6 doubtful chalcedony casts; Mt. of Observation, lower horizon, 1 1 sp. ; Upper Rio Chalia, about 20 sp. (mostly casts) ; 30 miles north of Upper Rio Chalia, about 40 sp. (mostly casts); Arroyo Gio, 18 sp. ; Lake Pueyrredon, base of Tertiary, 2 fragments ; Lake Pueyrredon, 600' above base, about 22 sp., and numerous fragments. Note: Some of the casts from 30 miles north of upper Rio Chalia may belong to T-fiatagonica, since this species is also represented at this locality. Distribtttion : San Julian (Sow.); Santa Cruz (Sow., Roch. & Mab., v. Ih.); La Cueva andjegua quemada (v. Ih.); Patagonian and Suprapata- gonianbeds (v. Ih.). — Navidadbeds: Navidad (Sow., Phil., Moer.); Ypun Isl., Chonos Arch. (Sow.) ; Matanzas, Lota, Chiloe (Phil.). — New Zealand: Pareora beds (Miocene, Hutt.) and Wanganui beds (Pliocene Hutt. ; Chatham Isl. (Hutt). Affinities: This Turritella is a characteristic type of the Patagonian and Chilian Tertiary, and continues, in Chili — through the Pliocene T. cingulatiformis Moer. — into Recent times, where it is represented by T. cingulata (see Moericke). In Tertiary deposits of the northern hemis- phere this type of Turritella is quite rare, but it is represented neverthe- less. There is one species in the Miocene beds of Europe, which has some resemblance to it: T. bicarinata Eichw. (Hoernes, 1856, p. 426, pi. 43, f. 8—12). Especially what Hoernes calls the first and second vari- eties (fig. 10, u, 12) much resemble our form in the deep suture and the two strong spiral ribs. In this species, however, these two ribs are situ- ated closer together (the upper one being more remote from the suture), and there is no trace of an intermediate third principal rib. On the other hand, the young shell of T. bicarinata is quite different from the young T. ambulacrum, having only one principal rib. . But then again, T. bicari- nata agrees in the lack of granulations -on the ribs. ORTMANN : TERTIARY INVERTEBRATES. 195 In the American Tertiary we have one species that is apparently very closely allied to ours : T. apicalis Heilprin from the Pliocene beds of Florida. Especially in what Dall (1892, p. 316, pi. 16, f. 10) calls the typical form of T. apicalis, there are two principal ribs, on the upper and lower part of each whorl, with a third and smaller intermediate one, and besides, a number of fine spiral striae: a type of ornamentation that agrees completely with that of T. ambulacrum. (I have compared and verified this character in 6 specimens of this form from the Caloosahatchie beds in the Princeton Museum). In T. apicalis, however, the principal ribs are distinctly and regularly granulated, and the suture is less deep. Thus T. apicalis corresponds very closely to T. cingulatiformis of Moe- ricke (Pliocene of Chili), which is, according to Moericke, the Pliocene descendant of T. affinis = ambulacrum of the Navidad beds. This comparison of the morphological characters of T. ambulacrum with those of T. apicalis would accordingly, for T. ambulacrum, point to an age a little older than that of the Pliocene T. apicalis, i. e., to Miocene. T. aldingce Tate (1893, p. 336, pi. 8, f. i) from the so-called "Eocene" of South Australia (Aldinga Bay) comes very near to T. ambulacrum, but the suture is not so deep. 125. TURRITELLA BREANTIANA d'Orbigny. PI. XXXI, Fig. 14°' ». 1847 T- breantiana d'Orbigny, in: Voy. Astrolabe et Zelee, Geol. Atlas, pi. 5 (Paleont, pi. 2), f. 36, 37. 1887 T. breantiana Philippi, Tert. & Quart. Verst. Chiles, p. 77, pi. 9, f. ib. 1889 T. couteaudi Rochebrune & Mabille, in: Miss. Cap Horn, v. 6, p. 44 (no locality). 1897 T. tricincta v. Ihering, in: Rev. Mus. Paul., v. 2, p. 287, pi. 3, f. 3 (non T. tricincta Hutton, 1873, p. 13). 1898 T. iheringi Cossmann, in : Rev. crit. Paleozool., v. 2, p. 109. 1899 T. iheringi Ameghino, in: Seg. Cens. Nac. Rep. Argent. Supl., p. 4. 1899 T. breantiana var. indecussata v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 26. Shell large, very elongate, forming an angle of about 12 to 16°. Suture not very deep, whorls flat with 3 thick principal revolving ribs, the upper- most the strongest. Ribs, especially the uppermost, crossed by lines of 196 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. growth, and appearing distinctly crenulated, but not granulated. Be- tween the principal ribs are i to 3 fine striae, becoming more numerous \vith age. Measurements of largest fragment : Height, 58mm; diameter, 14 mm. Remarks: Of the principal ribs, the uppermost is always the strongest, forming a distinct angulation. In most of the cases, the lowermost rib is the second in thickness, and the middle one is the finest. But in some cases the lowermost is finer than the middle one, and such individuals represent v. Ihering's T. tricincta. In young individuals the suture is very shallow. Large individuals of this species are easily recognizable by the more elongated and more slender form, and by the thick revolving ribs. But young individuals and fragments, especially if a little worn, are almost indistinguishable from T. ambulacrum, since the slender form is not so evident. It is possible that some of the individuals recorded under T. ambulacrum belong really to this species. Record of specimens: Mouth of Santa Cruz River, 7 sp.; Paso del Rio Santa Cruz, i sp.; Las Salinas, 3 sp. (one of them a cast); Mt. of Obser- vation, lower horizon, 2 sp. (one of them a typical T. tricincta}. Distribution: Chiloe (Phil.); Santa Cruz (Phil., v. Ih.); Jegua quemada (v. Ih.): Patagonian and Suprapatagonian beds (v. Ih.). Affinities: T. perattenuata Heilprin (see: Dall, 1892, p. 316, pi. 16, f. 5, 9), from the Pliocene of Florida has a general resemblance, but seems to be more slender. There are (according to Dall) Miocene species, which resemble T. perattenuata, especially T. terebriformis Dall (p. 31 1), but since there is no figure published, I cannot say what are the relations to T. breantiana. At any rate, we must take T. breantiana as a species of Neo- gene relations. 126. TURRITELLA PATAGONICA Sowerby. PI. XXXI, Fig. is"'6. (?) 1846 T. pat. Sowerby, in: Darwin, Geol. Observ. South America, p. 256, pi. 3, f. 48. 1887 T. darwini Philippi, Tert. & Quart. Verst. Chiles, p. 75, pi. 9, f.7. 1887 T. patago nica Philippi, ibid., p. 76 (after Sowerby). 1889 T. patag. Rochebrune & Mabille, in: Miss. Cap Horn, v. 6, p. 43. ORTMANN : TERTIARY INVERTEBRATES. 197 1897 T. Patag. v..Ihering, in : Rev. Mus. Paul., v. 2, p. 287 (after Sowerby), 1899 T. patag. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 26. Shell elongated, forming an angle of about 17 to 24°. Suture simple, not in a furrow. Whorls flat, with a number of finer or stronger striae, three of which are usually stronger and granulate. Height, 28 mm (not complete); diameter, u mm; height, 38 mm (not complete); diameter, 14 mm. Remarks: This species resembles T. ambulacrum in its more rapidly increasing whorls, but is distinguished at once by the lack of a sutural depression. The three larger revolving ribs are less pronounced, and in well preserved individuals they show distinct granulations, which are inde- pendent of the lines of growth. There is no doubt that the form mentioned by v. Ihering in 1899 under the name of T. patagonica agrees with our individuals ; but there is some doubt whether it is really T. patagonica of Sowerby, since the figure given by the latter shows a distinct sutural furrow. But in this respect the figure does not correspond to Sowerby's diagnosis, which says: "sutura indistincta." Perhaps — as v. Ihering suggests — this figure is not accurate. I have not the slightest doubt that Philippi's T. darwini belongs here, since diagnosis as well as figure correspond closely, with the exception that granulations are not mentioned : but their apparent lack may be due to fossilization, as is the case in most of our specimens. Young fragments are hard to distinguish from T. ambulacrum, since in young ones of the latter species the suture is much less deep than in older ones. Record of specimens : Mouth of Santa Cruz River, 6 sp. ; Paso del Rio Santa Cruz, i sp. ; San Julian, Oven Point, 1 1 sp. ; 30 miles north of upper Rio Chalia, 2 sp. Distribution: Port Desire (Sow.), Santa Cruz (Roch. & Mab., v. Ih.), Navidad beds of Chili: Navidad (Sow., Phil.), Lota, Tubul, Lebu (Phil.). Philippi erroneously says that Darwin found this species at Puerto del Hambre (Port Famine). Affinities: T. chipolana Dall (1892, p. 312, pi. 22, f. 24), from the Miocene of Florida seems to be closely allied in form and sculpture, but the sculpture seems to be more complex and more strongly developed. 198 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. 127. TURRITELLA INNOTABILIS Pilsbry. PI. XXXI, Fig. 16"'". 1897 T. i. Pilsbry, in: Pr. Acad. Philad., p. 330. "Shell long-conic, of about a dozen slowly increasing whorls, which are but slightly convex, but become decidedly so below, the latter two or three being well rounded. Sculpture on the lower whorls of 5 rounded and subequal spiral cords separated by intervals of about the same width, traversed by one to three (generally two) sharp threads. Earlier whorls have three primary spirals parted by intervals bearing a single strong thread, and still earlier the threads disappear from the intervals." (Pilsbry) Remarks : This species is closely allied to the foregoing ( T. patagonicd] : the external form, the suture, and the ornaments are essentially the same. The only difference I can discover is found in the more rounded lower whorls, which produce a more distinct suture, and in the increase of the spiral ribs to 5 on the last whorl. Whether there were any granulations on the principal spiral cords, is hard to say; the external casts are too imperfect to render it certain, but sometimes there is the appearance of granulations. T. cingulatif omits of Moericke (1896, p. 556, pi. 11, f. 4) from the Pliocene Coquimbo beds of Chili seems to be closely allied, but it has only 3 principal cords, and the suture is more depressed. Record of specimens : Cape Fairweather, numerous internal and exter- nal casts. Fam. VERMETIDsE Ad. Gen. VERMETUS Adams. 128. VERMETUS cf. INTORTUS (Lamarck). PI. XXXII, Fig. i. 1848 V. i. Wood, Crag Moll., v. i, Univ. p. 113, pi. 12, f. 8. 1856 V. i. Hoernes, in: Abh. K. K. geol. Reichsanst, v. 3, p. 484, pi. 46, f. 16. 1 86 1 V. i. Moerch, in: Pr. Zool. Soc. London, p. 353. ORTMANN I TERTIARY INVERTEBRATES. 199 1885 V. i. Zittel, Handb. Palaeont, v. 2, p. 212, textf. 285. 1900 V. cf. i. Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Shell generally gregarious, tubular, subquadrate, closely and regularly spiral in the young state, with the whorls in close contact. The extrem- ity suddenly reflected, straightened and free. Aperture subcircular. Surface transversely rugose and often with longitudinal ribs. Diameter of tubes in our specimens : 2 mm. Remarks: Our specimens are not well preserved and only fragmentary, but they agree in general form closely with the figure given by Hoernes. Longitudinal ribs are present at the suture, where the whorls touch each other, and further, there seems to be a single rib in the middle of the whorls, but this rib is visible only on the uppermost whorl of the figured specimen. In size (diameter), our specimens agree best with Wood's figure 8a, and differ considerably from that of the Italian Pliocene form figured by Zittel. Record of specimens : Shell Gap, Rio Chico, upper horizon : i sp.; Lake Pueyrredon, 600' above base : i sp. Distribution: V. intortus is found in Oligocene, Miocene, and Pliocene deposits of Europe. Affinities: Our specimens agree best in the surface characters with the Central-European Miocene form figured by Hoernes, in size with the English Pliocene form figured by Wood, while the Italian Pliocene form is larger, and has more, and more distinct longitudinal ribs. According to Moerch's diagnoses, it would correspond best to the French Miocene form of this species. There remains, however, some doubt, whether we really have to deal here with this European species, but the material at hand is too incomplete to decide this question. 129. VERMETUS (?) INCERTUS sp. nov. PI. XXXII, Fig. 2. Tubes fragmentary, elongate-cylindrical, very slightly and irregularly curved, almost straight. Walls thick. Outer surface transversely rugose, in one specimen indistinctly flattened on one side. Diameter of tube : 5—8 mm. Remarks: There is considerable doubt whether these tubes belong at all to Vermetus, and I cannot find any described species, with which to 200 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. compare them. But I describe and figure them in order that they may be recognized if found again. Record of specimens : Mouth of Santa Cruz River : 3 fragments ; San Julian, Darwin Station : 2 fragments. Fam. APORRHAIDA1 Phil. Gen. APORRHAIS da Costa. 130. APORRHAIS ARAUCANA (Philippi). PI. XXXIII, Fig. 9. 1887 Chenopus a. Philippi, Tert. & Quart. Verst. Chiles, p. 35, pi. i, f. i, 1900 Aporrhais a. Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Shell fusiform, smooth. Upper whorls carinato-angulated, last whorl bicarinate ; upper keel indistinctly nodulose or merely waved. Outer lip dilated, produced into two fingers, and a short process appressed to the spire and directed toward the apex. Height of fragment: 17 mm; diameter, 10 mm. Remarks: I have at my disposal only one single incomplete individ- ual ; the lower digit of the outer lip is broken away, as well as the lower canal. Otherwise it agrees completely with Philippi's species, with the exception that the upper carina is slightly waved, thus giving a suggestion of granulations or tuberculations. Record of specimens : Mouth of Santa Cruz River, i sp. Distribution: Navidad beds of Chili : Lebu (?) (Phil.). Affinities: Species of Aporrhais with carinated whorls (type: A. pes pelecani L., Miocene-Recent, see Hoernes, 1856, p. 194, pi. 18, f. 2-4) begin in the Oligocene beds (A. speciosa Schloth., see Speyer, 1864 a. p. 166, pi. 31, f. 1—5) of Europe, and continue up to recent times, and it is to this group that A. araucana bears the closest resemblance, as has al- ready been pointed out by Philippi. In the lack of distinctly developed nodules, and in the lack of a third (lower) carina on the last whorl, our species differs strikingly from these. ORTMANN I TERTIARY INVERTEBRATES. 2OI Fam. STROMBID^E. d'Orb. Gen. STRUTHIOLARIA Lmck. 131. STRUTHIOLARIA HATCHERI Ortmann. PI. XXXIII, Fig. 10°' ". 1899 5. h. Ortmann, in: Amer. Journ. Sci., v. 8, p. 431. Shell ovato-pyramidal, spire scalariform. Whorls with revolving ribs, which number 20 to 22 on the last whorl, and are all equal in size and distance from each other. Upper part of the whorls oblique, not canalic- ulate, rendered subangular by a series of 10-11 blunt, conical, subcosti- form nodes. Of the spiral ribs, about 5 or 6 are in the region of the nodes, the rest (14-15) are below the nodes, on the lower part of the whorls. Measurements: Height, 22 mm, diameter, 13 mm; another: height, 21 mm, diameter, 14 mm. But it grows a little larger, as is shown by a fragment that is 16 mm in diameter. Remarks: This species differs from all the other South American spe- cies of the genus in the spiral ribs, which are of uniform size, in the upper part as well as in the lower part of the whorls. In all the following spe- cies there are at least a few ribs on the lower part of the last whorl, which are distinctly and considerably stronger than those on the upper part. Also the small number of nodes, and the suture, which is not canaliculate, serve to distinguish this species. ' Record of specimens : Punta Arenas, horizon II (lower Magellanian), 8 sp. 132. STRUTHIOLARIA AMEGHINOI v. Ihering. PI. XXXIII, Fig. n". 1897 S. ameghinoiv. Ihering, in: Rev. Mus. Paul, v. 2, p. 289, textf. 14. 1900 S. chilensis Ortmann, in: Amer. Journ. Sci., v. 10, p. 380 (non S. chilensis Phil., 1887). Shell ovato-pyramidal, spire scalariform. Whorls spirally ribbed, on the last whorl about 20 to 25 ribs, which are very unequal. Upper part of whorls oblique near the suture, not canaliculated, angulated, angulation formed by a series of 12 to 16, rarely up to 18, costiform, conical, blunt 2O2 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. or subacute nodes. In the upper part of the whorls, in the region of the nodes, the spiral ribs (8 to 10 of them) are fine and subequal ; in the lower part, below the nodes, are, on the last whorl, 5 to 6 stronger ribs, alter- nating with finer ones, followed by about 5 finer ones in the lowermost part of this whorl. Height of largest complete individual : 56 mm, diameter, 35 mm ; a fragment has a diameter of 40 mm. Remarks: V. Ihering does not mention the finer ribs on the lowermost part of the last whorl, which are not shown in casts. Well-preserved specimens of medium size agree completely with Philippi's figure of 5". chilensis, having the nodes more conical and subacute, and thus I was led to believe that S. ameghinoi and chilensis are identical. But v. Ihering writes to me, that the true 5. chilensis has 19 very fine spiral threads in the region of the nodes. If that is the case, it is impossible to unite these two species. Record of specimens: Mouth of Santa Cruz River, 43 sp.; San Julian, Oven Point, i sp.; 30 miles north of Upper Rio Chalia, 4 sp.; Lake Pueyrredon, base of Tertiary, 5 sp.; Lake Pueyrredon, 600' above base, i sp. Distribution: La Cueva and Santa Cruz, Suprapatagonian and Pata- gonian beds (v. Ih.). In 1899 (p. 37) v. Ihering doubts the occurrence of this species in the "Patagonian" beds. Affinities: S. chilensis Phil, is the representative of this species in the Navidad beds of Chili (Matanzas and Navidad). i32a. STRUTHIOLARIA AMEGHINOI VAR. MULTINODOSA var. nov. PL XXXIII, Fig. ii6. Not so high, more globular. Whorls convex, hardly angulated, nodes 18-19, more elongated and distinctly costiform. Only 4 to 5 stronger ribs on the lower part of the last whorl, without intermediate finer ones, followed by about 5 finer ribs on the lowermost part of this whorl. Height, 32 mm, diameter, 21 mm; another: height, 41 mm, diam- eter, 27 mm. Remarks: I first believed that this was v. Ihering's S. ornata var. dense- striata (1897, P- 29r> textfig. 15) ; but after having sent a specimen to v. Ihering he informs me that it is not his S. densestriata, the latter being merely a S. ornata without the two larger spiral ribs. ORTMANN : TERTIARY INVERTEBRATES. 203 I think this form is only a variety of .9. ameghinoi, since we possess individuals, which are in some degree intermediate, and especially the number of nodes sometimes increases in S. ameghinoi to 17 and 18. The other differences, number of spiral ribs, smaller size and more costi- form appearance of the nodes, smaller size and more globular form of the shell, are only differences in degree of development ; but at any rate, this form is a very distinct variety. Record of specimens : Mouth of Santa Cruz River, 5 sp.; San Julian, Oven Point, 24 sp. (most of them poor) ; San Julian, Darwin Station, 2 sp. (poor) ; Shell Gap, Rio Chico, upper horizon, i sp. 133. STRUTHIOLARIA ORNATA Sowerby. PI. XXXIII, Fig. i2a-». 1846 S. o. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 260, pi. 4, f. 62. 1887 5. o. Philippi, Tert. & Quart. Verst. Chiles, pi. i, f. 5 (after Sowerby). 1889 S. o. Rochebrune & Mabille, in: Miss. Cap Horn., v. 6, p. 40. 1897 S. o. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 291. 1897 S- °- var- densestriata v. Ihering, ibid., p. 291, textfig. 15. 1899 S. o. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 27. Shell ovate, apex acuminate. Whorls convex, with unequal revolving ribs to the number of about 20 on the last whorl. Upper part of whorls deeply canaliculate at the suture, not angulated, with a series of about 15 costiform, elongated nodes. In the region of these nodes are about 12- 13 fine, subequal spiral ribs, in the lower part, just below the nodes, are 2 very strong ribs, followed by about 5-6 finer ones ; the uppermost of the latter is sometimes a little stronger than the rest, and sometimes a fine rib is intercalated between the two large ones. Height, 25 mm; diameter, 16 mm. Remarks: The canaliculate suture, and the two strong ribs just below the nodes serve to distinguish this species at once. But it is to be re- marked, that in rare cases only one of the larger spiral ribs is developed, and even none at all. The latter form has been called by v. Ihering var. densestriata. Sowerby's figure is very poor ; it represents a cast, and does not bring out the most characteristic features of the shell. 204 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Record of specimens: Mouth of Santa Cruz River, about 270 sp. ; Paso del Rio Santa Cruz, i sp. ; Las Salinas, 4 sp. Distribution: Santa Cruz (Sow., Roch. & Mab., v. Ih.), La Cueva (v. Ih.), Patagonian beds (v. Ih.). Sowerby mentions casts of a large variety from San Julian ; our collec- tions show (see above) that these casts belong to S. ameghinoi, and chiefly to the variety multinodosa. Affinities of the genus Struthiolaria : The genus Struthiolaria is restricted to the southern hemisphere, and is found — aside from the Patagonian, Navidad and Magellanian beds — only in the Tertiary beds of New Zealand, and living in New Zealand and Australia, and, further, it has been dis- covered in the lower Miocene of northern Peru (Zorritos, see : Grzybowski, 1899, p. 647). It begins in New Zealand, according to Hutton (1873, p. x), in the lower Miocene, but it is not represented in the Oamaru or Oligocene beds. The New Zealandian species differ considerably in sculpture from the South American forms, only the form described by Zittel (1864, p. 35, pi. 15, f. 3) without specific name resembles slightly the Patagonian type of this genus. Fam. DOLIIDsE Ad. Gen. DOLIUM Lmck. 134. DOLIUM OVULUM Ortmann. PI. XXXIII, Fig. I3°'». 1900 D. o. Ortmann, in: Amer. Journ. Sci., v. 10, p. 374. Shell ovato-globular, spire short, conical, acute, last whorl large. Sur- face with fine and crowded revolving striae, which are subequal, but in the lower part of the shell finer striae are intercalated. Mouth large, elon- gated-oval, canal very short, truncated, straight and comparatively nar- row. Inner lip without callus, tubercles or folds. Outer lip slightly thickened. Height, 34 mm; diameter, 25 mm. Remarks: I do not see any crenulations on the inside of the outer lip, but the latter is partly broken away or obscured by hard matrix. ORTMANN : TERTIARY INVERTEBRATES. 20$ Record of specimens : Mouth of Santa Cruz River, 2 sp. Affinities: The genus Dolium is preeminently recent and tropical. Fossil representatives — aside from a doubtful Upper Cretaceous species — have been found from Miocene beds upward, so that the presence of this species in the Patagonian beds points distinctly to Neogene age. Gen. PYRULA Lmck. (= Ficula Sw.). 135. PYRULA CAROLINA d'Orbigny. PI. XXXIII, Fig. I4"-6. 1847 P. c. d'Orbigny, in : Voy. Astrolabe & Zelee, Geol. Atlas, pi. 5 (Pale- ontol., pi. 2), f. 34, 35. 1887 Ficula c. Philippi, Tert. & Quart. Verst. Chiles, p. 52, pi. 4, f. 2. 1897 F- c- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 293, pi. 4, f. 19. 1899 F. c. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 30. Shell fusiform, elongated, slender. Spire very short, apex acute. Sur- face with about 22-25 revolving ribs, which are equidistant from each other and equal, crossed by fine, crowded, longitudinal striae. In old shells the revolving ribs become more distant from each other, and at two or three places a single finer one is intercalated. Mouth elongated, canal long and slender. Measurements of a complete individual: Height, 52mm; diameter, 31 mm. Record of specimens: Mouthwof Santa Cruz River, 19 sp.; San Julian, Darwin Station, i sp.; Lake Pueyrredon, 600' above base, i sp. Distribution: Santa Cruz (Phil., v. Ih.), Jegua quemada (v. Ih.); Pata- gonian and Suprapatagonian beds (v. Ih.). Navidad, Chili (Phil.). Affinities: This species is closely allied to a Neogene or recent group of species which are closely connected with one another, and differ chiefly in the development of the spiral sculpture. Ficus pyriformis of Gabb (1869, p. 48, pi. 14, f. 4) from the Miocene of California is very near in external form and sculpture (smaller ribs intercalated between the larger ones, are rare), but the number of the spiral ribs is much larger (about 40), and they are, accordingly, more crowded. F. concinna Beyr. (see Speyer, 1864, p. 184, pi. 33, f. 15, especially fig. I5c), from the Oligocene of Germany is also closely allied : the secon- 206 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. dary spiral ribs are wanting, exactly as in P. Carolina, but the ribs are more numerous (in a much smaller specimen figured by Speyer 30 are present), and the longitudinal striae are less crowded. All other species differ more considerably, especially those forms desig- nated under the name of/*, condita Brongn. (Hoernes, 1856, p. 270, pi. 28, f. 4-6), from Miocene to Recent, and P. reticulata Lmck. (Hoernes, ibid., p. 268, pi. 28, f. 1-3, and Speyer, 1864, p. 185, pi. 33, f. 12), from Oligocene to Recent, in which between the principal spiral ribs one or more secondary ones are regularly intercalated, and in which the longitudinal striae are rib-like, stronger, and more distant from each other. It is extremely significant, that the present species compares better with the Miocene P. pyriformis from California, than with any other species, and especially that it does not exhibit the characters of those forms (P. condita, reticulata} which continue to the Recent time. Fam. TRITONID^ Ad. Gen. TRITONIUM Lmck. 136. TRITONIUM BICEGOI v. Ihering. PI. XXXIII, Fig. 15. 1899 T. b. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 29, pi. i, f. 8. Shell ovato-conical, swollen below, with three varices. Whorls with fine spiral striae, and large tubercles, the latter, on the last whorl, in three spiral rows, those of the upper row larger (6-7 between two varices), those of the lower rows smaller, situated on two indistinct spiral ribs. Columella smooth, canal short, a little twisted and oblique. Outer lip subdentate, near the upper end with a distinct canaliform emargination, opposite to which, on the upper part of the inner lip, there is a dentiform fold. Height of incomplete individual : 76 mm ; diameter, 49 mm. Remarks: In our individuals the canal appears to be long, but the last whorl is almost completely gone. Record of specimens : Mouth of Santa Cruz River, 2 sp. Distribution: Santa Cruz, Patagonian formation (v. Ih.). ORTMANN : TERTIARY INVERTEBRATES. 2OJ Affinities: The type of ornamentation is essentially the same as in the following species (T. morgam], but the external form is quite different, being much broader and comparatively shorter, and less elongated. 137. TRITONIUM MORGANI Ortmann. PI. XXXIII, Fig. 1 6. 1900 T. m. Ortmann, in: Amer. Journ. Sci., v. 10, p. 374. Shell subfusiform, elongated, with three varices. Whorls with fine, unequal, spiral striae and large tubercles, the latter, on the last whorl, in three spiral rows, those of the upper row large, about 7 between two varices ; those of the middle row ( 5-6 ) small, and those of the lower row (3-4) very indistinct, and indicated only by a slight spiral rib. Colu- mella smooth, with a few indistinct crenulations in the lower part ( on the canal). Canal comparatively long, narrow. Outer lip distinctly crenu- lato-dentate within in the lower part, with an indistinct canaliform emar- gination in the upper part, opposite to which, on the upper part of the inner lip, there is a distinct dentiform fold. Height, 63 mm; diameter, 28 mm. Remarks: The specific name is given in honor of Mr. J. Pierpont Mor- gan. Record of specimens : Mouth of Santa Cruz River, i sp. Affinities: This species comes very near T. verruculosiim (Sow.) (1846, p. 260, pi. 4, f. 63, and Philippi, 1887, p. 57, pi. 4, f. 10) from Navidad, but it differs in the much more slender form, and more numerous tuber- cles (in T. verruculosum there are only 2-4 between two varices). The two species of Tritonium known from Patagonia offer a dis- tinctly Neogene feature. The genus is found from Eocene up to Recent times, but the Eocene and Oligocene species differ considerably in sculp- ture from our species, and it is in Miocene deposits where we first find this type of sculpture. In external form as well as in ornamentation, T. morgani — as regards the number of spiral rows or tubercles on the last whorl — comes nearest to T. tarbellianum (Grat.) (see Hoernes, 1856, p. 203, pi. 20, f. 7-12), and especially to the more nodulose variety of this species from the Miocene of Europe. It differs, however, in the more slender form, the longer canal, and the slighter development of the spiral ribs. 2O8 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. Species of Tritonium offering a similar structure to these two Patago- nian species are found in the so-called "older" Tertiary beds of Austra- lia (see Tate, 1888, p. 116, ff.), but they require further investigation. Fam. BUCCINID^. Trosch. Gen. BUCCINUM L. 138. BUCCINUM (COMINELLA) ANN^E Ortmann. PL XXXIII, Fig. 17. 1900 B. a. Ortmann, in: Amer. Journ. Sci., v. 10, p. 374. Shell subfusiform, elongated-oval. Spire long. Whorls 7-8, angu- lated, the angulation with a series of tubercles, 12-14 on tne last whorl which are continued downward as irregular longitudinal ribs. Upper part of whorls (above angulation) slightly concave, appressed toward the suture. Exposed part of upper whorls, below angulation, subcylindrical. Whole surface of shell covered with numerous revolving striae, which are somewhat unequal. Last whorl large. Mouth ovate, elongate, upper end subcanaliculate, lower end truncate, and with a short, reflected canal, forming a varix on the columella. Inner lip a. little expanded, thin. Outer lip thin, smooth within. Height, 66 mm ; diameter, 30 mm. Remarks : The tubercles of the angulation become somewhat irregular on the last whorl, and indistinct near the mouth. The longitudinal ribs are irregular on the last whorl, sometimes two of them starting from one tubercle, sometimes being quite indistinct. This species belongs to the subgenus Cominella. The specific name is given in honor of Mrs. Anna Ortmann. Record of specimens : Mouth of Santa Cruz River, 4 sp. Affinities: This species comes near B. veneris Basterot (1825, p. 47, pi. 2, f. 15) from the Miocene of Southern and Western Europe. The general form is essentially the same, only the canal is a little longer, and the ornaments of the shell are slightly different : in B. veneris the angu- lation has more numerous and more closely set tubercles, and the longi- tudinal ribs are indistinct or wanting. There is no other Bitccinum, to my knowledge, that resembles our species so much as this one. ORTMANN : TERTIARY INVERTEBRATES. 209 139. BUCCINUM (COMINELLA) OBESUM VAR. MINOR (Philippi). PI. XXXIII, Fig. 18. 1887 Fus2ts obesus var. minor Philippi, Tert. & Quart. Verst. Chiles, p. 48, pi. 3, f. 4b. 1900 Buccinttm obesuni Ortmann, in : Amer. Journ. Sci., v. 10, p. 379. Shell broadly oval, swollen. Spire short, conical. Whorls 5, convex, slightly depressed in the upper part, near the suture, with about 14-15 longitudinal ribs, which disappear suddenly on the upper part of the whorls, some distance from the suture, and run down, on the last whorl, almost to the canal. Surface of shell with numerous revolving striae, which are somewhat unequal. Last whorl large. Mouth ovate, upper end slightly canaliculate, lower end truncate, with a broad and very short, reflected canal, forming avarix on the columella. Inner lip thin, outer lip smooth within. Height, 19 mm; diameter, 14 mm. Remarks: This is also a Cominella, and I have no doubt that it is con- generic with the foregoing. The short reflected canal forming a varix is quite unlike the long canal of Fusus, and agrees well with that of B. annce, and the genus Buccinmn in general. It may be remarked that — if belonging to Fusus — the specific name would be preoccupied by Fusus obesus Michelin (subgenus Euthria, see Zittel, 1885, p. 272). Record of specimens : Mouth of Santa Cruz River, 3 sp. Distribution: Chili : Matanzas and Cucao (Phil.). The typical form of B. obesum is found at Navidad. Gen. CHRYSODOMUS Sw. 140. CHRYSODOMUS CANCELLATUS (Ortmann). PI. XXXIV, Fig. 2"-". 1900 Fusus c. Ortmann, in : Amer. Journ. Sci., v. 10, p. 375 (non F. c. Sowerby). 1901 F. ortmanni Cossmann, in: Rev. crit. Paleozool., v. 5, July, 1901, p. 151, footnote (3). Shell small, fusiform, elongate. Spire a little shorter than the last whorl. Whorls convex, surface ornamented by revolving and longi- 2IO PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. tudinal ribs, cancellated. Besides, there are distinct and regular lines of growth. Spiral ribs, on the upper whorls, to the number of 4-5, 12-13 on the last whorl. They are sharp, but flat, equidistant, narrower in the intervening spaces between the longitudinal ribs, but on the points of intersection with the latter, they are slightly broadened, giving the appearance of low tubercles. Longitudinal ribs 12-13 on eacn whorl, rounded (not sharp), but distinct, running from suture to suture, but dis- appearing on the canal. The lines of growth are very distinct, fine and sharp, and very numerous. Mouth elliptical, canal comparatively short. Outer lip crenulated within. Height, 16 mm; diameter, 6.5 mm. Record of specimens: Mouth of Santa Cruz River, 5 sp. Affinities: This species seems to be very closely allied to the Euro- pean Miocene Fusus glomus Gene (Hoernes, 1856, p. 279, pi. 31, f. 2), but the latter is less slender, larger, and the longitudinal ribs are less devel- oped. The character of the spiral sculpture is essentially the same. Another closely allied form is F. nexilis Ball. (1890, p. 128, pi. 8, f. 4) from the Miocene Silex-beds of Florida, but it is less slender, and the outer lip has no crenulations. Another similar form, Chrysodomus glyptus Verr., is found living in the West Indies, but this one has a larger mouth and longer canal (see Dall, 1889, pi. 61, f. 82). 141. CHRYSODOMUS PILSBRYI (Ortmann). PI. XXXIV, Fig. 3. 1900 Fusus p. Ortmann, in: Amer. Journ. Sci., v. 10, p. 375. Shell thick, elongated, fusiform ; spire a little shorter than the last whorl. Whorls 7-8, convex, slightly appressed in the upper part near the suture, ornamented with 8-9 strong, rounded, longitudinal ribs, which are slightly oblique and curved. On the upper whorls these ribs reach from suture to suture, on the last whorl they disappear at a short distance below the middle. All of the surface of the shell is covered by very fine, numerous, but distinct and subequal spiral striae. Lines of growth fine and indistinct. Mouth comparatively small, continued into a compara- tively short canal. Inner lip expanded, smooth; outer lip thick and ap- parently without crenulations. Height, 36.5 mm (not quite complete at upper end); diameter, 12 mm. ORTMANN I TERTIARY INVERTEBRATES. 2 1 I Remarks: There are slight variations in the external form ; some speci- mens are less slender: Height, 30 mm (not complete, but damaged to about the same extent as the specimen given above) ; diameter, 12.5 mm. The number of longitudinal ribs is 1 1 in one individual, in all others 8-9. I am not quite satisfied as to the generic position of this shell. The specific name has been given in honor of Mr. H. A. Pilsbry, of Philadelphia. Record of specimens : Mouth of Santa Cruz River, 5 sp. Gen. SIPHONALIA Ad. 142. SIPHONALIA DOMEYKOANA (Philippi). PI. XXXIV, Fig. 4. 1887 Fusus domeykoanus Philippi, Tert. & Quart. Verst. Chiles, p. 45, pi. 2, f. 10. 1896 F. d. Moericke, in: N. Jahrb. Miner., etc., Beil. Bd. 10, p. 569. 1899 Siphonalia dilatatavzx. stibrecta v. Ihering, in : N. Jahrb. Miner., etc., v. 2, p. 30. 1900 Fusus domeykoanus Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell large, biconically-fusiform, with 7-8 whorls ; whorls angulated, last whorl large. Angulation with a spiral series of large tubercles, 10— ii in one volution, tubercles blunt, conical, situated, on the upper whorls, at a little distance over the suture, continued downward, on .the last whorl, as short longitudinal ribs. Upper part of whorls (above the tubercles) oblique, flat. Whole surface with numerous, crowded, strong spiral striae. Mouth ovate, angulated on the outer lip, and slightly canal- iculate at the upper end. Canal about as long as the mouth, open, slightly curved. Measurements of a complete individual : Height, 98 mm, diameter, 48 mm ; of another : height, 94 mm, diameter, 53 mm ; of an individual with an upper and lower end incomplete: height, 114 mm, diameter, 63 mm. Remarks: V. Ihering refers this form to the New Zealandian living and fossil (Miocene upward) species Fusus dilatatus Quoy & Gaimard (1832, p. 498, pi. 34, f. 15, 1 6), and this species is, no doubt, closely related, but, 212 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. according to the original figure, the living form is broader and less elon- gated. F. domeykoanus, of Philippi, is mentioned by v. Ihering as re- sembling the Patagonian species, but he says that Philippi's sentence: "apertura superius subcanaliculata " does not apply to it. But this cana- liculation is well shown in our specimens, and agrees perfectly with Phi- lippi's figure. On the other hand, I do not think that F. subreflexus of Sowerby (1846, p. 259, pi. 4, f. 57), from Navidad belongs here. The latter has the upper part of the whorls concave, and the general outline, especially of the last whorl, and the ornaments are a little different. Since Philippi does not give any comment on Sowerby's figure or diagnosis (1887, p. 45, pi. 2, f. 8), but simply copies the former, we are to suppose that really another species agreeing with Sowerby's F. subreflexus exists at Navidad. Our specimens, on the average, differ from Philippi's figure of F. domey- koanus only in being a little more slender, but there is variation in this respect among our material, as is shown by the measurements given above. Philippi's figure still more approaches, in this respect, the living F. dilatatus than our specimens do. F. encodes Philippi (p. 45, pi. 2, f. n), also from Navidad, seems to be only a variety, and agrees in outline better with our individuals. F. steinmanni Moericke (1896, p. 570, pi. n, f. 18, 19), from the'Plio- cene Coquimbo-beds of Caldera, Chili, is very closely allied, but the tuber- cles are less developed. Partly exfoliated individuals of our species, where the tubercles are more or less gone, resemble F. steinmanni very closely, so that I was inclined at first, when I had only such poor mate- rial, to take it for that species. Record of specimens : Mouth of Santa Cruz River, 24 sp.; Las Salinas, i cast. Distribution: Santa Cruz, Patagonian beds (v. Ih.); Navidad, Chili (Phil., Moer.). Affinities: As has been demonstrated above, S. domeykoana is repre- sented in the Pliocene beds of Chili by S. steinmanni, and in New Zea- land by S. dilatata, which has been found in Miocene and Pliocene beds (see Hutton, 1873, p. 3, and 1886, p. 348), and still lives there. Among the European species the Miocene F. virgineus Grat (Hoernes, 1856, p. 286, pi. 31, f. 10-12) might be compared with this one, but it is more slender, and the sculpture, although of the same type, is a little dif- ORTMANN : TERTIARY INVERTEBRATES. 2 1 3 ferent. There are no species of this type in Eocene deposits of the north- ern hemisphere. • 143. SlPHONALIA NOACHINA (Sowerby). PI. XXXIV, Fig. 5. 1846 Fusus noach. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 259, pi. 4, f- 58, 59- 1897 Siphonalia noach. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 298. ? 1899 S. n. Cossmann, in: Journ. Conchyliol., p. 19 (of sep. cop.), pi. u, fig. 2, 3 (perhaps jun.?). Shell ovato-fusiform, spire much shorter than the last whorl, subconical. Whorls 5-6, convex, last whorl very large and swollen. Surface orna- ments consisting of spiral grooves, which are deep and quite distant from each other, separated by low and rounded ribs, which are about twice as broad as the sulci. The bottom of the sulci is finely pitted (by lines of growth). Upper whorls with very indistinct longitudinal ribs (8 or 9). Mouth oval, large. Canal of medium length, broad, open, slightly curved. Measurements of an almost complete individual : Height, 61 mm, diam- eter, 35 mm ; another one, slightly damaged, measures : height, 97 mm, diameter, 55 mm. Remarks: This species grew to a considerable size : an individual, with the larger part of the spire broken away, measures : Height, 93 mm ; diam- eter, 59 mm. In large specimens the spiral sculpture of the last whorl becomes very strong, consisting of a number of strong, rounded, sub- equal ribs, separated by narrower, deep and flat grooves, the pitted appear- ance of which is not so strongly exhibited. It seems doubtful whether the small specimen described by Cossmann really belongs here. Record of specimens: Mt. of Observation, upper horizon, 2 sp.; San Julian, Oven Point, 7 sp.; San Julian, Darwin Station, i sp.; Canon near Sierra Oveja, Rio Chico, i sp.; Lake Pueyrredon, 600' above base, i sp. Distribution: San Julian (Sow.); ? Jegua quemada, Suprapatagonian beds (Cossm.). 214 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Fam. MURICID^. Tryon. Gen. MUREX L. 144. MUREX HATCHERI Ortmann. PI. XXXIV, Fig. 6. 1900 M. h. Ortmann, in: Amer. Journ. Sci., v. 10, p. 375. Shell ovato-subfusiform. Whorls 5-6, rapidly increasing. Spire short, conical. Upper whorls angulated by a prominent, but blunt carina, which is situated below the middle of the whorls ; this carina also forms an angulation on the last whorl, and, below it, there are 4-5 other carinae, of which the first, or the first and second, are strong, resembling the upper carina, while the others are smaller, becoming indistinct toward the canal. Upper part of whorls, above the upper carina, flat and obliquely descending from the suture, with a few revolving striae becoming indis- tinct on the last whorl. Varices 5—6, lamelliform, strong and thick, at the points of crossing with the spiral carinae produced into short leaf- or ear-like lobes, strongest on the uppermost carina, and decreasing in size toward the canal. On the upper whorls only the upper row of lobes is visible, which become indistinct toward the apex. Mouth large, ovate, prolonged into an open canal of medium length, hardly as long as the mouth. Outer lip ornamented with 5—6 ear-like lobes, identical with the lobes of the varices. Height, 63 mm ; diameter, 44 mm. Remarks: The development of the spiral carinae differs in the two indi- viduals at hand. The larger one has only two of them strongly devel- oped, the others are small and indistinct, and indicated chiefly by the lobes of the varices, which are quite large. In the other individual, these two larger carinae are also developed, but below them are about four dis- tinct ribs decreasing in size downward. The lobes of the varices are indistinct and small in this individual (partly worn off on the whorls), but very distinct, although smaller than in the first one, on the margin of the mouth. As to the resemblance of this species to Urosalpinx pyriformis (v. Ih.) see below. Record of specimens : San Julian, Darwin Station, 2 sp. ORTMANN : TERTIARY INVERTEBRATES. 215 Affinities: This is the first true Murex known from the Patagonian beds, and, indeed, the first that is referable without doubt to this genus from any South American Tertiary deposits, and it belongs in the sub- genus Phyllonotus Montf. Species of this type are hardly found before Miocene times, but are quite abundant in Miocene, Pliocene, and Recent beds. I cannot compare it with any known form, none having a particu- larly close affinity to it, but on the whole it has a Neogene character. Gen. TROPHON Montf. 145. TROPHON PATAGONICUS (Sowerby). PI. XXXIV, Fig. 7"-". 1846 Fusus patagoniciis Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 259, pi. 4, f. 60. 1897 Trophon lac^niat^^s var. santacruzensis v. Ihering, in: Rev. Mus. Paul., v. 2, p. 294, pi. 3, f. 4. 1897 T. patag. v. Ihering, ibid., p. 296. 1899 T. p. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 31. 1900 T. p. Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell ovato-oblong, with lamelliform varices. Whorls angulated, upper part, near the suture, flat, varices not extending upon this flat part, or only represented by growth-lines. Number of varices from 8— 16, on the last whorl sometimes quite crowded, on the upper whorls more or less distant, elevated, and produced, on the angulation, into acuminate, often recurved lobes. Mouth subcircular or subovate, canal about as long as the mouth or a little shorter, umbilicus larger or smaller. Whole surface of shell, except upper flat part, with spiral striae, which are more or less distinct, often entirely obliterated. Measurements (not quite complete individual) : Height, 72 mm, diame- ter (varices included) 60 mm ; another one, with the upper end gone : Height, 82 mm, diameter, 75 mm ; of a complete individual : Height, 63 mm, diameter, 38 mm. Remarks: v. Ihering's T. laciniatus var. santacruzensis differs, at the first glance, considerably from Sowerby's T. patagonicus: nevertheless, both are connected by numerous intermediate forms, so that it is impos- 2l6 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. sible to draw a line between them. According to v. Ihering, the chief characteristics of T. santacruzensis are : i. It has only 8 varices. 2. The flat part of the whorls is slightly ascending toward the suture (in T. patagoniciis it is said to be descend- ing or excavated). 3. The form of the shell is more elongated and the canal longer. I would make the following remarks on these three points : 1. The number of varices increases with age, but it is already variable in the young shell. Indeed, we have specimens of the same size as v. Ihering's figure, which have only 8 varices ; but many others have more at the same size, 9-11. The average number, in individuals a little larger than v. Ihering's, is between 10 and 12, but sometimes, on the last whorl, the number increases rapidly, reaching 16. Our largest individual, from Darwin Station, however, has only 12. Sowerby's figure represents an individual of a little more than medium size (height, 69 mm), possessing very numerous varices. 2. There is a slight variation as to the flat upper part of the whorls, it being more or less ascending toward the suture, but in most cases it is almost horizontal. In no case, even in individuals corresponding closely to Sowerby's figure, it is descending toward the suture. The excavated appearance is due to the strongly elevated varices. 3. The external form is very variable. As a rule, younger individuals are more slender, older ones comparatively broader, but there are many exceptions to this rule in young ones. Our largest individuals, however, are all short and broad. With the external form the length of the canal varies, and there is considerable variation as to the size of the umbilicus and the develop- ment of the spiral striae, which in many individuals are entirely absent, and in a few very strongly developed. They are best developed in com- paratively young specimens. This species differs from T. laciniatus in the following points: i. The upper flat part of the whorls is not crossed by the varices, and this flat part is broader. 2. The lobe formed by each varix on the angulation is more strongly developed, more acuminate and recurved. In the occasional presence of spiral striae this species approaches also the recent T. geversiamis (Pall.) (see : Kuester & Kobelt, 1878, p. 275, pi. 72, f. 1—3, pi. 73, f. i), which is probably nothing but a variety of T. laciniatus. ORTMANN I TERTIARY INVERTEBRATES. 21J Record of specimens : Mouth of Santa Cruz River, 13 sp.; San Julian, Oven Point, 21 sp.; San Julian, Darwin Station, 5 sp.; 30 miles north of upper Rio Chalia, i sp.; Lake Pueyrredon, base of Tertiary, 4 sp.; Lake Pueyrredon, 600' above base, I sp. Distribution: San Julian (Sow., see Darwin, 1846, p. 112); Santa Cruz, La Cueva, and Jegua quemada, Patagonian beds (v. Ih.). Affinities: This species is apparently the ancestral form of both, T. laciniatus, which is found in the Cape Fairweather beds, and is also recent, and T. geversianus, which is recent. The genus Trophon, according to Zittel (1885, p. 278) is a character- istic Tertiary genus, and hardly found before Oligocene times. The present species has no closely allied forms in deposits of the northern hemisphere. 146. TROPHON LACINIATUS Martyn. PI. XXXIV, Fig. 8°-". 1847 Fusus /. Reeve, Conch. Icon., v. 4, pi. 4, f. 14. 1878 Trophon I. Kuester & Kobelt, in : Martini & Chemnitz, System. Conch.-Cabin., v. 3, pars 2, p. 280, pi. 72, f. 6, 7. 1880 T. /. Tryon, Man. Conch., v. 2, p. 143, pi. 31, f. 330. 1897 T, 1. Pilsbry, in : Pr. Acad. Philad., p. 329. Shell ovato-oblong, with lamelliform varices ; whorls more or less angulated, upper part, near the suture, flat, narrow, crossed by the varices. Varices quite numerous, elevated into ear-like lobes on the angulation. Mouth suboval, canal moderately long, umbilicus larger or smaller. Surface of shell, between the varices, without spiral sculpture. Height (not quite complete), 62 mm, diameter, 36 (varices 'included) ; height (not quite complete), 72 mm, diameter, 38 mm. Remarks : The ear-like lobes in this Cape Fairweather fossil are larger, but less acuminate than in the recent form, but I do not think that this warrants the creation of a new species, especially if we take into consid- eration the enormous variability of the recent form. Record of specimens : Cape Fairweather, 16 sp. Distribution: Known living from the Straits of Magellan and Pata- gonia. 2l8 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. 1463. TROPHON LACINIATUS VAR. INORATUS (Pilsbry). PI. XXXIV, Fig. 8". 1897 ?• varians v. Ihering, in: Rev. Mus. Paul., v. 2, p. 296. 1897 ^ inornattis Pilsbry, in: Pr. Acad. Philad., p. 330, textfig. This form, found — at Cape Fairweather — associated with the typical form of T. laciniattis, is hardly anything more than a variety of the latter. Its external form is more or less slender, sometimes quite swollen. Sur- face without lamellose varices, or only with slight traces of them, smooth except for lines of growth. Height, 60 mm, diameter, 37 mm. Remarks: We possess individuals that show slightly developed vari- ces, in some parts of the shell, which fact makes it the more certain that it is only a variety of T. laciniatus. The upper part of the whorls has sometimes a distinct angulation and a distinct, but narrow, flattened space near the suture (var. gradata v. Ih.) ; in other cases no trace of this angulation is seen, the whorls being evenly convex. This variety much resembles some of the varieties generally classed with T. geversianus, especially : T. geversianus var. calva ( Kuester & Kobelt, 1878, p. 305, pi. 75, f. i, and Tryon, 1880, pi. 32, f. 338), and T. geversianus var. varians (Tryon, pi. 32, f. 346). Some of our specimens, for instance, that figured by Pilsbry, which are more obese- and have no angulation, are indistinguishable from T. varians as figured by Tryon. On the other hand, we have specimens that are more elongate, and the complete lack of spiral sculptures, as well as the fact that this form is found associated with T. laciniatus, is in favor of the course adopted, to leave it with T. laciniatus. The fact that T. laciniattis offers the same variations as T. geversianus is very interesting, and would bring these two supposed species still closer together. Record of specimens: Cape Fairweather, n sp.; San Julian, Darwin Station, above Patagonian beds, 3 sp. Distribution: T. varians, mentioned by v. Ihering from Santa Rosa (or Punta Raza, see pp. 112, 119 and 177), between Santa Cruz and San Julian, and from between San Jorge and Deseado, from the Tehuelche beds, is no doubt this form. ORTMANN : TERTIARY INVERTEBRATES. 2 19 Gen. UROSALPINX Stps. 147. UROSALPINX ELEGANS Ortmann. PI. XXXIV, Fig. 9. 1900 U. e. Ortmann, in : Amer. Journ. Sci., v. 10, p. 376. Shell ovato-fusiform ; whorls 5-6, convex, with spiral striae and 7-8 longitudinal variciform costas, which are rounded. Mouth oval, elongated into an open, but narrow canal, which is about as long as the mouth. Outer lip distinctly crenulated within. Height, 16.5 mm ; diameter, 8 mm. Remarks: This species closely resembles Triton leucostomoides of Sowerby (1846, p. 260, pi. 4, f. 64 = Fusiis sowerbyanus Philippi, 1887, p. 48, pi. 3, f. 1 6), but appears to be more slender than the latter, and the number of ribs is smaller (8 against 12 in T. leucostomoides, according to Philippi), and the ribs are accordingly more distant from one another. As to the genus Urosalpinx szz: Ball, 1890, p. 147, v. Ihering, 1897, p. 297, and Cossmann, 1899, p. 18. U. leucostomoides Cossmann is different from Sowerby's species. Record of specimens : Mouth of Santa Cruz River, 3 sp. Affinities: U. leitcostomoides (Sow.) is closely allied; it is found, according to Sowerby and Philippi, in the Navidad beds of Chili, and according to v. Ihering (1897, p. 321), in the Suprapatagonian beds. V. Ihering does not give any particular locality, nor does he give any additional description or figure, so that it is impossible to control his identification. Murex lamelliferoi Philippi ( 1 887, p. 56, pi. 3, f. 22) from the Navidad beds of Matanzas, Chili, also resembles this species, but it has spiral striae only in the upper part of the whorls, and the external form is a little different. 148. UROSALPINX COSSMANNI Ortmann. PI. XXXIV, Fig. lo"". 1899 U. cf. leucostomoides Cossmann, in: Journ. Conchyl., p. 17 (of sep. cop.), pi. 10, f. 7. 1900 U. cossmanni Ortmann, in: Amer. Journ. Sci., v. 10, p. 380. Shell small, fusiform. Whorls 6, convex, suture deep; surface orna- mented with spiral cords, 5 of which are exposed on the upper whorls ; 220 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. they are crossed by numerous longitudinal, sometimes variciform, ribs. Mouth ovate, elongated into an open, slightly curved canal of medium length, a little shorter than the mouth. Outer lip thickened and crenu- lated on inner side. Height, 10 mm, diameter, 4.5 mm; a larger, but incomplete, individual has a diameter of 6.5 mm; Cossmann gives: height, 13 mm, diameter, 6 mm. Remarks: This species differs from T. leucostomoides in the more numerous (30 and more), and finer longitudinal ribs, and further, in the more slender form. Record of specimens: Mouth of Santa Cruz River, 14 sp. Distribution: Jegua quemada, Suprapatagonian beds (Cossm.). • 149. UROSALPINX PYRIFORMIS (v. Ihering). PI. XXXIV, Fig. ii. 1897 Trophon p. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 295, pi. 3, f. 5. Shell ovato-pyriform, spire short. Whorls with spiral ribs, 2—3 of which are stronger, the uppermost of them forming an angulation on the whorls. Varices lamellar, 7-9. Mouth ovate, canal short, straight. Height, 14 mm; diameter, 9.5 mm. Remarks: Our specimen is smaller than that figured by v. Ihering, and poorly preserved, v. Ihering's figure is very indistinct, and shows only the external form, which agrees completely with our individual, v. Iher- ing mentions 3 stronger spiral ribs, which are not seen in his figure. Our individual has only 2 stronger ribs ; and further, he gives 9 varices, while I see only 7. There is a very striking resemblance -to Murex hatcheri: the chief dif- ference is, that M. hatcheri has only 5 or 6 varices, which are distinctly lobate. In our individuals of M. hatcheri, the varices on the upper whorls are so much obscured, that it is impossible to count them correctly, but they seem to be more numerous there. Possibly, T. pyriformis is only the young stage of M. hatcheri, but the lack of more material prevents me from determining this question. Record of specimens : Lake Pueyrredon, base of Tertiary, i sp. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). ORTMANN : TERTIARY INVERTEBRATES. 221 Fam. FUSIDsE Tryon. Gen. FUSUS Lmck. 150. Fusus SUBSPIRALIS spec. nov. PI. XXXIII, Fig. 19. Shell fusiform, elongated and slender. Eight whorls are preserved, but the shell is defective below. Whorls very sharply angulated, suture deep. Angulation forming a sharp and simple (not nodulose) carina on the upper 6 whorls ; on the yth and 8th whorl there are small, remote, but distinct tubercles, not continued as ribs below or above the angulation. Upper surface of whorls, above the carina, and lower surface slightly concave, receding toward the suture. Surface of shell apparently smooth on the upper whorls, but with fine revolving striae on the lowermost whorl in its upper as well as in its lower part, continued downward as far as can be seen. Striae not much crowded, crossed by very fine and indistinct lines of growth. Columella straight, slender, elongate, indicating a long and straight canal. Height, 31 mm; diameter, 12 mm. Remarks: Only one incomplete individual, imbedded in hard, refrac- tory matrix — apparently the upper part of a rather large species — is pres- ent. Last whorl not preserved, but the elongated columella indicates a species of the genus Fusus. The number of nodes of the carina cannot be ascertained, since only part of the lower volutions is exposed, and this part lacks most of the shell. I have, however, no doubt that it is possible to recognize this species, if better material should be found. Record of specimens : Punta Arenas, horizon II (lower Magellanian); i sp. Affinities: F. subspiralis resembles F. oxytropis Philippi from the Navidad beds of Chili (see under F. archimedis], but differs in the larger size and much more slender spire ; also the form of the upper whorls is different, the carina being situated, in F. oxytropis, nearer to the lower suture, in F. subspiralis, nearer to the upper one, although only very slightly above the middle. 222 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. 151. Fusus ARCHIMEDIS Ortmann. PI. XXXIII, Fig. 20--». 1900 F. a. Ortmann, in: Amer. Journ. Sci., v. 10, p. 374. Shell fusiform ; spire shorter than the last whorl, scalariform. Whorls over 5 (upper part of spire missing), very prominently angulated, suture very deep. Upper part of whorls, above angulation, flat, obliquely descending from the suture, lower part of upper whorls (below angula- tion) very slightly convex, obliquely receding toward the lower suture. Angulation blunt, with a number (10—13) °f blunt, often indistinct tuber- cles. Sometimes these tubercles resemble indistinct longitudinal ribs, running for a short distance downward. Surface of shell with fine revolv- ing ribs on the lower part of the whorls and on the angulation, but these ribs disappear on the upper part of the whorls at a short distance from the angulation. Whole surface with distinct lines of growth, which have a squamulose appearance where they cross the revolving ribs. Last whorl large. Mouth triangular, continued into a long and straight canal. The revolving ribs of the last whorl become indistinct on the canal. Height, 50 mm (but defective at upper end), diameter, 25 mm ; diam- eter of a fragment, 31 mm. Remarks: Characterized by the strongly angulated whorls and deeply receding suture. The larger part of the upper flat portion of the whorls is quite smooth, except for growth-lines. Only near the angulation 3 to 4 revolving ribs begin to appear, and these ribs continue downward over the angulation toward the canal, where they become indistinct. Record of specimens : San Julian, Darwin Station, 3 sp. Affinities: Only one species of this characteristic, strongly angulated form is known from the South American Tertiary : F. oxytropis Philippi (1887, p. 50, pi. 3, f. 15) from Navidad and Tubul, but this one is much smaller, and the angulation much sharper, cariniform. I know only one other species that may be compared with ours : F. hector Whitfield (1892, p. 199, pi. 25, f. 3-6) from the Eocene Marls of New Jersey. But the latter is distinguished at once by the spiral sculp- ture: the upper part of the whorls, above the angulation, does not possess any spiral lines, and those below the angulation are much more distant from each other and fewer in number. But, on the whole, the type of sculpture is very similar in both species. ORTMANN I TERTIARY INVERTEBRATES. 223 152. Fusus TOROSUS Ortmann. PI. XXXIV, Fig. i. 1900 F. t. Ortmann, in: Amer. Journ. Sci., v. 10, p. 375. Shell subturbinate-fusiform. Spire short, rather depressed. Whorls 4, last whorl very large. Surface with numerous fine spiral ribs, which are rather crowded and somewhat unequal, crossed by very fine, squamiform lines of growth. Whorls strongly convex, swollen, with about 7 strong, variciform, longitudinal ribs, which begin at the suture, and become thick and swollen in the middle of the last whorl, attenuating again toward the lower end of the shell. Mouth ovate, continued into a canal of moderate length, which is slightly curved. Height, 31 mm; diameter, 20 mm. Remarks: In external form this species closely resembles F. pyruli- formis of Sowerby (1846, p. 258, pi. 4, f. 56) from Navidad. I should not hesitate to identify my individual with Sowerby's species, if it was not for the account of the Navidad form given by Philippi (1887, p. 43, pi. 2, f. i) and Moericke (1896, p. 569, pi. 11, f. i, 2). According to these authors, Sowerby's figure is very poor, and both give better figures of the Navidad species (Philippi in an unnumbered figure in the upper right hand corner of plate 2). According to these figures, the Navidad species differs from ours in the following particulars : (i) The upper part of the last whorl, as well as the upper whorls, is smooth, the longitudinal ribs not being continued upward to the suture : in our species these ribs continue to the suture, being very distinct on the spire. (2) The last whorl, from the middle downward, is occupied by swollen longitudinal ribs, which are cut up into a number of tubercles, which form spiral rows : in our species these ribs are crossed by fine spiral cords, but are not cut up by them into tubercles. (3) The canal is perfectly straight, while it is slightly curved in our individual. I entertain some doubt, whether Philippi's and Moericke's species is really the F. pyndiformis of Sowerby. Sowerby's figure, no doubt, is poor, but his diagnosis applies perfectly to our individual, since he calls the spire "rudis," by which expression he may have intended the rough, tuberculated appearance given to the spire by the upper continuations of the ribs toward the suture, and, further, since he describes the sculpture 224 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. of the last whorl as tubercles continuing downward as ribs, crossed by furrows ( " anfractibus . . . medio tuberculatis, tuberculis transversim sul- catis, in costas subdecurrentibus"), which agrees better with our species than with Philippi's and Moericke's species. If our species should prove to belong really to Sowerby's species, the specific name of Pyruliformis is to be retained, and that of Philippi's species is to be changed. Record of specimens : Mouth of Santa Cruz River, i sp. Affinities: According to the considerations given above, there is no doubt that the Navidad form F. pyruliformis is very closely allied. The latter has been compared by Moericke with F. burdigalensis Bast, from the Miocene of Europe (see Hoernes, 1856, p. 296, pi. 32, f. 13, 14). This species is remarkable for the Pyrula-\\\Lt form of the shell, and this character is still more strongly expressed in F. pyruliformis as well as in F. torosus. The sculpture, however, is different, F. burdigalensis having only a row of small tubercles, but no costiform tubercles, and no acces- sory rows of tubercles as F. pyruliformis. Fam. VOLUTIDA1 Gray. Gen. MARGINELLA Lmck. 153. MARGINELLA GRACILIOR v. Ihering. PI. XXXV, Fig. i. 1897 M. g. v. Ihering, in: Rev. Mus. Paul, v. 2, p. 308, textfig. 18. Shell ovato-oblong, subcylindrical, solid, smooth. Spire short, mucro- nate. Upper whorls with a series of indistinct tubercles, wanting com- pletely on the last whorl. Columella with 4 folds. Height, 20 mm; diameter, 10.5 mm. Remarks: The obtuse tubercles mentioned in v. Ihering's diagnosis are not visible in his figure, and in our individual only very slight traces of them are discernible. This species differs from M, quemadensis and con- finis v. Ih. in the more slender form. Record of specimens : Mouth of Santa Cruz River, i sp. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: This species, in its external form, recalls M. bella (Conr.) and M. faimula Dall (1890, p. 53, pi. 4, f. 8, 9), the former from the Mi- ORTMANN : TERTIARY INVERTEBRATES. 225 ocene, Pliocene and Recent of the southern states of North America, the other from the Miocene of Florida, but M. gracilior is very; much larger, more than double their size, and the nodules of the upper whorls are not present in the North American species. 154. MARGINELLA PLICIFERA v. Ihering. PL XXXV, Fig. 2. 1897 Af. p. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 308, textfig. 19. Shell ovato-elongate, solid. Spire more elongated than in any other Pata- gonian species, conical. Whorls with rib-like, longitudinal folds, above, near the suture, slightly concave and with spiral striae. Columella with 4 folds. Height, 22 mm, diameter, 1 1 mm, length of mouth, 15 mm ; v. Ihering gives: height, 31 mm, diameter, 16 mm, mouth, 20 mm. Remarks: v. Ihering, in the diagnosis, calls the spire "breviuscula," but says farther on that it is more elongated in this species. His speci- mens were badly preserved and partly damaged. In our complete indi- vidual the spire is almost intact, and accordingly, the form appears more slender than in v. Ihering's figure. v. Ihering further says, in his diagnosis and description, that an im- pressed line accompanies the suture. Nothing of this kind is seen in his figure. In our specimens there is a slight depression of the upper part of the whorls near the suture, and in one of them there are distinct revolving striae (5—6) on this part of the shell. Record of specimens : Mouth of Santa Cruz River, 6 sp. (most of them greatly damaged). Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities : The sculpture of this species is very remarkable, and I can- not find any other species that might be compared with it in this respect. 155. MARGINELLA OLIVELLA nom. nov. PL XXXV, Fig. 3a'». 1900 M. olimformis Ortmann, in: Amer. Journ. Sci., v. 10, p. 376 (non M. oliviformis Tuomey & Holmes, 1857). Shell elongate, subcylindrically-fusiform. Spire conical. Surface of shell smooth and shining. Suture quite indistinct. Mouth long and 226 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. narrow, canal very short, represented only by a rounded sinus. Colu- mella with 4 subequal folds. Outer lip thickened, smooth within. Height, 1 1 mm ; diameter, 5 mm ; length of mouth, 6.5 mm. Remarks: This species differs from the other Patagonian species in the perfectly smooth surface and very elongated form. Record of specimens : Mouth of Santa Cruz River, 5 sp. Affinities: The Pliocene and recent M. styria Dall (1890, p. 54, pi. 5, f. i) resembles this species in form, but it is smaller, still more slender, and the spire is longer. Gen. VOLUTA L. 156. VOLUTA TRIPLICATA Sowerby. PI. XXXV, Fig. 4—. 1846 V. t. Sowerby, in: Darwin, Geol. Observ. S. America, p. 262, pi. 4, f. 74. 1887 V. t. Philippi, Tert. & Quart. Verst. Chiles, p. 70, pi. 7, f. 8-12. 1897 V- dorbignyana v. Ihering, in: Rev. Mus. Paul., v. 2, p. 303 (teste v. Ihering). 1899 V. triplicata v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 33. Shell moderately elongated, subfusiform. Whorls 6, surface with numerous, distinct, and rather crowded spiral cords, and a number of longitudinal ribs (8-16), which are rather prominent; their upper end terminates abruptly some distance from the suture, being often tuberculi- form ; the uppermost part of the whorls, near the suture, is more or less distinctly concave. Last whorl moderately inflated, mouth (including canal) about half as long as the whole shell. Columella mostly with 3 plaits, which are sometimes subequal, and sometimes the uppermost is weaker than the others and may disappear; in one case there are 4 dis- tinct plaits, the uppermost the smallest. Measurements: Height, 106 mm; diameter, 46 mm (large part of canal gone). Height, 92 mm ; diameter, 41 mm (large part of upper end gone). Height, 87 mm ; diameter, 34 mm (almost complete, only apex wanting). Height, 44 mm; diameter 21 mm (complete, young). Remarks: The most important characters of this species have already been pointed out by v. Ihering, and are found in the relation of the length ORTMANN : TERTIARY INVERTEBRATES. 22y of the mouth to the whole shell, and in the longitudinal sculpture. The longitudinal ribs end abruptly before reaching the suture, forming a dis- tinct shoulder, and sometimes this shoulder is marked by a distinct tuber- culiform development of the ribs. In all other characters this species is extremely variable. The external form is more or less slender : there are some individuals, which are as slender as the following species (V. grac- ilior}; the upper whorls are sometimes about half as high as broad, some- times almost as high as broad ; the spiral cords are more or less devel- oped ; the number of longitudinal ribs is very variable, between 8 and 16, and, further, the number of columellar plaits varies from 2 to 4. Even the character of the longitudinal ribs is not quite constant ; some- times their upper termination is less sudden, and, indeed, there are indi- viduals which approach the following species also in this respect. v. Ihering says (under V. alto], that the transition from the mouth into the canal is well marked by an obtuse angulation : I cannot see anything like that in any of our specimens : the outer lip of the mouth passes in a regular curve into the canal. I possess two young individuals, which show the nucleus of the shell. It corresponds closely to what Ball (1890, p. 68) calls the Caricella-nncletts (see plate XXXV, Fig. 4rf>'), with a distinct small point or apical spur. This species is, accordingly, not to be classed with the Volntoid-series, as Dall(p. 69) does, but with the Scaphelloid-series (ibid., p. 70). I may men- tion here that we have the same nucleus in V. dorbignyana, and it is quite probable that the other Patagonian species also belong to the same type, which fact would bring them into closer relation with the living Volutce (Gen. Scaphelld] of Patagonia (V. ancilla, magellanica, etc.). Record of specimens : Mouth of Santa Cruz, 1 3 sp. Distribution: Santa Cruz, Patagonian beds (v. Ih.); Navidad beds of Chili: Navidad (Sow., Phil.), Matanzas (Phil.). 157. VOLUTA GRACILIOR v. Ihering. PI. XXXV, Fig. s-«. 1887 V. gracilis Philippi, Tert. & Quart. Verst. Chiles, p. 70, pi. 7, f. 13 (non V. gracilis Lea, 1833). 1896 V. gracilior v. Ihering, in: Nachr. deutsch. malakozool. Ges., p. 96. 1896 V. queniadensis v. Ihering, ibid., p. 97. 228 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. 1897 ^ quemadensis v. Ihering, in : Rev. Mus. Paul., v. 2, p. 304, pi. 3, f. 7. 1897 V- philippiana v. Ihering, ibid., p. 305 (non V. pliilippiana Dall., 1890). 1899 V. philippiana v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 34. 1899 V. quemadensis v. Ihering, ibid., p. 34. 1900 V. gracilior Ortmann, in: Amer. Journ. Sci., v. 10, p. 381. This species is extremely near the preceding, and it may be only a variety of it. According to our material it differs in the following points: 1. The shell is more elongated, and the whorls are higher. 2. The longitudinal ribs are more numerous (about 18-20), and do not end abruptly before reaching the suture; they disappear gradually, and traces of them are continued to the suture. They do not form a series of tuberculiform prominences. 3. The whorls are more evenly convex, with hardly a trace of a shoulder. The upper part of the whorls is only slightly depressed. 4. There are, as a rule, only two columellar plaits, although traces of a third (upper) one are sometimes developed. Meastirements : Height, 135 mm; diameter, 66 mm (large part of spire missing). Height, 119 mm; diameter, 46 mm (not quite complete). Height, 88 mm ; diameter, ca. 36 mm (only uppermost part of spire miss- ing, but last whorl damaged). Remarks: I unite the two species called by v. Ihering V. philippiana and V. qtiemadensis, respectively. V. philippiana is said to possess a longer spire, and longitudinal ribs, which do not terminate in tubercles, and the whorls are said to be evenly convex : in all other respects it resembles V. triplicate. V. quemadensis is said to differ from V. pJiilip- piana in the still more elongated spire, with higher whorls, the larger number of ribs, and the presence of only two plaits on the columella. Among our material, I find that the external form is very variable. Although all specimens registered under this form are more slender than the average of V. triplicate, there are, among the latter, individuals which approach this form closely. Those with the most slender spire, and with the highest whorls (almost as high as broad), which would correspond, in this respect, to V. quemadensis, possess a distinct third columellar plait, thus uniting characters of V. quemadensis and philippiana; as to the number of longitudinal ribs, there is so much variability, that it is impos- sible to draw any line. Indeed, in the more elongated individuals, the number of these ribs is larger than in the typical V. triplicata, but some ORTMANN I TERTIARY INVERTEBRATES. 229 individuals, belonging undoubtedly to V. triplicata, possess 16 or even 17 ribs, while sometimes in the elongate form only 17 or 1 8 are present. As to the height of the upper whorls, there is no less variability : some spec- imens correspond to the proportions given by v. Ihering for V. quema- densis, while others correspond to those of V. philippiana, but many intermediate individuals are found. Even the development of the longitudinal ribs is not quite uniform, some specimens showing traces of an angulation near the suture, giving a suggestion of the series of tubercles found usually in V. triplicata. Thus, I can separate only a more slender form from V. triplicata, but I doubt very much that it is really a good species. Dall (1890, a, p. 314) and v. Ihering (1897, P- 3°5> an<^ ^99, p- 32) have already suggested that all these forms may belong as varieties to one and the same species (V. triplicata}. The recent Volutilithes philippiana Dall (1890, a, p. 303, pi. 9, f. 4) from the coast of Chili (677 fath.) is quite different from this fossil species, as is seen at once by the fact that it already has 6 whorls at a size of only 36 mm, while in the fossil form of the same size hardly more than 3 whorls are present. And, further, this recent form is a Volutilithes, while our fossil — in analogy with V. triplicata — seems to belong to Scaphella. Accordingly, the specific name of philippiana cannot be used for this fossil, and since gracilis of Philippi is preoccupied, we must adopt the name gracilior proposed by v. Ihering in 1896. A cast of this species has been sent to us by v. Ihering under the name of V. triplicata. •Record of specimens : Mouth of Santa Cruz River, 7 sp.; San Julian, Oven Point, i sp.; San Julian, Darwin Station, 2 casts; Upper Rio Chalia, i cast; Arroyo Gio, i cast. Distribution: Santa Cruz (Phil., v. Ih.), Patagonian beds (v. Ih.); Jegua quemada, Suprapatagonian beds (v. Ih.). 158. VOLUTA PETERSON: Ortmann. PI. XXXV, Fig. 6. 1900 V. p. Ortmann, in: Amer. Journ. Sci., v. 10, p. 376. Shell elongate, fusiform. Whorls 5 (aside from the apex, which is broken off). Spire slender, conical, mouth apparently not much longer 230 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. than half of the shell (lower end of shell damaged). Upper whorls quite high, about two-thirds as high as broad (the measurements are : height of penultimate whorl, 24 mm ; width, 38 mm). Whorls almost evenly con- vex, only slightly appressed and concave toward the suture, without a dis- tinct angulation. Surface beautifully cancellated by spiral and longitudi- nal ribs. Spiral ribs strongly developed, equidistant, sharp, a little more crowded on the upper whorls than on the last one. Longitudinal ribs a little stronger than the spiral ribs, sharp, running from suture to suture, about 30 on the last whorl. Cancellations rectangular, about twice as broad as high on the last whorl, and about three or four times as broad on the upper whorls, traversed by fine lines of growth. Mouth elongated (lower end not preserved). Columellar plaits not clearly visible, but there are at least two which seem to be quite weak. Height, 148 mm (not complete); diameter, 65 mm. Remarks: The sculpture of this species is quite unique, and charac- terizes it sufficiently. Although the sculpture can be compared with that of V. triplicata and gracilior to which this species is apparently related, the large number of longitudinal ribs, which are developed as sharp and nar- row carinse, and the cancellations produced by the stronger development of the spiral ribs are quite unlike what we see in the other species named. The specific name is given in honor of Mr. O. A. Peterson, Mr. Hatcher's assistant, who collected this species. Record of specimens : Mouth of Santa Cruz River, i sp. 159. VOLUTA DORBIGNYANA Philippi.1 PI. XXXVI, Fig. I "-'. 1887 V. d. Philippi, Tert. & Quart. Verst. Chiles, p. 70, pi. 7, f. 7. 1899 V. d. v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 33. Shell fusiform. Whorls 6 (and two apical whorls). Spire conical, moderately long. Mouth distinctly longer than half of the shell, almost two-thirds of it. Whorls convex, with a more or less distinct shoulder, above which the upper part of the whorls is depressed or slightly concave (more distinctly so on the last whorl), and appressed to the suture. Sur- llt is to be remarked, that there already exists a Valuta orbignyana Mueller (Mon. Aachen. Kreideverst, v. 2, 1851, p. 50). The latter species, however, is brought by Holzapfel (Palason- tograph., 34, 1868, p. 97) into the genus Volutolithes. ORTMANN : TERTIARY INVERTEBRATES. 231 face with spiral striae and longitudinal ribs, this sculpture becoming quite indistinct on the last whorl. In large individuals the last whorl is often quite smooth except for growth-lines. Spiral sculpture very variable, sometimes strongly developed, in other cases weak, or entirely disappear- ing. Longitudinal ribs also variable ; in most cases distinct on the upper whorls, but not reaching the upper suture. On the lower whorls these ribs disappear, being represented, in many cases, by slight and indistinct swellings on the shoulder, and disappearing in very large shells entirely on the last whorl. Mouth elongate. Columella with three plaits, the uppermost sometimes very slightly developed. Measurements: Height, 186 mm; diameter, 79 mm (complete, except for apex). Height, 70 mm ; diameter, 27 mm (complete, young). Remarks: As v. Ihering points out, this species is well characterized by the elongated mouth, which is comparatively much longer than that of V. triplicata and allied forms, and, further, by the tendency of the longi- tudinal ribs to disappear on the lower whorls, the last whorl, in large individuals, being quite destitute of ribs. V. Ihering has sent us the lower half of a large specimen of this spe- cies, which agrees well with our large individuals. We possess a number of small individuals, which agree in form. Two of them show the apex: it is distinctly of the Caricella-type of the Sca- phelloid-series (Ball, 1890, b, p. 70). (See our plate XXXVI, fig. !*•«.) One individual (plate XXXVI, fig. i') shows the spiral striae very strongly developed, even on the beginning of the last whorl. Specimens like this, when only parts of the surface of the lower whorls are preserved, which do not possess any longitudinal ribs, may have been taken by v. Ihering for V. alta. V. Ihering himself says that his V. alta and dor- bignyana are similar in form (which is really not correct, the true V. alta of Sowerby from the Navidad beds having a much shorter mouth), and that he had only very poor material of his V. alta (only one individual showing remains of the shell). Since the absence of longitudinal ribs in V. alta is the only difference between v. Ihering's V. alta and V. dor- bignyana, and since these ribs also tend to disappear on the lower part in V. dorbignyana, and, indeed, do so complefely on the last whorl, it is quite possible that v. Ihering's V. alta corresponds to such specimens of V. dorbignyana as is represented in our figure. 232 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. It is probable that the doubtful casts from Santa Cruz referred to V. alta by Sowerby also belong to this species. We do not possess the true V. alta from Santa Cruz. Record of specimens: Mouth of Santa Cruz River, 17 sp.; Upper Rio Chalia, i cast. Distribution: Santa Cruz, Patagonian formation (Phil., v. Ih.). Affinities: In general form and size this 'species corresponds closely to the living V. and/la Sol. (Reeve, 1851, pi. 17, f. 39; Tryon, 1882, p. 97, pi. 29, f. no; Lahille, 1895, p. 21, especially pi. i, f. 9, pi. 8 and n), but it differs chiefly in the presence of longitudinal ribs on the upper whorls. It is quite possible that this is the ancestral form of V. ancilla. In Australia, this species is represented by V. halli Pritchard (1896, p. 101, pi. 2, f. 1-3) from supposed Eocene, but probably Miocene beds of Tasmania and Victoria. This species has the same mamillate (Scaphel- loid-) apex, but seems to possess a shorter mouth and longer spire. 1 6O. VOLUTA DOMEYKOANA Philippi. PL XXXVII, Fig. ia'6. 1887 V. d. Philippi, Tert. & Quart. Verst. Chiles, p. 70, pi. 8, f. 4. 1899 V- pilsbryi v. Ihering, in : N. Jahrb. Miner., etc., v. 2, p. 34, pi. 2, f. 9. 1900 V. domeykoana Ortmann, in: Amer. Jour. Sci., v. 10, p. 381. Shell ventricoso-fusiform ; whorls 5 (besides the apex). Spire short, conical. Mouth considerably longer than half of the shell, about as long as three-fifths of it. Whorls convex, with a distinct shoulder, upper part concave, and appressed toward the suture. Surface with spiral striae, which become indistinct on the last whorl, and with longitudinal ribs, which form distinct nodes on the shoulder. Last whorl inflated, large. Mouth wide, elongated. Columella with tv/o plaits, and sometimes with a suggestion of a third (upper) one. Height, 149 mm; diameter, 72 mm (almost complete individual, only apex gone). Remarks : V. Ihering figures the lower part of a very large individual, and has sent to us the same part of another, large one, which both agree well with our specimens, two of which are almost of the same size. On the other hand, our fine individual figured on pi. XXXVII, fig. ia, agrees so closely with the description and figure of Philippi's V. donieykoaua, ORTMANN I TERTIARY INVERTEBRATES. 233 except for its larger size, that I do not entertain any doubt as to their specific identity. In a letter, v. Ihering maintains the difference of both, saying that the mouth is much wider in V. pilsbryi. But according to our material, the width of the mouth increases with age, and since our indi- viduals are smaller than v. Ihering's, and agree better with V. domeykoana in the size of the mouth, I believe this supposed character of y. pilsbryi is only a character of age. The development of the nodes is variable, especially on the last whorl. In one of our specimens these nodes are much less distinct (see fig. i*). In all other respects our four specimens much resemble one another, and may be recognized at once by the characteristic shape of the shell, which is, in some degree, intermediate between V. dorbignyana and y. ameghinoi. Record of specimens : Mouth of Santa Cruz River, 4 sp. Distribution: Santa Cruz, Patagonian beds (v. Ih.); Chiloe, Navidad, and Quinquina, Chili (Phil.). As to the occurrence of this species in the Cretaceous beds of the island of Quinquina, compare Philippi, 1. c., and Steinmann, 1895, p. 24. Affinities: As v. Ihering points out, this species is closely related to the following species, V. ameghinoi, and to the living V. magellanica Lmck. (see Reeve, 1851, pi. 14, f. 33; Tryon, 1882, p. 98, pi. 29, f. 107, 108; Lahille, 1895, p. 25, especially pi. i, f. 11, pi. 8), and it seems to be es- pecially close to the variety figured by Reeve in fig. 33a, which has nodes on the shoulder, and which seems to be included in Lahille's V. tubercu- lata Wood (Lahille, 1895, pi. i, f. 12, 13 and pi. 7, f. 140-145); we may take it for the ancestral form of V. tubercttlata and magellanica, and the latter would represent the same relation to V. domeykoana, as V. ancilla does to V. dorbignyana. 161. VOLUTA AMEGHINOI v. Ihering. PI. XXXVI, Fig. 2. 1896 y. a. v. Ihering, in: Nachr. deutsch. malakozool. Ges., p. 97. 1897 ^ a- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 302, textfig. 17. Shell globoso-ovate. Whorls about 3-4 (besides the apical part). Spire very short, conical. Mouth very large, over ^ of the length of the shell. Whorls convex, with a distinct shoulder, which is ornamented by a series of strong tubercles. Upper part of whorls oblique, appressed toward the 234 PATAGONIAN EXPEDITIONS 1 PAL/EONTOLOGY. suture, and covering the tubercles of the preceding whorls. Surface smooth, only with lines of growth. Last whorl large and inflated. Mouth wide. Columella with three plaits, the uppermost of which is very indistinct. Height, 74 mm (apex damaged); diameter, 56 mm. Remarks: This is distinguished from the other Patagonian species by the very short spire and the strong nodes on the last whorl. Our indi- viduals are considerably smaller than v. Ihering's (height, 156 mm; diam- eter, 100 mm), hardly half as large, and the slight differences that may be noticed between our figure and that of v. Ihering are, no doubt, differ- ences of age. The cast from Lake Pueyrredon is very poor, but the general form agrees, and there are also indications of the nodes. It cannot be united with any other species, but compares well with this one. Record of specimens : Mt. of Observation, upper horizon, 2 sp. (one of them very small); Lake Pueyrredon, 600' above base, I cast. Distribution: La Cueva, Suprapatagonian beds (v. Ih.). Affinities: V. brasiliana Sol. (Reeve, 1851, pi. 5, f. 34; Tryon, 1882, p. 98, pi. 29, f. 113; = V. colocynthis Chemn., Lahille, 1895, P- IO> pi- !» f. 3, 4, pi. 5) seems to be the descendant of K ameghinoi, as has been pointed out already by v. Ihering. V. pacifica Sol. (Zittel, 1864, p. 38, pi. 15, f. 4, and Hutton, 1873, p. 7) also resembles this species, but is more slender. It has been found from the Oamaru (Oligocene) beds to Recent times in New Zealand, and apparently V. atkinsoni Pritchard (1896, p. 100, pi. 3, f. i) comes very near to the latter. It is from the Table Cape beds of Tasmania. NOTE. — It is an extremely interesting fact that the three types of fossil Patagonian Volutce, V. dorbignyana, V. domeykoana, and V. ameghinoi are still represented on the Patagonian coast by very closely allied forms, namely: V. ancilla, V. rnagellanica, and V. brasiliana. This brings the fauna of the Patagonian beds into close relationship with the present Patagonian fauna, and makes it probable that the interval of time is not very great. The type of K triplicata is no longer represented in the Recent South American waters, at least not by very closely allied forms. But we have seen that V. triplicata is linked to this recent group through V. dorbig- nyana, so that we may say that all the known Tertiary and living Volutas from Patagonia belong to one and the same stock. ORTMANN : TERTIARY INVERTEBRATES. 235 In Australian Tertiary deposits we have representatives of the group of V. triplicata in V. serissa Tate and V. tateana Johnst. (see Tate, 1889, p. 129, pi. 2, f. i, and p. 132, pi. 2, f. 5), the first from the Miocene beds of the River Murray Cliffs, the second from the supposed Eocene (?) beds of Tasmania. Fam. CANCELLARIID^E Ad. Gen. CANCELLARIA Lmck. 162. CANCELLARIA GRACILIS v. Ihering. PI. XXXVI, Fig. 3"-*. 1897 C- g- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 310, pi. 3, f. 11. 1899 C. g. var. major v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 35, pi. 2, f. 10. Shell ovato-fusiform, elongate, not umbilicated. Whorls ^]/2 to 8. Spire acuminate. Whorls convex, near the suture, in the upper part, indistinctly angulated, suture deep. Surface with 10 or 11 longitudinal ribs, crossed by spiral cords. Mouth oval. Outer lip crenulated within. Columella with two subequal plaits. Canal short, slightly curved. Height, 12 mm; diameter, 6 mm. Remarks : Our complete specimen is a little smaller than the original one described by v. Ihering in 1897, but agrees with it completely, with the exception that it has only 45^ whorls (5^ in v. Ihering's specimen). But it seems that this species attains a very much larger size ; the individual described by v. Ihering in 1899 as var. major has 8 whorls, and is 43 mm high, but otherwise it much resembles this species, with the excep- tion that there are 4-5 smaller plaits on the columella in addition to the two larger ones ; but this may be due to age. This species is very closely allied to the following, but differs in the more elongate form, less distinctly angulated whorls, and number of longitudinal ribs. Record of specimens : Mt. of Observation, upper horizon, i sp.; San Julian, Darwin Station, i cast. Distribution: Patagonian beds of Santa Cruz, and Suprapatagonian beds of La Cueva (v. Ih.). 236 PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. 163. CANCELLARIA cf. MEDIN/E Philippi. PL XXXVI, Fig. 4«-». 1887 C. m. Philippi, Tert. & Quart. Verst. Chiles, p. 68, pi. 7, f. i. 1900 C. cf. m. Ortmann, in: Amer. Journ. Sci., v. 10, p. 379. Very closely allied to C. gracilis, but differing in the following partic- ulars: Shell broader and shorter, whorls distinctly angulated near the deep suture; number of longitudinal ribs 14-15. Plaits of columella less strongly developed. Measurements: of specimen from Mt. Observation: height, 10.5 mm, diameter, 6 mm ; specimen from Santa Cruz : height, 1 2 mm, diameter, 7.5 mm. Remarks : If our individuals really belong to C. medince, they are very young. Philippi gives: Height, 44 mm; diameter, 29 mm. But since in all other respects they agree with Philippi's description and figure, I identify them, although with some doubt, with this species. They cannot belong to Philippi's C. mdali and the other allied form mentioned by him but not described, both from Santa Cruz, since both possess an open umbilicus, which is not seen in our individuals. Record of specimens : Mouth of Santa Cruz River, 3 sp. ; Mt. of Obser- vation, upper horizon, i sp. Affinities: There are many species of the genus Cancellaria known from Tertiary deposits of the northern hemisphere that may be compared with those mentioned here, but I am unable to point out any close affinities to any particular one. Fam. TEREBRID^E Ad. Gen. TEREBRA Lmck. 164. TEREBRA COSTELLATA Sowerby. PL XXXVI, Fig. 5. 1846 T. c. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 262, pi. 4, f. 70, 71. ORTMANN : TERTIARY INVERTEBRATES. 237 1887 T. c. Philippi, Tert. & Quart. Verst. Chiles, p. 67, pi. 7, f. 3 (after Sowerby). 1897 T. c. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 311. Shell turrite, elongate. Whorls numerous, convex in the middle, in the upper part depressed, with an indistinct, depressed line running par- allel to the suture. 1 1-15 longitudinal ribs on each whorl ; these ribs are slightly curved, and extend from suture to suture, although less distinct in the upper, depressed part. Mouth elongate, with a short canal. Colu- mella smooth. Height (of fragment), 29 mm; diameter, 10 mm. Remarks: V. Ihering distinguishes two varieties: var. quemadensis, with the ribs extending from suture to suture, and more elongate mouth, and var. santacruzensis, with the ribs indistinct just below the suture, and shorter mouth. Our specimens seem to belong to the first variety, since the ribs, although less strongly pronounced on the upper part, still are distinctly discernible, and since the mouth is a little more elongate than in Sowerby's figure. Recently, v. Ihering has informed me, that the Chilian T. costellata is different from both Patagonian forms, but does not say in what char- acters. Record of specimens : Mouth of Santa Cruz River, 3 sp. ; Lake Pueyr- redon, 600' above base, i sp. (jun.). Distribution : La Cueva and Jegua quemada, Suprapatagonian beds (v. Ih.); Navidad, Chili (Sow., Phil.). Affinities: The genus Terebra is rare in Eocene deposits, but the num- ber of species increases rapidly from the Miocene upward. It is a char- acteristic tropical genus. Hoernes (1856, p. 134, pi. 11, f. 30) records this species (T. costellata] from the Miocene of the Vienna basin ; comparing his figure and descrip- tion with our individuals, I do not believe that they are identical ; in our species the ribs are much stronger, less numerous, and the whorls are more distinctly convex. Nevertheless, there is a close resemblance to this Miocene species. (The second Navidad species of Sowerby, T. imditlifera, has been identified by Hoernes, p. 130, with T. acuminata of Borson from the Miocene and Pliocene, of Europe and, indeed, I cannot see any differences between them.) 238 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. Fam. PLEUROTOMID^. Stol. Gen. PLEUROTOMA Lmck. 165. PLEUROTOMA SUB^EQUALIS Sowerby. PI. XXXVI, Fig. 6. 1846 P. s. Sowerby, in: Darwin, Geol. Observ. S. Amer., p. 257, pi. 4, f. 52. 1887 P. s. Philippi, Tert. & Quart. Verst. Chiles, p. 38, pi. i, f. 9. 1899 P. discors v. Ihering, in: N. Jahrb. Miner., etc., v. 2, p. 35 (non P. discors Sowerby). 1900 P. subcequalis Ortmann, in: Amer. Journ. Sci., v. 10, p. 381. Shell turrite, elongate, subfusiform. Whorls seven, convex, upper part (above carina) slightly concave. Upper whorls with a tuberculiferous carina in the middle and a number of spiral threads. Number of tuber- cles on the carina, in the last whorl, about 12. Last whorl large, about as long as the spire, or a little longer, with 3 prominent ribs below the carina, which are slightly and indistinctly nodulose and finer spiral threads on the canal. Mouth large, canal of medium length. Sinus of outer lip situated on the upper tuberculiferous carina. Height, 23 mm (not complete); diameter 10 mm. Remarks: Our individual is larger than Sowerby's and Philippi's fig- ures, but agrees well with them and with the descriptions, although Sowerby mentions 5 ribs on the last whorl, which is, according to Philippi, probably a mistake. Our specimen has, below the upper tuber- culiferous carina, two strong ribs, followed by a very weak one, but the latter is still stronger than the fine threads of the canal. I have no doubt that v. Ihering has made a mistake in the identifica- tion of his P. discors. What he describes, in 1897, as P. discors var. unifascialis is the following species, and what he mentions, in 1899, under the name of P. discors is apparently the present one. For he states expressly that the latter possesses three stronger ribs below the carina, a character that belongs to P. subceqtialis, while in the diagnosis of P. discors (Sowerby, 1846, p. 258), nothing of this kind is mentioned and no traces of it are seen in the figure (1. c., pi. 4, f. 54). And further, Philippi expressly states that P. discors is a Fusus, having no sinus on the outer lip, while v. Ihering's P. discors has a sinus. ORTMANN : TERTIARY INVERTEBRATES. 239 In addition we should consider the fact that we do not possess in our collection any form that might be taken for Fusus discors and so it is quite probable that this Navidad species is not found at Santa Cruz, and that the specimen referred to it by v. Ihering, although his description is very meagre, really belongs to our species, the P. subczqualis. This species grows much larger. Philippi gives: Height, 33 mm, diameter, u mm, and v. Ihering: Height, 45 to 50 mm. Record of specimens : Mouth of Santa Cruz River, i sp. Distribiition: Patagonian beds of Santa Cruz (v. Ih.); Navidad beds of Chili: Huafo Island (Sow.), Lebu, Matanzas, Navidad (Phil.). Affinities: Sowerby compares this species with an undescribed living species from South America, but the identification of the latter is impos- sible and the differences he mentions do not make the relation appear to be a very close one. Among fossil forms there is one that much resembles our species : the Miocene P. monilis Br. of Europe (see Hoernes, 1856, p. 353, pi. 38, f. 14-16) : especially the presence of a tuberculiferous carina, and a few stronger ribs below this on the last whorl are significant. 1 66. PLEUROTOMA UNIFASCIALIS v. Ihering. PI. XXXVI, Fig. 7"'". 1897 P- discors var. unifascialis v. Ihering, in: Rev. Mus. Paul., v. 2, p. 312. 1900 P. tmifascialis Ortmann, in: Am. Journ. Sci., v. 10, p. 381. Shell turrite, subfusiform. Whorls seven. Upper whorls covered with a number of fine spiral threads and with a tuberculiferous carina below the middle ; tubercles on this carina about 20 on the last whorl. Upper part of whorls slightly concave. Last whorl large, longer than the spire, sculptured as the upper whorls, but in addition there is, some distance below the carina, a single, rib-like angulation. Canal moderately long, sinus of outer lip situated on the tuberculiferous carina. Height, 17 mm; diameter, 7 mm. Remarks: The single, more strongly developed rib below the carina on the last whorl, seems to be characteristic. V. Ihering says that this is the first of the striae on the lower part of the shell, and this is apparently the case in our specimen also. But a closer investigation reveals the fact that 240 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. this stronger rib is not the first below the carina, but is preceded by a number of very fine striae, which are easily overlooked, and, indeed, in some parts of our specimen there seems to be a smooth space in their place. V. Ihering regards his specimen as a variety of P. discors. But, as has been demonstrated above, his P. discors is really the P. subcequalis of Sowerby, and, further, according to our specimens, there are other differ- ences between both, that make it advisable to separate P. unifascialis from P. subcequalis as a distinct species. The chief differences are : ( i ) The existence of only one spiral rib below the carina. (2) The position of the tuberculiform carina on the upper whorls, which, in the present species, is considerably below the middle, nearer to the lower suture, which gives to the whole shell a different appearance. (3) The number of tubercles on the carina, which is, in P. unifascialis, about 20 in the last whorl, against only 12 in P. subcequalis. Record of specimens : Mouth of Santa Cruz River, i sp. Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Gen. GENOTA Ad. 167. GENOTA CUEVENSIS v. Ihering. PI. XXXVII, Fig. 2. 1897 G- c- v- Ihering, in: Rev. Mus. Paul., v. 2, p. 313, pi. 3, f. 10. Shell subfusiform, biconical. Whorls seven, in the upper part concave, with longitudinal ribs and spiral cords. Last whorl large, higher than the spire. Longitudinal ribs about 12-16, terminating in small nodes above, these nodes forming an angulation on the last whorl ; on the upper whorls the nodes are close to the lower suture. Spiral cords unequal on the lower part of the last whorl (below the angulation), the larger ones are slightly nodulose, the nodes situated at the points of crossing with the longitudinal ribs. Mouth narrow, elongate, canal short, slightly recurved. Sinus of outer lip very slight, situated on the concave part of the whorl, near the suture. Measurements: Height, 16 mm; diameter, 8 mm (complete individ- ual). Height, 22 mm; diameter, 10 mm (apex gone). V. Ihering gives : Height, 32; diameter, 14 mm. ORTMANN I TERTIARY INVERTEBRATES. 241 Remarks: There are slight variations in sculpture. The number of longitudinal ribs varies between 12 and 17 in our individuals; v. Ihering gives 1 8 for the last whorl, but his specimen is larger than any of ours. The nodulose appearance of the upper end of the ribs is more or less dis- tinct: sometimes the ribs only end suddenly without forming a node. The spiral cords are unequal, with from i to 4 smaller cords between the stronger ones. The nodulose appearance of these cords is sometimes quite distinct, sometimes hardly indicated. Record of specimens : Mouth of Santa Cruz River, 6 sp. Distribution: Santa Cruz and La Cueva, Suprapatagonian beds (v. Ih.). Affinities: The European Miocene species P. intorta Brocchi (Hoernes, 1856, p. 331, pi. 36, f. i, 2, especially fig. i) resembles the Patagonian form so closely that it is difficult to point out the difference ; but it seems that in P. intorta the longitudinal ribs are less distinct, and their upper noduliferous termination is more strongly developed. The Pliocene form of P. intorta (Wood, 1848, p. 53, pi. 6, f. 4) differs more from our species. Gen. DRILLIA Gray. 1 68. DRILLIA SANTACRUZENSIS Ortmann. PI. XXXVII, Fig. 3">s. 1900 D. s. Ortmann, in: Amer. Journ. Sci., v. 10, p. 376. Shell turrite, subfusiform. Last whorl hardly half as long as the shell. Whorls 8, convex, but depressed and slightly concave in the upper part, near the suture. This depression forms a shallow groove, following the suture, and is sharply separated from the rest of the whorl, which is orna- mented by oblique longitudinal ribs, which end abruptly at the sutural depression. The number of these ribs is from 12 to 15. Besides, there are very fine lines of growth, but no trace of spiral sculpture. Mouth elongate, canal short. Sinus of outer lip semicircular, situated in the sutural depression, close to the suture; at the point of junction of the outer lip with the columella (in the upper angle of the mouth) there is a distinct nodulose, callous swelling. Height, 13 mm; diameter, 4.5 mm. Remarks : The most characteristic features of this species are : ( i ) The callosity of the upper end of the inner lip. (2) The sharply defined sutural depression. (3) The complete lack of spiral sculpture. 242 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Record of specimens: Mouth of Santa Cruz River, 3 sp. Affinities: Within the genus Drillia we know a few species in which a collosity of the inner lip is combined with the character of slight develop- ment of spiral sculpture, but in most of them spiral lines are present at least in the lowermost part of the shell (on the canal). A complete lack of this sculpture is given for D. perpolita Ball (1890, p. 36, pi. 2, f. 2), Pliocene and Recent, West Indies and southern United States, but the latter has no callosity. Species, which have a callosity, but have traces of spiral ribs, are, for instance, D. kceneni Speyer (1867, p. 203, pi. 22, f. 6, 7), Oligocene, Germany, D. sigmoidea Bronn, Pliocene, Europe. The most closely allied species seems to be D. limatula Conrad. In the Princeton Museum there are 7 specimens of this species from the Miocene of St. Mary's River, Maryland, which agree in the complete lack of spiral sculpture and the presence of a callosity, as well as in the presence of a sutural depression with our Patagonian fossil. The differences, however, are: (i) the number of longitudinal ribs is less (9-11); (2) the sutural depression is less distinct, the ribs not ending so abruptly ; (3) the shape of the shell is less elongate. Gen. BORSONIA Bellardi. 169. BORSONIA PATAGONICA Ortmann. PI. XXXVII, Fig. 4-*. 1900 B. p. Ortmann, in Amer. Journ. Sci., v. 10, p. 377. Shell subfusiform, biconical ; whorls, about 6 ; last whorl a little larger than half of the shell, Whorls convex, depressed in the upper part, with a slight swelling just below the suture ; depressed part smooth (except for lines of growth), the rest ornamented by 10-12 longitudinal, rib-like swellings, which are slightly tuberculiform on the upper whorls; on the last whorls they are rib-like, but less distinct, and tend to disappear toward the mouth. Besides the ribs, there are spiral cords on the lower part of the whorls, which are wanting on the depressed part, but continue on the last whorl upon the canal. Mouth elongate, canal of medium length. Outer lip with a moderately developed sinus, which is situated in the sutural depression. Columella with two plaits, of which the lower one is sometimes quite indistinct. ORTMANN : TERTIARY INVERTEBRATES. 243 Height, 19 mm; diameter, ca. 9 mm (apex gone). Height, 17 mm; diameter, ca. 9 mm (larger part of apex gone). Remarks: The spiral sculpture is in most of our specimens quite indis- tinct, worn off, only in one of them, a small one, it is well preserved. It consists of a number of spiral cords, which, toward the canal, become finer. The upper plait of the columella is stronger and longer than the lower one, the latter being in one case so indistinct as to be hardly perceptible. Record of specimens : Mouth of the Santa Cruz River, 4 sp. Affinities: The genus Borsonia is quite characteristic of Eocene deposits, being very rare in the Neogene. Our species agrees in shape, size, and sculpture very closely with the European Oligocene species : B. dehicii Nyst (see Speyer, 1867, p. 205, pi. 23, f. 34). This agreement is so close that it is hard to point out a difference ; but it seems that the spiral sculpture in B. dehicii is a little different, the cords being finer, with smaller ones intercalated. Among the Eocene species of the Paris basin, there is none that comes so near our species, although the differ- ences between the species are not very marked. Fam. ACTA^ONIDAI d'Orb. Gen. ACTION Montf. 170. ACT/EON CHILENSIS Philippi. PI. XXXVII, Fig. 50'6. 1887 A. c. Philippi, Tert. & Quart. Verst. Chiles, p. in, pi. 13, f. 16. 1899 A. c. Ortmann, in: Amer. Jour. Sci., v. 8, p. 431. Shell ovate, rather broad ; spire short, conical, about y^ of the length of the shell ; whorls 5^ , convex, the upper ones much broader than high (3 to 4 times as broad as high) ; suture distinct. Whole surface covered with fine spiral furrows, which are punctate, subequal, and separated by broad, flat intervals, crossed by extremely fine lines of growth. Mouth elongate-ovate, columella with a fold below. Height, 8.5 mm; diameter, 5 mm; Philippi gives: Height, 10 mm; diameter, 6.5 mm. Remarks: Philippi's enlarged figure appears more swollen than the outline sketch in natural size ; the latter corresponds completely to our 244 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. individual. In all other respects there is no difference between Philippi's figure and description, and our specimen, so that I am entirely satisfied that we have to deal with this species. Record of specimens : Punta Arenas, horizon II (lower Magellanian), i sp. Distribution: Navidad and Matanzas, Chili (Phil.). Affinities: Philippi and Moericke (1896, p. 596) compare this species with A. tornatilis (L.), Neogene and Recent, Europe (see Hoernes, 1856, p. 508, pi. 46, f. 24), but in my opinion, there is no closer relation to this species, A. tornatilis being much more slender and differing in sculpture. Among the Eocene species of the Paris basin there is A. turgidtts (Des- hayes) (Deshayes, 1864, p. 594, pi. 37, f. 14-16), which agrees very closely with our species in form and sculpture, and, indeed, there is hardly any difference, except the much more considerable size (15 mm) of A. turgidus. Cossmann (1897, p. 2, pi. i, f. 6, 9) has lately described another, but much smaller (height, 4.5 mm; diameter, 3 mm) species from supposed Eocene beds of South Australia, A. subscalatus, which is also closely allied to our species in form and sculpture. 171. ACTION SEMILEVIS Ortmann. PI. XXXVII, Fig. 6"-». 1900 A. s. Ortmann, in: Amer. Journ. Sci., v. 10, p. 377. Shell elongate-oval, rather slender; spire short, conical, about one- fourth of the length of the shell. Whorls 4, convex, the upper ones 2-3 times as broad as high. Suture distinct, a slight carina running close to the suture and parallel to it on the uppermost part of the whorls. Below this carina, there is an indistinct spiral groove. Below the latter, the sur- face of the shell is smooth ; but in the lower third of the last whorl there are 5-7 spiral furrows, which are rather broad, almost as broad as the flat intervals. Mouth elongate, wider below, columella with a distinct fold below. Height, 7 mm; diameter, 3.5 mm. Remarks: The smooth part of the whorls exhibits in one specimen very slight traces of spiral structure, but only close to the edge of the mouth. The furrow running close to the upper sutures appears sometimes bifid toward the mouth. • Record of specimens : Mt. of Observation, upper horizon, 3 sp. ORTMANN: TERTIARY INVERTEBRATES. 245 Affinities: In general form and sculpture this species is closely allied to A. semistriatus (Per.) (Hoernes, 1856, p. 507, pi. 46, f. 22, 23), Mio- cene, Europe. Fam. BULLIDAi Pilsbry. Gen. BULLA Klein. 172. BULLA REMONDI Philippi. PI. XXXVII, Fig. ;«.». 1887 B. r. Philippi, Tert. & Quart. Verst. Chiles, p. 109, pi. 13, f. 7. 1899 B. r. Ortmann, in: Amer. Journ. Sci., v. 8, p. 431. Shell subcylindrical. Apex deeply umbilicated, funnel-shaped, with blunt edge, whorls hidden. Whole surface covered with fine, spiral, im- pressed lines. Mouth elongate, narrow above, not produced upward, dilated below the middle. Height, 13 mm; diameter, 5 mm. Remarks : The measurements given by Philippi in the text (p. 109) do not agree with his figure, but would indicate a much thicker shell (height, 19 mm; diameter, 9 mm; rel. 2.1:1), while his figure is: Height, 21 mm; diameter, 8 mm; rel. 2.6: i, which agrees well with our specimens, although the latter are much smaller. Philippi says that there is a fold on the columella ; I do not see it, but the preservation of our specimens (filled with and imbedded in hard matrix) does not permit a closer investigation. Philippi figures no fold. B. chilensis d'Orb. (Philippi, 1887, p. 109, pi. 13, f. 23), from Port Famine (supposedly Cretaceous) somewhat resembles this species, but it is broader and shorter and the spiral lines are wanting in the lower part of the shell. B. patagonica v. Ih. comes very near in form, but the spiral lines are distinct in the lowor part of the shell and indistinct or wanting in the upper part. B. triticum Philippi (1887, p. no, pi. 13, f. 9) from Navidad has also a similar outline, but is distinguished at once by the sharp angulation on the apical part, the latter being only slightly concave, not funnel-like. Record of specimens : Punta Arenas, horizon II (lower Magellanian) ; 6 sp. 246 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Distribution : Navidad, Matanzas, Tubul and Lebu, associated with Pyrula hombroniana, said to be from the Navidad beds (Phil.). Re- corded also from the Cretaceous of Tumbez. Moericke (1896, p. 394) questions this fact. Affinities : Among the numerous Tertiary species of Btilla, it is hard to point out a particular one to which this one shows the closest affinities. It seems to me that B. striatissima Deshayes (1864, p. 636, pi. 38, f. 20-22), from the Eocene of the Paris basin might be compared with B. remondi, although the form of B. striatissima is slightly shorter, the mouth a little more produced upward and the spiral sculpture much finer. An- other species that might be compared is B. conoidea (Deshayes, p. 622, pi. 39, f. 24-26), from the Oligocene of Europe, but the latter is a little more inflated and less cylindrical and the apex is less distinctly excavated. 173. BULLA PATAGONICA V. Ihering. PI. XXXVII, Fig. 8°-'. 1897 B. p. v. Ihering, in: Rev. Mus. Paul., v. 2, p. 271, textfig. 8. Shell subcylindrical, a little narrowed above. Apex deeply umbilicated, funnel-shaped with a blunt edge ; whorls hidden. Surface with spiral striae in the lower part of the shell ; the striae are wanting in the upper part. Mouth elongate, narrow above, very slightly produced upward, di- lated below the middle. No fold on the columella. Height, 12 mm, diameter, 6 mm; another one: Height 10 mm, di- ameter, 4.5 mm. V. Ihering gives : Height, 1 1 mm, diameter, 5.5 mm. Remarks : This species is closely allied to the preceding in general form, although it seems to be a little shorter (rel. of H. : D. = 2 : i or 2.2:1). The chief difference is the development of the spiral lines, which are distinct only in the lower half or one-third of the last whorl. In most individuals there is no trace of the spiral lines in the upper part, but in some of them, especially young ones, there is a slight indication of these lines, but they are always much less distinct than in the lower part. Sometimes there are a few of these lines (1-3), more distinct from one another, near the upper extremity. The striae (impressed lines) are more remote from one another than in B. remondi. Record of specimens: Mouth of Santa Cruz River, 109 sp; Arroyo Gio, 2 casts. ORTMANN : TERTIARY INVERTEBRATES. 247 Distribution: Jegua quemada, Suprapatagonian beds (v. Ih.). Affinities: In external form, this species might be compared with the same as the foregoing species. In the peculiar development of the spiral sculpture it differs, however, and in the latter respect it recalls the Euro- pean Eocene B. glaphyra Deshayes (1864, p. 639, pi. 39, f. 6-18). The external form of the latter is less cylindrical, and more ovoid. CRUSTACEA. ClRRIPEDIA. Fam. LEPADID^E Darw. Gen. SCALPELLUM Leach. 174. SCALPELLUM JULIENSE Ortmann. PI. XXXVII, Fig. 9"-'. 1900 S.j. Ortmann, in: Amer. Journ. Sci., v. 10, p. 377. Only the carina known. Carina narrow, elongate, strong and solid, curved ; basal margin bluntly pointed ; surface smooth, but with lines of growth. Tectum strongly arched in its upper part, only slightly so in its lower; upper part solid, its cross section almost quadrangular; this form is brought about by the presence of a prominent ridge on the concave side, formed by the junc- tion of the inflected parietes on each side. Parietes very narrow, separated from the tectum by distinct, but blunt ridges. Length, 40 mm ; width, 8 mm. Remarks: The only carina at hand resembles so closely that of S. solid- ulum Steenstrup (see Darwin, 1851, p. 42, pi. i, f. 8) from the Upper Cretaceous of Scania, that I have no doubt that it belongs to the genus Scalpelhim as well as that species. S. solidiilum differs from all other species of the genus in the solid, almost quadrangular, section of the upper part of the carina, brought about by the peculiar conformation of the "parietes," and this very character is exhibited in our fossil, as may be seen at once on comparing our figure 9' on plate XXXVII with Dar- win's figure 8C. Our species differs from S. solidulum in the absence of 248 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. radial ribs on the tectum, and the distinct separation of the .parietes from the tectum by a blunt ridge. Record of specimens : San Julian, Darwin Station, i carina. Affinities : The affinities to the Tipper Cretaceous S. soliduhim have been explained above. Fam. VERRUCID^. Darw. Gen. VERRUCA Schum. 175. VERRUCA LEVIGATA Sowerby. PL XXXVIII, Fig. I"". 1854 V. 1. Darwin, Mon. Cirrip. Balan., p. 520, pi. 21, f. 3. 1900 V. I. Ortmann, in: Arner. Journ. Sci., v. 10, p. 379. Walls of the shell smooth. Movable scutum with the lower articular ridge broader than the short upper articular ridge; movable tergum broader than high, with the upper articular ridge produced into a point. Diameter of largest individual, 5.5 mm. Remarks: I cannot find any difference between our fossil form and the living V. l&vigata. The walls of the shell are smooth, with concentric lines of growth, without radiating ribs. The movable scutum has the lower articular ridge broad (decidedly broader than in y. strcemia), and the upper articular ridge short. The movable tergum ( T] is broad. In our figure \c on plate XXXVIII, the scutum appears to be narrower than in fig. 3 of Darwin, but one must bear in mind, that our figure represents only the exposed parts of scutum and tergum ; of the occludent margin (the convex one) a part is covered by the occludent margin of the fixed scutum, and is not shown. The same is the case with the basi-carinal angle of the tergum. Of the three complete specimens, two have the left hand scutum and tergum fixed, and the third is reversed, i. e., with the right hand scutum and tergum fixed (compare Darwin's figure ia and \d on pi. 21). Darwin says that the latter case (left opercular valves movable) appears to be more common in the recent form. The reversed specimen is the largest. The figure i* was drawn from a combination of the two others, each 4 mm in diameter. The fourth in- dividual, with the movable opercular valves missing, is still smaller. ORTMANN : TERTIARY INVERTEBRATES. 249 Record of specimens: Upper Rio Chalia, 4 sp. (on a specimen of Tnr- ritella ambulacrum}. Distribution: Recent: Tierra del Fuego, Eastern Patagonian, Chili, Peru (Darw.). Fam. BALANID& Darw. Gen. BALANUS List. 176. BALANUS cf. PSITTACUS (Molina). PI. XXXVIII, Fig. 2. 1854 B. p. Darwin, Mon. Cirrip. Balan., p. 206, pi. 2, f. 3. 1887 B. p. var. minor Philippi, Tert. Quart. Verst. Chiles, p. 223, pi. 51, f. 4, 5- 1896 B. p. var. minor Moericke, in: N. Jahrb. Miner., etc., Beil. Bd. 10, P- 59i- 1897 B. p. Weltner, in: Arch. f. Naturg., v. i, p. 261. "... Orifice hexagonal. Scutum with the articular ridge very small, confluent with the very prominent adductor ridge, forming a tubular cavity, which extends up to the apex of the valve. Tergum with apex produced, needle-like, purple ; spur placed at less than its own width from the basi-scutal angle." (Darwin.) Remarks: From the base of the Patagonian beds at Lake Pueyrredon we possess two specimens of a large Balanus, which seems to be dis- tinct from B. varians. The one measures : Height, 52 mm, diameter, at base, 40 mm; the other: height, 48 mm, diameter, 51 mm. Both are poorly preserved, and do not possess opercular valves. The parietes are porous, and the orifice is subhexagonal, which would correspond, together with the large size, to B. psittacus. This view is supported in some degree by the fact that this species has been recorded and figured by Philippi from the Navidad beds of Chili, and our specimens agree closely with the figure 4 of Philippi. Philippi figures also (fig. 5) the scutum, which corresponds to that of this species (Darwin, pi. 2, f. 3. Ihering. Following Ameghino, who furnished the material, v. Ihering distinguishes two marine horizons, the Patagonian and Supra- patagonian (or marine Santacruzian) beds, a division, which is not ac- cepted by us. For the present, however, we shall give the list of species added to these faunas, as v. Ihering has given it. The following are the so-called Patagonian species added by him : 1. Rhynchonella plicigera. 5. Pecten prcenuncius. 2. Magellania globosa (M. lenticularis). 6. Mytilus cf, chorus. 3. Bouchardia zitteli. 7. Venus volckmanni. 4. Perna quadrisulcata. 8. Struthiolaria ameghinoi. All these 8 species are represented in our collections. A few more species given by v. Ihering are to be dropped as synonyms (so Cucullcsa dalli and multicostata = C. alta; Turritella argentina and steinmanni = T. ambulacrum; Trophon laciniatus var. santacruzensis = T. patagonicus}. Pecten patagonensis and Siphonalia cf. nodosa are very doubtful, and Tro- phon varians has been dropped by v. Ihering himself in 1899. The following are the new Suprapatagonian species : 1. Schizaster ameghinoi. 21. Eulima subventricosa. 2. Area patagonica. 22. Odostomia suturalis (Odontostomia s.). 3. Pectunculus puhinatus cuevensis (Glyci- 23. Turbonilla cuevensis. meris ibari). 24. Trochita magellanica (Jnfundibulum cly- 4. Pecten centralis (P. proximus). peoluni), 5. Modiola ameghinoi. 25. Natica consimilis. 6. Crassatella longior (Crassatellites /.). 26. Natica subtenuis. 7. Amathusia angulata. 27. Triton dautzenbergi. 8. Venus striatolamellata ( V. navidadis). 28. Trophon pyriformis (Urosalpinx p.~). 9. Cytherea splendida (Meretrix iheringf). 29. Trophon leucostomoides (Urosalpinx l^). 10. Dosinia meridionalis. 30. Marginella quemadensis. 1 1 . Tellina perplana. 3 1 . Marginella confinis. 12. Tellina patagonica. 32. Marginella gracilior. 13. Tellina jeguaensis. 33. Marginella plicifera. 14. Mactra indistincta. 34. Valuta ameghinoi. 15. Solen elytron. 35. Valuta patagonica. 1 6. Glycimeris quemadensis (Panopea q.). 36. Cancel/aria ameghinoi. 17. Pholas paucispina. 37. Cancellaria gracilis. 1 8. Dentalium octocostatum (D. octocostellmn). 38. Pleurotoma discors (P. unifascialis). 19. Fissurella sp. 39. enota Gcuevensis. 20. Gibbula dalli. 40. Bulla,patagonica. ORTMANN : TERTIARY INVERTEBRATES. 263 The Trochita mentioned by Philippi is identified as T. corrugata, and the Terebra, given as undulifera by Rochebrune and Mabille, as T. costellata. Other species mentioned by v. Ihering as new, have been recognized as synonyms of known forms ; they are : Pecten quemadensis, Cucullaria tri- dentata, Nucula tricesima, Gibbula fracta, T^lrritella tricincta, Valuta phi l- ippiana and quemadensis. Thus v. Ihering adds to what he calls Patagonian and Suprapatagonian fauna: 48 species, which brings up the total number of species known from these beds to 95. Of the 48 species, 36 are represented in our collection. In a second contribution, in 1899, v. Ihering added the following spe- cies, all from the typical Patagonian beds : 1. Pinna semicos tails var. magellanica. 7. Tritonium bicegoi. 2. Crassatella kokeni (Crassatellites /£.). 8. Siphonalia dilatata (S. domeykoana). 3. Lucina ortmanni. 9. Valuta triplicata. 4. Tellina santacmzensis. i o. Valuta pilsbryi ( V. domeykoana). 5. Tellina tehuelcha. 11. Pleurotoma discors (P. subcequalis). 6. Glycimeris nucleus (Panopea ».). Of these 1 1 species, 8 are found in our collection. Venus cf. uncinata, Pyrula hombroniana are extremely doubtful, while Pecten fissicostalis is a synonym of P. geminatus. This contribution brings up the number of well-known Patagonian species to 106. Finally Cossmann, in 1899, describes the following "Suprapatagonian" species : 1. Lcptothyra philippii. 8. Trichotropis patagonica. 2. Solar ie I la dautzenbergi. 9. Fossarus pillula. 3. Calliostoma pararatum. 10. Odontostomia euryope. 4. Calliostoma santacruzense. 1 1 . Triton obliteratus. 5. Gibbula diametralis. 12. Urosalpinx leucostomoides (U. cossmannf). 6. Gibbula iheringi. 13. Peratotoma ihcringi. j. Gibbula margaritoides. He gives to the Fissurella already mentioned by v. Ihering the name of F. eurytreta. Odontostomia synarthrota and Turbonilla iheringi are synonyms. Of these 13 species, onty 6 are represented in our collection. 264 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. This adds 13 species, giving the total number of the Patagonian fossils known up to this date as ng. Comparing our list of species (see pp. 257, 258) with that of the species reported previously from the Patagonian beds, we see that of the 119 species known, 95 are represented in our collection, while 56 are added, our list containing altogether 151 species. The following is the list of 24 species, reported previously, which are more or less well established, but not represented in our collection : 1. Leda glabra (Sow.), 1846, and v. Ih., 12. Gibbula margaritoides Cossm., 1899. 1897. 13. Eulima subventricosa v. Ih., 1897. 2. Pinna semicostata var. magellanica v. Ih., 14. Odontostomia euryope Cossm., 1899. 1899. 15. Fossarus pillula Cossm., 1899. 3. Venus patagonica Phil., 1887, and v. Ih., 16. Trichotropis patagonica Cossm., 1899. 1899. 17. Tritonium dautzenbergi v. Ih., 1897. 4. Tellina perplana v. Ih., 1897. 18. Tritonium obliteratum Cossm., 1899. 5. Tellina patagonica v. Ih., 1897. 19. Marginella quemadensis v. Ih., 1897. 6. Tellina santacruzensis v. Ih., 1899. 20. Marginella confinis v. Ih., 1897. 7. Solen elytron Phil., v. Ih., 1897. 21. Valuta patagonica v. Ih., 1897. 8. Mactra indistinc to, v. Ih., 1897. 22. Cancellaria vidali Phil., 1887. 9. Panopea nucleus (v. Ih.), 1899. 23. Cancellaria ameghinoi v. Ih., 1897. 10. Pholas paucispina v. Ih., 1897. 24. Peratotoma iheringi Cossm., 1899. 11. Gibbula iheringi Cossm., 1899. This would bring up the total number of well established Patagonian species to 775, to which we have to add the following doubtful forms : 1. Pecten nodosoplicatus v. Ih., 1897. — Too incompletely known. 2. Fimbria {t} patagonica Phil., 1887. — Too incompletely known. 3. Venus cf. uncinata Phil., v. Ih., 1899. — Casts only known. 4. Mactra (?) rugata Sow., 1846. — Too incompletely known, and never found again. 5. Plwtinula detecta Roch. & Mab., 1885. — Description unsatisfactory. 6. PJwtimda resurrecta Roch. & Mab., 1889. — Description unsatisfactory. 7. Natica omoia Roch. and Mab., 1885. — Perhaps identical with one of the other Naticce. 8. Turritella elachista Roch. & Mab., 1889. — Description unsatisfactory. 9. Pyrula cf. hombroniana Phil., v. Ih., 1899. — Material very poor. 10. Fusus cf. nodosus Mart, v. Ih., 1897. — Cast only. 11. Trophon leucostomoides Sow., v. Ih., 1897. — Identification beyond control. 12. Valuta alta Sow., 1846, and v. Ih., 1899. — Identification probably incorrect. 13. Terebra undulifera Roch. & Mab., 1899. — Perhaps = T. costcllata. Other species, especially some of those given by d'Orbigny, Sowerby, and in Philippi's list (1887, p. 251) do not belong to the Patagonian for- ORTMANN : TERTIARY INVERTEBRATES. 265 mation, but to much younger deposits. I mention here the following, in which this fact has been demonstrated : 1. Area bonplandiana d'Orb. Parana formation, see v. Ih., 1897, p. 329. 2. Ostrea patagonica d'Orb. (and Phil.). Tehuelche formation, see v. Ih., p. 329, and our collections, p. 112. 3. Ostrea ferrarisi d'Orb. Synonym of 0. patagonica. 4. Pecten dandnianus d'Orb. (and Sow.). Parana formation, see v. Ihering, p. 329. 5. Pecten paranensis d'Orb. (and Sow.). Tehuelche and Parana formations, see v. Ihering, pp. 226 and 328. 6. Pecten actinodes Sow. Tehuelche formation, v. Ih., and our collection, p. 119. 7. Cardium platense d'Orb. Parana formation, see v. Ihering, p. 330. 8. Venus muensteri d'Orb. (and Phil.). Tehuelche and Parana formation, see v. Ihering, pp. 328 and 330. THE IDENTITY OF PATAGONIAN AND SUPRA- PATAGONIAN BEDS.1 i. THE TYPE-LOCALITY AT SANTA CRUZ. Ameghino (1898 and 1899), and, following him, v. Ihering distinguish two marine deposits underlying the non-marine Santacruzian beds (con- taining Mammalian remains) : the lower division is called by the term 1 The general results to which the following detailed investigations lead have been published by Mr. Hatcher and the present writer in preliminary notes (J. B. Hatcher, 1900 a, Ortmann, 1900). Although it was distinctly stated that the observed facts which furnish the base for our conclusions would be given in the present final report, F. Ameghino did not deem it necessary to wait for its publication, but, from the start, rejected them altogether as having no value at all (see, for instance, F. Ameghino, Notices preliminaires sur des Ongules nouveaux des terrains Cretaces de Patagonie, in : Boletin de la Academia Nacional de Ciencias de Cordoba, vol. 16, 1901, p. 349). This is the more astonishing and throws a very peculiar light upon the character of Ameg- hino's work, since he must have been aware of the fact that Mr. Hatcher, as well as the present writer, tried faithfully, at the beginning, to verify his views on the geology of Patagonia, and that it was during the progress of their studies that they ventured to express opinions contrary to those of Ameghino (compare Hatcher, 1897, pp. 334-338, and 1900 a, p. 99 ff. ; Ortmann, 1898, p. 482, and 1899, p. 432). Of course, then it was impossible to give the bulk of the evi- dence supporting our views, but the mere fact that we arrived at our conclusions by slow steps ought to have been sufficient proof to Ameghino that they were not hastily formed, and in this connection it is well to remember that Ameghino himself has never published any geological or stratigraphical facts in the form of sections or the like. 266 PATAGONIAN EXPEDITIONS '. PALAEONTOLOGY. "Patagoniany^r;//c?//b;/," and Ameghino subdivides this into a lower part, " Piso Juliense," and an upper part, " Piso Leonense." The upper division he calls by the name of "Suprapatagonian " beds, and separates it from the Patagonian formation, making it a subdivision, "Piso Suprapatagonico," of the Santacruzian formation. V. Ihering follows him in this arrange- ment, and gives "Santacruzian formation" as the horizon for all marine "Suprapatagonian" fossils. (It is well to bear in mind that v. Ihering's "Santacruzian" means the same as Ameghino's "Suprapatagonian.") As Hatcher (1900 a, p. 99, ff., and i9Oob, p. 263, ff.) has repeatedly indi- cated in his preliminary publications, he has failed entirely to verify these subdivisions of the marine beds underlying the typical (non-marine) San- tacruzian beds, and he was unable to find any stratigraphical evidence for them, and I shall endeavor here to show that the palczontological material collected also fails to warrant any subdivisions of this marine series. In order to demonstrate this, we must choose the type-locality at the mouth of the Santa Cruz river and its fauna as a starting point. Accord- ing to Hatcher (1900 b), the deposits at this locality consist of several hundred feet of soft sand and clays, often filled with hard concretions, and interrupted occasionally by a stratum of hard sandstone. The fossils are found both in the loose material and in the harder concretions. The following is the list of all fossils recorded from this locality : 1 . Cidaris antarctica. 2. Schiz aster ameghinoi. 3. Aspidostoma giganteum. 4. Reticulipora patagonica. 5. Tennysonia subcylindrica. 6. Rhynchonella plicigcra (v. Ih.). 7. Terebratella dorsata var. 8. Terebratella patagonica. 9. Nucnla patagonica. 10. Nucula reticularis. 11. Leda glabra (Sow., v. Ih.). 1 2. Leda oxyrhyncha. 13. Leda errazurizi. 14. Malletia ornata. 15. Cucullcea a/ fa. 1 6. Cucullcea darwini. 17. Limopsis insolita. 1 8. Area patagonica. 19. Glycimeris ibari. 20. Pcrna quadrisulcata. 2 1 . Ostrea ingens. 22. Pec ten proximus. 23. Pecten prcemmcius. 24. Pecten geminatns. 25. Pinna semi'costata (v. Ih.). 26. Mytilus cf. chorus (v. Ih.). 27. Crassatellites lyelli. 28. Crassatellites kokeni. 29. Crassatellites quartns. 30. Cardita elegantoides. 3 1 . Cardita inaqualis. 3 2 . Cardita patagonica. 3 3 . Litcina promaitcana. 34. Lvcina ortmanni. 3 5 . Cardium philippii. 36. Cardium puclclnnn. 37. Cardium pisinii. 38. Amathusia angnlata. ORTMANN I TERTIARY INVERTEBRATES. 267 39. Venus mcridionalis. 40. Venus volckmanni. 41. Venus patagonica (Phil., v. Ih.). 42. Venus darwini. 43. Venus navidadis. 44. Dosinia meridionalis. 45. Dosinia l&viuscula. 46. Tellina tehuelcha (v. Ih.). 47. Tellina jeguaensis. 48. Tellina santacrusensis (v. Ih.). 49. Psammobia patagonica. 50. Mactra darwini. 5 1 . Corbula hatcheri. 52. Panopea regularis. 53. Panopea quemadensis. 54. Panopea nucleus (v. Ih.). 5 5 • Martesia patagonica. 56. Martesia pumila. 57. Dentalium sulcosum. 58. Liotia scotti. 59. Leptothyra philippii. 60. Solariella dautzenbergi. 6 1 . Calliostoma peraratum. 62. Calliostoma cossmanni. 63. Calliostoma santacruzense. 64. Calliostoma garretti. 65. Calliostoma iheringi. 66. Gibbula l&vis. 67. Gibbula dalli. 68. Gibbula diametralis. 69. Odontostomia suturalis. 70. Scalaria rugulosa. 71. Infundibulum corrugatum. 72. Infundibulum clypeolum. 73. Sigapatella americana. 74. Crepidida gregaria. 75. Natica ovoidea. 76. Natica secunda. 77. Natica darwini. 78. Natica subtenuis. 79. Natica consimilis. 80. Turritella ambulacrum. 8 1 . Turritella breantiana. 82. Turritella patagonica. 83. Aporrhais araucana. 84. Struthiolaria ameghinoi. 85. Struthiolaria ornata. 86. Dolium ovulum. 87. Pyrola Carolina. 88. Tritonium bicegoi. 89. Tritonium morgani. 90. Buccimim anna. 91. Buccinum obesum. 92. Chrysodomus cancellatus. 9 3 . Chrysodomus pilsbryi. 94. Siplwnalia domeykoana. 95. Trophon patagonicus. 96. Urosalpinx elegans. 97. Urosalpinx cossmanni. 98. /*««« torosus. 99. Marginella gracilior. 100. Marginella plicifera. 101. Marginella olivella. 1 02. Valuta triplicata. 103. Valuta gracilior. 104. Valuta petersoni. 105. Valuta dorbignyana. 1 06. Valuta domeykoana. 107. Cancellaria gracilis. 1 08. Cancellaria cf. medince. 109. Cancellaria -vidali (Phil.). 1 10. Terebra costellata. in. Pleurotoma subcequalis, 112. Pleurotoma unifayialis. 113. Genota cuevensis. 1 1 4. Drillia santacnizensis. 115. Borsonia patagonica. 1 1 6. Bulla patagonica. 1 1 7. Geryon peruvianus. The question arises, which one of Ameghino's subdivisions is repre- sented in this fauna. According to Mr. Hatcher, this fauna is a unit stratigraphically, and no subdivisions are possible; he tried at first, in collecting, to distinguish different horizons, but soon found that this was 268 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. impossible. That this is not due to defective observation, will be shown further on, and it will be demonstrated, that the palaeontological charac- ters given by Ameghino do not hold good in the face of a careful com- parison with our material. Ameghino (1899) gives the following characteristic fossils for his sub- divisions. Piso Juliense : Hypechinus patagonensis. Rhynchonella plicigera. Echinarachnius juliensis ( = Scutella pata- Bouchardia zitteli. gonensis). Pecten geminatus. Schizastcr ameghinoi. Pecten prammcius. Terebratula patagonica (=• Terebratella p.). Siphonalia noachina. Piso Leonense : Ostrea percrassa (= 0. ingens). Turritella argentina (= T. ambulacrum). Perna quadrisulcata. Struthiolaria ornata. Cucullaa alta. Piso Siiprapatagonico : Ostrea patagonica (= 0. ingens). Amathusia angulata. Pecten quemadensis (= P. geminatus). Dentalium octocostatum (= D. octocostellatuni). Cytherea splendida (= Meretrix iheringt). Valuta ameghinoi. It is most important to know, where we are to look for the type-locali- ties of these divisions. In this respect Ameghino is remarkably careless ; indeed, he does not say in any case, what we are to take for typical rep- resentations of his divisions, and thus we have to guess, or rather to infer from the names given, which one is the locality intended in each case. The Piso Juliense is apparently called after San Julian, and this we must take for the type-locality of it (see below). The Piso Leonense has been named apparently from Mt. Leon at the mouth of the Santa Cruz River, and thus those beds, of which the list of fossils is given above, are to be taken as representation of this division. And indeed, all five species, mentioned by Ameghino as characteristic of this subdivision, have been found at the mouth of the Santa Cruz River, and further, since two of them, Cucullcea alta and Struthiolaria ornata, have been found nowhere ORTMANN : TERTIARY INVERTEBRATES. 269 else, it is beyond doubt fhat we must take the mouth of the Santa Cruz River as the type-locality for the Leonense beds.* As regards the Suprapatagonian beds, Ameghino does not take the trouble to give the slightest hint as to a type-locality. Species coming from this subdivision are given by v. Ihering chiefly from two places called "La Cueva" and "Jegua quemada." The geographical position of these places is not given, and Mr. Hatcher — although trying to do so — was not able to locate them, when he was at Santa Cruz. Thus the only way left is to try to identify the Suprapatagonian beds according to the characteristic fossils given, but — as we shall see below — this has proved to be a complete failure. V. Ihering recognized the importance of the establishment of a type- locality, and the careful registering of the fossils found there. He sent his collector, Mr. Bicego, to Santa Cruz, and tried to get as many fossils as possible from this old type-locality of Patagonian beds. The result of these investigations was published in the paper of 1899. V. Ihering here gives a list of the fossils found at Santa Cruz, and by comparing them with those recorded by him — on the authority of Ameghino — from the Suprapatagonian beds, gives (1. c., p. 38) a list of the characteristic fossils from both the Patagonian and Suprapatagonian beds. According to him, the following species found at Santa Cruz must be taken as characteristic of the Patagonian formation, since they never have been found in what Ameghino calls Suprapatagonian beds) :2 Cucullcza alta. Pecten fissicostalis (= P. geminatus). Cardita patagonica (= C. ituequalis). Lucina ortmanni. Ca.rd.ium puelchum. Venus patagonica. Dosinia Icevitiscula. Struthiolaria ornata. Siphonalia dilatata (= S. domeykoana\ For the Suprapatagonian beds he gives the following characteristic fos- sils, which — according to him — have never been found at Santa Cruz : 1 The manner in which Ameghino distorts facts in order to suit his preconceived theories is simply astonishing. In 1899 (p. 12) he says — in discussing the section of Punta Arenas discov- ered by Hatcher and described by the present writer — that the fauna of his Piso Leonense is known only in small part, while, as has been demonstrated above, this fauna must be identical with that of the type-locality of the whole of the Patagonian beds at Santa Cruz, and this fauna, indeed, is the best knmvn part of the Patagonian fauna. 2 1 omit Siphonalia noachina, since this has never been found at Santa Cruz, and also Valuta alta, since this species is altogether doubtful (see p. 231). 270 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. Cucullaria tridcntata (= Cucullaa darwini). Natica solida (= N. darwini). Area patagonica. Struthiolaris ameghinai. Pecten centralis (= P. prciimus). Marginella quemadensis. Lucina promaucana. Marginella confinis. Amathusia angulata. Marginella gracilior. Dosinia meridionalis. Marginella plicifera. Scalaria rugulosa. Valuta ameghinoi. Of the 1 1 7 species found at the type-locality of Santa Cruz (see our list, p. 266 f.) the following are mentioned in v. Ihering's list of character- istic Patagonian fossils : Cucullcea alta. Lucina ortmanni. Dosinia Iceviuscula. Pecten geminatus. Cardium puelchum. Strnthiolaria ornata. Cardita incequalis. Venus patagonica. Siphonalia domeykoana. These are nine species (all of those mentioned by v. Ihering), eight of which also occur in our collections, only Venus patagonica not being rep- resented. On the other hand, of v. Ihering's characteristic Suprapatagonian fossils the following have been found at Santa Cruz, according to our collection : Cucullcea darwini. Amathusia angulata. Struthiolaria ameghinoi. Area patagonica. Dosinia meridionalis. Marginella gracilior. Pecten proximus. Scalaria rugulosa. Marginella plicifera. Lucina promaucana. Natica darwini. That is to say, all the characteristic Suprapatagonian species, with only three exceptions (Marginella quemadensis and confinis , Valuta ameghinoi'] have been found here. This result shows at a glance, that v. Ihering's list of characteristic fos- sils for both supposed formations does not hold good. Comparing Ameghino's characteristic fossils with our list of the Santa Cruz fauna we find the following species represented there : Jtdiense species : 4 out of 9. Schisaster ameghinoi. Terebratella patagonica. Pecten prcenunchts. Pecten geminatus. ORTMANN : TERTIARY INVERTEBRATES. 271 Leonense species: Ostrea ingetis. Perna quadrisulcata. Cttcull&a a/fa. Turritella ambulacrum. Struthiolaria ornata. all of them. Suprapatagonian species : Ostrea ingens. \ Pec ten geminatus. \ 3 out of 6. Amathusia angulata. This gives about the same result, although the fact is remarkable that of the Leonense species (5) all are found at Santa Cruz, of the others only a part. Thus we see, that — taking together all the forms mentioned by both, Ameghino and v. Ihering — there are, at the type-locality of the Pata- gonian beds, 75* Patagonian species, namely : Schizaster ameghinoi. Pecten prcenuncius. Venus patagonica. Terebratclla patagonica. Pecten geminatus. Dosinia l&viuscula. Cucullcea a/fa. Cardita incequalis. Turritella ambulacrum. Perna quadrisulcata. Lucina ortmanni. Struthiolaria ornata. Ostrea ingens. Cardium puelchum. Siphonalia domcykoana. At the same time, ij Suprapatagonian species are found there, namely Cucullcea darwini. Lucina promaucana. Struthiolaria amegltinoi. Area patagonica. Amathusia angulata. Marginella gracilis. Ostrea ingens. Dosinia mcridionalis. Marginella plicifera. Pecten proximus. Scalaria rugulosa. Pecten geminatus. Natica darwini. These facts would make the locality at Santa Cruz as well Patagonian as Suprapatagonian, and suggests strongly the identity of both. On the other hand, we have seen that the so-called "Leonense" species of Ameghino are all (5) found at Santa Cruz, while only a part of the "Juliense" (4 out of 9) have been found there. Of the remaining 5 Juliense species, however, 3 have been found at San Julian (Hypechinus Patagonensis, Scutella patagonica, Siphonalia noachina] by Mr. Hatcher. We shall discuss this fact later. 272 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. 2. COMPARISON OF OTHER LOCALITIES WITH THE TYPE FAUNA AT SANTA CRUZ. (In the following P means characteristic Patagonian, according to v. Ihering ; S means characteristic Suprapatagonian, according to Ameghino or v. Ihering ; / means characteristic Juliense, according to Ameghino ; L means characteristic Leonense, according to Ameghino.) Pescadores, Rio Santa Cruz ; ca. 50' above high tide. Only Ostrea ingens has been found here. Paso del Rio Santa Cruz : 2 miles above Las Salinas (see below), at about high tide level. J Schizaster ameghinoi. L & S Ostrea ingens. Turritella breantiana. J Terebratella patagonica. S Lucina promaucana. Turritella patagonica. Malletia ornata. L Turritella ambulacrum. L Stmthiolaria ornata. All of these 9 species are found at Santa Cruz, and 5 of them are Pata- gonian, 2 of which are Juliense, and 3 Leonense, while 2 are Suprapata- gonian. Since this locality represents a very low horizon, close to the water's edge of the Santa Cruz River and the sea, the presence of Leon- ense and Suprapatagonian species is very significant. Las Salinas ; 30 miles above mouth of Santa Cruz River, ca. 200' above high tide. J Rhynchonclla plicigera. P Cardium puelchum. L Turritella ambulacrum. S Cucullaa danvini. Venus meridionalis. . Turritella breantiana. Limopsis insolita, Psammobia patagonica. L Stmthiolaria ornata. P Cardita intequalis. Marlesia patagonica. P Siphonalia domeykoana. All of these 12 species were found at Santa Cruz. This locality is at a much higher level than Paso del Rio Santa Cruz, but has 3 fossils in common with it ( Turritella ambulacrum and breantiana, Stmthiolaria or- nata], two of which are Leonense according to Ameghino. 6 specimens are Patagonian, i of which is Juliense, while 2 are Leonense. Ctictillcea darwini is a Suprapatagonian fossil according to v. Ihering. Thus we have here a mixture of fossils from all three subdivisions. Mount of Observation, lower horizon ; from between tides to 25' above tides. ORTMANN : TERTIARY INVERTEBRATES. 273 L & S Ostrca ingens. L Turritella ambulacrum. Geryon pennianus. L Cucullcea alta. Turritella breantiana. The prevalence of Leonense species is opposed to the low level to which this locality belongs, especially when we come to compare it with the upper horizon at the same locality. All 5 species are found at Santa Cruz. Mount of Observation, upper horizon ; 25-150' above tides. / * Terebratclla patagonica. * Venus meridionalis. S *Scalaria nigulosa. *Nucula rcticularis. *Mactra darwini. * Infundibulum corrugatum. S *Arca patagonica. Mactra garretti. J SipJionalia noacliina. L & S * Ostrea ingens. * Corbitla hatcheri. S Valuta ameghinoi. Modiola ameghinoi. *Martesia patagonica. * Cancellaria gracilis. * Cardita elegantoides. S Dcntalium octocostcllatum. * Cancellaria cf. medina. P * Cardita incequalis. Calliostoma observationis. Action semilavis. P * Cardium piiclchum. Turbonilla cuevensis. Balanus varians. Of these 24 species, 15 are found at Santa Cruz (those marked*). This horizon would correspond to about the middle of the series at Santa Cruz; nevertheless it contains 2 distinctly Juliense, and no less than 5 Suprapat- agonian species, while Leonense species are hardly represented. On the other hand, 9 species are found here, which have not been found at Santa Cruz. These, however, cannot be taken as representing a "Suprapatagonian " fauna. 6 of them are either new species, or have not been found elsewhere ; of the rest 2 are found at San Julian (see be- low) at a much lower level (Siphonalia noachina and Balanus vartans), and V. ameghinoi has been found at Lake Pueyrredon, 600' above base of Tertiary. That this latter locality cannot be regarded as Suprapata- gonian will be demonstrated below. According to stratigraphical evidence, we are to expect that this local- ity should belong to the upper Juliense or Leonense division of the Pata- gonian beds : instead of that, Suprapatagonian species prevail, while a few Juliense are intermingled with them. Thus it is impossible to say, to which of Ameghino's and v. Ihering's subdivisions the beds of this local- ity are referable. All the foregoing localities resemble more or less in the state of pres- ervation of the fossils and in the matrix the type-locality at Santa Cruz, and, indeed, geographically, they are not far distant from it. The first 274 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. locality that differs in the character of the deposits is the next, near San Julian. San Julian, Oven Point; near the water's edge, not over 10' above high tide. This horizon is distinctly lower than any of the beds at Santa Cruz. Facies : sandy, with shell-fragments. * Cidaris antarctica. Heteropora pelliculata. * Turritella patagonica. Toxopneustes precursor. J * Terebratella patagonica. S * Struthiolaria ameghinoi. J Scutella patagonica. L & S * Ostrea ingens. J Siphonalia noachina. J *Schisaster ameghinoi. J & S Pec ten geminatus. * Trophon patagonicus. Serpula patagonica. Mytilus magellanicus. * Valuta gradlior. Terebella magna. * Gibbula l&vis. Balanus varians. * Aspidostoma giganteum. * Infundibulum corrugatum. Although this locality represents the lowermost horizon of the whole series known on the coast of Patagonia, we have here no less than 3 species, which are Suprapatagonian, according to Ameghino and v. Iher- ing. For the rest, Juliense species prevail. Of the 20 species, 12 (marked *) have been found at Santa Cruz. San Julian, Darwin Station; at a higher horizon than Oven Point. Matrix resembling that of Santa Cruz. This horizon seems to be about the same as the lowermost part of the Santa Cruz section. * Cidaris antarctica. *Panopea quemadensu. Murex hatcheri. J Hypechinus patagonemis. *Dentalium sulcosum. * Trophon patagonicus. Serpula patagonica. S * Sea/aria rugulosa. Fusus archimedis. J * Terebratella patagonica. *Natica ovoidea. * Valuta gradlior. Heteropora pelliculata. * Turritella sp. * Cancellaria gracilis. J *Pecten pranuncius. Vermetus incertus. Scalpellumjuliense. J& S *Pecten geminatus. S * Struthiolaria ameghinoi. Balanus varians. * Corbula hatcheri. *Pyrula Carolina. * Geryon peruvianus. *Panopea regularis. J Siphonalia noachina. Of these 26 species, 17 are also found at Santa Cruz (marked *). Again here, Juliense species prevail ; typical Leonense species are wanting, but 3 characteristic Suprapatagonian species are present. It is impossible to decide which one of these two localities near San Julian is to be regarded as the type-locality of Ameghino's " Piso Juli- ense." Both contain a number of Juliense fossils (each 5), of which 3 ( Terebratella patagonica, Pecten geminatus, Siphonalia noachina] have been found at both. The fact that the matrix at Darwin Station ap- ORTMANN : TERTIARY INVERTEBRATES. 275 proaches more that of the mouth of the Santa Cruz River, is expressed in the higher percentage of species that are also found at Santa Cruz, although this difference from Oven Point is very slight. Both San Julian localities have 9 species in common, but the abundance of the respective species at each locality is very different. Taking to- gether these two localities, we may regard them as the type of the Juli- ense beds, since 7 of the 9 species mentioned by Ameghino have been found here. But the fact, that i Leonense (Ostrea ingens], and 4 Supra- patagonian species (Ostrea ingens, Pecten geminatus, Sea/aria ritgulosa, Stntfhiolaria anieghinoi] have been found here, strongly points to the conclusion, that these beds also represent what Ameghino calls "Supra- patagonian." Two more localities on the coast are: Port of Deseado (Port Desire), which has yielded only Pecten cf. centralis, and Port Madryn (New Bay, Terr. Chubut), which has yielded Ostrea ingens from two horizons : be- tween tides, and 25' above high tide. The following localities are situated more or less inland : they all have a sandy facies, and often the matrix is composed of fragments of shells. Shore of Salt Lake, 10 miles north of the mouth of Rio Chico ; near the base of the series, within 50' above barren Cretaceous sandstones. J Scutella patagonensis. J * Terebratella patagonica. L &• S * Ostrea ingens. Even among this small number of species, a Leonense and Suprapata- gonian is associated with two Juliense. Upper Rio Chalia ; beds immediately underlying the Santacruzian beds, containing mammals. Top of marine series. J Scutella patagonensis. * Venus navidadis ( ? ). L * Turritclla ambulacrum. Melicerita triforis. S *Dosinia meridionalis ( ? ). * Valuta gradlior. J * Terebratella patagonica. * Panopea quemadensis. * Valuta dorbignyana. * Glycimeris ibari. Fissurella eurytreta. Verruca lavigata. L&S* Ostrea ingens. * Gibbula dalli. Balanus vatians. *Mytilus cf. chorus. S *Scalaria rugulosa. S *Lucina promaucana. * Ctcpidula gregaria. These beds which are undoubtedly near the extreme top of the marine series, and which ought to be, according to stratigraphical evidence, Supra- patagonian, contain — among 19 — no less than 14 species, which have been found also at Santa Cruz (marked*). 4 of the species are charac- 276 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. teristic Suprapatagonian, while 4 are Patagonian (2 Juliense and 2 Leon- ense). The presence of 7 species in these beds (Scutella patagonensis, Terebratella patagonica, Ostrea ingens, Panopea quemadensis, Scalaria rugulosa, Valuta gracilior, Balanus varians] is most significant, since these have been found also in the lowermost beds of the whole series at San Julian. Thus the palaeontological evidence — if we follow Ameghino's divisions- is in conflict with the stratigraphical. jo miles north of upper Rio Chalia; beds corresponding to the last locality : immediately below Santacruzian beds, top of marine series. * Cidaris anatarctica. *Psammobia patagonica. L * Turritella ambulacrum. J Scutella patagonensis. * Panopea quemadensis. * Turritella patagonica. J '* '• Terebratella patagonica. S * Scalaria rugulosa. S * Struthiolaria ameghinoi. * Glycimeris ibari. * Infundibulum corrugatum. * Trophon patagonicus. L & S * Ostrea ingen s. S * Natica darwini. Balanus varians. Although this list differs a little from that of the last locality (8 species in common), there is much resemblance between both as regards matrix, etc., indeed, both belong apparently to about the same level in the marine series. Of these 15 species, 13 have been found at Santa Cruz (marked*); 4 are characteristic Suprapatagonian, while 4 are Patagonian (2 Juliense and 2 Leonense). The conclusions are identical with those drawn from the last locality. Canon near Sierra Oveja, Rio Chico ; extreme top of the series: these beds are interstratified with Santacruzian beds containing Mammalian remains. J ^Terebratella patagonica. J &• S*Pecten geminatus. S * Scalaria rugulosa. L&S* Ostrea ingens. P * Cardium puelchum ( ? ). J Siphonalia noachina. These beds, which ought to be, by all means, Suprapatagonian, contain only a single form that is characteristic of the Suprapatagonian beds, while 2 species are found in both Patagonian and Suprapatagonian, and 2, Terebratella patagonica and Siphonalia noachina, at the base of the series, in the "Piso Juliense" of Ameghino. The presence of these two species at this locality is entirely opposed to Ameghino's conception of Patagonian and Suprapatagonian beds. ORTMANN : TERTIARY INVERTEBRATES. 277 At the same locality 100-150' below the fossils mentioned above, have been found Mammalian bones associated with Leda errazurizi(>}. Shell Gap, Rio Chico, lower horizon. Toxopneustes praecursor. L & S * Ostrea ingens. Cellaria fistulosa. J &• S *Pecten geminatus. The list is too small to permit any conclusions. Toxopneustes pre- cursor is also from the lowermost beds of the marine series at San Julian. Shell Gap, Rio Chico, iipper horizon; 150-200' above lower horizon. J Scntella patagonensis. * Venus volckmanni ( ?). Galerus araucanus. J *Rhynchonella plicigcra. * Tellina tehuelcha. Vermetus cf. intortus. *Terebratella dorsata. *Mactra danvini. S * Struthiolaria ameghinoi. J * Terebratella patagonica. *Martesia patagonica. Balanus varians. P * Cardium puelchum ( ? ). *Gibbula dalli. * Venus meridionalis ( ? ). * Infundibulum corrugatum. Of these 16 species, 12 have been found at Santa Cruz. 4 are charac- teristic Patagonian (mostly Juliense) fossils, while one is Suprapatagonian. This locality offers in matrix, etc., a striking resemblance to that of the upper Rio Chalia and 30 miles north of it, and 6 of the species have also been found there. Petrographically and stratigraphically, these beds would belong near the top of the series, while palaeontological evidence points partly to the identity with the beds of the upper Rio Chalia, partly to a lower position, near the base of the series. Mayer basin; upper Lignites, below Patagonian beds. Ostrea ingens has been found here, and indeterminable remains of other Bivalves, possibly of a Mytihts. Arroyo Gio ; ca. 100 feet marine beds, between barren Cretaceous sandstones and Santacruzian beds ; sandy facies. Heteropora pelliculata. * Cardium philippii (?). Crucibulum dubium. *Leda oxyrhyncha. * Cardium pisum. * Crepidula gregaria. *Leda errazurizi. S *Dosinia meridionalis. L *Turritella ambulacrum. S * Area patagonica. * Tellina jeguaensis. * Valuta g racilior. * Glycimeris ibari. *Psammobia patagonica. *Bulla patagonica. L&- S* Ostrea ingens. S Dentalium octocostellatum. J& S *Pt>cten geminatus. * Gibbula dalli. Casts of indeterminable species of Natica, Marginella and Bivalves. 278 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Of these 19 species, 16 (marked *) are found at Santa Cruz, which would make these beds identical with those of the type-locality. The fact, that Juliense, Leonense, and Suprapatagonian species — although only a few of each category — are associated here, is in so far interesting, as it shows that these three supposed different divisions are condensed here into only 100 feet. LAKE PUEYRREDON. East end of Lake Piieyrredon; horizon not ascertained. J * Rhynchonella plicigera. L & S * Ostrea ingens, * Cardium philippii. * Terebratella dorsata. J& S *Pecten geminatus. All of these 5 species have been recorded from Santa Cruz, and, although the number of species is very small, we have a mixture of Patagonian and Suprapatagonian forms. According to the matrix and the species found, this horizon seems to correspond to the lowermost of the Rio Tarde section (see below). High bhtffs, S. IV. of Lake Pueyrredon; horizon not ascertained, but about 1000' below Santacruzian beds. J * Rhynchonella plicigera. Magellania lenticitlaris. Grypluza cf. tarda. Rhynchonella squamosa. J * Terebratella patagonica. This horizon also seems to correspond to the lowermost of the Rio Tarde section. LAKE PUEYRREDON, Rio TARDE SECTION, 700' thick (see Hatcher, 1 900 a, p. 100). Base of marine Tertiary : The lowermost beds of the Rio Tarde sec- tion, immediately overlying a basaltic deposit that lies on top of the barren Cretaceous sandstones. *Cidaris antarctica. L *Perna quadrisulcata. *Dentalium sulcosum. J Scutella patagonensis. L & S * Ostrea ingens. * Gibbula lizvis. Cyrtoma posthumum. J& S *Pecten geminatus. S *Scalaria rugulosa. J * Rhynchonella plicigera. Modiola andina. L * Turritdla ambulacrum. Rhynchonella squamosa. Crass ate llites longior. S * Strut] dol aria ameghinoi. Magellania lenticularis. * Cardium philippii. * Trophon patagonicus. * Terebratella dorsata. * Cardium pisum. Urosalpinx pyriformis. J * Terebratella patagonica. * Venus volckmanid. Balanus cf. psittacus. J Bouchardia zitteli. *Panopea regularis. Balanus varians. * Glycimeris ibari. *Panopea quemadensis. ORTMANN I TERTIARY INVERTEBRATES. 279 Of these 29 species, 19 have been found at Santa Cruz. This horizon, the lowermost in this section, ought to be Juliense, and, indeed, it con- tains 5 Juliense species ; but this conclusion is entirely upset by the fact that 3 Leonense and 4 Suprapatagonian species are* also represented here. 400 above base. Only Modiola andina has been collected here. 600 above base (or 100' below top of marine series). * Terebratella dorsata. J * Terebratella patagonica. *Nucula patagonica. *Leda errazurizi. * Glycimeris ibari, L *Perna quadrisulcata. L & S * Ostrca ingens. J& S *Pecten geminatus. Modiola andina, * Crassatellites quartus. Cardita volckmanni. * Cardium philippii. * Venus meridionalis. * Venus volckmanni. Mactra garrctti. *Parwpea regularis. *Panopea quemadcnsis. *Martesia patagonica. *Solariclla dautzenbergi. *Gibbula lavis. *Gibbula dalli. * Infundibulum corrugatum. Galerus araucanus. L *Turritella ambulacrum. Vermetus cf. intortus. S * Struthiolaria ameghinoi. *Pyrula Carolina. J Siphonalis noachina. * Trap/ton patagonicus. S Valuta ameghinoi. * Terebra costellata. * Geryon peruvianus. The comparison of this locality with others is very important and inter- esting. About half (15) of the number of species are identical with those of the lowermost horizon of this section, although both are separated from each other by almost 600 feet of deposit. In this locality, out of 32 species, 25 are common to the type-locality at Santa Cruz, and this shows conclusively that the three (Santa Cruz, lower and upper horizon of Rio Tarde section) cannot be separated, and further, it shows that at Santa Cruz no two or more horizons can be represented, since the latter section comprises only about 250 feet, while here, at 600 feet above the base, still an unmistakable Patagonian fauna (as found at Santa Cruz) is present. Comparing our list with the subdivisions of Ameghino and v. Ihering, we see that we have here again, at 600' above the base, and 100' below the top, where we should expect a characteristic Suprapatagonian fauna, a mixture of 3 Juliense, 3 Leonense, and 4 Suprapatagonian elements, a relation that has hardly changed from that found at the base of the Rio Tarde section. Extreme top of marine series (Rio Tarde section). Ostrea ingens has been found here in a form (see p. 109) that corre- sponds to O. hatcheri, which is found, according to v. Ihering and Ameghino, exclusively in the Leonense beds. 280 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. Punta Arenas, horizon V (uppermost). See the description of the Punta Arenas section, given by the writer, according to Hatcher's observa- tions, 1898, p. 481. Ameghino (1899, p.? 13) has completely misunderstood, and arbitrarily changed the account the writer has given of this horizon. Hatcher labelled some oysters collected here: "below" and "above" V. This means only that this horizon begins, at the base, with an almost solid layer of oysters, then follows a layer of dark-greenish sand that contains the other fossils, which ends again, at the top, in a similar layer of oyster shells. This whole horizon, including both oyster beds, is a unit, and there is no break at all. Nevertheless, Ameghino constructs an hiatus between the upper oyster layer and the rest, and correlates the former with his Tehuelche formation, the latter with the Suprapatagonian part of the Santacruzian formation. This assumption of an hiatus within our horizon V is entirely unwarranted, and characterizes Ameghino's manner of trimming facts to suit his theories, even without having seen the origi- nal locality. We must take together all the fossils found here, and may mention only that Ostrea ingens occurs everywhere in this horizon (also in the sand be- tween the two oyster beds), and that Sigapatella has been found only in- side of oyster shells of the upper oyster bed. The latter consists of a form of Ostrea ingens, that resembles much the Cape Fairweather variety : *Glydmeris ibaria. Venus chiloensis. * Sigapatella amcricana. L & S * Ostrea ingens. S Meretrix iheringi. S *Lucina promaucana. * Crepidula gregaria. Although very small, this list contains 5 species that are found at the type-locality at Santa Cruz (marked *). For the rest, so-called "Supra- patagonian" species prevail. Ostrea ingens is a Patagonian (Leonense) and Suprapatagonian species, and it is to be remarked, that the form hatcheri, which is said to be characteristic of the Leonense beds, has also been found here. In conclusion, I add here a list of those species which have been ascer- tained to be present both near the base and near the top of the Patagonian series (as understood by us). (base : San Julian. I. Ciaans antarctica. . - ~. «_,. (top : 30 miles north of Rio Chalia. ORTMANN TERTIARY INVERTEBRATES. 28l 2. Scutella patagonensis. 3. Rhynchondla plicigera. 4- TerebrateUa dorsata. 5. TerebrateUa patagonica, 6. Glycimeris ibari. 7- Perna quadrisulcata. 8. 9. Pec ten geminatus. i o. Modiola andina. 1 1 . Lucina promaucana. 1 2. Cardium philippii. 1 3 . Cardium puelchum. 14- Venus volckmanni. 15. Panopea regularis. *6. Panopea qncmadensis. 1 7. Gibbu/a Icevis. 1 8. Scalaria rugulosa. 19. Iiftmdtbtthtm corrugatum. 20. Turritella ambulacrum. 2 1 . Turritella patagonica. J base (top: f base (top: fbase (top: f base : (top: J base : (top: J base (top: [base (top: Jbase (top: J base : (top: fbase (top: J base (top: jbase (top: J base : (top: f base : (top: f base : (top: fbase (top: fbase (top: (top: fbase (top: Jbase (top: 22. Strutliiolaria ameghinoi. (top: San Julian; Salt Lake; Lake Pueyrredon. Lhaha ; 30 miles north of Rio Chalia. Lake Pueyrredon. Las Salinas ; Shell Gap. : Lake Pueyrredon. Shell Gap. : San Julian; Salt Lake; Lake Pueyrredon. Kio Chaha ; Lake Pueyrredon. : Lake Pueyrredon. Rio Chalia ; Lake Pueyrredon. ' Lake Pueyrredon. Lake Pueyrredon. : San Julian; Salt Lake ; Lake Pueyrredon. -haha; Sierro Oveja; Lake Pueyrredon. San Julian ; Lake Pueyrredon. Sierra Oveja ; Lake Pueyrredon. : Lake Pueyrredon. Lake Pueyrredon. : Paso del Rio Santa Cruz. Rio Chalia. : Lake Pueyrredon. Lake Pueyrredon. : Santa Cruz. Sierra Oveja. : Santa Cruz ; Lake Pueyrredon. Lake Pueyrredon. : San Julian ; Lake Pueyrredon. Lake Pueyrredon. : San Julian ; Lake Pueyrredon. Rio Chalia ; Lake Pueyrredon. : San Julian ; Lake Pueyrredon. Lake Pueyrredon. San Julian ; Lake Pueyrredon. Rio Chalia ; Sierra Oveja. : San Julian. Rio Chalia ; Lake Pueyrredon. : Santa Cruz ; • Lake Pueyrredon. Rio Chalia ; Lake Pueyrredon. : Santa Cruz ; San Julian. Rio Chalia. San Julian ; Lake Pueyrredon. Rio Chalia ; Lake Pueyrredon. 282 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. (base : San Julian. 23. Pyrula Carolina. T , „ , [top: Lake Pueyrredon. (base : San Julian. 24. Siphonaha noachma. ~. ~ . T i r> (top: Sierra Oveja ; Lake Pueyrredon. f base : San Julian ; Lake Pueyrredon. 25. Troplion patatromcus. ~. ,,, ,. T , „ [ top : Rio Ghana ; Lake Pueyrredon. f base : San Julian. 20. Volnta granhor. T,. „, .. ( top : Rio Chaha. (base: San Julian ; Lake Pueyrredon. 27. Balamis vanatis. „. _, .. „, .. „ (top: Rio Chalia ; Shell Gap. ("base: San Julian; Santa Cruz. 28. Gery on peruvianus. [ top : Lake Pueyrredon. We add the following species, which are either given as Suprapata- gonian by v. Ihering, Cossmann or Ameghino, but found in our collec- tion only near the base, or are given as Juliense by Ameghino or Patagonian by v. Ihering, but found in our collection at the top. Suprapatagonian species, found at the base : 29. Crassatellites longior, Lake Pueyrredon, base. 30. Amathusia angulata, Santa Cruz, at water's edge. 3 1 . Urosalpinx pyriformis, Lake Pueyrredon, base. Patagonian species, found at the top : 32. Mytilus cf. chorus, Rio Chalia. 33. Tellina tehuelcJta, Shell Gap. 34. Psammobia patagonica. Rio Chalia. 35. Martesia patagonica, Shell Gap, Lake Pueyrredon. 36. Valuta dorbignyana, Rio Chalia. Finally, I give here the species which are recorded by v. Ihering as Patagonian and Suprapatagonian. 37. Niicula patagonica. 43. Crcpidula gregaria. 38. Limopsis insolita. 44. Natica secunda. 39. Cardita incequalis. 45. Natica consimilis. 40. Cardita patagonica. 46. Turritella breantiana. 41. Venus mcridionalis. 47. Cancellaria gracilis. 42. Infundibulum clypcolum. Of these 47 species, for which a distribution from top to bottom of the series is very probable, the following belong to Ameghino's or v. Ihering's characteristic fossils of their different subdivisions. ORTMANN I TERTIARY INVERTEBRATES. J x Scutclla patdgonensis. J Rhynclwnella plicigcra. J x Terebratella patagonica. L Perna quadrisulcata. 5. L &• S x Ostrea ingens. 6. J & S x Pecten geminatus. 7- -S Liicina promaucana. i. 2. 3- 4- 8. P Cardium puelclmm. 9. 5 Amathusia angulata. I O. S x Scalaria nigulosa. 11. L x Turritella ambulacrum. 12. 5 x Struthiolaria ameghinoi. 13. J Siphonalia noachina. This fact is the more important since just these species belong to the most striking and characteristic features of this series, especially those marked x. Finally, I offer here a list of the so-called characteristic species of Juliense, Leonense, and Suprapatagonian beds, which have been found associated in one and the same block : Darwin Station, San Julian: J & S Pecten geminatus J Pecten prcenuncius. Mouth of Santa Cms River: Block i : 6" Area patagonica. J Pecten prcenuncius. J Terebratella patagonica. S Area patagonica. Block 4: P Cardita incequalis. Block 5: 5 Cucullpe . Recent est Indies . Recent West Indies Crepidula praerupta and allied forms Miocene . . . - - Pliocene Recent N. callosa, California Grot Miocene North Pacific ip of Natica h . Pliocene. . eros . Recent North America Natica heros Miocene ... - Pliocene Recent I T. chipolana, Florida . . Miocene . . •forth America T. apicalis, Florida T. peratten- uata, Florida . Pliocene L . Recent Europe Genus : Aporrhais . . Miocene Pliocene Europe Genus : Dolium Miocene Pliocene Recent P. pyriformis, California T. tarbel- lianum, Europe B. veneris, Europe C. glomus, Europe Europe V 294 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. CRETACEOUS. " EOCENE. OLIGOCENE. MIOCENE. PLIOCENE. RECENT. 56. Murex Subgenus : Phyllonotus hatcheri Miocene Pliocene Recent Tropical seas 57. Fusus F. hector, archimedis N. Jersey 58. Fusus F. burdiga- torosus lensis, Europe 59. Margi- M. faunula, nella Florida gracilior Marginella be\ a Miocene . . . Pliocene . . . . Recent • North America 60. Margi- Marginella styria nella Pliocene Recent olivella North America 61. Terebra Terebra sp. costellata (costellata) Europe 62. Pleuro- P. monilis, toma Europe subaequalis 63. Genota G. intorta, cuevensis Europe 64. Drillia D. limatula, santacru- N. America zensis 65. Borsonia B. delucii, patagonica Europe 66. Actaeon A. semistri- semilaevis atus, Europe 67. Bulla pat- B. glaphyra, agonica Europe 68. Scalpel- S. solidulum, lum juli- Europe ense 69. Balanus Bal. unguiformis, varians Eocene Oligocene Europe Among these 69 species available for comparison, we have the follow- ing relations : Cretaceoiis relations : 3 species = 4 per cent. Cyrtoma posthumum. Gryphaa cf. tarda. Scalpellum juliense. ORTMANN : TERTIARY INVERTEBRATES. 295 Eocene relations: 5 species = 7 per cent. Cuculliza alta. Crassatella kokeni. Psammobia patagonica. Fusus archimedis. Bttlla patagonica. Eocene and Oligocene relations: \ species =1.5 per cent. Balanus varians. Oligocene relations: 7 species = 10 per cent. Nucula patagonica. Nucula reticularis. Cardita patagonica. Mactra garretti. Corbula hatcheri. Martesia patagonica. Borsonia patagonica. This gives the sum of all Eogene relations (Eocene and Oligocene, but not passing up into the Neogene) as 13 species = 18.5 per cent. Intermediate (Eogene as well as Neogene] species : 8=12 per cent. Cellaria fistulosa (Olig.-Rec.). Scalaria rugulosa (Olig., Mioc.). Cucullcea darwini (Eoc. & Plioc.). Natica secunda (Olig., Mioc.). Cardium puclchum (Olig.-Rec.). Vermetus intortus (Olig. -Plioc.). Dentalium sulcosum (Olig., Mioc.). Aporrhais araucana (Olig.-Rec.). Miocene relations: Cidaris antarctica. Leda errazurizi. Ostrea ingens. Crassatellites longior. Venus chiloensis. Venus meridionalis. Venus darwini. Tellina jeguaensis. 22 species = 32 per cent. Calliostoma pararatum. Calliostoma santacruzense . Calliostoma garretti. Turritella patagonica. Pyrula Carolina. Tritonium morgani. Buccinum anna. Pliocene relations : 4 species = 6 per cent. Melicerita triforis. Turritella ambulacrum. Pecten prcenuncius. Recent relations : 3 species = 4 per cent. Toxopneustes precursor. Heteropora pelliculata. Chrysodomus cancellatus. Fusus torosus. Terebra costellata. Pleurotoma sub&qualis. Genota cuevcnsis. Drillia santacrusensis. Actaon semil&i'is. Turritetla breantiana. Liotia scotti. Neogene relations (Miocene- Recent] : 13 species = 19 per cent. Scutella patagonensis. Odontostomia suturalis. Natica subtenuis. Area patagonica. Crucibulum dubium. Dolium ovulum. 296 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. Glycimeris ibari. Crepidula gregaria. Murex hatcheri. Lucina promaucana. Natica danvitti. Marginella gracilior. Dosinia meridionalis. Miocene and Pliocene relations: i species =1.5 per cent. Calliostoma iheringi. Pliocene and Recent relations : 2 species = 3 per cent. Gibbula dalli. Marginella olivella. To sum up, among the species which may be compared with known ones, the percentage of the relations with particular beds is the following : Cretaceous : 4 % • ••• 4 % Eocene 7 Oligocene 10 Eoc. and Olig 1.5 Eogene 18.5.... 18.5 % Intermediate 12 .... 12 % Miocene 32 Pliocene 6 Recent 4 Mioc.-Rec 19 Mioc. and Plioc 1.5 Plioc. and Rec 3 Neogene 65.5....* 65.5 % 100% The above figures speak for themselves. We see a constant increase of the percentage from the Cretaceous to the Miocene, and then again quite a sudden decrease from Miocene to Recent The percentage of all Neogene forms is 65.5, considerably more than half of the number, while that of the Eogene and Cretaceous together is only 22.5 ; the rest (12 per cent.) is intermediate between Eogene and Neogene. These facts bring the Patagonian beds undoubtedly into the Neogene, and when we consider the fact that after the Miocene there is a marked decrease in the percentage, we must put these beds in the beginning of the Neogene times, that is to say, in the Miocene. This course is ren- dered the only possible one by the fact that 32 per cent, of the whole number of species shows Miocene relations. No other group has a per- ORTMANN : TERTIARY INVERTEBRATES. 297 centage like this. Taking together all species that might possibly be Miocene, we would have : Miocene species .................... 22 Mioc.-Rec. species ................. 13 Mioc. & Plioc. species ............... i Eogene & Mioc. species ............. 8 Directly opposed to Miocene age are : Older than Miocene. Cretaceous species .................. 3 Eocene species ..................... 5 Oligocene species ................... 7 Eogene species ..................... i "16=23^. Younger than Miocene. Pliocene species ................... 4 Recent species ..................... 3 Plioc. & Rec. species ................ 2 ~~9= 13 $• Thus 64 per cent, of the whole number may be taken safely as Miocene specimens, while 23 per cent, point to an older age, and only 13 per cent. to a younger age. This latter fact has apparently the following meaning : there are, in these Miocene beds, more relations with the underlying than with the overlying beds, and, accordingly, we are to consider this fact by placing the Patagonian beds in the Lower Miocene.1 2. COMPARISON OF THE PATAGONIAN BEDS WITH TERTIARY DEPOSITS OF THE SOUTHERN HEMISPHERE. This result, that the Patagonian beds are Lower Miocene, has been obtained by comparing them exclusively with deposits of the northern hemisphere. Now it will be very interesting to compare them with other beds of the southern hemisphere, and we shall see that there are extremely significant connections. 'According to Ball (18980) some of the North American and West Indian beds classed here with the Miocene are really Oligocene ; this would, however, affect our conclusions only in so far as it would increase slightly the relations with the older Tertiary, and would thus place the Patagonian beds more decidedly in the lower Miocene. 298 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. The Navidad beds of Chili are undoubtedly the best known Tertiary deposits of the southern hemisphere, at least as regards their Palaeontol- ogy, and Darwin, d'Orbigny, Philippi, and Moericke held the opinion that they are about contemporaneous with the Patagonian beds. According to our collections, the following Navidad fossils have been found in the Patagonian beds : A. Identical species : Leda oxyrhyncha. Dentalium sulcosum. Aporrhais araucana. Leda errazurizi. Gibbula lavis. Pyrula Carolina. Limopsis insolita. Scalaria rugulosa. Buccinum obcsum minor. Glycimeris ibari ( ? ). Galerus araiicanus. Siphonalia domeykoana. Mytilus cf. chorus ( ? ). Crepidula gregaria. Valuta triplicata. Cardita volckmanni. Natica ovoidea. Valuta domeykoana. Lucina promaucana. Natica secunda. Cancellaria medina ( ? ). Amathusia angulata. Natica darwini. Terebra costellata. Venus chiloensis. Turritella ambulacrum. Pleurotoma subczqualis. Venus meridionalis. Turritella breantiana. Balanus cf. psittacus ( ? ). Venus volckmanni. Turritella patagonica. Balanus varians. Venus navidadis. B. Closely allied species are the following : Schizaster ameghinoi S. valdivianus. Nucula patagonica .TV. araucana. Malletia ornata M. volckmanni. Cucullcea alta C. chilensis. Area patagonica A. oxytropis. Pecten proximus P. caracolensis. Cardium philippii C. multiradiatum. Cardium pisum. . -. C. spharidium. Tellina jeguaensis T. promaucana. Mactra garretti M. truncatula. Sigapatella americana S. colchaguensis. Strut] dol aria ameghinoi S. chilensis. Tritonium morgani T. verruculosum. Urosalpinx elegans U. leucostomoides. Fusus torosus F, pyruliformis, Thus, out of 151 species, 34 are identical with species found in the Navidad beds, that is to say, 22 per cent, or almost one quarter. This is certainly to be regarded as a high percentage, considering the consid- erably more northern situation of the Navidad beds, and undoubtedly ORTMANN I TERTIARY INVERTEBRATES. 299 establishes the identity at least of a part of the Navidad series with the Patagonian beds. Whether of all of it, remains doubtful, since we do not possess any stratigraphical observations on the Navidad beds, and we shall become acquainted, further on, with facts, which point to the possi- bility that these Chilian beds are not a unit, but contain horizons of dif- ferent age. Moericke (1896, p. 593), following Steinmann, included the Patagonian beds of Santa Cruz in the "Navidad stage," and makes (p. 597) it Miocene, with a suggestion of Oligocene, which agrees well with our re- sults, which make the Patagonian beds Lower Miocene. Relations to New Zealand were discovered first by Zittel (1864): he directly identifies some New Zealand species with Patagonian. Accord- ing to the list of New Zealand fossils published in 1873 by Hutton, and his subsequent writings on this fauna (1885), I have been able to compile the following list of identical and closely allied species. (In the following O means Oamaru (Oligocene) ; P means Pareora (Miocene); W means Wanganui (Pliocene); R means Recent. A. Identical species : Cellaria fistulosa. P Heteropora pelliculata. P Rhynchonella squamosa. 0 °Magellania lenticularis. 0, P, R ° Tercbratella dorsata. P Terebratella patagonica. 0, P 0 Cucullcea a/fa. 0, P °Limopsis insolita. P °0strea ingens. Gryphcea tarda. Mytilits magellanicus. ° Scalaria rugulosa. ° Crepidula gregaria. ° Natica daruuini. Turritella ambulacrum. 0,P 0, /'(Chatham Isl.). P, W 0,P P P P, W B. Closely allied species : Melicerita triforis M. angustiloba. Rhynchonella plicigera R- nigricans. Leda oxyrhyncha L. sp. (Zittel). Malletia ornata M. australis. Pecten proximus P- athleta. Dentalium sulcosiim D. mantelli. Solariella dautzenbergi 5 stoliczkai. Sigapatella americana •$". maculata, Genus Struthiolaria Genus Struthiolaria. Siphonalia domeykoana S. dtlatata. Valuta ameghinoi V. pacifica. P 0, P, W, R P P, W,R 0 P 0 0, P, W, R P, W, R P, W,R 0, P. W, R 300 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Thus, out of 151 species, 15, or 10 per cent, are identical with New Zealandian Tertiary fossils, while 1 1 more are closely allied, making the total number of relations between Patagonian and New Zealandian Ter- tiary 1 7 per cent. This is indeed a large percentage, considering the wide separation of the two localities. Of these New Zealand species, 4 go through from the Oamaru beds to Recent times, and 3 from the Pareora to Recent. Five are found in both the Oamaru and Pareora, and 2 in the Pareora and Wanganui. Of the rest (12), 3 belong to the Oamaru, and 9 to the Pareora beds. Thus the bulk of the affinities points directly to a comparison with the New Zea- landian Pareora beds, which are, according to Hutton (i885a), Miocene, and this result again corroborates our identification of the Patagonian beds as Miocene.1 We cannot, however, disregard the fact that there is considerable dis- cussion as to the age of these New Zealandian beds, especially Hector, in opposition to Hutton, considers them to be much older (Cretaceous-Ter- tiary). But in this respect I should say that our investigations tend to confirm Hutton's opinion, which makes these beds Oligocene and Miocene (Oamaru and Pareora). On the other hand, v. Ihering (1899, p. 40, footnote) tries to minimize the evidence furnished by the comparison of Patagonian and New Zea- landian fossils, and thinks that a closer inspection of the New Zealand species will prove their specific difference in many cases. In my opinion, the question of specific identity is of secondary value, although I firmly believe that in those cases marked ° in the list it is well established. But even if there should be no identical species, the fact remains that a num- ber of Patagonian fossils very closely resemble New Zealandian species, and this fact is the more important, since some of these forms are ex- tremely characteristic, for instance : Ciicullcea, Limopsis insolita, Scalaria rugtilosa, Malletia, Sigapatella, Struthiolaria, Siphonalia domeykoana, which types are hardly represented elsewhere. Indeed, it is the association of forms like these which gives to the Patagonian and New Zealandian faunas their striking similarity, not the fact that a few species are really identical. 1 1 have disregarded Cardita patagonica and Venus meridionalis, which are found, according to Hutton (1886, p. 362 and 364) in the Pareora and Wanganui beds, since I have no means of deciding their identity with the respective New Zealand species (C. intermedia and V. vellicata). ORTMANN : TERTIARY INVERTEBRATES. 30! I have further tried to compare our Patagonian fossils with those of the Tertiary beds of Tasmania and southern Australia, and am able to give the following list of relations : Melicerita triforis M. angustiloba, S. Australia (? Miocene). Reticulipora patagonica R. transennata, S. Australia (? Eocene). Rhynchonclla squamosa same species in Tasmania (? Miocene). Nucula patagonica N. tumida, Australia, Tasmania (? Eocene). Leda oxyrhyncha and errastttrisn.taa&xr species in Australia. Limopsis insolita same species, S. Australia (? Eocene). Area patagonica A. pscudonavicularis, Australia (? Eocene). Ostrca ingens 0. stnrtiana, River Murray Cliffs (? Miocene). Gryplicea cf. tarda G. tarda, S. Australia (? Eocene). Pecten pnenundus P. palmipes, Australia (? Miocene). Venus meridionalis V. multitceniata and hormophora, Australia, Tasmania (? Eocene). Dosinia meridionalis D. denselineata, Tasmania, Victoria (? Miocene). Psammobia patagonica P. hamiltonensis, Victoria, Tasmania (? Eocene). Dentalium sulcosum D. mantelli, Australia, Tasmania (? Eocene). Turritella ambulacrum T. aldingce, S. Australia (? Eocene). Valuta triplicata V. sarissa and tateana, Australia (? Miocene, Eocene). Valuta ameghinoi V. atkinsoni, Tasmania. No attempt has been made to correct or to control the age given for these Australian species. This list is very defective, since it was impos- sible for me to make a closer comparison, especially because the figures of Australian species given by Tate (1886-1893) are in most cases very poor. Nevertheless the fact is apparent that a few identical species are found, which are in part also recorded from New Zealand, and that a larger number of species show close affinities with Patagonian forms. The latter number will undoubtedly be increased considerably, after a careful examination of the Australian fossils has been made. For the present, it is sufficient to call attention to the fact that not only in New Zealand, but also in Australia and Tasmania, Tertiary deposits are found, which yield a fauna that shows unmistakable affinity to the Patagonian fauna.1 'According to Harris (1897), whose Australasian Tertiary Mollusca were not consulted until the above was written, I can add the following striking cases : Liotia scotti L. roblini Johnst. (Harris, p. 284, pi. 8, f. 4) "Eocene," Muddy Creek. Fissurella eurytreta Fissurellidea malleata Tate (ibid., p. 287, pi. 8, f. 5) "Eocene," Muddy Creek. 302 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. i Unfortunately our information on Australian stratigraphy is very defec- tive, and especially as to their age we are again confronted with the same contradictory opinions that have been expressed in the case of New Zea- land and Patagonia. Most of the Australian Tertiary deposits belong, according to Tate (see Tate, Correlation of the marine Tertiaries of Australia in Trans. Roy. Soc. S. Australia, vol. 17, 1893, vol. 19, 1895, vol. 20, 1896), to the Eocene, and the same opinion is held by others. On the other hand, most of these beds have been classed with the Miocene by the Geological Survey of Victoria, while Harris (1897, P- I5)> although following Tate, expresses doubts as to the correct correlation of his "Eocene." Tate's determination of the age of these beds relies exclusively on the percentage of recent forms found in them.1 As has been said above, we consider this line of evidence as entirely inadmissible, and since Tate has not tried to introduce any other proofs, we may safely say that there is no evidence at all warranting the reference of these beds to the Eocene. And, indeed, the writer is of the opinion that these Australian and Tas- manian beds are also to be regarded as Miocene, simply because, in that case, we would not have any discrepancies in the stratigraphical position of these species, which have been found both in Australia and New Zea- land. If we regard — as we actually do — the New Zealandian Pareora beds as Miocene, the Australian beds containing Pareora species (for instance Limopsis insolita, Dentalium mantelli] must be correlated with them, un- less other evidence points to a contrary conclusion ; but no proof of the latter kind has been offered so far. Of course, we do not claim that all of the Australian Tertiary is Mio- cene, but we should expect to find that other deposits are also represented there. All we wish to say is that beds corresponding in age to the Mio- cene Patagonian beds must be present in Australia and Tasmania, and that there are apparently faunistic relations between both continents. It is left for future investigation to ascertain how far this parallelism ex- tends, and we entertain no doubt that the faunal relations between Pata- gonia and Australia, as well as New Zealand, will prove a very fruitful and interesting subject for research. 1 The same is true of the determination of the Eocene age of these beds by other writers ; for instance, Hall and Pritchard (see : Proc. R. Soc. Victoria, v. 7, 1894, p. 180, ff. and v. 8, 1895; p. 151, ff.) do not use any other method than that used by Tate. ORTMANN : TERTIARY INVERTEBRATES. 303 The result of the foregoing considerations is : We regard tlic Patago- nian beds as of Lower Miocene age ; contemporaneous deposits are found, in the southern hemisphere, not only in Chili (within the Navidad series], but also in New Zealand (Pareora beds of Htitton] and Australia, and the fatmas of these three localities (Sotith America, New Zealand, and Aus- tralia] show unmistakable affinities with each other. We shall return to this fact below. THE MAGELLANIAN BEDS. The " Magellanian beds," discovered by Mr. Hatcher near Punta Arenas, were first described by the present writer in 1898, and the term "Magellanian beds" was introduced by him in 1899, and accepted by Hatcher (1900 a, p. 97). The stratigraphical position of these beds has been ascertained positively by Hatcher: they are several hundred feet below the Patagonian beds, and separated from them by the Punta Arenas coal (Upper Lignites of Hatcher, 1. c., p. 99). Ameghino (1899, p. 12) has referred to this Punta Arenas section, and has attempted to correlate it with his subdivisions of the Patagonian for- mation, and, indeed, practically identifies our Magellanian beds with his " Patagonian formation." It is hardly necessary to pay any attention to this entirely unwarranted opinion (see : Ortmann, 1899, p. 427, first footnote). In what Ameghino calls his Piso Juliense in the Punta Arenas section (our horizon I), not a single Juliense species is present, but only plant remains have been found. What he calls Piso Leonense (our horizon II) does not contain a single Leonense fossil ; what he calls transitional beds between Patagonian and Suprapatagonian formation (our horizon III) contains only a single Pata- gonian species (Cardita elegantoides spec, nov.), but no other Patagonian or Suprapatagonian fossils. Such correlations are simply ridiculous, not to mention the fact that Ameghino's subdivisions, as has been demon- strated above, have no reality at all. The following is the fauna of the Magellanian beds. (II = lower hori- zon, 111 = upper horizon.) RELATIONS. . x Ostrea torresi (III) 0. bellovacina, Low. Eocene, Europe. Cardita elegantoides (III) identical species in the Patagonian beds. x Lucina neglecta (II) L. promaucana, Patagonian beds. 304 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. x Venus difficilis (II & III) V. subsulcata, Cretaceous, Chili. x Venus arenosa (III) V. landbccki, Cretaceous, Chili. x Meretrix pseudocrassa (III). . . .M. alta, Cretaceous, Chili ; M. crassa, Pliocene, Chili. x Dosinia magdlanica (II) D. semilavis, Navidad. x Lutraria undatoides (II) L. undata, Chili (Navidad beds ?). x Panopea ibaria (II). x Panopea subsymmetrica (III). x Patella pygmcea (III). x Calliostoma philippii (III) C.fricki, Navidad ; C. observations, Patagonian beds. Infundibulum merriami (II identical species (/. costellatum Phil.) in the Navidad beds ; and III) /. filosum, Miocene, California. Natica chiloensis (II and III) . . identical species in Chiloe ; N. venusta, Eocene, Europe. x Turritella exigita (II) T. granulosa, Eocene, Europe. x Struthiolaria hatcheri (II) possibly ancestral form of the Patagonian species of the genus. x Fusus subspiralis (II) F. oxytropis, Navidad. Act&on chilensis (II) identical species in the Navidad beds ; A. turgidus, Eocene, Europe. Bulla remondi (II) identical species in the Navidad beds ; B. striatissima, Eocene, Europe. Of these 19 species, 3 have been found in both horizons. One is found in the Patagonian beds of Santa Cruz, 4 have been found in the "Navi- dad beds" of Chili, while the rest (14, marked x] are restricted to this locality. Of these, 4 show relations to Navidad species, 2 relations to Patagonian species. One of the Navidad species (Infundibulum merri- ami] shows affinities to a Miocene California species, 5 species show re- lations to European Eocene species, and even Cretaceous affinities are not wanting (3 species). Although the number of species known from these deposits is quite small, the comparatively more numerous affinities to deposits of older age (Eocene and even Cretaceous) agree well with what we know of the stratigraphical conditions ; the Magellanian beds are older than the Pata- gonian, and further, one fact needs special mention : while the lithological character of the deposit, especially of horizon II, agrees strikingly with that of the type locality of the Patagonian beds at Santa Cruz (hard con- cretions, filled with shells, imbedded in looser material), some genera found in both localities (Punta Arenas and Santa Cruz) are represented by different species, for instance : Venus difficilis V. darwini (Santa Cruz). Dosinia magellanica D. meridionalis. Infundibulum merriami /. corrugatum. ORTMANN : TERTIARY INVERTEBRATES. 305 Natica chiloensis N. ovoidea. Turritella exigua T. ambulacrum. Struthiolaria liatcheri S. ameghinoi (and ornatd). Bulla remondi B. patagonica. Since these genera belong to the most characteristic forms of these de- posits, it is very significant that they are represented by different species, and this fact affirms the difference in age of both series indicated by the stratigraphical evidence. The fact that 4 Navidad species have been discovered in the Magel- lanian beds suggests that this latter horizon is also represented within the Navidad series. In this respect it is significant that these Navidad species have never been found in Patagonia, and it is quite possible that the Navidad fauna as described by Philippi contains elements of different age, including the Magellanian beds. As regards the age of the Magellanian beds we must depend in the first line on the stratigraphy, since palaeontology — although suggestive of a slightly older age than Patagonian — is altogether insufficent to permit any definite opinion. The Magellanian beds are several hundred feet be- low the Patagonian, and are separated from them by a coal deposit ; since the Patagonian beds are Lower Miocene, this would bring the Magellanian beds into the Oligocene, or perhaps — taking into consideration the changed conditions under which the Upper Lignites were deposited — into the Upper Eocene. It would be very interesting and important to get more material from the Magellanian beds. This first indication of this fauna was given by Philippi (1887), who described a number of fossils from Punta Arenas and Skyring Water. From his list of 16 species (leaving out Turritella patagonica, which has been inserted by mistake) the following are repre- sented in our collection : Haliotis imperforata = Crepidula gregaria. Venus chiloensis. Panopea ibari. Pectunculus ibari and magellanicus (both identical and = Glycimeris ibari). Ostrea bourgeoisi and patagonica (both identical and = 0 . ingens). Ostrea torrcsi. Two of these belong in the Magellanian beds (Panopea ibari and Ostrea torresi], while the rest has been found in horizon V, which represents the Patagonian beds. 306 PATAGONIAN EXPEDITIONS ! PAL/liONTOLOGY. The Punta Arenas section (Ortmann, 1898, p. 481) has been mentioned twice before in literature. First Mallard and Fuchs (1873, p. 67, ff.) have given a profile taken at 6-7 kilometers from Punta Arenas on the left bank of the river (Rio de las Minas). There does not seem to be any agreement with Mr. Hatcher's observations, but the fact that these writers mention at the base of their section, a glauconitic sand, which contains " Ostrea patagonica" and a large Pectuncuhis (= Gtycimeris], renders it beyond doubt that this bed corresponds to Hatcher's horizon V, which consists of a dark green sand containing a .large oyster (O. ingens] associated with a large Pectnnculns (Glycimeris ibari}. Thus Mallard and Fuchs' section begins just where Hatcher's section ends. A second time this section has been mentioned by Nordenskjoeld (1898, p. 24, footnote). His account agrees fairly well with Hatcher's, and the comparison is as follows (beginning at the top) : Sand, Sandstein und Geroell in maechtigen Schichten, unten mit etwas Lignit includes probably horizon V (Patagonian). Schieferthon mit Lignit und Pflanzenresten (Araucaria) undoubtedly = horizon IV (Upper Lignites). Sandstein mit einer muschelfuehrenden Bank : reichliche Schalen von Ostrea bourgeoisi und torresi. = horizon III (Ostrea torresi\ Sand mit kalkigen Einlagerungen (mit Stein- kernen schlecht erhaltener Mollusken) probably beds separating horizons II and III. Muschelfuehrende Bank (Ostrea fehlt ; Turri- tella und andere Gasteropoden vorhanden) = horizon II (Turritella eiigud). Sand und Sandsteine mit kalkigen Konkre- tionen, die schlecht erhaltene Pflanzenver- steinerungen enthalten (Fagus) = horizon I. Lignitische Schicht. The plant remains of our horizons I and IV have been described by Dusen from the collections made by the Swedish expedition (Dusen, 1899). He provisionally refers the upper Lignites (horizon IV ; slmucaria-beds) to the Miocene, and the horizon I (Faults-beds) ! to the Oligocene (p. 93), although it may be Eocene (p. 91). 1 1 have been able to identify some of the plant remains collected by Hatcher in horizon I, namely, Fagus subferruginea Dus. (p. 94), Nothofagus variabilis forma inicropliylla Dus. (p. 97), and others, which are identical with forms mentioned by Dusen from the Fagus beds of Punta Arenas. This establishes beyond doubt the identity of our horizon I and the "Fagus beds" of Dusen. ORTMANN : TERTIARY INVERTEBRATES. 307 This would make the Magellanian beds Oligocene in age. THE CAPE FAIRWEATHER BEDS ( ? MARINE TEHUELCHE BEDS). The Cape Fairweather beds were first described by Hatcher (1897 a» P- 342). They are of marine origin, and their stratigraphical position is un- conformably on top of the Santacruzian beds (see section, 1. c., p. 344) : this separates them at once stratigraphically from the Patagonian beds, which are below the Santacruzian beds. A first, preliminary account of the fauna of these beds has been given by Pilsbry (1897 a)- Further investigations have increased the number of species, and have made necessary a few corrections, so that the list of species known from the Cape Fairweather beds stands at present as follows : RELATIONS AND DISTRIBUTION. 1. Terebratclla gigantca Mutation of T. patagonica, Patagonian beds. 2. Ostrea ingens Patagonian beds. 3. Pectcn actinodes Mutation of P. geminatus, Patagonian beds; distribution north- ward from San Julian, in the Tehuelche beds (v. Ih.). 4. Mytilus cf. chorus Patagonian beds and Recent (Chili). 5. Meretrix rostrata Recent (Brazil and Uruguay). 6. Dosinia meridionalis Patagonian beds and Entrerios beds (v. Ih.). 7. Panopea pilsbryi. 8. Golems mamillaris Recent, Chili to California. 9. CrepiduladUatata Recent, southern Patagonia. 10. Turritella innotabilis Mutation of T. patagonica, Patagonian beds; closely allied to T. cingulatifonnis, Pliocene, Chili. 11. Trophon laciniatus Mutation of T. patagonicus, Patagonian beds; Recent, Pata- gonia. T. 1. var. inornatns also at Darwin Station ; Tehuelche beds (T. varians v. Ih.). 12. Balanus cf. psittaats Patagonian beds, and Recent, Patagonia. 13. Balanus cf. trigonus Recent, almost Cosmopolitan (Patagonia). 14. Balanus Icevis Pliocene, Chili ; recent, Patagonia, Chili to California. Indeterminable remains of Pinna, Area, Lucina (or Diplodonta ?), Cardita. Of these 14 species, 4 (Ostrea ingens, Mytilus chorus, Dosinia merid- ionalis, Balanus psittacns] are also represented in the Patagonian beds, while we may consider 4 others ( Terebratella gigantea, Pecten actinodes, Turritella innotabilis, Trophon laciniatus} as mutations of Patagonian 308 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. forms. On the other hand, we have in 5 species (Meretrix rostrata, Galerus mainillaris, Crepidula dilatata, Balanus trigonus, Balanus Icevis] new elements in this fauna, and it is very significant that these are all still living species. This introduction of new, recent elements into the Cape Fairweather fauna, as compared with the Patagonian, is the most impor- tant character, agreeing completely with the stratigraphical position. Altogether, the Cape Fairweather fauna contains 8 living species (57 per cent.), 5 of which are still found in Patagonia (Crepidula dilatata, Trophon laciniatus, Balanus psittaciis, B alarms trigomis, Balamis lcems], while 2 are found in Chili (Mytilus chorus and Galerus mamillaris), and i in Uruguay (Meretrix rostrata}. This fact approximates this fauna so closely to the recent one, that we may safely regard the Cape Fairweather beds as Pliocene. Only a few members of this fauna have been found elsewhere. Pecten actinodes was recorded long ago from various localities in Patagonia (chiefly the northern part), and has been given by v. Ihering (according to Ameghino's material) for the Tehuelche formation. The same is true in the case of Trophon laciniatus var. inornatus, which is mentioned by v. Ihering (as Trophon varians] from the Tehuelche beds (see below). Dosinia meridionalis has been recorded (aside from the Patagonian beds) from the Entrerios beds, and Turritella innotabilis finds a closely allied form in the Pliocene T. cigulatiformis of Chili. Thus it would seem, that the Pliocene Coquimbo beds of Chili, the Entrerios beds of Parana, and the "Tehuelche" beds are in some degree correlated with the Cape Fairweather beds.. But since a large number of Coquimbo- and Entrerios- species are known (Philippi, Mcericke, v. Ihering), the above relations are not satisfactory, and if these beds are really contemporaneous, we should expect a larger number of affinities. On the other hand, we must take into account the much more southern location of the Cape Fairweather beds than any of the other beds, and if they all really belong to the Pliocene, we should expect considerable climatic differences in their fauna. For the present it is impossible to say, whether the few relations on the one side, and the differences on the other, indicate a difference in age, or a difference in geographical position. The solution of this question may be obtained by a closer examination of corresponding beds in the north- ern parts of Patagonia (from San Julian northward to Entrerios) : we have numerous indications that such beds really exist there. ORTMANN : TERTIARY INVERTEBRATES. 309 Mr. Hatcher's observations indicate the existence of such beds near San Julian. He discovered marine beds unconformably overlying the Pata- gonian beds at Darwin Station (Hatcher, 1900 a, p. 108). Here he col- lected only two species : Ostrea patagonica and Trophon laciniatus var. inornatus. The latter form most distinctly points to the Cape Fair- weather beds, but the oyster is different. It is this the southernmost locality at which the true O. patagonica has been found, and the associa- tion of this Entrerios oyster with the Cape Fairweather Trophon suggests very strongly that both deposits may be identical in age, and that their difference may be due to their geographical location : then the Miocene Patagonian Ostrea ingens would remain the identical species in Pliocene times in the south, while it changes into O. patagonica in the Pliocene farther north. It is not impossible that our locality at Darwin Station is identical with one of the type localities of Ameghino for the marine Tehuelche beds. V. Ihering (1897) mentions four species ( O. ferrarisi = patagonica, Pccten actinodes, Scalaria rngulosa var. obsoleta, and Trophon variant = laciniatns var. inornatus] from a locality between Santa Cruz and San Julian, which he spells : Santa Rosa (pp. 225, 277, 296), Punta Rosa (p. 227) and Pta. or P. Rasa (pp. 322 and 323). Since the latter form is given for the same species, for which Santa or Punta Rosa is quoted, there is no doubt that the same place is intended. Mr. Hatcher informs me that he has the vague impression that the peninsula between the bay of San Julian and the sea is called " Punta Raza" by the sailors. If that is true, it is very probable that our locality at Darwin Station is not very far from, if not identical with Ameghino's Punta Rasa, since it is situated near the base of this peninsula. Punta Rasa is said to represent "Tehuelche" beds, and of the four species mentioned by v. Ihering, we possess 2 from Dar- win Station (Ostrea patagonica and Trophon inornatus], and 2 from Cape Fairweather (Pecten actinodes and Trophon inornatus}. This fact also is much in favor of the view that Hatcher's locality at Darwin Station as well as the Cape Fairweather beds belongs to the same horizon as v. Ihering's and Ameghino's Punta Rasa. Another locality, which corresponds stratigraphically with the Cape Fairweather beds, has been discovered by Mr. Hatcher at Lake Pueyr- redon, where marine beds again overly the Santacruzian beds, and cap the whole "Rio Tarde section" (Hatcher, 1900 a, p. 108). Only two 3IO PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. species have been collected here ; the one is Ostrea ingens in a form which approaches distinctly the Cape Fairweather type of this species, the other is a Pecten, apparently P. gcminatiis, but it is to be remarked that only casts have been found which render the identification doubtful : we may have to deal with P. actinodes. While thus the correlation of the Cape Fairweather beds with other deposits still remains somewhat doubtful, we may safely say that they themselves are of Pliocene age. Further investigations of corresponding beds of other localities, especially of the marine "Tehuelc.he" formation of Ameghino are very desirable, and will probably throw much light upon the Cape Fairweather beds.1 ORIGIN AND DEVELOPMENT OF THE PATAGONIAN MARINE FAUNAS. i. THEORY OF "ANTARCTICA." We may take the marine fauna of the Miocene Patagonian beds as the standard for all fossil Patagonian faunas, since it is the only fauna that we may call "well-known." Of the Magellanian fauna only a small part is known, and the Cape Fairweather beds also contain only a compara- tively small number of species. One of the most striking characters of the Patagonian fauna is, as we have seen above (pp. 299—302), the presence of a number of species which show distinct affinities with New Zealandian and Australian fossils. This rela- tion of Patagonia to New Zealand and Australia is no new feature : it has been observed before in land and fresh-water animals, and also in marine animals and in plants2 by numerous authors, and we possess various theories for the explanation of this remarkable zoogeographical fact. A very good — although not quite complete — review of the theories connected with these relations between the southern continents has been 1 The paper of Borchert (Die Molluskenfauna und das Alter der Parana-Stufe. Stuttgart, 1901) was received after the above was written. The dissimilarity of the Parana and Cape Fairweather faunas is very striking, and the relations between them are still unsettled. 1 A partial list of animal and plant groups, in which coexistence of allied forms in Australia and South America has been observed, is given by Hedley (1895, p. 3, footnote i). ORTMANN : TERTIARY INVERTEBRATES. 311 given by Hedley (1895) who has formulated his own views in the follow- ing words (1. c., p. 6) : "... during the Mesozoic or older Tertiary, a strip of land witli a mild climate extended across the South Pole from Tas- mania to Tierra del Fuego, and . . . Tertiary New Zealand then reached sufficiently near to this Antarctic land, without joining it, to receive by flight or drift many plants and animals!' This theory, which has been worked out more especially in its bearing upon the Australian and Pacific faunas in a later paper by the same author (Hedley, 1899), differs in important points from all theories hitherto advanced, as it demands only a minimum of land extension, and further, as he states expressly (p. 7) that this Antarctic continent ("Ant- arctica") was probably "an unstable area, at one time dissolving into an archipelago, at another resolving itself into a continent." He admits further the existence of certain facts that suggest a former connection of South Africa also with Antarctica. The facts leading to this and the older theories were observed long ago, and consist of a marked similarity in the animal and plant life of the respective continents, a similarity which is also recognizable, as we have seen above, among the fossil marine animals. With the exception of the theory of Wallace (1876, pp. 287 and 461), who believes that the common elements of the southern faunas have been derived from a gen- erally distributed stock, which was pushed by the competition of other animals into the southern ends of the continental masses, where it alone survived, all explanations of this zoogeographical fact have started from the fundamental idea that there must formerly have existed a connection between the respective parts by a land bridge, and opinions differ only as to the location and probable extent of it. As to the time of its existence there is a fairly complete unanimity among the writers on this subject, provided that they have given any expression at all of their opinions on this point; if they construct this bridge for any particular time, it is for the end of the Secondary or the beginning of the Tertiary. Only Forbes (1893) makes an exception by putting his Antarctica into the "Ice age." To my knowledge,1 Hooker was the first to hold the opinion that, with reference to plant life, there may have existed a connection of the differ- ent parts of Antarctic and Subantarctic continents and islands by land. 1 See Ortmann, "The Theories of the Origin of the Antarctic Faunas and Floras" (American Naturalist, v. 35. February, 1901). 312 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. This opinion is the more remarkable, since it was first expressed at a time (Hooker, 1847, P- 2II)> wnen Darwin's "Origin of Species" had not yet been published. Although Hooker hints at this possibility very cau- tiously, he returns to this point in 1853 (p. xxiii ff.) more emphatically, and again in 1859 (pp. xvii and civ), and in this case from a Darwinian point of view (he refers here to Darwin's unprinted "Origin of Species"). His general idea was, that the southern floras indicate one great vegeta- tion, which may once have covered a larger southern area of land; but he leaves it uncertain where was the position of this southern continent, especially he does not connect it with the polar lands of the southern hemisphere. Some of his remarks even indicate that he was in favor of placing this land connection in lower latitudes (about that of Tierra del Fuego and Kerguelen Islands). Among zoologists this theory of former connections of the southern lands was not taken up, until Ruetimeyer (1867, pp. 15 and 23) — but without reference to Hooker — expressed the opinion that the Antarctic continent is to be regarded as a center of a separate development of a certain stock of animals, from which the inhabitants spread northward, and that we should regard the faunal elements common to Australia, South America and South Africa as remnants of this Antarctic fauna. He expresses no opinion on the probable extent and configuration of this southern center, but only says that the assumption of a connection of the three southern land masses with the Antarctic continent would explain many facts of present distribution. The next to discuss this question was Hutton (1873 and 1874). He has practically the same idea as Ruetimeyer, and assumes a former greater extension of land in the southern hemisphere, South America, New Zea- land, Australia and South Africa were connected by a continent, which in its largest extension existed at the beginning of Cretaceous times, but which was not necessarily a single, completely continuous mass at one and the same time. Accepting Wallace's opinion (1876) mentioned above, Hutton subse- quently changed this view (1884), and abandoned the connections of these regions by an extension of the Antarctic continent, especially he no longer believes that South Africa had a connection with it. But he still main- tains that there was a land connection between Australia and South America, and he constructs this bridge across the middle part of the Pa- ORTMANN : TERTIARY INVERTEBRATES. 313 cific Ocean by way of a now submarine plateau (p. 433) "from Guinea and North Australia, through the Fiji and Tonga Islands to Samoa, spreading South to New Zealand and North to the Ellice, Gilbert, Mar- shal, Caroline and Pelew Islands;" another plateau "extends from Chili in a northwest direction to the Society Islands and Cook's Islands, in- cluding Juan Fernandez, Easter Island, the Paumotus and the Marquesas Islands." Thus these two plateaus closely approached each other, if they were not actually connected. Shortly after Hutton's first publication, Gill (1875) presented another somewhat similar view, but this was given in a very vague form. Con- sidering the distribution of fishes, he divided the land masses in two large sections, an Eogcea, comprising Africa, South America and Australia, and a Gznogcea, comprising the rest of the present continental masses. He does not introduce the Antarctic continent at all, and does not give any details of the connection, simply intending this as a zoogeographical di- vision. But the fact that he calls these two sections "areas of derivation or gain from more or less distant geological epochs," and that he refers to them again later (Science, 8 June, 1900), calling them "hemispheres," makes it apparent that he understood his Eogaea as a large continental mass. Thus we have to distinguish, practically, three different theories, aside from Wallace's: (i) The Ruetimeyer-Hutton theory of the connection through an Antarctic continent (1867, 1873); (2) Gill's Eogaea theory (1875); (3) Hutton's theory of 1884, constructing a connection across the mid-Pacific. In all these, the fundamental idea first expressed by ' Hooker, that there must once have been a connection by land, serves as a basis. Gill's theory has never been taken up by anybody else, while the two other theories have been taken into consideration by subsequent writers. Among them we should mention in the first line H. O. Forbes (1893). He practically accepts the first and oldest theory of Ruetimeyer and Hutton, in assuming the former existence of a larger Antarctic continent ; but on the other hand, he goes far beyond Ruetimeyer's and Hutton's ideas, in constructing this continent on a very large scale: his "Antarc- tica," in its coast line, follows nearly what is now the 2000 fathom line, and extends in broad stretches over Australia and New Zealand to the Fiji Islands, to the Mascarene Islands and South Africa, and to South 314 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. America ; in other words, it is an exaggeration of Ruetimeyer's and Hut- ton's conception of it. As to the time of existence of this huge conti- nental mass, Forbes differs from all previous writers in placing it in the "Ice age" of the northern hemisphere. Von Ihering (see: 1891 and 1894) has accepted both theories of Hut- ton with a slight modification, assuming a connection running from South America (Archiplata, see below) by way of Antarctica to Australia, which was in turn connected with a "Pacific continent." This Pacific continent does not correspond exactly to Hutton's (1884) bridge from Australia to Chili, since v. Ihering does not assume a direct connection of it with Chili, and thus v. Ihering's theory conforms more to the Ruetimeyer- Hutton theory. This is shown principally by the fact that for Patagonia v. Ihering (1897) urges chiefly the Antarctic origin of a part of its fauna, not a Pacific origin. * It is not necessary to quote here the large number of other writers, who have pointed out — in connection with their studies in special groups of animals or plants — the allied elements in the faunas and floras of the southern continents, since none of them has materially added to or changed the existing theories ; it may be sufficient to say that all of them — if they have expressed any opinion at all — hold the view that there once existed an Antarctic continental connection between the respective parts, without venturing into a closer discussion of the question as to the probable extent and location of it. We must, however, mention especially the writings of Hedley (1895 and 1899). The main idea of Hedley has been reproduced above (p. 311), and it remains to point out its relation to the theories set forth above. There is no doubt that Hedley's view keeps close to the old Ruetimeyer- Hutton theory in assuming an ancient Antarctic continent. But while Forbes enlarged this continent to an incredible size, Hedley chooses the safest and most conservative way in not extending the Antarctic land beyond its present limits unless absolutely necessary. Thus he leaves the known parts of the Antarctic continent as they are, and extends them only in narrow strips so as to join Australia, South America, and (pos- sibly) South Africa. x 1 The same opinion that we have to restrict the land connections of the southern continents was expressed again by Lydekker (1896, p. 134), but apparently without knowledge of Hedley's article of 1895. For the rest, Lydekker does not favor any particular theory, and even leaves it uncertain whether there was an Antarctic or a Pacific land bridge. ORTMANN : TERTIARY INVERTEBRATES. 315 Finally Osborn (1900, p. 565 and map on p. 566) accepts fully the theory of Antarctica, and, in the main, follows Forbes, although his recon- struction of this old continent by elevation to the 3O4o-meter sounding line is not quite as extensive as that of Forbes. In this respect Osborn's view is intermediate between Forbes' and Hedley's, and decidedly approaches our conception. Thus we see that of the various theories advanced for the explanation of the similarity of the southern faunas, the theories of Gill, Wallace, and also the second theory of Hutton have not been considered seriously by subsequent writers, while the oldest one, formulated by Ruetimeyer, has furnished the fundamental idea for them. One of them, Forbes, has- pushed this idea to an extreme, which we cannot accept by any means, while Hedley has attempted to restrict it to reasonable proportions, and to reconcile it with the zoogeographical facts as well as with the present conditions of distribution of land and water in the southern hemisphere. In this sense, Hedley's specification of the Ruetimeyer-Hutton theory is the most conservative, especially as compared with Forbes' fancies, and it is only natural that we should accept his ideas as the most probable of all, that is to say, we accept the first theory of Ruetimeyer and Hutton, with the restrictions put upon it by Hedley. For our present purpose, this acceptance of the theory of the former existence of an "Antarctica" means that we are of the opinion that the elements of the fossil Patagonian fauna resembling certain forms in New Zealand and Australia are to be regarded as an additional proof of the former connection of South America with Australia and New Zea- land. Since the respective shells are all preeminently inhabitants of the littoral, of shallow water, and since it is very probable that they were unable to cross over large extents of deep sea, a region of shallow water must have formed a connection between both parts, and nothing is more natural than to assume that this shallow water accompanied the coast line of ancient "Antarctica." It does not necessarily follow that this coast line was a continuous line along the uninterrupted shores of a truly con- tinental mass, but it may have consisted of a chain of islands, at least in part. As Hedley urges, we should not regard the Antarctica as a solid continent, but probably it was broken up at certain times into parts, which were united again in one or another direction. This assumption seems to be chiefly supported by the evidence furnished by land animals, and will be discussed elsewhere in this work. 316 PATAGONIAN EXPEDITIONS : PALEONTOLOGY. We wish only to emphasize here the fact that the marine fossil Pata- gonian fauna materially strengthens the theory of "Antarctica" by giving evidence for the former existence of a coast line, at any rate of shallow water, between Australia and New Zealand on the one side, and South America on the other. As to the connection of Africa with Antarctica, hardly any evidence is found among our material ; we should, however, call attention to the fact that the Bryozoan Tennysonia subcylindrica of the Patagonian beds is extremely closely allied to the only known species of the genus, T. stellata, which is recorded from the Cape of Good Hope. This instance would hardly have any value if it was an isolated one. But other groups of animals have furnished similar cases, and, although these are less pronounced and less frequent than the cases of relations between South America and Australia, we must take them into account, and grant a former extension of Antarctica in the direction toward South Africa. In the map (pi. XXXIX) accompanying this report we have tried to reconstruct ancient Antarctica : it has been assumed that the Antarctic portions of land known at the present time (the region around Graham Land ; Victoria and Wilke's Land ; Enderby and Kemp Land) form parts of a still existing Antarctic continental mass ; we have not tried to enlarge the boundaries of this continent, except only to such a degree that a con- nection is formed with the present southern ends of the continents of Australia, South America and South Africa. As regards Australia (and New Zealand), we have followed Hedley's idea, as expressed in his map in his second paper (1899, p. 404); as to the connection with South America we have followed chiefly the tectonic relations known to exist between Tierra del Fuego, South Georgia, South Sandwich and Graham Land, as represented by Fricker (1900, chiefly p. 140 ff.); and as to the much more doubtful connection with South Africa, we have taken into account chiefly the results of the German Valdivia Expedition, as pub- lished by Chun (1900, Lieferung 4).1 'As to the tectonic configuration of Antarctica, and the evidence thus furnished for its former connections with Australia, New Zealand and South America, compare the article of Gregory (Gregory, T. W. The work of the National Antarctic Expedition in : Nature, vol. 63, No. 1643, 25th April, 1901, pp. 609-612), and the sketch map given by him (p. 611). Unfortunately this very important note was published after the above was written ; it supports, however, in a large part the ideas set forth above. ORTMANN : TERTIARY INVERTEBRATES. 317 Although in our map Antarctica has been drawn as a continental mass, we have mentioned above that it was — at certain times — possibly broken up into archipelagoes. Antarctica may have been a continent once, but it is hard to say at what time. We may safely say — and all authors ex- cept Forbes agree in this — that it existed about the close of the Cretaceous and the beginning of Tertiary time, but we do not know anything beyond that. In this respect it is interesting to see what evidence is furnished by the geological configuration of southern Patagonia. Through Mr. Hatcher and others we know that the Cretaceous ends with a series of deposits, called Guaranitic beds (see Hatcher, 1900 a, p. 93), which indicate a gen- eral upheaval of the land. After the deposits of these beds the respective parts were land, and no deposits were formed till the beginning of Oligo- cene times (Magellanian beds). From this time on we have a slow subsi- dence, which reaches its maximum in Patagonian time (lower Miocene), and then follows, in Santacruzian time (upper Miocene): another upheaval, which culminated, possibly, in the final formation of the Cordilleras at the close of the Miocene. Within these general movements, there were a number of smaller oscillations, for instance like that indicated by the Upper Lignites (Hatcher, 1900 a, p. 99). It is beyond the scope of this report to go more into detail ; but we may say here that the geology of southern Patagonia points to a maximum extent of land at the end of Secondary time and during the Eocene, and to a large — if not maximum — extent of water during lower Miocene time. If it is permitted to draw any conclusions from this, we should put the largest extent of Antarctica at the end of the Cretaceous and in the Eocene, while a marked, if not final, interruption was brought about in the lower Miocene. Within this time, smaller, and more or less important oscillatory movements took place. This refers, however, only to the history of the Antarctic continent in the first part of the Tertiary period. In Cretaceous times similar move- ments may have taken place, so that the connection of Antarctica with the present continents (or parts of them) may have been established and destroyed repeatedly. And indeed v. Ihering (1894, pp. 405 and 425) dates some connections of America and Australia with Antarctica far back in Mesozoic times. (Compare v. Ihering's Mesozoic " Archinotis," 1893. P- 9-) And further, the above refers only to South America. It is not at all necessary that the connection between Antarctica and the Austral lands 3l8 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. should have coexisted with that between Antarctica and South America ; indeed, it is quite possible that the one was interrupted when the other was established. The same refers to the connection of Antarctica with South Africa. As regards the latter point, there is no doubt that some evidence in favor of this connection has been found.1 But this evidence is much less distinct than in the case of the other two continents. Possibly the junc- tion of South Africa with Antarctica is to be sought for far back in Meso- zoic time, or it was, in the Tertiary, only of short duration. As to the reconstruction of this bridge, we must pay due attention to the fact that great depths of the sea have been discovered by the "Valdivia" to the south of the Cape of Good Hope (see Chun, 1900, p. 180 and map by G. Schott, ibid. Lieferung 4). Although great depth of the present sea is by no means a decisive argument against the former existence of land (as for instance Chun believes), it is better, if no other evidence is forth- coming, to be as conservative as possible, and not to interfere with these great depths. In our map we have given two indications for this land bridge : the one going from Enderby, or possibly Wilke's Land, by way of the Kerguelen, Crozet and Prince Edward Islands, the other following the submarine ridge in the South Atlantic indicated by Schott in his map, and connecting southwest Africa with the Falkland Islands by way of Tristan da Cunha. Which one of these bridges, or whether either of them, is correct, we have at present no means of deciding.2 Although there is still much room for speculation, we wish to emphasize the fact that the fossil marine animals of Patagonia distinctly point to this old connection of South America with Antarctica at the end of Cretaceous and the beginning of Tertiary times, and that Antarctica in turn was at some time connected with Australia and possibly with New Zealand. As Hedley maintains, there was no continental connection with the latter ; we cannot decide this question, since we treat only of marine literal ani- mals, and for them a close vicinity of the respective literal waters is suffi- cient. The Patagonian fauna demands a theory that assumes a compara- 1 1 should like particularly to call attention to the presence of the freshwater fish Galaxias capensis at the southwestern corner of Africa (see Weber, 1 897, p. 1 97). 2 The extremely uneven and rugged bottom of the sea between Enderby Land and the Ker- guelen Islands, as described by Chun, is in favor of the first assumption ; as regards the second, I refer the reader to what Weber (1897, p. 198) says about a direct communication of south- west Africa with v. Ihering's " Archiplata." ORTMANN : TERTIARY INVERTEBRATES. 319 tively shallow sea, where there is now deep water, and when we make this assumption, there is no difficulty in constructing a continental connection of the same parts, as soon as other lines of evidence force us to do so. This communication of the Patagonian seas with Antarctica through shallow water persisted through a large part of the Tertiary, probably almost up to the recent time. V. Ihering has already assumed a repeated and continuous immigration of marine Antarctic forms into the South American literal (v. Ihering, especially 1897 b, pp. 532 and 533), and there is no reason to reject this theory. That during Tertiary times a direct communication was present with the nearest parts of Antarctica, is quite probable after the discovery of marine fossils on Seymour Island, Dirck Gherritz Archipelago (Graham Land). The fossils found there, especially the presence of a species of Cucullcea, strongly suggest Patagonian beds1). 2. RELATIONS OF THE PATAGONIAN DEPOSITS TO OTHER PARTS OF SOUTH AMERICA, AND TO THE REST OF THE WORLD; THEORY OF "ARCHIPLATA" AND "ARCHHELENIS." We have seen above that the most characteristic feature of the Pata- gonian fauna is the dissimilarity to other faunas of about the same age, the resemblance to the Antarctic faunas of Australia and New Zealand 1 It may be well to state here the facts about these Tertiary fossils, since they have been re- peatedly mentioned lately, but without proper quotations, so that it is difficult to keep track of the literature. These fossils were collected in the season 1 892-3 on Seymour Island, Dirck Gherritz Archi- pelago (northeast of Graham Land) in 64° 24' S. Lat, by Captain Larsen of the "Jason," and given to Dr. C. W. Donald of the " Active," who brought them back to Scotland. Dr. J. Mur- ray was the first to notice them in the Geographical Journal, vol. 3, January, 1894, p. 1 1, foot- note, and he says that — according to Messrs. G. Sharman and E. T. Newton, of the Geological Survey, — they belong to the genera Citcull&a, Cytherea, and Natica (besides pieces of Coniferous wood). He points to the probable Tertiary character of them, and compares them with lower Tertiary fossils of England and Patagonia, which would especially refer to the Cucullaa. Dr. Murray again refers to these fossils in the Scottish Geographical Magazine, vol. 10, April, 1894, p. 195, and they are also mentioned in Peterson's report on Captain Larsen's discoveries (Peterson, Die Reisen des " Jason " und der " Hertha," in : Mitteilungen der Geographischen Gesellschaft in Hamburg. 1891-92. 1895, p. 273). Again they have been mentioned by Hedley(i89S, p. 7), by Heilprin (Science, Febr. 28, 1896, p. 306), by Ohlin (Ymer, 1898, h. 4, p. 301), and by Fricker (1900, p. 182), but with insufficient and partly incorrect quotations. 32O PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. excepted. Among the contemporaneous deposits in South America, only the Navidad fauna of Chili is related to the Patagonian, and, indeed, in such a degree that we are able to draw valuable conclusions from the comparison with it. If we come to compare the Patagonian deposits with other American deposits, the close affinity disappears. In this respect one of the most important points is the dissimilarity of the Miocene fauna of northern Peru (Payta and Tumbez), as described by Grzybowski (1899). This fauna possesses only a few features in common with the Navidad fauna, but hardly any with the Patagonian, and the most important is the pres- ence of a species of Struthiolaria in these beds. For the rest, this fauna (called Ecuadorian province by Grzybowski) shows close affinities to the Caribbean province, /'. e., the Miocene deposits of the West Indies, and even in some respects to the European Miocene.1 We have seen above that it was possible to compare our Patagonian material to some extent with Miocene faunas of Europe and North America, and the latter is in a certain degree a dependency or offshoot of the West Indian fauna, but these relations are only very general, consisting in a more or less close affinity of species, but hardly in identity. (The fact that we did not find a larger number of relations to the West Indies is probably due to the chiefly Oligocene, not Miocene, age of the latter beds, according to Ball.) Thus we have on the one hand a close relation of the Patagonian beds to the Chilian, which extends to an identity of a num- ber of species, while, on the other hand, with the Caribbean and European provinces only remote relations can be established. The deposits of northern Peru are closely connected with the Caribbean province, although a few closer relations with the Chilian beds are recognizable. For the explanation of these facts we have a theory propounded by v. Ihering, which we may conveniently call his Archiplata-Archhelenis-theory (v. Ihering, 1891, pp. 434, 437, and especially : 1893 ; here on p. 9 a list of other publications by the same author referring to the same subject ; and 1894 p. 404 ff.). Von Ihering maintains that South America is not a genetic unit, but consists of two parts, which became united subsequently : a southern part 'According to Ball (1898 b and in : Science, Novemb. 23, 1900, p. 808) the so-called Miocene of the West Indies is really Oligocene, and this would possibly affect slightly Grzybowski's deter- mination of the age of the Peruvian beds. ORTMANN : TERTIARY INVERTEBRATES. 321 which comprises what is now Chili, Argentina, and southern Brazil, and which he calls " Archiplata," and a northern part comprising chiefly northern Brazil and Guiana, which he calls "Archibrazil," resp. "Archi- guyana," or " Archiamazonas." This latter part was connected — in Meso- zoic times — with West Africa by way of St. Helena, and he calls this continental mass "Archhelenis." Archiamazonas or Archhelenis were separated completely from Archi- plata by a broad stretch of sea, which extended across the present conti- nent, where is now the valley of the Amazon River, the Cordilleras not being yet formed, and thus a broad communication existed between what is now the Atlantic and Pacific Oceans.1 Since Archiamazonas was connected with Africa, and this in turn in a certain degree with India and Europe, the assumption of this old conti- nent Archhelenis (which, by the way, differs only in the supposed geo- graphical position from the "Atlantis" of previous writers) explains sat- isfactorily the relations of the marine faunas of the West Indies (and southern North America) with Europe. The connection of Archiplata with Antarctica explains its relations to Australia and New Zealand, and the existence of a broad connection of the Atlantic and the Pacific sepa- rating Archiplata and Archhelenis (or Archiamazonas) explains the dis- similarity of the southern and northern faunas of South America. Nevertheless, as we have seen, there are some remote affinities between Patagonia and the West Indies, and even Europe, and apparently this is due to the fact that — confining our view to marine animals — communica- tion was possible in a certain degree between the shores of Archiplata and Archiamazonas. This fact is most plainly seen in the presence of Navidad fossils in corresponding deposits of northern Peru : the Navidad beds were apparently deposited near the northwestern extremity of Archi- 1 A former connection of the Atlantic and Pacific oceans up to Miocene times has been gener- ally accepted, but this connection was placed chiefly, where is now the Isthmus of Panama. Ac- cording to Hill, however (The Geological History of the Isthmus of Panama, etc. in : Bull. Mus. Comp. Zool. Harvard Coll., vol. 28, 1898), a Central American barrier separating both oceans has existed since Jurassic times, which was only temporarily interrupted at the close of the Eocene. On the other hand, Hill admits that there must have been a broad and important con- nection of oceans somewhere, but not at Panama. It seems to me that this broad connection is to be found in the sea across South America mentioned above, and, personally, I am much in favor of the theory that the communication of the Atlantic and Pacific took place, not at Panama, but farther south in the sea separating v. Ihering's Archhelensis and Archiplata (see : Science, December 14, 1900, p. 929). 322 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. plata, and the Payta-Tumbez beds near the southwestern extremity of Archiamazonas : at these places we have probably these two continents at their points of nearest approach to each other, and an exchange of marine forms may have been possible. This exchange, however, was rendered more difficult by climatic conditions, and especially, although possible in a certain degree between the Ecuadorian and the Chilian prov- ince, it was hard for the Caribbean and Ecuadorian fauna to migrate far- ther south, into the Patagonian region. This leads us to investigate the probable climatic conditions of Pata- gonian times. There is no doubt that in Miocene time, when the Pata- gonian beds were deposited, a considerably warmer climate prevailed in these regions, a fact that has been recognized by all previous writers on this question (see for instance v. Ihering, 1897 b, P- 535)- Of the fossils of the Patagonian beds the following point to a warm, tropical or sub- tropical climate: Scutella, Perna, Cucullcea, Dolium, Ficula, Mtirex sub- gen. Phyllotonus, Terebra, Drillia, Borsonia. A part of these genera has also been found in the Navidad beds. On the other hand, a number of characteristic tropical genera are missing in the Patagonian beds, which have been found in the corresponding Chilian beds, for instance : Conus, Mitra, Oliva, Cyprcea, Cassis, Columbella. Of these, Mitra, Oliva, and Columbella are represented in the Miocene of northern Peru, and the pres- ence of a Strombus in the latter beds (upper Miocene) points still more to a truly tropical character. Thus we may say that the Patagonian beds — although still retaining a subtropical character — differ distinctly from the Navidad beds in the lack of some features, which depend probably on the climatic conditions, and that the Navidad beds approach in just these features the tropical depos- its of the Caribbean province, where genera like Conus, Strombus, Mitra, Oliva, Columbella, Cyprcea form a very prominent element of the fauna. In this connection we may also mention the complete and very singular lack of the genus Cerithium in Patagonia. The Navidad beds contain only a single species of this genus, which is in striking contrast to the contemporaneous deposits of the northern hemisphere. While we thus should claim for the Patagonian beds at least a sub- tropical climate, we have, on the other hand, in them the beginning of a differentiation of an "Antarctic" fauna, or, more properly, speaking, of an extratropical fauna of the southern hemisphere, which is a pendant to, but ORTMANN I TERTIARY INVERTEBRATES. 323 entirely different from the extratropical fauna to the north of the tropics. This fauna, of course, developed as soon as the climatic differentiation of the South Pole offered the necessary conditions, and we may say that for a large part of this fauna the shores of the supposed Antarctic continent formed the center of origin. Since Miocene Patagonia was apparently a part of this center, it is only natural that we should find here indications of this Antarctic fauna. Some elements of the latter in the Patagonian beds are no doubt the ancestors of corresponding forms still living in these regions, and a few of them have not changed at all, so that they must be regarded as iden- tical species, for instance (compare p. 289): Aspidostoma giganteum, Tere- bratella dorsata, Mytilus magellanicus, Infundibulum corrugatum, Infun- dibulum clypeohim, Verruca Icevigata, Balanus psittacus. One species, Mytilus chorus, is no longer found in southern Patagonia, but has retreated a little northward, to Chili. Other species are no longer found in South America: Cellaria fistiilosa (otherwise almost cosmopolitan), Heteropora pelliculata (New Zealand and Japan), Rhynchonella sqtiamosa (Kerguelen Islands), Magellania lenticularis (New Zealand). In other cases, the fossil and living species are to be regarded as dif- ferent, but they are apparently genetically connected. This is the case in the following genera (see v. Ihering, 1897 b, p. 532): Valuta, Trophon, Turritella, Natica, Vemis, Meretrix, Dosinia, Pecten, to which we may add Bouchardia. This is especially remarkable in the genus Valuta, where three chief types of living Patagonian Volute, V. ancilla, magel- lanica, and brasiliana, have their prototypes in the following species : V. dorbignyana, dotneykoana and ameghinoi. While all these forms are characteristic of the American part of the Antarctica, some of them, for instance the Volute, reappear in similar types in Australia and New Zealand. The same is true of Siphonalia domeykoana, which does not seem to exist at present in South America, but it is still represented in the recent seas of New Zealand in S. dila- tata, and analogous are the cases of Struthiolaria, Malletia, Sigapatella. Struthiolaria is found fossil in South America (Oligocene of Patagonia, Miocene of Patagonia, Chili, and northern Peru) and in New Zealand (Miocene upward) and still lives in New Zealand waters. Malletia is known fossil from the Miocene of Patagonia, Chili, and New Zealand, and living from Chili and New Zealand. Sigapatella is found in the Miocene 324 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. of Patagonia and Chili, from the Oligocene upward in New Zealand, and lives in New Zealand waters. Perhaps we may put into this category the subgenus Cominella of Buccinum, which, although known fossil in the northern hemisphere, is at present confined to the southern, the metrop- olis of the typical species being New Zealand (see Tyron, 1881, p. 201). Finally, in Terebmtella dorsata, we have a species that in the fossil state is common to New Zealand and Patagonia, while at the present time it is extinct in New Zealand, but survives in Patagonia. Thus we see that, in the Miocene Patagonian beds, we must distinguish two chief faunal elements : a tropic-subtropical one, which shows relations to the tropical parts of the rest of the earth (and through these with the subtropical faunas of the northern hemisphere, in Europe and North America), and an antarctic element, which is peculiar to the southern hemisphere, and which shows relations only to the faunas belonging to or connected with ancient Antarctica. The first element was the chief factor that enabled us to compare the Patagonian beds with deposits of the northern hemisphere, and thus to ascertain their age, while the other has given us valuable hints for the comparison with New Zealandian and Australian beds. For the later history of the Patagonian marine fauna the Cape Fair- weather beds are valuable. While their fauna shows on the one side a continuation of Patagonian types (see p. 307), we have here, on the other side, an introduction of new forms. These new elements are possibly in large part new immigrants from the Antarctic shores, and, as v. Ihering (1897 b) urges, this immigration from the south must have continued on a large scale almost up to the present time, at least to the Pleistocene. This later introduction of Antarctic elements is the most important change in the general character of the fauna that has taken place since the time of the deposition of the Patagonian beds, and is — in connection with the retreat of the tropical elements toward the north — the most prominent feature that distinguishes the present Patagonian, Magellanian, and Chilian faunas from the fauna that lived on the shores of ancient "Archiplata." ORTMANN : TERTIARY INVERTEBRATES. 325 BIBLIOGRAPHY. Agassiz, A. 1872 Revision of the Echini (Illustrated Catalogue of the Museum of Comparative Zo- '73 °l°gy at Harvard College, No. 7, parts I and 2, 1872 ; part 3, 1873). Agassiz, L. 1841 Monographies d'Echinodermes vivans et fossiles, 2 ; Des Scutelles, 1841. Agassiz, L., and Desor, E. 1846 Catalogue raisonne des families, des genres et des especes de la classe des Echino- '47 dermes (Annales des Sciences naturelles, ser. 3, vol. 6, 1846; ser. 3, vol. 7, 1847). Ameghino, F. 1889 Contribucion al conocimiento de los Mammiferos fosiles de la Republica Argentina, 1889. 1894 Enumeration synoptique des especes de Mammiferes fossiles des formations Eocenes de Patagonia, 1894. 1898 Sinopsis Geologico-Palaeontologica. Segundo Censo Nacional de la Republica Argen- '99 tina, vol. i, 1898; Suplemento, 1899. Basterot, B. de. 1825 Memoire geologique sur les environs de Bordeaux, 1825. Bazin. 1884 Sur les Echinides du Miocene moyen de la Bretagne (Bulletin de la societe geologique de France, ser. 3, vol. 12, 1884). Berg, C. 1898 Substitucion de nombres genericos (Communicaciones del Museo Nacional de Buenos Aires, vol. I, 1898). Bcettger, 0. 1870 Neue Conchylien des Mainzer Tertiaerbeckens (Palaeontographica, vol. 19, 1870). Busk, G. 1854 Catalogue of marine Polyzoa in the collection of the British Museum, part 2, 1854. 1859 A monograph of the fossil Polyzoa of the Crag (Palaeontographical Society, 1859). 1875 Catalogue of marine Polyzoa in the collection of the British Museum, part 3, Cyclosto- mata, 1875. 1879 On recent species of Heteropora (The Journal of the Linnean Society, Zoology, vol. 14, 1879). 1884 Report on the Polyzoa collected by H. M. S. Challenger, part I, Cheilostomata (Re- port on the scientific results of the voyage of H. M. S. Challenger, Zoology, vol. 10, 1884). Chemnitz, J. N. 1785 Neues systematisches Conchylien-Cabinet, vol. 8, 1785. 326 PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. Chun, C. 1900 Aus den Tiefen des Weltmeeres, Schilderungen von der deutschen Tiefseeexpedition, 1900. Clark, W. B. 1896 The Eocene deposits of the middle Atlantic slope in Delaware, Maryland and Virginia (Bulletin of the U. S. Geological Survey, No. 141, 1896). Clessin, S. 1889 Die Familie der Mytilidae, in : Martini and Chemnitz, Systematisches Conchylien- Cabinet, vol. 8, Abteil. 3, 1889. Conrad, T. A. 1849 Fossils from northwestern America, 4, Mollusca in : Dana, U. S. Exploring Expedi- tion, vol. 10, Geology, 1849. 1856 Descriptions of the fossil shells in : Report of explorations and surveys to ascertain the most practicable and economical route for a railroad from the Mississippi River to the Pacific Ocean, vol. 5, 1856. Cossman, M. 1897 The Gasteropods of the older Tertiary of Australia, Les Opisthobranches (Transactions of the Royal Society of South Australia, vol. 21, 1897). 1898 Revue critique de Paleozoologie, vol. 2, No. 3, July, 1898. 1899 Description de quelques coquilles de la formation Santacruzienne en Patagonie (Jour- nal de Conchyliologie, 1899, No. 3). Dall, W. H. 1889 A preliminary catalogue of the shell-bearing marine Mollusks and Brachiopods of the southeastern coast of the United States (Bulletin U. S. National Museum, No. 37, 1889). iSgoa Preliminary report on the collection of Mollusca and Brachiopoda. Scientif. res. explor. Albatross (Proceedings of the U. S. National Museum, vol. 12, 1890). 18905 Contributions to the Tertiary fauna of Florida (Transactions of the Wagner Free Insti- '92, 'gSa tute of Science, vol. 3, part i, 1890; part 2, 1892; part 4, 1898). i8g8b A table of the North American Tertiary horizons, correlated with one another and with those of western Europe, with annotations (i8th Annual Report of the U. S. Geolog- ical Survey, part 2, 1898). 1900 Pelecypoda in : Zittel and Eastman, Text-book of Palaeontology, vol. I, 1900. Darwin, C. 1846 Geological observations on South America, 1846. 1851 A monograph on the fossil Lepadidae of Great Britain (Palaeontographical Society, 1851). A monograph on the fossil Balanidae and Verrucidse of Great Britain (Palaeontographical Society, 1854). A monograph on the Sub-Class Cirripedia, Balanidas, etc., 1854. Davidson, T. 1852 Descriptions of a few recent species of Brachiopoda (Proceedings of the Zoological Society of London, 1852). 1880 Report on the Brachiopoda dredged by H. M. S. Challenger (Report on the scientific results of the voyage of H. M. S. Challenger, Zoology, vol. i, 1880). 1886 A monograph of recent Brachiopoda, part i (Transactions of the Linnean Society of '87 London, ser. 2, Zoology, vol. 4, part I, 1886) ; part 2 (ibid., part 2, 1887). ORTMANN I TERTIARY INVERTEBRATES. 327 Deshayes, G. P. i860 Description des animaux sans vertebres decouverts dans la basin de Paris, vol. I, 1 860 ; '64, '66 vol. 2, 1864; vol. 3, 1866. Desor, E. 1846 Sur quelques oursins fossiles de la Patagonie (Bulletin de la Societe geologique de France, ser. 2, vol. 4, 1846). 1858 Synopsis des Echinides fossiles, 1858. Dcering, D. A. 1882 Geologia in : Roca, D. J. A., Informe oficial de la comision cientifica agregada al estado mayor general de la expedicion al Rio Negro, vol. 3, 1882. Dusen, P. 1899 Ueber die tertiaere Flora der Magellanslaender (Svenska Expeditionen till Magellans- laenderna, vol. I, No. 4, 1899). Forbes, H. 0. 1893 The Chatham Islands, their relation to a former southern continent (Royal Geographical Society, Suppl. paper, vol. 3, 1893); extract : Antarctica, a supposed former southern continent (Natural Science, vol. 3, 1893, pp. 54-57). Flicker, K. 1900 The Antarctic Regions, London, 1900. Gabb, W. M. 1869 Palaeontology of California, vol. 2, 1869 (Geological Survey of California). GUI, T. 1875 On the geographical distribution of fishes (The Annals and Magazine of Natural His- tory, ser. 4, vol. 15, 1875). Gray, J. E. 1867 Notes on the specimens of Calyptraeidae in Mr. Cuming's collection (Proceedings of the Zoological Society of London, 1867). Grzybowski, J. 1899 Die Tertiaerablagerungen des ncerdlichen Peru und ihre Molluskenfauna (Neues Jahrbuch fuer Mineralogie, Geologic und Palaentologie, Beilage Band 12, 1899). Harris, G. F. 1897 Catalogue of the Tertiary Mollusca in the department of geology, British Museum. Part I. The Australasian Tertiary Mollusca. London, 1897. Hatcher, J. B. The Cape Fairweather beds ; a new marine Tertiary horizon in southern Patagonia (The American Journal of Science, vol. 4, 1897). On the Geology of southern Patagonia (ibid.), igooa Sedimentary rocks of southern Patagonia (ibid., vol. 9, 1900). igoob In : Science, February 16, 1900, p. 363 ff. Hedley, C. 1895 Considerations on the surviving refugees in Austral lands of ancient Antarctic life (Proceedings of the Royal Society of N. S. Wales, August 7, 1895). 1899 A zoogeographic scheme for the Mid-Pacific (Proceedings of the Linnean Society of N. S. Wales, 1899, part 3). 328 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. Hincks, T. 1880 A history of the British marine Polyzoa, 1880. 1881 Contributions toward a general history of the marine Polyzoa. (The Annals and Magazine of Natural History, ser. 5, vol. 7, 1881). Hcernes, M. 1856 Die fossilen Mollusken des Tertiaerbeckens von Wien., Bd. i, Univalven (Anhandlungen '70 der K. K. geologischen Reichsanstalt, vol. 3, 1856) ; Bd. 2, Bivalven (ibid., vol. 4, 1870). Hooker, J. D. 1847 The Botany of the Antarctic voyage of H. M. discovery ships Erebus and Terror (Flora Antarctica), part 2, 1847. 1853 Introductory essay to the flora of New Zealand (Reprint from vol. i of the " Flora of New Zealand," 1853). 1859 On the flora of Australia, its origin, affinities, and distribution (Botany of the Antarctic expedition, part 3, Flora of Tasmania, vol. i, 1859). Button, F. W. 1873 Catalogue of the Tertiary Mollusca and Echinodermata of New Zealand, 1873. 1873 On the geographical relations of the New Zealand fauna (Transactions of the New Zea- '74 land Institute, vol. 5, 1873, reprint in: The Annals and Magazine of Natural History, ser. 4, vol. 13, 1874, pp. 25-39, 85-102). 1884 On the origin of the fauna and flora of New Zealand (New Zealand Journal of Science, January, 1884, reprint in : The Annals and Magazine of Natural History, ser. 5, vol. 13, June 1884, p. 435 ff.). i88sa Sketch of the Geology of New Zealand (The Quarterly Journal of the Geological So- ciety of London, vol. 41, 1885, pp. 191-220). l885b On the correlations of the " Curiosity-shop bed " in Canterbury, New Zealand (ibid., pp. 547-564). 1886 New species of Tertiary shells, and : The Wanganui System (Transactions and Proceed- ings of the New Zealand Institute, vol. 18, 1886, pp. 333 ff., and 336 ff.). Ihering, H. von. 1891 On the ancient relations between New Zealand and South America (Transactions of the New Zealand Institute, vol. 24, 1891). 1893 Das neotropische Florengebiet und seine Geschichte (Engler's botanische Jahrbuecher, vol. 17, Heft 5, 1893). 1894 Die Ameisen von Rio Grande do Sul (Berliner entomologische Zeitschrift, vol. 39, Heft 3, 1894). 1896 Zur Kenntnis der suedamerikanischen Voluta und ihrer Geschichte (Nachrichtsblatt der deutschen malakozoologischen Gesellschaft, 1896, nos. 7 and 8). Os Molluscos dos terrenes terciarios da Patagonia (Revista do Museu Paulista.vol. 2, 1 897). Zur Geschichte der marinen Fauna von Patagonian (Zoologischer Anzeiger, 110.548,1897). 1899 Die Conchylien der Patagonischen Formation (Neues Jahrbuch fuer Mineralogie, Geologic und Palaeontologie. Jahrgang, 1899, Bd. 2). Jullien, J. 1891 Bryozoaires in : Mission scientifique du Cap Horn, vol. 6, Zoologie, part 3, 1891. Kissling, E. 1896 Die Fauna des Mittel-Oligocasns in Berner Jura (Anhandlungen der Schweizerischen Palaeontologischen Gesellschaft, vol. 22, 1896). ORTMANN I TERTIARY INVERTEBRATES. 329 Kcenen, A. von. 1867 Das marine Mittel-Oligocaen Nord-Deutschlands. I Theil (Palaeontographica, vol. 16, '68 1867); 2 Theil (ibid., 1868). Kuester, C. H., and Kobelt, W. 1878 In : Martini and Chemnitz, Systematisches Conchylien-Cabinet, vol. 3, Abtheil. 2, 1878. Lahille, F. 1895 Contribucion al estudio de las Volutas Argentinas (Revista del Museo de la Plata, vol. 6, 1895. 1896 Variabilite et affinites du Monophora darwini (ibid., vol. 7, 1896). 1898 Notes sur le nouveau genre Iheringia (ibid., vol. 8, 1898). 1899 Notes sur Terebratella patagonica (ibid., vol. 9, 1899). Loriol, P. de. 1875 Description des Echinides Tertiaires de la Suisse (Abhandlungen der Schweizerischen Palaeontologischen Gesellschaft, vol. 2, 1875). Lydekker, R. 1896 A geographical history of Mammals. Cambridge, 1896. Mallard and Fuchs, E. 1873 Notes sur quelques points de la Geologic du Chili (Annales des Mines, ser. 7, vol. 3, 1873). McClelland, J. 1840 On Cyrtoma, a new genus of fossil Echinida (The Calcutta Journal of Natural History, vol. i, 1840). Medlicott, H. B., and Blanford, H. T. 1879 A manual of the Geology of India, part I, 1879. Mercerat, A. 1893 Contribucion a la Geologia de la Patagonia (Anales de la Sociedad cientifica Argen- tina, vol. 36, 1893). 1896 Essai de classification des terrains sedimentaires du versant oriental de la Patagonie australe (Anales del Museo Nacional de Buenos Aires, ser. 2, vol. 2, 1896). Middendorf, A. T. von. 1849 Beitraege zu einer Malacozoologia Rossica, 2 (Memoires de 1'Academie Imperiale des Sciences de St. Petersbourg, ser. 6, vol. 6, 1849). Milne-Edwards, A. 1860 Histoire des Crustaces Podophthalmaires fossiles (Annales des Sciences naturelles, ser. 4, vol. 14, 1860). Mcerch, 0. A. L. 1861 Review of the Vermetidae (Proceedings of the Zoological Society of London, 1861). Moericke, W. (and Steinmann, G.). 1896 Die Tertiaerbildungen des ncerdlichen Chile und ihre Fauna (Neues Jahrbuch fuer Mineralogie, Geologic und Palaeontologie, Beil. Bd., 10, 1896). Nicholson, H. A. 1880 On the minute structure of the recent Heteropora neozelanica (The Annals and Maga- zine of Natural History, ser. 5, vol. 6, 1880). 330 PATAGONIAN EXPEDITIONS \ PALEONTOLOGY. Nordenskjceld, 0. 1898 Ueber die posttertiaeren Ablagerungen der Magellanslaender, nebst einer kurzen Ueber- sicht ihrer tertiaeren Gebilde (Svenska Expeditionen till Magellanslaenderna, vol. i, No. 2, 1898). Orbigny, A. d'. 1842 Voyage dans 1'Amerique meridionale, vol. 3, part 4, Paleontologie, 1842 ; vol. 5, part '43 3. Mollusca, ,1843. 1847 Geologic. Atlas in : Voyage au Pole Sud et dans 1'Oceanie sur les corvettes 1'Astro- labe et la Zelee, 1847 (no text has been published). 1852 Paleontologie Francaise. Terrains Cretaces, vol. 5, Bryozoaires, 1852; id., vol. 6, '60 Echinoides irreguliers, 1 860. Ortmann, A. E. 1897 On some of the large oysters of Patagonia (The American Journal of Science, vol. 4, Preliminary report on some new marine horizons discovered by Mr. J. B. Hatcher near Punto Arenas, Chili (ibid., vol. 6, 1898). 1899 The fauna of the Magellanian beds of Punta Arenas, Chili (ibid., vol. 8, 1899). 1900 Synopsis of the collections of Invertebrate fossils made by the Princeton Expedition to southern Patagonia (ibid., vol. 10, 1900). Osborn, H. F. 1900 The geological and faunal relations of Europe and America during the Tertiary period and the theory of the successive invasions of an African fauna (Science, April 13, 1900). Philippi, R. A. 1845 Abbildungen und Beschreibungen neuer oder wenig gekannter Conchylien, vol. I, 1845. 1846 Verzeichnis der in der Gegend von Magdeburg aufgefundenen Tertiaerversteinerungen (Palaeontographica, vol. i, 1846). 1851 Abbildungen und Beschreibungen neuer odor wenig gekannter Conchylien, vol. 3, 1851. 1887 Die tertiaeren und quartaeren Versteinerungen Chiles, 1887. Pilsbry, H. A. '89, '91 Manual of Conchology, vol. n, 1889; vol. 13, 1891. 1897 Patagonian Tertiary fossils (Proceedings of the Academy of Natural Sciences of Phila- delphia, 1897). Pilsbry, H. A., and Sharp, B. 1897 Scaphopoda of the San Domingo Tertiary (ibid.). '97, '98 Manual of Conchology, vol. 17, 1897-1898. Pilsbry, H. A. 1900 Gastropoda in : Zittel and Eastman, Text-book of Palaeontology, vol. i, 1900. Pritchard, G. B. 1896 A revision of the fossil fauna of the Table Cape beds, Tasmania, with descriptions of the new species (Proceedings of the Roy. Society of Victoria, new ser., vol. 8, 1896). Quenstedt, F. A. 1875 Petrefactenkunde Deutschlands, vol. 3; Echiniden, 1875. Quoy and Gaimard, P. 1832 Voyage de 1'Astrolabe, Zoologie, vol. 2, 1832. ORTMANN : TERTIARY INVERTEBRATES. 331 Reeve, L. A. 1847 Conchologia Iconica, vol. 4, 1847. 1851 Conchologia Iconica, vol. 6, 1851. 1858 Conchologia Iconica, vol. 10, 1858. 1859 Conchologia Iconica, vol. u, 1859. 1873 Conchologia Iconica, vol. 18, 1873. Rochebrune, A. T. de, and Mabille, J. 1885 Diagnoses de Mollusques nouveaux recueillis par les membres de la mission du Cap Horn (Bulletin de la Societe Philomathique de Paris, ser. 7, vol. 9, 1885). 1889 Mollusques in : Mission scientifique du Cap Horn, vol. 6, Zoologie, 1889. Ruetimeyer, L. 1867 Ueber der Herkunft unserer Thierwelt, Basel, 1867. Sandberger, C. L. F. 1863 Die Conchylien des Mainzer Tertiaerbeckens, 1863. Sowerby, G. B. 1846 Descriptions of Tertiary fossil shells from South America, in : Darwin, 1846. Speyer, 0. i864a Die Conchylien der Casseler Tertiaerbildungen, I, 2 (Palaeontographica, vol. 9, 1864). i864b Die Tertiasrfauna von Soellingen bei Jerxheim im Herzogthum Braunschweig (ibid.). 1866 Die ober-oligocaenen Tertiaergebilde und deren Fauna im Fuerstenthum Lippe-Detmold (ibid., vol. 1 6, 1866). 1867 Die Conchylien der Casseler Tertiaerbildungen 3 (ibid., vol. 16, 1867). Steinmann, G. 1881 Zur Kenntnis der Jura und Kreideformation von Caracoles. (Neues Jahrbuch fuer Mineralogie, Geologic und Palaeontologie, Beil. Bd. i, 1881.) Steinmann, G. (and Deeke, W., and Moericke, W.). 1895 Das Alter und die Fauna der Quiriquina-Schichten in Chile (ibid., Beil. Bd. 10, 1895). Stoliczka, F. ' 1864 Fossile Bryozoen aus dem tertiaeren Gruensandstein der Orakei-Bay bei Auckland (Novara Expedition, Geologischer Teil., Bd. i, Abteil. 2, 1864). 1873 The Echinodermata in : Cretaceous fauna of southern India, vol. 4, part 3, 1873 (Memoirs of the Geological Survey of India. Palaaontologia Indica). Suess, E. 1864 Brachiopoden in : Zittel, 1864. Tate, R. 1880 On the Australian Tertiary Palliobranchs (Transactions and Proceedings of the Roy. Society of South Australia, vol. 3, 1 880). '86, '87 The Lamellibranchs of the older Tertiary of Australia, i. (Transactions of the Roy. Society of South Australia, vol. 8, 1886); 2. (ibid., vol. 9, 1887). 1887 The Scaphopoda of the older Tertiary of Australia (ibid., vol. 9, 1887). '88, '89 The Gastropods of the older Tertiary of Australia. I. (ibid., vol. 10, 1888); 2. (ibid., '90, '93 vol. 11, 1889); 3. (ibid., vol. 13, 1890); 4. (ibid., "vol. 17, 1893). 332 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Tenison-Woods, J. E. 1878 On the Tertiary deposits of Australia (Journal and Proceedings of the Roy. Society of N. S. Wales, vol. n, 1878). Tryon, G. W. 1880 Manual of Conchology, vol. 2, 1880. 1 88 1 Manual of Conchology, vol. 3, 1881. 1882 Manual of Conchology, vol. 4, 1882. 1886 Manual of Conchology, vol. 8, 1886. Wallace, A. R. 1876 The geographical distribution of animals, vol. I, 1876. Waters, A. W. 1879 On the occurrence of recent Hetropora (Journal of the Roy. Microscopical Society, vol. 2, 1879). 1882 On fossil Chilostomatous Bryozoa from Mount Gambier, South Australia (The Quar- terly Journal of the Geological Society of London, vol. 38, 1882). 1884 Fossil Cyclostomatous Bryozoa from Australia (ibid., vol. 40, 1884). Weber, M. 1897 Zur Kenntnis der Suesswasserfauna von Sued-Afrika (Zoologische Jahrbuecher, Abteil. f. System., vol. 10, 1897). Weltner, W. 1897 Verzeichnis der bisher beschriebenen recenten Cirripedienarten (Archiv fuer Naturge- schichte, Jahrgang 1897, Bd. i). Whitfield, R. P. 1885 Brachiopoda and Lamellibranchiata of the Raritan Clays and Greensand Marls of New Jersey (Monographs of the U. S. Geological Survey, vol. 9, 1885). 1892 Gasteropoda and Cephalopoda of the Raritan Clays and Greensand Marls of New Jersey (ibid., vol. 18, 1892). 1894 Mollusca and Crustacea of the Miocene formations of New Jersey (ibid., vol. 24, 1894). Wood, S. F. 1848 A monograph of the Crag Mollusca of England, vol. I, Univalves, 1848; vol. 2, '56 Bivalves, 1856 (Palaeontographical Society). 1861 A monograph of the Eocene Mollusca of England, part i, Bivalves (Palaeontographical Society, 1861). Woodward, S. P. 1854 A manual of the Mollusca, part 2, 1854. Zittel, K. A. 1864 Fossile Mollusken und Echinodermen aus Neu Seeland (Novara Expedition, Geolog- ischer Teil., Bd. i, Abtheil. 2, 1864). '80, '85 Handbuch der Palaeontologie, vol. i, 1880; vol. 2, 1885. PART III. MAMMALIA OF THE SANTA CRUZ BEDS. MARSUPIALIA. BY WILLIAM J. SINCLAIR, PRINCETON UNIVERSITY. INTRODUCTION. THE first descriptions of marsupials peculiar to the Santa Cruz for- mation of Patagonia appeared in a brief dissertation by Dr. Florentine Ameghino, issued at Buenos Aires in 1887. It soon became apparent from the publication of figures in later papers by the same author (Amegh., 1889, 1894) that some of these peculiar genera resembled closely the pouched wolf or thylacyne of Tasmania, while others were more or less like the smaller diprotodont marsupials of Aus- tralia, and still others appeared to be related to the opossums of North and South America. These resemblances to existing genera, so far as could be determined from the figures and descriptions available, were confined mainly to the teeth and to the shape of the jaw with its strongly iriflected angle so characteristic of, although not entirely restricted to, marsupials. A large amount of material illustrating this group was secured by Messrs. Hatcher and Peterson and we are now able, for the first time, to ascertain what these animals were like and whether or not they were related to existing forms. It was originally planned that Mr. Hatcher should contribute this part to the series of volumes describing the magnificent collections which his energy and devotion to science have brought together in the museum of Princeton University. His untimely death necessitated a transfer of the work which was to have been his to other hands, and, at the request of Professor Scott, it was undertaken by the writer. 333 334 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. Before proceeding farther, the writer desires to express his indebted- ness to Professor Scott for permission to study this important group and for much helpful criticism and encouragement during the progress of the work. A number of photographs of type specimens in the Ameghino col- lection taken by Professor Scott have proved of great value in the study of the diprotodont forms. Indeed, without their help, it would have been impossible to be at all certain regarding the determination of many of the species. The American Museum collection of Santa Cruz marsupials, especially rich in diprotodonts, was placed at the writer's disposal by Professor H. F. Osborn, who has also contributed much valuable informa- tion regarding marsupial characters in general. Several important sug- gestions by Dr. W. D. Matthew and Mr. Charles Knight have been incorporated in the drawings of the restored skeletons of Frothy lacynus, Borliycena and Cladosictis (PI. LXI). Dr. J. A. Allen has kindly per- mitted the examination and illustration of a skull of Ccenolestes obscurus in the collection of the American Museum of Natural History (PI. LXIII, figs. 14-14^). To Dr. Oldfield Thomas, of the British Museum, the writer is indebted for the loan of valuable osteological material illus- trating the skeleton of Thylacynus, without which it would have been im- possible to work out in a satisfactory manner the relationships of the Santa Cruz carnivorous marsupials. Liberal use has been made of this material in the figures presented in the accompanying text and plate (PI. LXV, figs. \-\b}. CLASSIFICATION OF THE SANTA CRUZ MARSUPIALS. As among recent marsupials, two suborders may be recognized, agree- ing in every respect with the Polyprotodontia and Diprotodontia. It has seemed advisable to retain the major subdivision of the order based on dental characters rather than to follow the recent classification proposed by Bensley (1903) in which dental characters are subordinated to foot structure, because practically nothing is known of the feet in the Santa Cruz diprotodonts, and what little is known of the feet in Ccenolestes, their nearest living ally, adds to the confusion already existing in the classifi- cation of the Marsupials by combining a non-syndactylous foot with a diprotodont dentition. The Santa Cruz representatives of the Polyprotodontia are carnivorous, SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 335 and, with the exception of Microbiotlieriitm, a minute opossum, have been placed by Ameghino (1894, p. 108) in a suborder termed by him the Sparassodonta, a group which he regards as referable neither to the creo- donts, the placental carnivores, nor the carnivorous marsupials. A com- parison of the so-called Sparassodonta with existing carnivorous marsu- pials shows that they possess in common a large number of characters, either confined entirely to marsupials, or peculiar to but few additional orders. An examination of the following list will, it is believed, convince the reader that the Sparassodonta are true carnivorous marsupials and not worthy of subordinal rank. MARSUPIAL CHARACTERS OF THE SO-CALLED SPARASSODONTA. i. A typical marsupial dental for- mula ±2 i IHUld, 3 » x» l * 3» 4- 2. The number of successional teeth is reduced below that char- acteristic of the placentals. 3. The nasals are broad posteriorly, excluding from contact the fron- tals and maxillae. 1. The presence of three incisors above and below is exceptional among marsupials. A fourth molar occurs in Otocyon and Centetes among placentals. 2. According to Ameghino, the number of teeth having decidu- ous predecessors is greater in the Santa Cruz forms than among existing marsupials but less than in the placental Carni- vora. 3. A posterior broadening of the nasals is characteristic of most existing marsupials. The con- tact between the maxillary and frontal is more or less extensive but may be reduced to zero as in some specimens of TricJio- sunts vulpecula. Certain of the Creodonta (Harpagolestes, Dromocyon, Mesonyx] also show the posterior expansion of the nasals. 336 PATAGONIAN EXPEDITIONS '. PALAEONTOLOGY. 4. Lachrymal spreading out on the face. Lachrymal duct within the orbital rim. 5. Antero-posterior shortening of basis cranii. 6. The mandibular angle is strong- ly inflected. 7. An alisphenoid bulla present in some genera, absent in others. Tympanic annular and unfused with the adjacent elements in genera with bulla, unknown in those without alisphenoid bulla. 8. Basisphenoid perforated by in- ternal carotid artery. 9. Posterior extension of the malar bar to form the preglenoid process. 4. In most marsupials the perfor- ation for the lachrymal duct is either on or external to the or- bital rim. In Tliylctcynus and in some specimens of Borhycena there are perforations both with- in and without the orbit. The facial expansion of the lachry- mal is common to marsupials and also to a number of the Creodonta and Insectivora. 5. Characteristic of carnivorous marsupials. 6. Mandibular angle more or less strongly inflected in all marsu- pials except Tarsipes. This character is also developed in some of the Insectivora and in Creodonts of the Mesonychid phylum. 7. The alisphenoid bulla is a mar- supial-insectivore character. 8. Characteristic of the Marsupialia and Monotremata. Acrobates Pygmceus is exceptional in lack- ing this perforation, the internal carotid entering the skull through a foramen between the petrosal and the basisphenoid (Weber, Die Saugetiere, p. 46). 9. A marsupial character. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 337 10. Optic foramen confluent with sphenoidal fissure. 11. Basisphenoid and alisphenoid ridged as in existing marsupial carnivores and unlike the struc- ture of this region in the pla- centals. 1 2. Posterior border of palate thick- ened. 10. Not exclusively marsupial, al- though characteristic of that order. 11. "In the marsupial Carnivora the basisphenoid is relatively longer than in the placental Carnwora, and, at its posterior part, it sends a ridge down- ward from that part of each lateral margin which is not overlapped or covered by the base of the alisphenoid, the suture of which long continues distinct. These ridges, with the alisphenoid, render the whole under surface of the basisphenoid canaliculate, or concave transversely ; the basi- sphenoid is flat beneath in the placental Carnivore*, and that part of the base of the skull is made canaliculate by the de- velopment of the ectopterygoid plate from the alisphenoid ; these plates exist likewise in the marsupials, but, as they extend backwards to join the alisphenoid bullse, they diverge from the basisphenoid ridges and are external to them." (Owen, Phil. Trans. Royal Soc. London, Vol. 149, p. 315, 1860.) 1 2. Characteristic of the majority oi existing marsupials. Present also in some of the Creodonta (Mesonychidae). 338 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. 1 3. Posterior border of palate perfor- ated by a large foramen on either side of the posterior nares. 14. Premaxillae excavated for re- ception of the tip of the lower canine. 15. Palatal vacuities absent. 1 6. Presence of a vascular foramen (the post-zygomatic of Cope) perforating anteriorly the baseof the zygoma below or within the lip of the post-glenoid foramen. 17. Presence of a large vascular foramen (the sub-squamosal of Cope) perforating the squa- mosal on or above the crest which connects the base of the zygoma with the inion. 1 8. Epipubic ossifications absent. 19. Sutures of the skull and epi- physial elements of the skele- ton distinct in fully adult indi- viduals. 20. Transverse process of seventh cervical perforated by arterial canal. 13, A marsupial character. 14. The same structure occurs in Dasyurus, Tkylacynus and Di- delphys, but is not present in Perameles and Sarcophilus. 15. Present in the majority of liv- ing marsupials. Absent in some species of Macropiis and Petaurus. Usually thought of as primitive, but more probably to be regarded as a secondary character. Found also among the Insectivora. 1 6. Confined to the Marsupialia and Monotremata. (Cope, Proc. Am. Phil. Soc., Vol. 18, p. 453, 1880.) 17. Characteristic of the Marsu- pialia, but not confined to that order. Absent in placental carnivores. 1 8. Cartilaginous and vestigial in Thylacynus. 19. Not strictly a marsupial char- acter, but indicative of marsu- pial affinities, when considered in connection with the other characters presented. 20. Imperforate in nearly all pla- cental carnivores and also in Sarcophilus. Perforate in most marsupials. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 339 Two families are represented among the Santa Cruz marsupial carni- vores. The smaller forms, comprising the genus Microbiotlicrititn, arc opossums comparable in size and, to a certain extent, in dental structure to existing South American representatives of the Didelphyidae, to which, however, they are not ancestral. They have been grouped by Ameghino in a separate family, the Microbiotheridae, but this is hardly warranted in view of their very apparent didelphid affinities, a discussion of which will be found on a later page. The larger forms have long been known to resemble in dental struct- ure the Tasmanian genus Thylctcymts, and various hypotheses have been formulated to account for the observed similarity. A comparison of the dentition, skull and skeleton of Tliylacynus and the Santa Cruz genera shows a much closer relationship between the Tasmanian and South American forms than has previously been supposed to exist, so much so that the propriety of referring them to the same family seems beyond question. A number of families for the reception of these South American genera have been proposed, among which may be mentioned the Borhya?nidae, Acyonidae, Amphiproviverridas, Hathlyacynidae, Prothylacynidae and Spar- assodontidse. In the present paper the existing Tasmanian and extinct Santa Cruz forms are referred to the family Thylacynidae (Bonaparte, 1838). The following is a tabulation of the characters on which this classification is based : Family:^ THYLACYNIM:. Incisor formula A~l ; protocone of upper molars variable, external styloid cusps vestigial ; premolar dentition unreduced, posterior premolar well developed ; metaconid absent ; hallux opposable (arboreal adaptation), reduced or absent (cursorial modifications) ; non-syndactylous. A. Skull dolichocephalic. (a) Alisphenoid bulla present. 1. Dental formula £, \, f, |. Protocone well developed on all the upper molars. M± with small but distinct metacone. Posterior premolar exceeding in size the anterior and median premolars in both series. Talonid of Mf supporting a single blunt cuspule. Palate perforate. Mandibular symphysis ligamentous. Hallux absent. Terminal phalanges blunt with slight clefts'. Thylacynus. (Recent T. cynocephalus, Tasmania. Pleistocene, T. spelceus, Queensland, New South Wales.) 2. Dental formula \, -\, f , \. Protocone well developed on Ml"-*. M± with small, conical protocone, large paracone and antero-external style ; metacone reduced to the merest vestige, or absent. Premolars increasing regularly in size poste- 340 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. riorly in both upper and lower series. Talonid of M? inclosing a small basin- shaped area with a single high point on its posterior rim. Palate imperforate. Mandibular symphysis ligamentous. Hallux reduced. Terminal phalanges uncleft, laterally compressed and pointed. Cladosictis. (Miocene, Santa Cruz formation, Patagonia, C. lustratus, C. petersoni.) 3. Dental formula $, \, |, |. Protocone well developed on all the upper molars. MA with protocone inclosing a basin-shaped area, paracone and antero-external style large, metacone vestigial or absent. Upper premolars increasing regularly in size posteriorly ; median and posterior lower premolars subequal. Talonid of Mj large and strongly bicuspidate. Palate imperforate. Mandibular sym- physis ligamentous. Hallux large and opposable. Terminal phalanges later- ally compressed and pointed, without clefts. Ampliiproviverra. (Miocene, Santa Cruz formation, Patagonia, A. niazaniana, A. minuta.) (b) Alisphenoid bulla absent. I. Dental formula A?, \, %, A. Protocone well developed on Ml, and M^, absent on Mi. MA with vestigial protocone and metacone. Posterior premolar not greatly enlarged, in the inferior series not exceeding the median premolar in size. Talonid of MT small and basin-shaped, with a single high cusp on its posterior rim. Palate imperforate. Mandibular symphysis fused. Hallux reduced, not supporting phalanges. Terminal phalanges laterally compressed, sharply pointed, and slightly cleft at tips. Prothylacynus. (Miocene, Santa Cruz formation, Patagonia, P. patagonicus.') B. Skull brachycephalic. (a) Alisphenoid not dilated to form a bulla. I. Dental formula f , \, |, A. Protocone on upper molars reduced. MA bicuspidate with paracone and antero-external style. Posterior premolars greatly enlarged. Talonid of Mj with single conical cusp. Palate imperforate. Mandibular sym- physis fused. Hallux unknown. Terminal phalanges round, blunt, and broadly fissured at the tips. Borhyeena. (Miocene, Santa Cruz formation, Patagonia, B. tuber ata, B. excavata.) The determination of the marsupial affinities of the Santa Cruz dipro- todonts rests almost entirely on their close resemblance to Ccenolestes. The discussion of the classification adopted is more conveniently post- poned to a later chapter, but it may be remarked in passing that all are referred to one family, the Caenolestidae, so named from its best known and only surviving representative, Ccenolestes. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. POLYPROTODONTIA. THYLACYNID&. The Santa Cruz thylacynes are short-legged animals, with large heads, long necks and heavy tails. These characters are well shown in the accompanying restorations of Prothylacynus patagonicus and Cladosictis lustratus (PI. LXI, figs, i, 2). In order to show clearly the points 01 FIG. i. >. .• • •: \ a b c d * Thylacynus cynocephalus. a, right humerus, front view ; b, right radius and ulna from the outer side ; c, right femur, front view ; d, right tibia, front view ; c , right fibula from the inner side. All figures x £. agreement between the various Patagonian genera and Tkylacynus, and the respects in which they differ, the following summary has been intro- duced : i. In all, the facial region of the skull is short in proportion to the 342 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. length of the cranium. The brain case is small in the Santa Cruz genera and greatly constricted postorbitally, and the orbits are placed much farther forward than in the Dasyuridce, opossums, or Thylacynus. In the latter genus (PI. LXV, fig. i), the capacity of the brain case has increased considerably, with a corresponding expansion of the postorbital region, both of which, together with the posterior shifting of the orbit, may be regarded as progressive characters. In the extinct members of the family the palate lacks the vacuities present in all existing carnivorous marsupials, but is perforated by a number of accessory palatine foramina. Between the molars, the margin of the palate is depressed into deep hemispherical fossae for reception of the tips of the lower teeth when the mouth is closed. The jugal arches are robust and rather broadly expanded, and the sagittal and lambdoidal crests well marked, but not very high. In the Santa Cruz forms, the occiput is semicircular in outline, in contrast with its triangular shape in the dasyures, SarcopJiilus and Thylacynus. The lachrymal canal opens well within the orbital rim. In the majority of living marsupials, the opening of the lachrymal duct is placed either on or external to the orbital rim. Thylacynus is transitional between these two types of structure in that it possesses a double lachrymal perforation, one branch of the canal opening without and the other within the orbit. The number and position of the cranial foramina in the existing and extinct members of the family are practically the same with one important exception, namely, that in Thylacynus the basisphenoid is perforated by two large foramina, as in DidelpJiys (cf. PI. LXV, fig. \d] whereas in the Patagonian forms there is but is but a single perforation. Borhycena and Prothylacynus resemble SarcopJiilus in the fusion of the mandibular symphysis. In the remaining genera the symphysial union is ligamentous. 2. The molars of the Santa Cruz genera are of the same type as in Thylacynus, differing principally in the greater reduction of MA, the loss of all the styloid cusps, except the antero-external, and the character of the heel of the last lower molar, which may be either small and conical, basin-shaped, or bicuspidate. The premolars are unreduced in number and usually increase in size posteriorly in both series. The canines are long, sharply pointed and slightly curved in the smaller genera. In Bor- hyczna the fang is swollen and the point short and blunt. The incisors in Borhyczna are reduced to §, an exceptional formula among marsupials SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 343 FIG. 2. in that the number above and below is the same. In Amphiprwiverra the median pair are conical and approximated at the tips, as in Dasynms and Didelphys. According to Ameghino, the number of teeth having deciduous predecessors is greater in the Santa Cruz forms than among existing marsupials (see pp. 348, 378). Unfortu- nately it has not been possible to check this impor- tant observation by the material in the Princeton collection. It would not be surprising if Mio- cene marsupials retained some trace of the fuller dental replacement characteristic of the placentals. An extreme is reached in Thylacynus, where the deciduous predecessors of the posterior premol- ars are minute triangular plates displaced before birth. In the Santa Cruz forms (cf. Cladosictis, PI. LIX, fig. 6) replacement does not occur until the animal is fairly mature. 3. The atlanteal intercentrum is unfused with the base of the neural arch in Borhycena and Amphiprovi'verra, as it is also in Thylacynus (text fig. 5, c\. In Frothy lacynus and Cladosictis 1 . fnylacynus cynocephalus, complete fusion has taken place, with oblitera- right scapula, x }. tion of the sutures. An atlanteal foramen for the transmission of the spinal nerve and vertebral artery is present in all the Santa Cruz genera, except Borhycena, which resembles Phascolomys in transmitting the nerve and artery through a groove in the anterior margin of the neural arch. The axis carries a large hatchet-shaped neural spine. The bases of the transverse processes of the second to the seventh cer- vicals are perforated for the transmission of the vertebral artery. The dorso-lumbar vertebral formula was probably the same as in Thylacynus : thirteen dorsals and six lumbars. As in that genus, the anticlinal verte- bra is the tenth dorsal. Two vertebrae are coossified in the sacrum. The tail was undoubtedly long, very heavy, and greatly thickened at the base. 4. The limbs are short in proportion to the length of the body. The shortening is especially noticeable in the bones of the fore arm and fore leg, which are elongated in Thylacynus (text fig. I, b, of, e), an adaptation to cursorial habits. In all, the radius and ulna are capable of some de- gree of pronation and supination. The tibia and fibula are unfused. The 344 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. inner humeral epicondyle is perforated by a large foramen in Thylacymts, Prothylacynus and Cladosictis; imperforate in AnipJiiproviverm. In Pro- thylacynus, the supinator ridge terminates in a hook-shaped extremity, which is wanting in Amphipro'viverra, Cladosictis and Thylacynus. In FIG. 3. Thylacynus cynocephalus, pelvis and sacrum, x \. a, from the left side ; b, from above ; c, from below. The epipubic cartilages are represented by the dotted outline. the latter genus the distal end of the humerus is much narrower trans- versely than in either Prothylacynus or Cladosictis, and both supinator ridge and inner epicondyle are smaller (cf. text fig. i, a and Pis. XLIX, figs, i, ia, \b ; LV, figs. 2, 20). SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 345 5. The patella is ossified in Amphiprovwerra and Protliylacynus. Among living marsupials the patella is ossified only in the Peramelidae. 6. The feet are small, with spreading toes. The degree of reduction of the hallux is variable. In the recent genus it has been entirely obliterated (text fig. 4, b]. In Prothylacynus a rudimentary metatarsal remains, while in Amphipro'viverra the hallux is large and opposable. The loss of the hallux is a cursorial adaptation, various stages in the perfection of which are illustrated by the forms just mentioned. These genera, however, have diverged in cranial and dental development and are not a true phyletic series. A still more peculiar cursorial modification of the pes in Thylacynus appears in the shifting of the ectocuneiform toward the outer side of the foot until it is supported almost entirely by the cuboid (text fig. 4, b]. In the Santa Cruz forms the shifting has progressed to about the same extent as in Sarcophilus. The pollex is known in Amphi- promverra and Cladosictis. In these genera, the phalanges of the pollex FIG. 4. Thylacynus cynoccphalus. Xf a, right fore foot, dorsum. b, right hind foot, dorsum. Both figures are deflected toward the inner side of the foot as a result of the enlarge- ment of the outer condyle of the metacarpal of the thumb. Indications of the same structure may be observed in Thylacynus. The manus ispenta- dactyl in Borhyana and Thylacynus, and probably also in Prothylacynus, 346 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. and both manus and pes are pentadactyl in Amphiprovi'verra and Clado- sictis. 7. The trochlear surface of the astragalus in the Santa Cruz forms is short and flat with feebly differentiated facets for the tibia and fibula. This has been interpreted as indicating a plantigrade gait. In the digiti- grade Thylacynus the astragalar trochlea is both longer antero-posteriorly and more deeply grooved. 8. There is no trace of syndactyly in either the living Tasmanian or extinct Patagonian thylacynes. 9. The pubes do not appear to have supported epipubic ossifications in Cladosictis. In the only specimen of Prothylacynus in the Princeton col- FIG. 5. Thylacynus cynocephalus. a, fifth and sixth lumbar vertebrae from the right side ; b, atlas from above ; c, atlas from below ; d, cervical series from the left side. All figures x \. lection the pubes are not preserved and the entire pelvis of Borhycena and Amphiproviverra is unknown. In Thylacymis the epipubic elements are mere vestigial cartilages (text fig. 3). SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 347 BORHY^ENA Ameghino. (Plates XL-XLVI ; LIT, Figs. 1,2,6; LIU, Figs. 2, 2«, 4-46, g,ga; LIV, Figs. 7, 13; LXI.Fig. 3.) Borhycena Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 8, Dec., 1887. Dynamictis Amegh.; Revista Argentina Hist. Nat. I, entr. 3^, pp. 148- 149, June, 1891. A genus of large marsupial carnivores containing the most powerful predatory mammals in the Santa Cruz fauna. In the Princeton collection two species are represented by good skulls, with one of which a consid- erable portion of the skeleton is associated. Dentition (Pis. XL; XLII ; XLIV; XLV, figs, i, 3). — In both spe- cies (B. tuberata, B. excavatd], the incisors are reduced to 3, a formula unknown among existing marsupials, with the exception of Notoryctes. The upper incisors here have been worn to such an extent that the pattern of the crown is entirely obliterated. The median pair are laterally com- pressed and show no tendency to assume a procumbent position or to be- come approximated at the tips, suggesting that in Borhycena the reduction of the superior incisor formula has been accomplished by the suppression of the teeth homologous with the conical, procumbent, median incisors of Didelphys and Dasynrus. The upper canine is large, with swollen root and thick blunt crown. The premolars are closely crowded and increase rapidly in size posteriorly. All are double-rooted. The anterior pre- molar is in contact with the canine, and is placed transversely to the tooth row. The crown is slightly compressed laterally, and the heel rounded, without heel cusp. The median premolar is similar to the preceding tooth, but carries a larger heel. The posterior premolar is greatly enlarged and the heel broad, extending around the inner side of the crown. The anterior margin of the tooth is more or less abraded by con- tact with the similarly enlarged lower posterior premolar. The worn molars bear a superficial resemblance to the teeth of Sarcopliilits, but their thylacyne structure is fully apparent, when it is remembered that in Sarcophilus the broadening of the upper molar cusps is produced by basal fusion with the outer row of styles, of which the antero-external alone is present in Boryhcena. The first, second and third molars increase regu- larly in size posteriorly. The protocone is small and ledge-like; disap- 348 PATAGONIAN EXPEDITIONS : PAL/EONTOLOGY. pearing entirely or almost entirely on worn teeth, although the inner root is still proportionately as robust as in Thylacynus, where the protocone is unreduced. The metacone spur is much less rotated outwardly than in the latter genus. The last molar shows greater reduction than in any other of the Santa Cruz genera. Its crown is composed of two cusps, the paracone and antero-external style, separated from each other by a sharp notch, forming a transversely placed shear. The anterior surface is greatly abraded by cutting against the postero-external margin of the pro- toconid of the last lower molar. M- is frequently single-rooted. In the inferior dentition (Pis. XL ; XLV, fig. 3), the incisors are closely crowded and the root of the second is displaced posteriorly with reference to the median and lateral teeth, as in Thylacynus and the Santa Cruz genera in general. The canine is large, with swollen root and slightly re- curved blunt crown, bearing a broad groove on its inner side. As in the superior series, the premolars are closely crowded and the anterior tooth, situated in contact with the canine, is placed obliquely to the long axis of the tooth row. The anterior and median premolars are like those of the superior series. The posterior premolar is similarly enlarged, but has a much smaller heel. The lower molars are double-rooted and increase in size posteriorly. In the heels of the first, second and third molars the hypoconid is reduced, while the hypoconulid and entoconid are represented by a single cusp. The heel of the last molar carries a single conical cusp. The cusps of the trigonid are high and sharply separated by deep notches. Milk Dentition. — Ameghino (1894, p. 109) states that in Borhycena " the milk dentition consists of a canine and a molar ; the latter has the form of a true molar and is replaced by the third tooth of the permanent dentition which follows behind the canine." The individuals in the Prince- ton collection are not sufficiently immature to permit of the confirmation of these important details. Skull (Pis. XL-XLIV; XLV, fig. i; XLVI, fig. 4).- -The skull is broad and depressed, with powerful, widely expanded arches and mode- rately elevated sagittal and lambdoidal crests. The upper border of the facial profile has but slight inclination fonvard. The cranium is depressed in the parietal region and is proportionately less constricted postorbitally than in the other Santa Cruz marsupial carnivores. The brain cavity is smaller than in Thylacynus and the cerebral hemispheres less convoluted, judging from their impression on the cranial walls (PI. XLII). The fossae SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 349 for lodgement of the vermis and lateral hemispheres of the cerebellum are proportionately as large as in Thylacynus. The olfactory sinuses are enormously developed, indicating the possession of keen powers of scent. The ascending processes of the premaxillae are shorter than in the other Santa Cruz genera. The nasals are greatly expanded posteriorly, occu- pying almost the entire width of the interorbital tract and excluding from contact the frontals and maxillae. The frontals between the orbits are plane in B. tuberata, slightly convex in B. excavata. Postorbital frontal processes are entirely wanting and the temporal ridges poorly defined. The sagittal crest is lower than in Thylacynus ; its free border is concave in profile. The supraoccipital is not exposed on the upper .surface of the skull, the parietal extending to the margin of the lambdoidal crest as in Prothylacynus. The orbits are smaller than in Thylacymis and are placed farther forward. A large lachrymal tubercle is present on the orbital rim. The lachrymal duct opens well within the orbit. A small foramen pierc- ing the facial expanse of the lachrymal in B. excavata (PI. XLV, fig. i) may possibly be homologous with the external opening of the lachrymal duct in Thylacynus. The zygomatic arches are heavier and more widely expanded than in Thylacymis. The postorbital jugal processes are smaller and the preglenoid processes larger than in the latter genus. The occiput (PI. XLVI, fig. 4) is semicircular in outline and in B. excavata does not project posteriorly beyond the condyles, which differ from Thylacynus in being wider superiorly and more obliquely placed. The areal extent of the mastoid on the lateral border of the occiput is proportionately less than in the recent genus. ' The base of the skull is fairly well preserved in the specimen of B. excavata (No. 15, 120, PI. XLIV). .The basioccipital and basisphenoid are almost flat. The paroccipital processes are short, resembling Prothyla- cynus rather than Thylacynus. The large condyloid foramen is preceded by an accessory foramen of approximately the same size. The auditory region resembles Prothylacynus in the absence of alisphenoid dilatation, but the foramen ovale does not pierce the alisphenoid opposite the glenoid cavity, as in the latter genus, the posterior branches of the fifth and seventh nerves probably emerging between the alisphenoid and the tym- panic. The latter element has been shed, exposing the petrous, which is smaller than in Pyothylacynus. The palate is without vacuities, but is pierced by a number of accessory palatal foramina. The anterior palatine 350 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. foramina are bisected, as in Thy lacy mis, by the premaxillo-maxillary suture. A pair of large foramina pierce the palate opposite the posterior margins of the canines. The neuro-vascular foramina at the posterior palatal border are enclosed by robust bars of bone, which are much more attenuated and often incomplete in Thylacymis (cf. PI. LXV, fig. ia). The narial border is thickened, much as in the creodont Mesonyx. The mandible (Pis. XL; XLV, fig. 3) is very heavy in proportion to its length. The rami are firmly coossified at the symphysis, but traces of the suture remain. The backward inclination of the coronoid process is about the same as in Prothylacynus, but the width is relatively less. The masseteric fossa is broader than in Thylacynus and the heavy flange bordering it inferiorly is produced to the outer extremity of the condyle, while in the latter genus this structure narrows abruptly just anterior to the condyle. The condyles are very wide transversely, decreasing in width toward their outer ends, while the reverse is true in Thylacynus. The angle is broad and less deeply notched posteriorly than in the latter. Six mental foramina are present in B. tuberata, varying in position on opposite halves of the same mandible. The largest of these is situated beneath the anterior premolar. Vertebral Column ; Ribs and Sternum. — The atlas (PI. LIII, figs. 2, 2*7, 4-46) is peculiar in lacking a foramen for the vertebral artery and first pair of spinal nerves, resembling in this respect Phascolomys. The nerve and artery are transmitted through a pair of deep notches in the anterior margin of the neural arch. The canal for the vertebral artery is small, entering the neural arch just above the condyles and emerging on the lower surface of the transverse process. A small foramen, possibly for a recurrent branch of the same artery, perforates the upper surface of the transverse process near its posterior border. The intercentrum (PI. LIII, fig. 4^) is separately ossified and unfused with the base of the arch. Its posterior border supports a small median styloid process. The transverse processes are semicircular in outline with thickened edges. The axis (PI. LII, figs, i, 2, 6) carries a large hatchet-shaped neural spine which overhangs the odontoid anteriorly. Posteriorly, the spine is extended to about the same degree as in TJiylacymis. The odontoid tapers less than in that genus, retaining about the same width through- out. Anteriorly, it is obliquely truncated. The transverse processes are perforated by the vertebral artery. Their extremities are broken off in both SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 351 specimens, but they were probably as long proportionately as in Thyla- cynns. The ventral surface of the centrum (PI. LII, fig. 6) has a strong median keel which increases in depth posteriorly. The concavities on either side are bounded externally by the inferior edges of the transverse processes. Traces of the suture uniting the axial centrum with the anterior cotyles and odontoid are visible in both species. The neural spine of the third cervical is proportionately larger and the median keel of the centrum stronger than in Thylacynus. In the fourth and fifth cervicals the diapophyses are well differentiated from the inferior lamella, unlike Thylacynus. The inferior lamella of the sixth cervical is less elongated antero-posteriorly than in the recent genus, but is much deeper. The transverse process of the seventh cervical is perforated by the vertebral artery. The neural spines of the cervicals increase in size and probably also in height posteriorly. The inferior keels on the centra decrease in depth on the fifth, sixth and seventh cervicals. The lateral surface of the neural arch above the canal for the vertebral artery is per- forated by a small foramen in the second to the seventh cervicals. The dorsal vertebrae associated with the skeleton of B. tuberata are larger than in Thylacynus, with heavier neural spines. The centrum of the first is keeled inferiorly, but keels are absent in the vertebrae inter- preted as the seventh (PI. XLV, fig. 6) and eighth dorsals. The dorso- lumbar vertebral formula was probably the same as in Thylacynus and has been so represented in the restoration (PI. LXI, fig. 3). Three caudals are preserved with the skeleton of B. tiiberata, which are interpreted as the third, fourth and fifth. The fourth caudal (PI. LI 1 1, figs. 9, 9«) has been selected for illustration owing to its better state of preservation. It is considerably larger than the corresponding vertebra in Thylacynus, but much smaller than the fourth caudal in Prothylacynus. The transverse processes are broadly expanded, but are rounded at the tips, in contrast with the antero-posterior extension of the tips of the trans- verse processes of the proximal caudals in TJiylacynus. The presence of chevrons on the fourth and fifth caudals is indicated by facets. The floor of the neural canal in all the cervicals, except the atlas, and in the dorsal and anterior caudal vertebrae is subdivided by a median ridge, on either side of which a foramen pierces the centrum. A similar structure is observable in Thylacynus and also in various placental carni- vores. 352 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. A peculiar feature noted by Ameghino (1894, pp. 112, 113) is the annular character of the vertebral epiphyses. A low round prominence from the centrum projects through the central perforation of the epiphysial ring. This structure is most typically developed in Borhycena, occurring less regularly in Frothy lacynus. Parts of several ribs are preserved with No. 15,701, of which the two most complete specimens are figured (Pis. XLV, fig. 4; XLVI, fig. i). The smaller of these, the second rib of the left side, is shorter than in Thylacynus, but slightly more robust and with thicker distal end. The larger rib (PI. XLV, fig. 4), from the cylindrical character of its shaft, evidently belongs near the middle of the thorax. The proximal end is more robust than any of the ribs in this region in Thylacynus. The shaft also is more cylindrical. Part of the sternum is preserved with the remains of both species of Borhycena in the collection. The presternal segment (PI. XLV, fig. 5) differs from the corresponding element in Thylacynus in having the dis- tal portion of the posterior bar narrower and deeper. The mesosternal segments are shorter and proportionately narrower than in the recent genus. Appendictdar Skeleton. The scapula (PI. XLVI, fig. 2), although smaller than in Thylacynus, has the neck of about the same length. The glenoid cavity is circular in outline and the coracoid process large, with the tip directed inwardly to about the same extent as in the Tasmanian genus. The high scapular spine divides the external surface into two unequal fossae, of which the anterior is the larger. Its surface is almost flat. The infraspinous fossa is deeply concave with elevated axillary border. The inferior angle is more acute than in Thylacynus. The scapular spine terminates in a long narrow acromion, the tip of which is missing. The supraspinous fossa is perforated by a large foramen. In Thylacynus several foramina pierce the base of the scapular spine, open- ing into the infraspinous fossa. These are represented by a single large foramen in Borhycena. In both genera a small foramen pierces the anterior margin of the neck near the middle of the suprascapular notch. The radius (PI. XLV, figs. 2, 2a) is of the same length as in Prothyla- cynus, but differs considerably in shape and is much less robust. The head is transversely flattened and elliptical in outline. The articular sur- face for the ulna is deeper than in Thylacymis and less confined to the SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 353 posterior side, indicating somewhat greater power of pronation and supi- nation. The bicipital tubercle is placed farther toward the outer side of the bone than in Thylacynus, corresponding rather with its position in Sarcophilus, and is much larger than in the former genus, although con- siderably smaller than in Frothy lacynus. The radial shaft is slightly curved and approximately circular in median transverse section, becoming oval in cross section toward the distal end, in striking contrast with the sharply triangular section of this portion of the bone in Frothy lacynus. The distal end is much deeper than in the latter genus, with the articular surface convex antero-posteriorly, as in Thylacynus, while in Frothylacynus it is concave. The styloid process is longer and heavier than in the last- named genus. The ulna (PI. XLV, figs. 2, 2.0} is much shorter and heavier than in Thylacynus. The posterior border is broadly concave. In this respect Borhycena differs from the other Santa Cruz thylacynes and resembles the existing genus. The olecranon is broad and heavy, comprising about one fifth the total length of the shaft. Its proximal end is greatly thick- ened and rugose. The greater sigmoid cavity is wider and deeper than in the recent genus and the coronoid process projects farther forward. In the lesser sigmoid cavity, the radial articular surfaces are more broadly connected proximally than in either Thylacynus or Frothy lacynus. The shaft is considerably flattened and broadly grooved on the outer side. The distal end is much heavier than in Thylacynus, with large hemi- spherical styloid process and broad radial tubercle. With the exception of the trapezoid and cuneiform, all the elements of the carpus (PI. LIV, fig. 7) are known. The radial surface of the sca- phoid is convex transversely and slightly concave in dorso-palmar section. Owing to the weathering of its palmar margin, the exact shape of the proximal articular surface cannot be determined. Distally, there is a broad contact with the magnum, sharply separated from the lunar facet. The trapezoidal facet is triangular in outline, deeply concave transversely and convex in dorso-palmar section. The facet for the trapezium is shaped like a capital B, with the invagination directed toward the inner side of the foot. It is almost plane in the dorso-palmar direction. In transverse section, the dorsal half of the facet is slightly convex and the palmar half concave. The inner posterior angle of the scaphoid supports a large hemispherical process, which is wanting in Thylacynus, but present 354 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. in Sarcophilus and Prottyilacynns. In shape, the scaphoid resembles more closely that of Sarcophilus than the corresponding element in Thylacymts. The lunar is a wedge-shaped bone, the proximal end of which is occu- pied by a broad, convex facet for the radius. Distally, the lunar articulates by a concave, oval facet with the magnum, unlike Thylacymts (cf. text fig. 4, a). The scaphoidal facet is semilunar in outline, almost plane in dorso-palmar section and plane or slightly convex at right angles to the former diameter. The facet for the cuneiform is triangular in outline and slightly convex in all diameters. It, is feebly differentiated from the semi- lunar facet for the unciform. The pisiform has a concave, elliptical facet for articulation with the cunei- form and a semilunar, dorso-palmarly convex facet for contact with the styloid process of the ulna. It is slightly more robust than in Thylacynus, terminating distally in a large hemispherical tubercle. The trapezium is much larger than in Thylacyrms. It is irregularly oblong in form, with a B-shaped facet for the scaphoid, a semilunar facet for the trapezoid, convex dorsally and slightly concave toward the palmar margin, and a large, concave, oval facet for the metacarpal of the pollex. The magnum resembles that of SarcopJiilus rather than the correspond- ing element in Thylacymts. Proximally, it supports a heavy crest with two facets for the scaphoid and lunar respectively. The former facet is irregularly oblong with a sigmoid curvature in dorso-palmar section. Transversely, it is concave dorsally and plane or slightly convex toward the palmar margin. The lunar facet is convex, becoming slightly concave dorsally. On the median side there is an irregularly quadrilateral, almost plane facet for the trapezoid. The facet for the unciform is confined to the dorsal margin of the magnum, unlike its position in Thylacymis. It is quite irregular in shape, with an uneven undulating surface. Distally, there is a broad triangular, concave facet for the third metacarpal. The unciform is partially broken and the facets for the magnum and cuneiform are incomplete. In shape it seems to have been irregularly tetrahedral. Distally this element bears a broad triangular facet for the fourth and fifth metacarpals. The surface for the fourth metacarpal is con- cave dorso-palmarly and plane transversely. That for the fifth metacarpal is also concave dorso-palmarly, but is convex transversely. The metacarpal of the pollex is missing, but, judging from the size of SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 355 its articular surface on the trapezium, it was larger than in Thylacynus. The second metacarpal overlaps proximally the third, and the fourth on the fifth. The proximal end of the third has not been preserved. The fourth is the longest element in the metacarpal series. Its proximal end is irregularly quadrangular in outline, wider dorsally than at the palmar margin. The head of metacarpal V is separated by a sharp keel into two convex facets for the overlapping metacarpal IV and the unciform respec- tively. Distally, the metacarpals are flattened, with well-developed keels confined to the palmar surfaces. The arrangement of the phalanges shown in the figure is arbitrary, as there was no possible way to determine the original association of these elements in the matrix. The phalanges of the proximal row are much shorter and heavier than in Thylacynus, with straighter shafts. Those of the second row are less flattened than in the recent genus. The distal trochleae of the phalanges of the first and second rows have no greater dorso-palmar extension than in Thylacynns and there is no reason to believe that the angulation of the digital elements was any greater. The unguals are stout, resembling the claws of Thy- lacynus, but differing from that genus in the broad Assuring of the tips. The subungual processes are large. Hoods are developed to about the same extent as in Thylacynus. An ungual foramen is present in all the claws. In proportion to the size of the skull, the femur (PI. XLVI, figs. 3, 3*7) is remarkably short. The head is of about the same size as in Thylacynus, but the neck is considerably longer. The great trochanter projects slightly above the level of the head, from which it is widely separated. The lesser trochanter is incompletely preserved, but was probably as large propor- tionately as in Frothy lacy nus. The shaft is straight. The condyles are narrower antero-posteriorly than in Thylacynus and are plane or slightly convex transversely. The inner condyle is considerably wider than the outer. Restoration (PI. LXI, fig. 3). — The disproportionately large size of the head and great length of the neck are at once apparent in the drawing of the restored skeleton. The length of the back has been determined by comparison of the lengths of the few dorsals preserved with the corre- sponding parts in Thylacynus. The lumbar vertebrae and pelvis are sup- plied from Prothylacynus. The length of the tail is largely hypothetical, but the size of the proximal caudals indicates that it was greatly thickened 356 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. at the base, less so, however, than in Protkylacyitus. Regarding the hind limb, two alternatives are possible. Either the tibia was short and the back sloped downward from the shoulders, or the hip and shoulder were equally elevated. The latter assumption seems preferable and an elon- gated tibia and fibula have been supplied, as in Thylacynus. An ossified patella has been introduced from analogy with Prothylacymts. The planti- grade pose is largely conjectural, as there is no certain means by which the gait of an animal may be determined from the skeleton of the fore foot alone. It is probable that Frothy lacy ttus, Cladosictis and Amphi- proviverra were plantigrade, and, in the absence of evidence to the con- trary, the same may be assumed tentatively for Borhyana. No trace of a clavicle is preserved, but one has been inserted from analogy with exist- ing carnivorous marsupials. Habits. — Some inference regarding the pugnacity of these animals may be drawn from the large cicatrice in the mandible of B. tuberata repre- sented in fig. 3, PI. XLV. Ability to trail by scent like the Tasmanian thylacyne may, perhaps, be inferred from the large olfactory sinuses. The blunt claws indicate adaptation to terrestrial progression. The animal undoubtedly preyed on the larger placental mammalia. BORHY/ENA TUBERATA AmeghinO. (Plates XL-XLII ; XLV, Figs. 2-6 ; XLVI, Figs. 1-3* ; LII, Fig. I ; LIII, Figs. 2, 20, 9, ga ; LIV, Figs. 7, 13; LXI.Fig. 3.) Borhycena tuberata Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 8, Dec., 1887. Borliyccna zitteli Amegh.; Enum. Syn. des Especes de Mam. Fos. des Formations Eocene de Patagonie, pp. 119-120, 1894; Bol. Acad. Cordoba, p. 375, 1894. A nearly complete skull and mandible (No. 15,701) associated with a considerable portion of the skeleton, secured by Mr. Peterson eight miles south of Coy River, has been identified with this species, which is one of the largest of the Santa Cruz thylacynes. Apart from characteristic measurements, B. tuberata may be recognized by the broad depressed skull, perfectly flat between the orbits, with robust, gradually expanding arches. The face is without antorbital constriction. The postzygomatic portion of the brain case is considerably elongated and SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 357 the lambdoidal crest expanded as a broad semicircular frill. The infra- orbital canal is double, neither branch being as large as the single canal in B. excavata. If constant, this may prove to be a good specific char- acter. The skull is a fourth larger than in B. excavata. It is not yet possible to separate generic from specific characters in the skeleton, as B. tuberata is the only species in which the skeleton is at all well known. Its important features have already been noted in the de- scription of the genus. MEASUREMENTS. Skull, length, prcmaxillae to lambdoidal crest ...... .230 " width across jugal arches ........ .160 " " between orbits ......... .062 Face, length to anterior orbital border ....... .076 Cranium, length to anterior orbital border . ...... .141 " least width of brain case . . . . . . ... .027 Palate, length from alveolus of median incisor to palato-nanal border . . . 104 " width between canines . . . . . . . . . .0245 " " " posterior premolars . . . . . . . .037 " " "Mi 075 Mandible, length . . 186 " transverse diameter of condyle ....... .0423 " length of symphysis ........ .060 " depth below posterior premolar . . . . . . . .031 " " " MT 038 Superior dentition, length, median incisor to MA. . . . . . .113 " " " of premolar series ...... -0333 " " " " molar " 045 Inferior dentition, length, base of median incisor to M? . . . .1025 " " " of premolar series ...... .035 " molar " 0515 Median upper incisor, width of crown ...... .002 Lateral " " .0036 Alveolus of upper canine, length ....... .0165 " " " " width .0135 Anterior superior premolar, antero-posterior diameter. . . . -0075 " transverse "... .0047 Median " " antero-posterior "... .010 transverse " . . .0063 Posterior " " antero-posterior "... -0145 " " " transverse .009 Mi, antero-posterior diameter .... -O12 " transverse "... -0082 M£, antero-posterior " -0126 " transverse " . °°9S 358 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. Mi, antero-posterior diameter . . . . . . . . . .015 " transverse " ......... .0105 MA, antero-posterior " ......... .0055 " transverse " ......... .0085 Lower canine, antero-posterior diameter at alveolar border . . . . .0163 " " transverse " " " 0112 Anterior inferior premolar, antero-posterior diameter ..... .009 " transverse "..... .005 Median " " antero-posterior " (approximate) . . .0117 transverse " 006 Posterior " " antero-posterior " 0143 " " " transverse "..... .0065 My, antero-posterior diameter . . . . . . . . . .012 " transverse " . . . . . . . . . .006 Mj, antero-posterior " . . . . . . . . .012 " transverse "......... .007 Mg-, antero-posterior "......... .013 " transverse "......... -0075 Mj, antero-posterior ".......... .016 " transverse "......... .009 Atlas, transverse breadth .......... .0805 " breadth across anterior cotyles ....... .0498 " width of neural arch ......... .022 " " intercentrum including median process . . . . .0145 Axis, length of centrum, including odontoid ...... .052 " width of posterior face of centrum ....... .0195 " depth" " " " 015 " length of odontoid .......... .01 1 " width across anterior cotyles ........ .039 " antero-posterior diameter of neural spine ...... -0745 " height of spine above floor of neural canal ..... .042 Third cervical, length of centrum ..... . . .026 " width across transverse processes (approximate) . . . .061 " prezygapophyses ...... -0335 Fifth cervical, length of centrum ........ .027 width of posterior face of centrum ..... -0195 " depth " " 0182 width across transverse processes . . . . . . .053 " " prezygapophyses ...... .040 antero-posterior diameter of inferior lamina of transverse process .0335 Sixth cervical, length of centrum ........ .028 width of posterior face of centrum . . . . . .0185 " " depth " " " " 0175 width across transverse processes (approximate) . . . .044 " " prezygapophyses . . . . . . .0335 antero-posterior diameter of inferior lamina of transverse process .029 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 359 Seventh cervical, length of centrum ........ .027 " width of posterior face of centrum . . . . . .0185 " " depth " " " " " 0185 " width across transverse processes . . . . . .053 " prezygapophyses ...... .034 First dorsal, length of centrum. ........ .024 " width of posterior face of centrum ...... .024 " " depth " « « ,, 0,73 " " width across diapophyses ....... .054 Seventh ? dorsal, length of centrum ........ .023 " width of posterior face of centrum ..... .023 depth " " " 0145 " " width across diapophyses. ...... .038 " " greatest width of neural spine ...... -0135 " " length of neural spine ....... .056 Eighth ? dorsal, length of centrum ........ .0226 " width of posterior face of centrum ..... .025 " " depth " " " " " 0145 Third ? caudal, length of centrum ........ .025 Fourth? " " " " 0237 " width of posterior face of centrum . . . . . .017 " " depth " " " " " 013 " " width across transverse processes ..... .056 Fifth? " length of centrum 0238 " " width of posterior face of centrum ..... .017 " " depth " " nun 0136 " " width across transverse processes (approximate) . . . .048 Scapula, length ........... -1378 " width of neck .......... .025 " antero-posterior diameter of glenoid cavity, including coracoid process .0345 " transverse diameter of glenoid cavity ...... .022 Radius, length ........... .1265 " width at proximal end . . . . . . . . . .018 " " « distal " 0228 Ulna, length 165 " greatest width of olecranon process ....... .024 " " " below sigmoid cavity ...... -0305 Femur, length ........... -1745 " width at proximal end ......... .049 " " distal end 036 Second rib, length .......... .066 Metacarpal II, width at proximal end . . . . . . . .012 " III, " " distal " 0115 IV, length .0435 " " width at proximal end ...... -Oi I " " " " distal " 012 360 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. Metacarpal V, length .......... .0245 " " width at proximal end ....... .0102 " " " " distal " . . . oil BORHY/ENA EXCAVATA Ameghino. (Plates XLIII; XLIV ; XLV, Fig. i ; XLVI, Fig. 4; LII, Figs. 2, 6; LIII, Figs. 4-4/7. ) Borhycena excavata Amegh.; Enum. Syn. des Especes de Mammiferes Fossiles des Formations Eocenes de Patagonie, p. 121, 1894; Bol. Acad. Cordoba, p. 377, 1894. A remarkably perfect skull (No. 15, 120), five cervical vertebrae, an incom- plete posterior dorsal, a fragment of the left scapula and two sternal seg- ments are referable to Borhycena excavata. This material was collected by Mr. Peterson from the Lower Santa Cruz beds ten miles south of Coy Inlet. B. excavata may be readily identified by the strong contraction of the facial portion of the skull just anterior to the infraorbital foramen. As a result of this contraction, the shape of the tooth row is rendered strongly crescentic, with the canine and anterior premolar offset externally to a greater degree than in B. tuberata. The palate retains the same width from the canine to the median premolar, while in B. tuberata its width steadily increases posteriorly. The skull is smaller than in the latter species, the arches more abruptly expanded and the sagittal and lamb- doidal crests lower. The interorbital tract is not as flat as in B. tuberata, the postzygomatic portion of the brain case is relatively less elongate and the infraorbital foramina are larger. The associated skeletal parts are smaller than in B. tuberata, but do not differ in structure sufficiently to require separate description. MEASUREMENTS. Skull, length on median basal line . . . . . . . . .187 " occipital condyle to tips of premaxillae . . . . . .195 " width across zygomatic arches ....... .140 " interorbital width .......... .056 Face, length .0655 Cranium, length to anterior border of orbit . . . . . . .133 width at narrowest part of brain case ...... .0245 Occiput, height ........... .051 " width . .062 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 361 Palate, width between canines . . . . . . . . .024 " "MA 065 " length from root of median incisor to palato-narial border . . .092 Nasals, length ........... .061 " width at anterior extremity ........ .0145 " " " posterior expansion ........ .049 Upper dentition, length from median incisor to M± . . . . . .100 " " " of space occupied by premolars .... .027 " " " " " " " molars 043 Median incisor, width of crown . . . . . . . . .0026 Lateral .<«<<<•< 003 Upper canine, antero-posterior diameter at alveolar border . . . . -0145 " " transverse " " " " oio " " length of crown ......... .027 Anterior premolar, antero-posterior diameter ...... .008 " " transverse " ...... .004 Median " antero-posterior " ...... .009 " " transverse " at widest part . . . .005 Posterior " antero-posterior " . . . . . . . .0107 " " transverse " at widest part . . . .0085 Ml, antero-posterior diameter . . . . . . . . . .oil " transverse " 008 M^, antero-posterior " . . . . . . . . . . .0125 " transverse ".......... .0083 M^, antero-posterior ".......... .014 " transverse ".......... .on MA, antero-posterior ".......... .006 " transverse ".......... .008 Atlas, transverse breadth across anterior cotyles ..... .043 " width of neural arch ......... .017 " " " intercentrum including median process . . . . .0115 Axis, length of centrum including odontoid ...... .045 " " " odontoid 0096 " width over anterior cotyles . . . . . . . . .0346 " height of neural spine above centrum ...... -0345 Third cervical, length of centrum ........ .0213 width across transverse processes ..... .046 " height of neural spine above centrum ..... -0255 Fourth " length of centrum ........ .0205 " " width across transverse processes ..... .044 " " height of neural spine above centrum . . . .023 Fifth " length of centrum .021 " " width across transverse processes .... .043 Sternal segments, average length ........ -0305 362 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. PROTHYLACYNUS Ameghino. (Plates XLVII-LI ; LII, Figs. 3, 5 ; LIII, Figs. 5-8 ; LIV, Figs. 2, za, 8, 9, 14 ; LXI, Fig. 2.) Prothylacynus Amegh.; Nuevos Restos. Mamif. F6s. Patagonia Austral, p. 26, Aug. 1891 ; Revista Argentina Hist. Nat. I, entr. 5^, p. 312, Oct., 1891. This genus, so frequently mentioned in previously published discussions of the relationships of South American and Australian marsupials, is represented in the Princeton collection by the remains of one individual referable to the species Protliylacynus patagonmis (No. 15,700), comprising an incomplete skull, mandible, parts of both scapulae, humerus, both radii, ulna, pelvis, both femora, tibia, fibula, patella, part of the hind foot, the scaphoid and a few phalanges of the fore foot, six cervicals, four dorsals, three lumbars, the sacrum, five caudals and fragments of the ribs and sternum. The -exceptionally perfect state of preservation of this material permits full comparison with living forms and indicates in no uncertain fashion the true relationships of the animal, a discussion of which will be found on a later page. Dentition (Pis. XLVII ; XLVIII, figs, i, 2). --The greater part of the facial region of the skull has been weathered away, invoking the incisors, canines and the anterior and median premolars. In the side view (PI. XLVII), these have been supplied from Borhycena. Prothylacynus dif- fers, however, in possessing an additional upper incisor (Ameghino, 1894, p. 121). The posterior superior premolar is a stout, double-rooted tooth, with a broad heel without heel cusp. It is considerably smaller than this tooth in Thylacynus. The first, second and third molars increase rapidly in size posteriorly. In M1 and M-, the protocone is a well-developed bunoid cusp, but is absent in M-, although the inner root supporting this part of the tooth crown is larger than in the preceding molars. The antero-external style, the only one of the outer row of styloid cusps remaining, is considerably larger than in TJiylacynns. With this excep- tion, the outer cingulum is entirely wanting. The fourth molar is more reduced than in Thylacynus. The protocone is represented by a broad smooth surface. The metacone is vestigial and the posterior root sup- porting it almost obliterated. The high conical paracone is connected by a sharp ridge with the antero-external style. In the mandible, the crowns of the three incisors have been broken off, SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 363 but the root of the second tooth in the series is seen to be displaced behind the median and lateral members. The tip of the canine is miss- ing, and the crown is broadly grooved on the inner side. The premolars are closely crowded. The anterior tooth is in contact with the canine and lies obliquely to the long axis of the jaw. The crowns of all the lower premolars are laterally compressed, with large, simple heels. The median and posterior teeth are of the same size. As in the superior series, the molars increase rapidly in size posteriorly. In the first molar the pro- toconid, paraconid and heel are in line. In the remaining teeth, the paraconid is deflected more and more internally, forming with the pro- toconid a powerful shear against the inner side of the metacone spur of the upper molars. The protoconid is high and conical, with a sharp anterior edge. It is separated from the lower and more blade-like para- conid by a deep cleft. The heel of M4 is far more reduced than in Tkylacynus, inclosing a shallow basin with a single high point on its posterior rim. On the anterior molars the heel is narrower and flatter than in Thylacynus, with a single prominent cusp at its posterior extremity. The hypoconid is greatly reduced, more so than in Cladosictis. The prominent cusp mentioned above occupies the position of the hypoconulid. The entoconid is smaller than in Thylacynus and scarcely distinguishable as a separate cusp. A small antero-external, shelf-like cingulum is de- veloped on the second, third and fourth molars. Skull (Pis. XLVII ; XLVIII, figs, i, la, 3).— As in all the Santa Cruz thylacynes, the skull is large in proportion to the size of the body, although relatively less so than in some of the smaller genera. The brain case is long, low and narrow, considerably depressed in the region above the tentorium and greatly constricted postocfritally. The facial region has been almost entirely destroyed by weathering, but was probably not unlike Borhycena and has been so represented in the figure (PI. XLVII). Enough is preserved to show the single, moderate-sized infra- orbital foramen, situated above the posterior prcmolar, and the imperforate facial expansion of the lachrymal. The zygomatic arches are robust and intermediate in degree of expansion between Borhyana and AmpkipfO- mverra. The orbits are not placed as far forward as in the latter genus, their anterior border coinciding with a line drawn through the posterior root of the second molar. Parietal and supraoccipital exhibit the same relationship as in Borhycena. 364 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. In the posterior view (PI. XLVIII, fig. 3), the occiput is seen to be quadrangular in outline, unlike the triangular occiput of Thylacynns. Its upper border projects considerably beyond the condyles. The exposure of the mastoid is relatively smaller than in Thylacynns. Unlike that genus, the condyles are wider dorsally, more obliquely placed and less sessile. The base of the skull is in excellent preservation, permitting full com- parison with the recent genus (cf. PL LXV, fig. 10). The basioccipital is broadly keeled below. The paroccipital processes are short, blunt and massive. The condylar foramen is large. It is preceded by a very small accessory foramen. The tubercles for the origin of the recti capitis muscles are about as large as in Thylacymis. The alisphenoid is without dilata- tion and does not form an auditory bulla. It is perforated by the foramen ovale directly opposite the glenoid cavity. The tympanic is unfused with the elements adjacent and has not been preserved. The petrous is larger and more completely hemispherical than in Thylacynns, the relative po- sition of the fenestrae remaining the same. The basisphenoid is broadly keeled and convex in cross section in con- trast with its plane, or slightly concave section in TJiylacynus. It is per- forated by a single canal, that for the internal carotid artery. The ali- sphenoid ridge, which is confluent with the auditory bulla in Thylacynns, is but slightly developed and is entirely wanting posteriorly. The post- glenoid processes are proportionately shorter than in Thylacynns, but the preglenoid processes are much larger. The foramina of the lateral sinuses, especially the post-glenoid and sub-squamosal, are larger than in the recent genus. A small foramen pierces the jugal process of the squa- mosal above the glenoid cavity, as in some specimens of Thylacynns. The palate is without vacuities, but is pierced by several accessory palatal foramina. Its posterior margin is perforated by a double neuro-vascular canal. The narial border is slightly thickened. The margins of the palate are depressed to accommodate the tips of the lower molars, as in most carnivorous marsupials. In proportion to its length, the mandible is much deeper than in Thy- lacynus. The anterior margin of the coronoid is convex, in contrast with its slight concavity in Thylacynns and is less steeply inclined posteriorly ; the masseteric fossa is more broadly rounded anteriorly and the posterior margin of the angle less deeply notched. Both rami are firmly coossified in the symphysis without trace of suture, while in Thylacynns symphysial SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 365 union is ligamentous. Three mental foramina are present, the largest of which lies beneath the anterior and median premolars. The others are situated beneath the anterior root of the first molar and posterior root of the second molar respectively. The condyles, unlike Thylacynus, in- crease in width externally. Vertebral Column; Ribs and Sternum. — The atlas (PI. LI 1 1, figs. 5, 5*2) differs from that of Borhycena, Amphiproviverra and Tliylacynus in having the intercentrum firmly fused with the base of the neural arch without trace of suture. The small neuro-arterial foramina are placed nearer to the anterior atlanteal margin than in the other Santa Cruz genera. The canal for the vertebral artery is small. It enters the neural arch within the spinal canal about midway between the neuro-arterial for- amen and the condyles, emerging on the lower surface (PI. LIII, fig. 5«) near the anterior margin of the base of the transverse process. The trans- verse processes are considerably thickened ; their antero-posterior basal constriction is much less than in Thylacynus. The posterior border of the inferior arch supports a small median tubercle. i^,The axis (PI. LIT, figs. 3, 5) resembles closely that of Borhycena, dif- fering from Thylacynus in the greater width of the neural spine anteriorly, the more attenuated extremities of the transverse processes and the greater depth of the posterior inferior keel. The posterior portion of the neural spine has been broken. It was probably as long and heavy as in Bor- hycena tuberata. The odontoid tapers slightly anteriorly, but less so than in Thylacynus. The anterior cotyles are extended to the same degree as the transverse processes. The lower surface of the centrum is almost the same as in Borhycena (cf. PI. LII, figs. 5, 6). The anterior portion of the inferior keel has been broken, which accounts for its apparently lower elevation than in Borhycena. The third, fourth and fifth cervicals differ from those of Thylacynus in having the diapophyses better differentiated from the inferior lamellae. The centra are strongly keeled inferiorly. In the fifth cervical the pos- terior portion of the keel is bifid. The upper surface of the neural arch of the third cervical is perforated by a pair of large foramina. Of the sixth cervical, an uncharacteristic fragment remains. Small foramina pierce the lateral walls of the neural arches of the third to the sixth cervicals, as in Thylacynus and Borhycena. The neural spines of the posterior cervicals have not been preserved. 366 PATAGONIAN EXPEDITIONS: PAL/GONTOLOGY. The seventh dorsal, if the position of the vertebra represented in fig. 4, PI. LI, has been correctly interpreted, is of about the same size as the corresponding vertebra in Thylacynus, but is considerably smaller than the seventh dorsal in Borhycena (PI. XLV, fig. 6). The neural spine is much wider transversely than in the former genus. The centrum is with- out inferior keel. The tenth dorsal, or anticlinal vertebra, (PI. LIII, fig. 7) lacks the tip of the neural spine, the anterior margin of which slopes steeply forward, while the posterior margin descends vertically. Promi- nent metapophyses and anapophyses are developed on the tenth dorsal and also on the succeeding dorsals, so far as preserved (PI. LXI, fig. 2). The centra of the tenth and eleventh dorsals are without inferior keels. The centrum of the twelfth is slightly keeled. The dorso-lumbar formula was probably the same as in Thylacynus — thirteen dorsals and six lumbars. Three of the latter are preserved (PI. LI, fig. 5). The centra are wider posteriorly than anteriorly, producing an hour-glass shape. Unlike Thylacynus, they are not keeled inferiorly. As in that genus, the lower surface of each centrum is perforated by one or more large foramina. The neural spines, although incompletely pre- served, are seen to be much heavier than in the recent genus. Meta- pophyses are less strongly developed than in the posterior dorsals. The anapophyses decrease regularly in size, until, in the sixth lumbar they are mere points. The transverse processes are heavier than in Thylacynus, and, as in that genus, curve forward and downward. The sides of the neural arches are pierced by small foramina. Two vertebrae are coossified in the sacrum (PI. L, figs. 3, 3^) which differs from that of Thylacynus (cf. text-fig. 3, b, c] in the greater size and more complete fusion of the sacral elements. The dorsal interverte- bral fenestra, present in the latter genus between the first and second sacrals, is wanting in Frothy la cynus. The proximal caudals (PI. LIII, fig. 8) are much larger than in either Thylacynus or Borhycena. The caudal centra rapidly increase in length, retaining functional postzygapophyses farther back in the series than in Thylacynus. The transverse processes have been more or less broken from all the caudals preserved, but were evidently very heavy (PI. LIII, figs. 6, 8). Facets for chevrons are present on all the caudals, beginning with the third. The proximal caudals are unkeeled. The seventh, eighth and ninth are doubly keeled inferiorly. Judging from the size of SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 367 the caudals, the tail of Prothylacynus must have been not only longer but much thicker at the base than in Thylacynus. It is possible that too much flexibility has been given to it in the drawing of the restored skele- ton. The floor of the neural canal in the cervical, dorsal, lumbar and anterior caudal vertebrae is keeled and perforated on either side of the keel by a large foramen, as in Thylacynus and Borhycena. The epiphyses, although more or less annular, as in Borhycena, do not have such promi- nent bosses from the centra projecting through the median perforations. Frequently the latter are absent. Fragments of several posterior ribs are preserved, but are too incom- plete to describe. They have about the same dimensions as the posterior ribs in Thylacynus. The incomplete presternal segment represented in fig. 6, PI. LI, is considerably smaller than that of either Thylacynus or Borhycena, differing from both in possessing a strong, inferior, median keel. Appendicular Skeleton. — The scapula (PI. XLIX, figs. 2, 2a) is shorter than in Thylacynus, with thicker neck and longer coracoid process. The surface of the supraspinous fossa is convex, in contrast with the approx- imately plane surface of this region in Thylacynus. In part, the con- vexity has been accentuated by crushing. The glenoid cavity is almost circular in outline. The coracoid process projects greatly below the glenoid margin. Its anterior border is strongly inflected. The infra- spinous fossa is somewhat wider than in either Borhycena or Thylacynus. As in those genera, the axillary border is strongly deflected outwardly. The spine has been broken, destroying the acromion. The border of the suprascapular notch is perforated by a small foramen. The humerus (PI. XLIX, figs. \-\b] is shorter than in Thylacymis but much heavier. The head is very broad and projects considerably beyond the posterior surface of the shaft. The greater tuberosity rises slightly above the head. The lesser tuberosity is low, not as sharply separated from the head as in Thylacynus. A third tuberosity rising from the inner side of the shaft below the lesser tuberosity probably marks the point of insertion of the coraco-brachialis muscle. The deltoid ridge is long and powerful, extending at least two thirds the length of the shaft. The distal end is exceedingly broad, with greatly enlarged supinator ridge and elon- gated inner epicondyle. The condylar surface is much narrower antero- posteriorly than in Thylacynus, but considerably wider transversely, with 368 PATAGONIAN EXPEDITIONS'. PALAEONTOLOGY. the trochlea and capitellum as sharply marked as in the recent genus. The supinator ridge is broad, terminating proximally in a hook-shaped process. A large entepicondylar foramen is present. The radius (PI. LI, figs. i-\b] may be readily distinguished from that of Borhycena by the great width of the distal portion of the shaft. The head is elliptical in outline and much broader antero-posteriorly than in that genus. The bicipital tubercle is much larger than in Borhycena and is situated on the postero-external margin of the shaft. The shaft has a slightly greater degree of curvature than in the last-named genus. It is sharply triangular in cross section distally, closely resembling in this respect the radius of Perame/es, but quite unlike Thylacynus. The styloid process is not as greatly elongated as in Borhyczna. The distal articular surface (PI. LI, fig. i£), unlike the condition in Thylacynus and Borhycena, is concave. The ulna (PI. LI, figs. \-\b] lacks the backward curvature characteristic of Borhycena and Thylacynus. The olecranon process is slightly shorter and somewhat heavier than in the latter genus, and is broadly grooved on the inner side. The greater sigmoid cavity is deep, as in Borhycena, but its proximal wall is much wider. In the lesser sigmoid cavity, the radial facets are more widely separated than in either Thylacymis or Borhycena. Beneath the sigmoid cavity, on the inner side of the shaft, is a deep rugose pit for insertion of the brachialis muscle. The shaft is straight, considerably compressed laterally and broadly grooved on the outer side. The styloid process is hemispherical in shape. The radial facet is smaller than in Thylacynus and is supported on a distinct pedicle. The scaphoid and a few phalanges of the fore foot are preserved. The proximal surface of the scaphoid is narrower in dorso-palmar section than is Borhycena; the magnum facet is smaller and the lunar facet more deeply concave than in the latter genus. The association of the phalanges represented in fig. 9, PI. LIV, is somewhat doubtful. That those of the first and second row were carried with respect to each other at a con- siderable angle (cf. PI. LXI, fig. 2) is plainly indicated by the restriction of the trochlear surface of the proximal phalanx to the distal and palmar surfaces, and by the extension dorsally of the proximal articular surfaces of the phalanges of the second row. The latter are much heavier than in Thylacynus. The ungual (PI. LIV, figs. 9, 14) is greatly compressed laterally, with a slight median cleft. The tip has been broken, but was SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 369 probably sharp. A hood is developed to about the same extent as in Borhycena. The subungual processes are large and an ungual foramen is present. The greater part of both halves of the pelvis is preserved (PI. XLIX, figs. 3; PI L, fig. i), but, unfortunately, the pubes are almost entirely missing. The peduncular portion of the ilium is considerably heavier than in Thylacynus, supporting a very large tubercle for the origin of the rectus femoris muscle. The gluteal surface is broadly expanded, but with little trace of the longitudinal grooving noticeable in the recent genus. The ilio-pectineal eminences are very large and rugose. The ischium is incomplete posteriorly, lacking the ischial tuberosity, but both acetabular and post-acetabular portions are heavier than in Thylacynus. The ischial spine is small. The pubis, so far as preserved, is narrower antero-pos- teriorly than in the recent genus. The acetabulum is large and deep. Beneath the acetabular notch, the border of the obturator foramen is exceedingly sharp. Posteriorly, it assumes the usual convex contour. The femur (PI. L, figs. 2, za], although of almost the same length as in Thylacynms, is much heavier. The large hemispherical head is sup- ported on a long neck. The greater trochanter rises above the level of the head, from which it is separated by a broad interval. The digital fossa is long and deep and the intertrochanteric ridge high and narrow. The lesser trochanter is slightly larger than in Thylacyims and is separ- ated farther from the head. The shaft is straight, circular in cross-section at the middle, but flattened antero-posteriorly at either end. Above the condyles, the distal end is semicircular in cross section. The condyles have about the same posterior extension as in Thylacynus, but are much flatter transversely, resembling BorJiyczna. The intertrochanteric fossa is wider and deeper, and the rotular groove broader and shallower than in Thylacynus. The patella (PI. L, figs. 4, 4«) is a stout, wedge-shaped element, broad at the proximal end, but tapering distally to a thin edge. The rotular surface is irregularly pentagonal in outline, concave in vertical section and convex transversely. The anterior surface is rugose for tendinous attachment. The tibia (Pis. XLIX, fig. 4 ; LI, fig. 2), unlike that of Thylacymts, is considerably shorter than the femur. The head supports two broad, flat surfaces for articulation with the femoral condyles. The spine is 370 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. much lower than in Thylacynus. Unlike the latter, the distal end of the shaft is curved backward. The articular surface for the astragalar trochlea is broadly concave, that for contact with the inner side of the astragalus convex. The internal malleolus is much heavier than in Thylacynus and the fibular facet larger. The shaft is triangular in cross section. The fibula (PI. LIII, fig. i) is much heavier than in Thylacymts. The proximal end is greatly thickened, supporting two approximately plane facets for articulation with the tibia and fabella respectively. The distal end carries three articular surfaces, a round tubercle for lateral contact with the tibia, a flat irregularly triangular facet for the outer margin of the astragalar trochlea, and a concave elliptical facet for the calcaneum. The peroneal groove is broader than in Thylacynus and less perfectly defined. The shaft is straight, with strongly marked interosseous ridge. In contrast with the robust femur and tibia, the pes is rather feeble. The astragalar trochlea is remarkably flat, with the articular surfaces for the tibia and fibula separated by a faint groove. The tibial surface is produced distally down the dorsal aspect of the neck, as in Amphiprom- •verra, but unlike Sarcophilus and Thylacynus. The head is less obliquely placed than in the latter genus and is supported on a long heavy neck. Its distal end bears a single convex facet for the navicular. The inner side of the neck is deeply grooved for articulation with the internal tibial malleolus (PI. LIV, fig. 2). In plantar aspect (PI. LIV, fig. 20), the semicircular ectal facet is seen to be deeply concave antero-posteriorly. The sustentacular facet is irregularly lobate in outline and strongly con- vex in dorso-plantar section. Several small foramina pierce the body of the astragalus at the point occupied by the astragalar foramen in certain primitive placentals. These are not visible in the dorsal view (PI. LIV, fig. 8). The navicular has much the same shape as in Thylacynus, but is considerably larger and differs also in supporting all three cuneiforms, whereas in the recent genus, the outer cuneiform is supported almost entirely by the cuboid (cf. PI. LIV, fig. 8, and text-fig. 4, b}. The cuboid has the same shape as that of Sarcophilus, from which it differs slightly in the arrangement of some of the facets. The cuboid in Thylacynus does not lend itself to comparison, owing to the outward shifting of the lateral cuneiform just mentioned. The proximal and distal surfaces of this element in Prothylacynus are almost the same as in Sarcophilus. The chief difference is in the facet for the outer cunei- SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 371 form, which does not extend to the distal margin of the bone, as in the latter genus. The facet is oval in outline and slightly concave ; that for the navicular, which immediately adjoins it, is triangular and convex. The cuneiforms resemble those of Sarcophilus rather than Thylacynus. The inner cuneiform is laterally flattened, supporting distally an oval, con- cave facet for the metatarsal of the hallux. Laterally it is in contact with the mesocuneiform by a narrow elliptical facet. The mesocuneiform is a small, oblong element, much shorter than the adjacent bones, so that the proximal end of the second metatarsal is received between them, articu- lating with the mesocuneiform by a concave oval facet, and with the ectocuneiform by a small plane facet confined to the dorsal margin of that bone. The outer cuneiform differs from that of Thylacynus and resem- bles Sarcophihis in lacking the large plantar process present in the former genus. It is in contact with the cuboid by a large triangular facet and has a very small facet for the fourth metatarsal. The facet for the meso- cuneiform is narrow and sigmoid in dorso-plantar section. The metatarsals are shorter and proportionately heavier than in Thyla- cynus. The hallux is vestigial, its distal end terminating in a rugose knob, which did not support phalanges. Proximally, the shaft bears a broad, plane facet for contact with metatarsal II. The third metatarsal is little more than half as long as the corresponding element in Thyla- cytms. The proximal end lacks the prominent plantar tubercle present in the latter and the articular surface for the ectocuneiform is more convex. The shaft is slightly curved and considerably flattened antero-posteriorly. The trochlear surface extends dorsally about as far as in Thylacynus. As in that genus, the keel is confined almost entirely to the plantar sur- face. The fourth metatarsal of the right pes is preserved. It is consid- erably longer than the third metatarsal. The proximal end is wider and more strongly convex in dorso-plantar section than in Thylacynus. The antero-posterior shortening and especially the transverse flattening of the trochlear surface of the astragalus are believed to indicate a planti- grade gait. Restoration (PI. LXI, fig. 2). — The restored skeleton shows effec- tively the large head, long neck and short limbs characteristic of all the Santa Cruz marsupial carnivores. The lack of proportion between the femur, tibia and hind foot is at once apparent. The length of the tail is conjectural, but the error is probably in underestimating rather than in 372 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. overestimating the lengths of the posterior caudals. It may have pos- sessed less flexibility than is indicated in the drawing. No trace of a clavicle has been preserved, and it may have been rudimentary, as in Thylacynus, (Cunningham, p. 2, 1882) and not attached to the acromion. It has been supplied in the restoration from analogy with existing car- nivorous marsupials. The length of the back has been determined by comparing the length of the dorsal series in Thylacynus with the lengths of the posterior lumbars in that genus and Prothylacymts. The depth of the chest is doubtful, as none of the longer ribs are completely preserved. Habits. — It may be inferred from the large pointed claws that Prothy- lacynus was more active in attacking prey than Borhycena. The greater degree of outward deflection of the metacone spur in the upper molars indicates a more completely carnivorous habit than in the latter genus. The reduction of the hallux is an adaptation to terrestrial progression. PROTHYLACYNUS PATAGONICUS Ameghino. (Plates XLVII-LI ; LII, Figs. 3, 5 ; LIII, Figs. 5-8 ; LIV, Figs. 2, 2«, 8, 9, 14; LXI, Fig. 2.) Prothylacynus patagonicus Amegh.; Nuevos Restos Mamif. F6s. Patagonia Austral, p. 26, Aug., 1891 ; Revista Argentina Hist. Nat. I, entr. 5^, p. 312, Oct., 1891. The preceding account of the osteology of Prothylacynus is based al- most entirely on the remains of a single individual of the classic species Prothylacynus patagonicus (No. 15,700) from the Upper Santa Cruz beds at Killik Aike. A reference to the discovery of this unique specimen will be found in Mr. Hatcher's Narrative of the expeditions (this series, Vol. I, pp. 53, 54). The American Museum collection contains a fragment of the right maxilla with the second, third and fourth molars in place (No. 9561 Am. Museum), also from Killik Aike. As a description of the anatomy of the hard parts has already been given, it remains only to tabulate the principal measurements. MEASUREMENTS. Cranium, length, condyles to anterior orbital border . . . . . .138 palato-narial border to condyles, inclusive . . . .109 least width of brain case ........ -0175 Skull, greatest width across arches . . . . . . . . . 1 20 Occiput, height .0485 " width at base .......... .057 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 373 Palate, width between M±. . . . . • Mandible, length .......... " " Mj to outer end of condyle . . . . . " height of coronoid process above angle .... " " " condyle above angle ...... " depth of horizontal ramus below MT . . . . . " " " " " posterior premolar . Upper dentition, length of space occupied by molars . Lower " " from anterior border of canine to M( inclusive. " " " of space occupied by molars . . . " " " " " " premolars Posterior superior premolar, antero-posterior diameter " " transverse M-L, antero-posterior diameter . . . . . " transverse diameter ......... M^, antero-posterior diameter ........ " transverse diameter ......... M£, antero-posterior diameter ........ " transverse diameter . . . . . MA, antero-posterior diameter ........ " transverse diameter ......... Lower canine, antero-posterior diameter at base . . . . " " transverse diameter at base . Anterior inferior premolar, antero-posterior diameter " " " transverse diameter . Median " " antero-posterior diameter " " " transverse diameter Posterior " " antero-posterior diameter " " " transverse diameter . . . . Mj, antero-posterior diameter ..... " transverse diameter ...... Mj, antero-posterior diameter .... " transverse diameter ..... Mj, antero-posterior diameter ..... " transverse diameter ..... M^, antero-posterior diameter . " transverse diameter .... Atlas, transverse breadth .... " width of neural arch .... " " " inferior arch .... Axis, length of centrum, including odontoid process " " " odontoid process .... " width across anterior cotyles . " depth of posterior face of centrum . " width " " " " . Third cervical, length of centrum .... .060 .1655 •0755 .075 •033 .030 .024 .042 .090 .048 .026 • .010 .005 .0105 .008 .0125 .0095 •0135 .012 .005 .009 .012 .008 .0078 .003 .0095 .0045 .0095 .0046 .010 .005 .011 .0055 .0125 .0065 .015 .008 .079 .017 .0115 •053 .012 .036 .013 .020 .028 374 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Third cervical, width across transverse processes ..... .050 " " " " prezygapophyses . . . . . . .0315 Fourth cervical, length of centrum ........ .025 " " depth of posterior face of centrum ..... .014 " " width " .<<«.< 0,95 " " " across prezygapophyses ...... .038 Fifth " length of centrum ........ .025 Seventh dorsal, length of centrum ........ .021 " depth of anterior face of centrum . . . . . .014 " " width " 0165 " " width across diapophyses . . . . . . .040 " " " of neural spine . . . . . . . .010 Tenth dorsal, length of centrum ........ -0235 " depth of anterior face of centrum . . ... . .014 " " width ".'"»« 020 Eleventh dorsal, length of centrum ........ -0235 " " depth of anterior face of centrum . . . . . .016 " " width " " 022 " " " " neural spine . . . . . . . .016 " " " across anapophyses ...... .027 Twelfth dorsal, approximate length of centrum ...... .0235 " depth of anterior face of centrum . . . . . .016 " width " " " 022 " " " of neural spine . . . . . . . .013 " " " across anapophyses ....... -0275 Fourth lumbar, length of centrum ........ .0345 " " depth of anterior face of centrum ..... .018 " width " " " " 0225 " " " " neural spine ....... .0165 " " " across prezygapophyses ...... .030 " " " of transverse process at base . . . . . .015 Fifth lumbar, length of centrum . . . . . . . . .0335 " " depth of anterior face of centrum . ..... .0185 width " ". 0235 " " " " neural spine . . . . . . . . .0155 " " " across prezygapophyses . . . . . . .033 " " " of transverse process at base . . . . . .015 Sixth lumbar, length of centrum ..... . . .033 depth of posterior face of centrum . . . . . .0185 " width " « « « 029 " " " " neural spine 015 " " across prezygapophyses ...... .034 " " " of transverse process at base ..... .016 Sacrum, length of centra .......... -0505 width at point of greatest expansion of auricular processes . . .061 depth of anterior face of centrum of first sacral .... .018 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 375 Sacrum, depth of posterior face of centrum of second sacral . . . .015 " width " " " ' " 022 Third caudal, approximate length of centrum ..... .0265 " " width of anterior face of centrum ...... .023 " " depth " " " " " . . . , . .015 Fourth caudal, approximate length of centrum ...... .0255 " " width of posterior face of centrum ..... .0225 depth ' " " . . . .016 Eighth caudal, length of centrum ........ .041 " " width of posterior face of centrum ..... .0205 " " depth " " •• ii ii o,jj Ninth caudal, length of centrum . ..... .044 " " width of posterior face of centrum . . . . . .02 1 5 " " depth "«"'<" ..... .016 Scapula, length - .132 " width of neck . . . . . . . . . . .03 1 5 antero-posterior diameter of glenoid cavity ..... .026 " transverse " " " "..... .022 Humerus, length . . . . . . . . . . . .155 " width of proximal end ........ -037 " depth " " 0445 " width of distal end 053 " length, tip of supinator ridge to capitellum, inclusive . . . .063 Radius, length . . . . . . . . . . . .130 " antero-posterior diameter of head . . 015 " transverse " " " . .0196 " antero-posterior " " distal end . . . . . . .015 " transverse " " " " .. • 0275 Ulna, length . .165 " greatest width below sigmoid cavity .031 " antero-posterior diameter of distal end .... .017 Pelvis, length (approximate) ...... .184 Ilium, width at greatest expansion ..... -0425 Acetabulum, antero-posterior diameter .... .029 " transverse " ... .026 Femur, length from great trochanter to outer condyle . .198 " width of proximal end ...... -0525 " " distal " -042 " depth of distal end -03 1 " diameter at middle of shaft ... .017 Patella, length . . . .0285 " width -0225 " greatest depth -OI2 Tibia, length ...... •I725 " width of proximal end ..... -°395 " " distal " 0235 376 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. Fibula, length 164 " antero-posterior diameter of proximal end ..... .020 " distal " 0175 Astragalus, length ........... .026 " width of trochlea . . . . . . . . . .019 " " " neck ......... .009 Metatarsal I, length 0245 " " width of proximal end ........ .008 " " " distal " 0065 Metatarsal III, length 037 " " width of proximal end ....... .0075 " " distal " . 012 CLADOSICTIS Ameghino. (Plates LII, Fig. 4; LIII, Figs. 3, 30, 10; LIV, Figs, i, 3, 4, 10, 12; LV-LVIII; LIX, Figs. '; LXI, Fig. i.) Cladosictis Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 7. 1887. Hathliacynus Amegh. ; Ibid., p. 7, 1887. Anatherium Amegh.; Ibid., p. 8, 1887. Proviverm (Rutimeyer) Amegh.; Revista Argentina, etc., I, pp. 149-150, 1891. Dolichocephalic thylacynes intermediate in size between Amphipro- viverm and Frothy lacymis. Dentition (Pis. LIV, fig. i ; LV, figs, i, 3; LVI, figs, i, 3; LIX, figs. 6-7*)- The median incisors (PI. LV, fig. 3) are laterally compressed, unlike the corresponding teeth in Dasyurus, DidelpJiys and Amphiproviverra. The remaining incisors, so far as preserved, have laterally compressed crowns, and increase regularly in size externally. The crowns of the lateral pair are triangular in cross section. In all the specimens examined, the incisors are badly worn, reducing the crowns to irregularly truncated, flattened columns. The canines are stout, laterally compressed blades, the posterior margins of which are almost straight, while the anterior margins are moderately convex. The anterior and median upper pre- molars are spaced in both species. The median and posterior premolars are spaced in C. petersoni, unspaced in C. lustratus. The anterior pre- molar is separated by a short interval from the canine. Its crown is a sim- SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 377 pie, laterally compressed blade, with a rounded heel, but without heel cusp. The median premolar is a replica of the anterior tooth on a larger scale. The posterior premolar is enlarged, with a prominent, piercing protocone and a broad heel, with distinct heel cusp. The prominence of the latter varies with the amount of wear to which this part of the tooth crown has been subjected. The first, second and third molars have the protocone well developed. It is slightly cupped in unworn teeth. On M1, the pro- tocone is reduced to a small conical cusp, while the paracone and antero- external style are enlarged, and the metacone is vestigial. In the anterior molars, the proportions of the dental cusps are much the same as in Pro- thylacymis, except that the protocone of M- is well developed. In the inferior series, the incisor crowns show considerable lateral com- pression, but are so badly worn in all the specimens examined that their original shape cannot be ascertained. As in many recent carnivorous marsupials, the root of the second tooth in the series is displaced pos- teriorly with respect to the roots of the median and lateral incisors. The canines are reniform in cross-section owing to the presence of a broad groove on the lingual side. The anterior surface of the crown is flattened (PI. LVI, fig. i ; LIX, figs. 7, fa) apparently to accomodate the tooth to the narrow groove in the premaxilla, into which its point is received when the jaws are brought into occlusion. The amount of spacing between the lower premolars varies somewhat in different individuals of the same species, the greatest amount of variation in this respect occurring in the width of the space between the median and posterior premolars. The anterior and median premolars closely resemble the corresponding teeth in the superior series. Each consists of an enlarged piercing central cusp and a round heel. In unworn teeth, the heel of the median premolar sup- ports a small conical heel cusp. A minute anterior basal cuspule is also present. In worn teeth, the heel is reduced to a broad convex area, and the identity of the anterior cuspule is lost The posterior premolar is enlarged like the corresponding tooth opposing it in the superior series. In unworn specimens a conical heel cusp is present. Characters of generic importance appear in the heels of the lower molars. The same rapid increase posteriorly in the size of the molar crowns is to be noted as in Borhycena and Prothylacynus. The cusps of the trigonid are iden- tical in shape and position with those of the latter genus. In the anterior molars, the heel supports a shallow, basin-shaped depression, bounded by a 378 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. narrow rim. A notch in the posterior margin divides this bounding ridge into two portions, the hypoconid and hypoconulid-entoconid. The latter cusps are not differentiated from each other. In worn teeth, the heel con- sists of two rounded eminences, separated by a shallow groove. In con- trast with the anterior molars, MT has a much smaller heel, ^enclosing a small basin with a single prominent cusp on its posterior rim. An antero-external cingulum is present on the second, third and fourth tooth in the series. Milk Dentition. — Ameghino states (1894, p. 109) that in Cladosictis the canine, median and posterior premolars have deciduous predecessors. The mandible represented in fig. 6, PL LIX, retains the deciduous tooth replaced by the posterior premolar, the germ of which is visible beneath the anterior root of the former tooth. Two large alveoli precede the de- ciduous tooth, evidently for the roots of the median premolar. No tooth germs were found beneath them. The region occupied by the anterior premolar has been destroyed by fracture, and in repairing this break the canine has been too closely approximated to the molars. The root of the canine is hollow, indicating, not that it is a deciduous tooth, but that it was not yet fully erupted and had not ceased its growth at the time of the animal's death. The tooth replaced by the posterior premolar differs from its permanent successor in the small size of the crown, which is greatly compressed laterally, carrying a central cusp preceded by an accessory basal cuspule. The narrow, ridge-like heel is subdivided by a shallow notch into two cuspules. The first and second molars are fully erupted and the third partly so. Skull (Plates LIV, fig. i; LV, fig. i; LVI; LXI, fig. i).-- Perhaps the most striking peculiarity of this remarkable genus is the greatly elon- gated, narrow skull, altogether disproportionate to the size of the body. The facial portion is short and slender and the cranium elongated. The post-orbital constriction is proportionately greater than in the much smaller genus Amphiprovi'verra. The inclination of the upper border of the facial profile varies with the species. The brain (PL LVI, fig. 3) was less convoluted than in Thylacymis, and the brain case proportionately much smaller. The ascending premaxillary processes have approximately the same degree of development as in Amphiprovi'verm, and the nasals are similarly expanded posteriorly. Well defined post-orbital processes are present in C. lustrattts. A short distance above these processes, SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 379 feeble temporal ridges converge to form a long, high sagittal crest, the top of which is practically horizontal. In C. petersoni, only a small por- tion of the anterior extremity of the crest is preserved, but this rises abruptly above the interorbital tract, with decidedly convex superior border. The crest terminates posteriorly in a semicircular lambdoidal frill, which does not project beyond the condyles. The supraoccipital is broadly expanded on the upper surface of the brain case. Between the supraoccipital and the squamosal, a narrow tongue of the parietal reaches the mastoid border of the occiput. The orbits are proportionally smaller than in Thylacymis and are placed much farther forward, their anterior border coinciding with a line drawn through the posterior half of the first molar. A large lachrymal prominence is situated on the orbital rim above the lachrymal duct, which lies wholly within the orbit. The occiput is semi-circular in outline, so far as can be judged from the crushed specimen (No. 15,170), in which this portion of the skull is preserved. The mastoid is exposed to about the same extent proportion- ately as in Protkylacynus. The basioccipital and basisphenoid (PI. LV, fig. i) are flat, unlike Frothy lacynus, resembling in this respect the recent genus. The paroc- cipital processes are broad, dome-shaped masses, lying considerably below the level of the auditory bullae. The latter are formed entirely from the dilated alisphenoids, the petrous not entering into the posterior wall of the bulla, as in some of the dasyures. A large foramen ovale pierces the alisphenoid opposite the glenoid cavity. The palate begins to increase in width back of the posterior premolar and is without vacuities, but is pierced by several accessory palatal foramina. The anterior palatine foramina project a short distance beyond the premaxillo-maxillary suture. A pair of large foramina pierce the maxillary on either side of the median line, either opposite or a short distance anterior to the posterior border of the canine. The usual thickening is observable about the border of the posterior nares. The foramina opening into the lateral venous sinus are large, sub-squamosal, post-glenoid and post-zygomatic foramina being present. The last mentioned opens well within the lip of the post-glenoid foramen. A small foramen pierces the outer side of the squamosal bar above the glenoid cavity. In contrast with the skull, the mandible is remarkably deep and heavy (Pis. LVI, fig. i ; LXI, fig. i). The anterior coronoid border is straight 380 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. as in Thylacynus, but less steeply inclined posteriorly than in that genus. The superior and posterior borders correspond in shape to the same region in Prothylacynus. As a whole, the coronoid is proportionately broader and higher and the masseteric fossa better defined anteriorly than in Thylacynus. The lower border of the horizontal ramus lacks the convexity observable in Prothylacyttus and Thylacynus. The rami are united in ligamentous suture at the symphysis, which extends as far back as the anterior half of the posterior premolar (PI. LIX, fig. 7 a). The mental foramina vary greatly in number and position in different indi- viduals of the same species and even on opposite sides of the same man- dible (cf. Pis. LVI, fig. i ; LIX, fig. 7), and are of no diagnostic importance. Vertebral Column and Ribs. — Cladosictis resembles Prothylacynus in having the atlanteal intercentrum firmly fused with the base of the neural arch (PI. LIII, fig. 3^). The intercentrum retains to a slight extent only the posterior emargination observable in Thylacynus and Borhycena. The foramina for the exit of the spinal nerves are separated from the anterior atlanteal margin by a broader osseous bar than in Prothylacynus, resembling rather in this respect the atlas of Thylacynus (text-fig. 5, 6, c) and Amphiproviverra (PI. LIII, fig. i). The canals for the vertebral artery enter the neural arch within the spinal canal on a level with and a short distance anterior to the superior border of the atlanteal condyles. They emerge on the lower surface of the atlas at the bases of the trans- verse processes (PI. LIII, fig. 3*7). From this point the arteries curved dorsally, traversing a groove in the outer wall of the neural arch just anterior to the transverse process (PI. LIII, fig. 3) and entered the neuro- vascular foramen. The margins of the transverse processes in the speci- mens studied have not been preserved. Their antero-posterior basal constriction is much less than in Thylacynus. The anterior margin of the neural arch supports a pair of tubercles, separated by a deep groove. In Thylacynus there is but a single median prominence in this region (cf. text-fig. 5, b). The axis (PI. LII, fig. 4) carries a powerful, hatchet-shaped neural spine, which projects anteriorly as far as the odontoid and posteriorly beyond the zygapophyses. The ventral surface of the centrum is strongly keeled posteriorly. On either side of the keel is a depressed area, bounded by the dependent edges of the transverse processes, as in Borhycena and Prothylacynus. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 381 In the remaining cervicals the neural spines increase regularly in robustness and probably also in height, although this can not be deter- mined, owing to the fracturing of their extremities. The inferior lamellae of the transverse processes are more wedge-shaped than in Thylacynus, resembling rather this region in Borhycena, and, as in that genus, the dia- pophyses are fully differentiated on the fourth cervical. The transverse processes of the seventh cervical are perforated by the arterial canal. The centra are all heavily keeled inferiorly, but, owing to imperfect preserva- tion, it can not be determined whether the keels decrease in depth in the posterior cervicals, as in Prothylacynus and Thylacynus. The upper sur- face of the neural arch of the third cervical is perforated on the left side by a rather large foramen. A smaller corresponding foramen occurs somewhat farther back on the right side. Small foramina may occupy similar positions in some of the other cervicals. The foramina piercing the lateral walls of the neural arches of the second to the seventh cer- vicals, which are so prominent in Thylacynus (text-fig. 5, d], are present in some of the cervicals of Cladosictis and absent in others. The dorso-lumbar formula is apparently nineteen, although this can not be accurately determined, owing to the incomplete state of preserva- tion of the column. The neural spines of the anterior dorsals are high and broad, but gradually decrease in both dimensions to the tenth, or anti- clinal vertebra. Beyond this point, the strong backward slope of the spines changes abruptly to a forward direction. Metapophyses are devel- oped on the tenth dorsal, and are prominent as far posteriorly as the sec- ond lumbar, beyond which they begin to decrease in height. Ana- pophyses also appear on the tenth dorsal and increase in size posteriorly until the fifth lumbar is reached, when they become smaller. The lumbars (Plate LVII, fig. 6) are six in number. They have heavy neural spines strongly inclined forward, with broad tips flattened superiorly, and wide, thin transverse processes, with a forward and downward curva- ture. The long, keeled centra increase regularly in size posteriorly. In the first, second and third lumbars, the keels are bifid posteriorly, inclos- ing a more or less flattened, wedge-shaped area. The prezygapophyses overlap the outer margin of the postzygapophyses, producing an inter- locking articulation. This applies also to the anterior zygapophyses of the thirteenth dorsal. The sacrum (Pis. LVII, figs i, \a\ LVIII, figs, i, 7, fa) is composed 382 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. of two vertebrae more completely fused than in Thylacynus and lacking the dorsal intervertebral fontanelle so conspicuous in the latter genus. The spines are remarkably feeble in contrast with the heavy spine of the last lumbar. The auricular processes are confined almost entirely to the first sacral. The centra are keeled inferiorly, the first having a single low median keel and the second a pair of parallel keels. The caudals are remarkably heavy. Those associated with No. 15,170, which have been reproduced in the figures (Pis. LVII, fig. 5 ; LXI, fig. i), are interpreted as the third to the tenth. The proximal ones are short, with stout, posteriorly directed transverse processes. The centra of the more distal caudals increase in length and the transverse processes become wider and shorter. The neural canal is complete as far back as the tenth caudal. It is not possible from the material at hand to deter- mine the length of the tail, but judging from the size of the caudals pre- served, it was probably long and heavy. From analogy with Thylacynus, it has been restored with a vertebral formula of twenty-three. Chevrons are present between the third and fourth caudals and are represented either by the small hatchet-shaped pieces themselves, or by surfaces for their attachment, as far back as the caudal series is preserved (PI. LXI, fig. i). A few ribs are associated with both specimens of C. lustratus. The anterior ribs (PI. LVIII, fig. 3) have more cylindrical shafts than in Thy- lacynus. The posterior ribs are slender and sub-round in section. Appendicular Skeleton. — The scapula (PI. LVII, figs. 2-3) corresponds more closely with that of Prothylacynus than with the corresponding ele- ment in Borhycena or Thylacynus. The neck is short, the glenoid cavity oval in outline and moderately deep and the coracoid prominent, with its anterior margin strongly inflected. In shape the scapular fossae are much as in Prothylacynus. The spine is high, but its free border has been somewhat fractured in both specimens, destroying the acromion. The foramen which perforates the margin of the suprascapular notch in 7Yiy- lacynus appears in Cladosictis some distance posterior to the coracoid border. No trace of a clavicle has been found and its introduction in the resto- ration (PI. LXI, fig. i) is conjectural. It may have been rudimentary and not attached to the acromion, as in Thylacynus. The humerus (PI. LV, figs. 2, za] is short and heavy, resembling de- cidedly that of Prothylacynus t but differing in the less prominent epicon- SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 383 dyle and the absence of a hook-shaped termination at the proximal end of the supinator ridge. In the specimen figured, this is not well shown, owing to fracturing of the extremity of the ridge, but it can be satisfac- torily determined from the distal end of the humerus of C. lustratus (No. 15,046). The head is broad and overhangs posteriorly, as in Prothyla- cynus. Distally, the trochlear and capitellar surfaces are better defined than in the latter genus and the capitellum is separated from the anterior margin of the external condyle by a deep groove quite unlike the condi- tion in Thylacynus (text-fig, i, a] and Prothylacynus (PI. XLIX, fig. i). The anconeal and antecubital fossae are deep, but the connection between them shown in the figures (PI. LV, figs. 2, 20) is due to accidental rupture of the thin lamina of bone separating them. The radius (PI. LVIII, figs. 4, 40, 5) is intermediate in shape between that of Borhycena and Prothylacynus. The head is oval in outline, as in the former genus. The shaft is arched, elliptical in cross section prox- imally, but becoming flattened and triangular in cross section toward the distal end. The large bicipital tubercle lies on the posterior surface of the shaft. The ulna is strikingly like that of Prothylacynus, but the sigmoid curva- ture of its posterior margin is somewhat more pronounced. The short, heavy olecranon is grooved on both sides. The shaft is considerably compressed laterally and deeply grooved on the outer side. The ex- tremity of the styloid process is hemispherical, but less sharply differen- tiated from the radial facet than in Prothylacynus. -One of the most interesting and important features in the anatomy of Cladosictis is the great size and decided opposability of the pollex. The feet are preserved with but one specimen (No. 15,046). Unfortunately the carpus is incomplete and the articular surfaces of the elements remaining have been partly destroyed by weathering. The arrangement shown in the figure (PI. LIV, fig. 4) is the same as that of the carpal elements in the matrix. The metacarpals are short and stout, interlocking proximally and spreading widely distally. The metacarpal of the pollex is a robust element with a broad proximal articular surface, convex transversely. Distally, the outer condyle is greatly enlarged, deflecting the proximal phalanx toward the inner side of the foot. The tip of the ungual phalanx of the pollex (PI. LIV, fig. 12) is incomplete, but, so far as preserved, shows no indication of a median cleft. Of the remaining metacarpals, the 384 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. fourth is the longest. The shafts are more slender than that of the pollex and are slightly arched and more or less compressed in the dorso-palmar plane. The structure of the pelvis (Pis. LVII, figs i, ia; LVIII, figs, i, 7, 7*7) is important from the bearing it has on the thylacyne affinities of Clado- sictis. The anterior margin of the pubic symphysis is slightly damaged in the otherwise nearly complete pelvis of C. liistratus (No. 15,170). The anterior pubic border is sharp, without trace" of supporting structures for epipubic bones. The pubic symphysis is closed, as in Thylacynus, so it seems probable that, if epipubic elements were present, they must have been vestigial cartilages, as in the latter genus (cf. text-fig. 3). The pe- duncular portion of the ilium is more attenuated than in Frothy lacynus, supporting heavy recti tubercles. The gluteal surface is broad and smooth with even less trace of muscular flutings than in Frothy lacynus. The iliopectineal eminences are large. The ischial tuberosities are less pro- nounced than in Thylacymis. The obturator foramina are large and oval in outline, with the posterior border emarginated by an anteriorly directed prominence. The same lack of proportion between the lengths of skull and femur, which was mentioned in the discussion of the genus BorJiycena, is observ- able to even a greater extent in the case of Cladosictis. The femoral shaft has a gentle sigmoid curvature, expanding at either end to about the same * extent and in much the same manner as in Borhycena. The greater tro- chanter and head reach the same elevation. The lesser trochanter is prominent. As in Borhycena, the inner condyle is somewhat wider than the outer one. The intercondylar notch is deep and narrow, leading an- teriorly into a wide, shallow rotular groove, the margins of which are more acute than in Thylacynus (cf. text-fig, i, c]. An ossified patella has not been found in association -with any of the skeletal material of Cladosictis in the collection, and has accordingly been omitted in the restoration. The tibia (PI. LVIII, figs. 2, 8) resembles that of Frothy lacynus, differ- ing, however, in being slightly narrower distally and in having the distal fibular facet less obliquely placed. The shaft is straight, in contrast with the curvature of the tibial shaft in Thylacymis (text-fig, i, d]. The cnemial crest is poorly differentiated, as in Frothy lacynus, extending more than half way down the shaft. Distally, the tibia exhibits a trochlear surface SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 385 similar to that in Prothylacynus (PI. XLIX, fig. 4). The internal malleolus is prominent. The fibula (PI. LVIII, fig. 9) is intermediate in shape between that of Prothylacynus and that of Thylacymts. Proximally, it is flatter than in the former genus, having the facets for the tibia and lateral sesamoid relatively longer. The shaft is curved sigmoidally, supporting a strong interosseous ridge. Distally the shaft becomes roughly circular in section. In shape, the articular surfaces for the astragalus and calcaneum are much the same as in Prothylacynus. In general, the pes (PI. LIV, fig. 3) resembles that of Dasyurus macu- latus. The trochlear surface of the astragulus (PI. LIV, fig. 10) is not as flat as in Amphipro'vi'verm and Prothylacymts and is somewhat narrower proportionately than in the latter genus. The tibial portion is not pro- duced distally on the dorsal surface of the neck, as in Prothylacynus. The articular surfaces on the plantar aspect are substantially the same as in the last-mentioned genus (cf. PI. LIV, fig. 2.0]. The calcaneum differs from that of Amphipromverra in lacking the deep groove for the accom- modation of the calcaneo-cuboidal ligament. The relative position of the tarsals is the same as in Prothylacymis. The cuboid lacks the notch in its dorsal border between the articular surfaces of the fourth and fifth meta- tarsals, which is so noticeable a feature in both Prothylacynus and Atnphi- proviverra. The shifting of the external cuneiform toward the outer side of the foot has progressed to about the same extent as in Prothylacynus. The hallux is not preserved, but from the small cup-shaped character of its articular surface on the entocuneiform, it is probable that it was propor- tionately no larger than in Dasyurus maculatus or D. viverrinus, having lost the opposable condition retained by Amphipro'vi'verm. The meta- tarsals interlock proximally and spread apart distally. Their distal ends have been destroyed in the right pes represented in the figure, but are well enough preserved in the left pes to show that the fourth is the longest, an arboreal character (Dollo, 1899 ; Bensley, 1903) retained alsojin the manus, where, it will be remembered, the pollex is large and opposable (cf. PI. LIV, fig. 4). Restoration of the Skeleton (PI. LXI, fig. i).- -The lack of proportion between head and body, which has been so often referred to in the discus- sion of the Santa Cruz thylacynes, is nowhere more marked than in Clado- sictis. The lengths of the few missing dorsal vertebrae were determined 386 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. approximately by comparison with adjacent portions of the column. The depth of the anterior portion of the thorax is conjectural, as none of the ribs in this region are preserved. The tail was probably long and heavy, judging from the weight of its proximal portion. The legs are remarkably short and the feet probably plantigrade or semi-plantigrade. Habits. — The elongation of the fourth digit and the opposabilityof the thumb point toward an arboreal habit. The loss of the opposable hallux is an adaptation toward terrestrial progression, which does not necessarily conflict with the view just stated, if we assume that these animals occupied a place in the economy of nature similar to that now filled by the dasyures and some of the smaller placental Carnivora. CLADOSICTIS LUSTRATUS (Ameghino). (Plates LII, Fig. 4 ; LIII, Fig. 10 ; LIV, Figs. 3, 4, 12 ; LV, Fig. i ; LVI ; LVII, Figs. I, la, 3, 5, 6; LVIII, Figs. i-4«, 6 ; LIX, Figs. 7-7l>; LXI, Fig. i.) Hathliacymis histratus Amegh. ; Enum. Sist. Especies Mamif. Fos. Pata- gonia Austral, p. 7, 1887. Anatherium defossum Amegh. ; ibid. p. 8, 1887. Hathliacymis defossus (Amegh.) Mercerat; Revista del Museo de La Plata, II, p. 53, 1891. Proviverra trouessartii Amegh. ; Revista Argentina, I, pp. 149-150, fig. 54. 1891. Cladosictis trouessarti Amegh.: Enum. Syn., etc., pp. 131-132, figs. 50, 51, 1894; Bol. Acad. Cordoba, pp. 386-388, figs. 50-51, 1894. Cladosictis lateralis Amegh.; Enum. Syn., etc., pp. 132-133, 1894; Bol. Acad. Cord., p. 388, 1894. Judging from the number of individuals represented in the collection, this must have been the most abundant of the Santa Cruz marsupial carnivores. The two more or less complete skeletons on which the pre- ceding account of the osteology of the genus is largely based (Nos. 15,046 ; 15,170) are from the Lower Santa Cruz beds, ten miles south of Coy Inlet. Three very fragmentary skulls associated with a small amount of skeletal material were collected by Mr. Hatcher at Lake Pueyrredon (see Narra- tive, this series, Vol. I, p. 173). Several additional specimens were obtained by Messrs. Hatcher and Peterson and Mr. Barnum Brown at Coy Inlet and along the coast to the south of the Coy (Nos. 15,015 ; 15,704), SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 387 the Canon de las Vacas (No. 9134, American Museum of Natural History), and North Gallegos (No. 15,703). Cladosictis lustratus may be readily recognized by the long, narrow, depressed skull, with tapering muzzle, moderately expanded arches and low inclination forward of the facial profile. The summit line of the long sagittal crest is practically horizontal. The anterior superior premolar is separated by a short diastema from the canine and by a larger interval from the median premolar. The median and posterior premolars are almost in contact at the alveolar border. The lower premolars are sep- arated from the canine and from each other by diastemata, which vary in width considerably in different individuals. The posterior premolar and first molar are separated in some specimens by a short interval. With the exception of smaller size and less robust build, the skeleton, so far as it has been possible to make comparisons, is exactly the same as in Cladosictis petersoni, and the description already given in character- izing the genus will apply equally to both. Within fairly well defined limits, there is considerable individual varia- tion in size. This is probably sexual, as the difference between extremes is not greater than between the sexes in Thylacynus (cf. Thomas, 1888, p. 261). The range of this variation is well brought out in the accom- panying measurements of the two individuals selected for illustration (Nos. 15, 170 ; 15,046), which are regarded respectively as male and female. MEASUREMENTS. No. 15,04.6. No. 15,170. Skull, extreme length, premaxillae to lambdoidal crest .142 .158 " widlh across jugal arches .... .066 approximately .076 " interorbital width 022 .0258 Face, length, premaxillae to anterior border of orbit . .052 -O57S Cranium, length to anterior border of orbit . . . .090 .1005 " width at postorbital constriction . . . .010 .0115 Nasals, length 052 " width anteriorly ...... .008 " " posteriorly 024 .023 Palate, length . .069 .079 " width between posterior premolars (approximate) .0125 .015 " M-i 030 .0323 Mandible, length .1105 -H95 " height of coronoid above angle . . . -039* .051 " transverse diameter of condyle . . . .0152 .0162 * Slightly decreased by fracture of the angle. 388 PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. No. 15,04.6. No. 75,770. Mandible, length of symphysis ..... -0345 .036 " depth of horizontal ramus below anterior pre- molar ....... .010 .012 " depth of horizontal ramus below posterior pre- molar . . . . . . . .0146 .017 depth of horizontal ramus below Mj- . . .015 .0205 Upper dentition, length from median incisor to MA . .0695 .078 " " " of premolar series . . . .025 .025 " " " " molar "... .027 .027 Lower dentition, length from median incisor to M^- . .0695 .072 " " " of premolar series . . . .024 .027 " " " " molar series . . . .032 -0314 Upper lateral incisor, width of crown .... .002 .002 Upper canine, antero-posterior diameter at alveolar border .0075 .009 " " transverse " " " " .005 -0055 Anterior superior premolar, antero-posterior diameter . -0055 .005 " " " transverse " . .002 .0022 Median " " antero-posterior " . -0073 .0065 " " " transverse " . .0026 .003 Posterior " " antero-posterior " . .0085 .008 " " " transverse " . .0034 .004 MJ-, antero-posterior diameter ..... .0078 .007 " transverse " ..... .005 .005 £, antero-posterior " ..... .0086 .008 transverse " ..... .0065 .0058 , antero-posterior " ..... .0086 .0085 " transverse " . . ' . . . .0067 .0066 M±; antero-posterior " ..... .003 .003 " transverse " ..... .006 .007 Lower lateral incisor, width of crown .... .00 1 3 .002 " canine, antero-posterior diameter at alveolar border ....... .007 .008 Lower canine, transverse diameter at alveolar border . .0045 .005 Anterior inferior premolar, antero-posterior diameter . -0055 -0055 " " " transverse " . .0023 .0025 Median " " antero-posterior " . .0077 .0085 " " " transverse " . .0028 .003 Posterior " " antero-posterior " . .008 .009 " " " transverse " . .003 -0035 M^-, antero-posterior diameter ..... .007 .0065 " transverse " ..... .003 .003 Mj, antero-posterior " . . . . . -0075 .007 " transverse " 0033 -0033 Mj, antero-posterior " . . . . . .0085 .0082 " transverse " ..... .004 .004 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 389 No. 15,04.6. No. 75,770. M?, antero-posterior diameter ..... .0095 .0096 " transverse " ..... .005 -0047 Atlas, width across anterior cotyles .... .0325 " " of neural arch ..... .0136 " " " inferior " (approximate) . '. . .008 Axis, length of centrum including odontoid . . .034 " width across anterior cotyles .... .0225 " approximate length of neural spine . . . -O44 " " height " " " above floor of neural canal ...... .022 Third cervical, length of centrum .... .019 " " width across prezygapophyses . . .022 Fourth cervical, length of centrum .... .017 width across prezygapophyses . . .023 Fifth cervical, length of centrum . .... -OI75 " " width across prezygapophyses . . .024 " " antero-posterior diameter of inferior la- mella ...... .017 Sixth cervical, length of centrum .... .017 " " width across prezygapophyses . . .024 " " antero-posterior diameter of inferior la- mella ...... .021 Seventh " approximate length of centrum . -°!55 " " width across prezygapophyses . . .023 Fifth dorsal, length of centrum . . . . . .on Sixth dorsal, " " " 0118 Tenth <•««•< 013 Eleventh dorsal, length of centrum .... .014 Twelfth " " 0145 " " width of posterior face of centrum . .0125 " " depth " " " " " . .008 First lumbar, length of centrum .... .017 " " width across prezygapophyses . . .019 " " height, including spine .... .025 Second lumbar, length of centrum .... -O2O " " width across prezygapophyses . . .021 Third lumbar, length of centrum ... .021 " " width across prezygapophyses . .018 Fourth lumbar, length of centrum . . . -O22 " " width across prezygapophyses . . .018 Fifth lumbar, length of centrum ... .021 Sixth lumbar, " ... .019 .021 Sacrum, total length .... .027 .03 1 " length of centrum of first sacral . . . .014 -OI55 " " " " " second sacral . . .013 390 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. No. 15,046. No. 15,170. Sacrum, greatest width across auricular processes . .025 -032 Third ? caudal, length of centrum .... -0135 Fourth? " «««««< .... .0137 Fifth? " " .... -0137 Sixth? " " .... .014 Seventh?" " .... .015 Eighth? " «..«.. .... .0178 Ninth? " " .... .019 Tenth? " " .... .021 Scapula, approximate length ..... .078 width of neck ...... .014 " antero-posterior diameter of glenoid cavity, including coracoid process . . . -O'/S " transverse diameter of glenoid cavity . .0105 Radius, length . . . . . . . . .0646 .0666* " antero-posterior diameter of head . . . .006 .0066 " tranverse " " " . . . .0102 .0114 " width of distal end ..... .01 1 .on Ulna, length 085 .087* " antero-posterior diameter at lower margin of sigmoid cavity . . . . . . .0136 .0166 " antero-posterior diameter of distal end . . .0088 .0085 Femur, length ........ .098 -1095 " width of proximal end ..... .022 .026 " " distal " 0175 .021 Tibia, length 0965 .102 " width of proximal end . . . . . .015 .019 " distal end . ' 0095 .010 Fibula, length ........ .0915 Pelvis, length . . . . . . . .'.1015 " width of ilium ...... .021 " across tubercles for origin oirectiisfemoris. .048 " length of obturator foramen .... .0264 " width of obturator foramen . . . . .0165 " antero-posterior diameter of acetabulum . . .013 " transverse " " " .0125 The following measurements of the manus and pes are from No. 15,046 : Metacarpal I, length . . . . . . . . . . .0105 " width of proximal end ....... .0066 " distal end 0055 II, length .017 " " width of proximal end ....... .007 * Exclusive of distal epiphysis. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 391 Metacarpal II, width of distal end 006 III, length 021 " width of proximal end ....... .0345 " " distal end . . . .0045 IV, length .020 " width of distal end 005 V, approximate length . . . . . ... . .013 Terminal phalanx of pollex, approximate length ..... .009 Length of calcaneum ...... ... .0226 Metatarsal II, approximate length ........ .027 " width of proximal end (approximate) ..... .0036 " " " distal end .0062 III, approximate length ........ .028 " width of proximal end ....... .0047 IV, width of proximal end ....... .0045 " V, " " " " . .005 CLADOSICTIS PETERSONI sp. nov. (Plates LIII, Figs. 3, 3«; LIV, Figs. I, 10; LV, Figs. 2-3* ; LVII, Figs. 2, 2a, 4, 4«; LVIII, Figs, 5, 7-9.) The type of this species (No. 15,702 Princeton University Museum) is the facial half of a skull associated with a large part of the skeleton, col- lected from the Santa Cruz beds ten miles south of Coy Inlet by Mr. Peterson, in whose honor the species is named. The skeletal material associated with the skull includes the right scapula, humerus, radius and ulna, the left femur, tibia, fibula and astrag- alus, the atlas, three dorsals, six lumbars, the first sacral and the pelvis. • Cladosictis petersoni may be recognized by its large size, exceeding in this respect the largest and most robust individuals of C. lustratus. The face is relatively shorter, and much deeper, than in that species and the upper margin of the facial profile is inclined forward more abruptly. The sagittal crest rises high above the interorbital tract. This portion of the skull has been crushed antero-posteriorly, approximating the origin of the crest and the posterior border of the nasals to a greater extent than is normal. The arches are more abruptly expanded than in C. lustratits. The median and posterior premolars are spaced to a greater extent than in the latter, but otherwise there is no difference in the dentition of the two species. The skeleton of C. petersoni presents no characters, apart from size, by which it may be distinguished from C. lustratus. 392 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. In the type skull, there are but three molars on the right side, where M- is wanting. This tooth is present on the opposite side, where it is proportionately no more reduced than in C. lustmtus. The broad, blunt points of the upper canines are the result of fracturing, and are not characters of specific importance. MEASUREMENTS. Skull, width across jugal arches (approximate) . . . . . .090 Face, length, premaxillae to anterior border of orbit ..... .060 Cranium, width at postorbital constriction ....... .oil Nasals, length (approximate) ......... -057 " width anteriorly .......... .009 " " posteriorly .......... .026 Palate, length 076 " width between canines . . . . . . . . . .014 " " " posterior premolars . . . . . ... .020 " " "MA .037 Upper dentition, length from median incisor to MA (approximate) . . -°79 " " " of premolar series . ...... .032 " molar " 0245 Upper median incisor, width of crown ....... .0012 " lateral " " " " 0018 Upper canine, antero-posterior diameter at alveolar border . . . . .010 transverse " " 007 Anterior superior premolar, antero-posterior diameter. .... .006 " " " transverse " . . . . . .0025 Median " " antero-posterior "..... .0085 " " " transverse "..... .003 Posterior " " antero-posterior " . . . . .0085 " " " transverse " 0038 Mi, antero-posterior diameter . . . . . ... . . .007 " transverse " ......... .005 , antero-posterior " ......... .0078 " transverse " ......... .006 , antero-posterior " ......... .0088 " transverse ......... .007 MA, antero-posterior " ......... .003 " transverse " ......... .007 Atlas, width across anterior cotyles . . . . . . . . .035 " " of neural arch ......... .016 " " " inferior " ; .009 Eleventh dorsal, length of centrum . . . . . . . . -0175 " " width of posterior face of centrum ..... .0165 " " depth " " " " 010 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 393 Eleventh dorsal, width across prezygapophyses ...... .034 Twelfth " length of centrum 019 " width of posterior face of centrum ..... .0164 " " depth " " " " 0104 " width across prezygapophyses . . . . . . .032 " height, including neural spine ...... .0275 Thirteenth " length of centrum 020 " width across prezygapophyses ...... .027 " height, including neural spine ...... .0275 First lumbar, length of centrum ........ .021 " width of posterior face of centrum . ..... .016 " " depth " " ....<< OI25 " " width across prezygapophyses . . . . . .028 " " height, including neural spine . . . . . . .031 Sixth lumbar, length of centrum ........ .022 " " width of anterior face of centrum ...... .017 " " depth " " " " . . ... .0125 width across prezygapophyses ...... .026 " " transverse processes ...... .045 First sacral, length of centrum ......... .018 " greatest width across auricular processes ..... .038 Scapula, length . . - . . . . . . . . .102 " width of neck .......... .018 " antero-posterior diameter of glenoid cavity including coracoid process .0228 " transverse " " " " . . . . . . .0135 Humerus, length . . . . . . . . . . . .112 " width of distal end 033 Radius, length exclusive of distal epiphysis ...... .0785 antero-posterior diameter of head ....... .0087 " transverse " " " 0135 " width of distal end ......... .013 Ulna, length, exclusive of distal epiphysis ....... -0995 " width at lower margin of sigmoid cavity ...... -0195 Femur, length . . . . . . . . . . . . -1255 " width of proximal end . . . . . . ' . . . .030 " " " distal end 024 Tibia, length . . . . . . . . . . . . .116 " antero-posterior diameter of proximal end ..... .02 1 5 transverse " " distal " OI2 Fibula, length 1095 " antero-posterior diameter of proximal end ..... .014 " transverse " " distal " .0115 Pelvis, length 123 " greatest width of ilium ......... -0275 " antero-posterior diameter of acetabulum . . . . . . -O'SS " transverse " " " 0145 394 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Pelvis, pubic symphysis, approximate length ...... -0495 Astragalus, length . . . . . . . . . . . .016 " greatest width of trochlea ....... .009 " length of trochlea . . . . . . . . . .010 " width of neck 0055 " " " head 0065 AMPHIPROVIVERRA Ameghino. (Plates LIII, Figs, i, ia; LIV, Figs. 5, 6, u ; LIX, Figs. 1-5 ; LX). Protopromverra Amegh.; Nuevos Restos Mamif. F6s. Patagonia Austral, pp. 26-27, Aug., 1891 ; Revista Argentina Hist. Nat, I, entr. 50, pp. 312-313, Oct., 1891. Preoccupied by Protoproviverra Lemoine. Amphiproviverra Amegh. Revista Argentina, etc., I, footnote p. 397, 1891. Small, highly carnivorous marsupials, in which the protocone on the last upper molar is basin-shaped and the heels of the lower molars are broad and strongly bicuspidate. Dentition (Pis. LIX, figs. i-ib, 3*2, 4; LX, figs, i, ia, 2-30). — Dental formula f, 1, f, t. The median upper incisors are styliform, and approxi- mated at the tips, as in the opossums and dasyures. The crowns of the lateral teeth are spatulate in shape. The incisor series is placed obliquely, so that the procumbent median pair are the most anterior. Its members increase regularly in size from the first to the fourth. The canines are long and slender, projecting below the lower border of the mandible when the jaws are closed (PI. LX, fig. 3). The anterior and median premolars are simple-crowned, double-fanged, piercing teeth, much compressed laterally. The median premolar supports a small heel cusp. The pos- terior premolar is enlarged, its crown projecting below the level of the molars. It is recurved to about the same degree as the tooth preceding it and also carries a small heel cusp. The anterior premolar is separated from the canine and median premolar by diastemata. The latter tooth is almost in contact with the posterior premolar in A. manzaniana and A. minuta (PI. LX, figs, i, 3). The anterior molars are of the characteristic thylacyne type, while the fourth resembles the last upper molar of Dasytirns. The first, second and third increase regularly in width, although retaining about the same antero-posterior diameter. In these teeth, the protocone is large, inclosing a basin-shaped depression, on both margins of which SINCLAIR : MARSUPIALIA OF THE SANTA CRUZ BEDS. 395 small intermediate cuspules may be observed in unworn teeth, as in Micro- biotherium and Dasyurus viverrinus. The outer cusps are high and coni- cal. The paracone is relatively higher than in Thylacynns and the meta- cone shear strongly developed, but less rotated outwardly on M1 than in the latter genus. A large antero-external style is always present. On M1 the protocone, paracone and antero-external style are functional. The protocone is large and basin-shaped, as in the anterior molars. The high conical paracone is connected with the antero-external style by a sharp ridge, producing a transverse shear, as in Dasyurus, but less perfectly so, owing to the relatively greater elevation of the paracone. In some indi- viduals of A. manzaniana a small metacone is present. With decrease in size of the metacone, the root supporting it is greatly reduced and may disappear entirely, producing a double-rooted tooth. The anterior molars are triple-rooted in both species. The lower incisors are similar to those of Thylacynus and Dasyurus with rather thick crowns divided by a transverse groove (PI. LX, fig. 30). The second tooth on either side is displaced behind the median and lateral pair. The canines are shorter and less robust than those of the upper series, with conical crowns curved to about the same extent as in Thylacynus. The lower premolars are simple-crowned, double-rooted, piercing teeth, of which the median and posterior are subequal in size. The heel cusp on the anterior premolar is small, becoming larger on the median and posterior pair (PI. LX, fig. 20.]. The anterior premolar is spaced on either side. The median and posterior premolars may be in contact or slightly spaced. The molars increase regularly, not only in size, but in the height of the external cusps. They are closely crowded, so much so that the heel of each is impressed into the anterior surface of the tooth next succeeding. In MT, the arrangement of the cusps is linear, but in the second, third and fourth the paraconid is more and more de- flected internally, producing a shear which cuts against the metacone spur of the upper teeth. The protoconid is high and conical, becoming flattened on the posterior side by shearing against the anterior face of the trigon of the upper molars. The lobate, blade-like paraconid is separated by a narrow slit from the protoconid. The heels are broad and strongly bi- cuspidate, the lingual cusp corresponding to the undifferentiated hypo- conulid-entoconid. A short antero-external cingulum is present on the second, third and fourth molars. 396 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Skull (Pis. LIX, figs. i-3«; LX, figs, i-ic, 3). — The facial portion of the skull is short and slender, the cranial portion elongate, with low sagittal and lambdoidal crests, widely expanded arches and exceptionally long, shallow brain-case. In profile (PI. LX, fig. i), the upper margin is almost horizontal, becoming slightly convex back of the orbits. The ascending premaxillary processes are short, proportionately less elongated than in Thylacynus and Dasytirus. The nasals are greatly expanded posteriorly and in broad contact with the lachrymals. The lachrymals are large, spreading out on the face and excluding the maxillae from the anterior border of the orbit. The lachrymal duct opens within the orbital rim, which is sharply defined, with a distinct lachrymal tubercle. The infraorbital canal is single, opening externally above the posterior pre- molar. The orbits are large and placed well forward, their anterior border lying above the middle of M1. The postorbital processes on the frontal and jugal are small. The zygomatic arches are robust and broadly expanded, the greatest width occurring at about the middle of the arch. The jugal bar bears a well-marked ridge, situated about a third of the dis- tance from its inferior border, for the origin of the anterior portion of the masseter. Posteriorly the jugal is continued to the glenoid cavity, of which it forms the anterior border. The postorbital constriction of the brain case is even more marked than in the opossum. A short distance anterior to the point of greatest con- striction the feeble temporal ridges unite to form a low sagittal crest. The supraoccipital, unlike its condition in Prothylacynus and Borhycena, has considerable anterior expansion on the upper surface of the skull, rather more proportionately than in Dasyunis niactilatus and the opossum. A broad bar of the parietal extends posteriorly between the supraoccipital and the squamosal to contact with the mastoid. The posterior view of the skull (PI. LX, fig. \c] shows the occiput to be almost semicircular in outline, in contrast with the triangular occiput of the dasyures, Sarcophihts and Thylacynus. It does not project beyond the condyles, which are of the same general shape as in Dasynrns. The foramen magnum is elliptical in outline. Its upper border is notched by an irregular vacuity, resembling a similar structure in some of the Macro- podidae. Owing to the scarcity of material for comparison, it can not be determined whether this peculiar feature is normal to the genus. The basioccipital is broad and flat. But one condyloid foramen is SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 397 present, in contrast with the double condyloid foramen so common among existing marsupials. The tympanic (PI. LIX, fig. 2) is . annular and unfused with the adjacent bones of the skull. The alisphenoid is dilated to form the auditory bulla. As this region is imperfectly preserved in all the specimens examined, it can not be satisfactorily determined whether the petrous was involved in the formation of the posterior portion of the bulla, as it is in Dasyums and Microbiotherium. The basisphenoid is ridged, as in existing carnivorous marsupial's. Prominent alisphenoid ridges extend posteriorly to the confluence with the auditory bullae. The pterygoids are not preserved in any of the specimens examined and were probably small and scale-like. The posterior nares terminate either opposite or slightly posterior to the last molar. The palato-narial border is thickened, resembling the corresponding region in the skull of Dasyums, and is more or less emarginate, varying slightly in the degree of develop- ment of the median process in different individuals of the same species. The palate is long and triangular in shape, increasing in width pos- teriorly. Palatal vacuities are conspicuously absent. The incisive fora- mina terminate a short distance posterior to the premaxillary suture. A large foramen perforates the palatal surface of the maxillary opposite either canine. Accessory palatal foramina are less numerous than in Borliyccna. The margins of the palate are depressed for reception of the tips of the lower molars, when the mouth is closed. The posterior border of the palate is perforated by a large foramen on either side of the nares, as in nearly all marsupials. With a few important exceptions, the cranial foramina are the same in number and position as in existing carnivorous marsupials. As already noted, the condyloid foramen is single. The basisphenoid has but one perforation, that for the internal carotid artery. The foramina of the lateral sinuses are especially well developed. The postglenoid and sub-squa- mosal foramina are the largest. The postzygomatic, which opens ante- riorly within the lip of the postglenoid foramen, varies in size in different individuals of the same species. A small foramen occasionally pierces' the jugal process of the squamosal just above the glenoid cavity. The mandible is slender, with moderately convex inferior border. The coronoid is less strongly inclined posteriorly than in Tliylacynus, resem- bling rather the condition in Dasynrus. The masseteric fossa is broad, with prominent borders. The condyles have about the same degree of 398 PATAGONIAN EXPEDITIONS 1 PALAEONTOLOGY. elevation as in Tkylacynus, but, unlike that genus, the condylar surfaces are wider internally than externally. The angle is broad and strongly inflected. The rami are unfused at the symphysis, which extends as far back as the anterior border of the posterior premolar. Four or five men- tal foramina are present, varying in number and position on opposite halves of the same mandible. The most anterior and also the largest of these is situated beneath the anterior premolar. Cervical Vertebra. — The atlas and third cervical are associated with the skull and feet of a specimen of A. mansaniana (No. 15,154). The atlanteal intercentrum (PL LIII, figs, i, \d] is separately ossified and un- fused with the neural arch. The canal for the vertebral artery pierces the inner surface of the neural arch above the condyles. A smaller foramen, possibly transmitting a recurrent branch of the same artery, penetrates the upper surface of the base of the transverse process near its posterior edge. The artery emerges on the lower surface of the atlas at the base of the transverse process. The neuro-arterial canal is large and widely sepa- rated from the upper margin of the cotyles by a broad bar of bone. The extremities of the transverse processes are lobate, and the upper surface of each is reenforced by a broad median rib. The neural spine of the third cervical (PL LIX, fig. 5) is proportionately larger than in Thy lacy mis. The centrum is strongly keeled inferiorly, and the posterior bar of the costal process supports a small diapophysis. Appendicular Skeleton. — The humerus of A. mansaniana (PL LX, fig. 4) is without internal epicondylar foramen. The supinator ridge is low and does not terminate proximally in a hook-shaped process. The deltoid crest is sharper than in Cladosictis or Protkytacynus. With the exception of the magnum, the carpus of No. 15,154 is want- ing. The magnum resembles that of Borhycena in shape and in the arrangement of the facets. The metacarpals (PL LIV, fig. 5) interlock proximally to about the same extent as in Sarcophilus. The third and fourth are equal in length. The proximal articular surfaces are convex in dorso-palmar section and concave transversely. The proximal sur- face of the fourth is irregularly quadrangular in outline like that of the third, instead of triangular, as in Thy lacy mis and SarcopJiilits. The dis- tal ends are transversely flattened, with moderately developed keels on the palmar surfaces. The metacarpal of the pollex is missing, but from the size of the proximal articular surface on the first phalanx of the pollex it SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 399 appears to have been proportionately larger than in Sarcopliilns or Tliy- lacynus. As in Cladosictis, this surface is divided by a sharp keel into a small inner and a large outer rotular groove, indicating an enlargement of the outer condyle of the metacarpal, which produced a deflection of the phalanges of the pollex toward the inner side of the foot. The remaining proximal phalanges are short and stout, with straight shafts and prom- inent tuberosities for the annular ligament. The distal trochleae are without trace of the median keel-like structures observable in Sarcopliilns, resembling in this respect Thy lacy mis. The proximal articular surfaces of the phalanges of the second row are prolonged dorsally by the devel- opment of tongue-like processes fitting between the condyles of the prox- imal phalanges, indicating that these two sets of elements were carried with respect to each other at a considerable angle. The unguals are la- terally compressed, sharp-pointed and without terminal clefts. Hoods are developed to about the same extent as in Sarcophilus and Dasynrus maculatus. With the exception of the claw of the second digit, the figure (PI. LIV, fig. 5) shows the original association of the phalanges of the manus. The ungual interpreted as that of the second digit lay in the matrix above the fifth metacarpal. In the pes (PI. LIV, fig. 6), the trochlear surface of the astragalus is short and almost flat transversely. Distally, the tibial trochlea is pro- duced on the upper surface of the neck. The neck is proportionately longer than in Sarcophilus and Thylacynns and the head is less obliquely placed than in those genera. The calcaneal and sustentacular facets are like those of Prothylacymis (PI. LIV, fig. 20) in shape and position. Two minute astragalar foramina are present. The calcaneum has the tubercle for the attachment of the calcaneo-cuboidal ligament greatly enlarged and grooved. The calcaneum carries a facet for articulation with the fibula, which is confluent with the ectal facet for the astragalus. The susten- tacular facet is concave and is broadly separated from the former, unlike the condition in Sarcophilus and Thylacynus. The tuber calcis is rela- tively shorter than in those genera, occupying somewhat less than one half the total length of the calcaneum. The remaining tarsals agree closely in shape and in the arrangement of the articular facets with those of Protkylacynus and do not call for separate description. The hallux, when brought into articulation with the entocuneiform, is 400 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. strongly deflected toward the inner side of the foot (opposable). Its proximal articular surface is convex in dorso-plantar section, fitting a large concave facet in the distal end of the entocuneiform. The distal end is, unfortunately, missing. The fourth metatarsal is the longest in the pes. The proximal articular surfaces of the second, third and fourth are irregu- larly triangular, wider on the dorsal than on the plantar margin. Distally, these bones are flattened, with moderate keels on the plantar surfaces. The fifth metatarsal is missing. The short, comparatively flat, astragalar trochlea, moderately interlock- ing metapodials and distal spreading of the toes indicate, in the writer's opinion, that Aniphipromverra was plantigrade. The opposable hallux and semi-opposable pollex point to arboreal habits. The proximal and distal ends of the left fibula are associated with the pes just described. Proximally, the shaft is triangular in cross section and carries a large facet for the fabella. Distally the fibula resembles the corresponding element in Prothylacymis. It carries the usual three facets for the calcaneum, astragalus and tibia. A fragment of the patella associated with the skull of A. minuta (No. 15,373) is hardly complete enough to describe, but is important in demon- strating the ossification of the patella in this genus as in Prothylacynus. Among living marsupials the patella is ossified only in the Peramelidae. The skull represented in figures i, la, Plate LIX (No. 15,154), shows an interesting pathological structure, which affords some suggestions of the pugnacious habits of these animals. The right upper canine has been torn out bodily and the wound has healed, leaving an irregular cavity in the side of the face. Wounds similarly received have been noticed in the discussion of the genus Borhyccna. AMPHIPROVIVERRA MANZANIANA Ameghino. (Plates LIII, Figs, i, ia; LIV, Figs. 5, 6, n ; LIX, Figs. 1-2, 4, 5 ; LX, Figs. \-2a, 4.) Protoproviverra manzaniana Amegh.; Nuevos Restos Mamif. Fos. Pata- gonia Austral., pp. 26-27, Aug., 1891 ; Revista Argentina, etc., I, entr. 50, pp. 312-313, Oct., 1891. Amphiproviverra manzaniana Amegh.; Enum. Syn., etc., p. 133, fig. 52, p. 134, 1894; Bol. Acad. Cordoba, p. 389, fig. 52, p. 390, 1894. Apart from size, there is hardly a character which will serve to distin- SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 401 guish A. manzaniaiia from A. niinnta. The species is represented in the Princeton collection by remains of three individuals : No. 15,148, a fragment of the right maxilla retaining in place the little- worn crowns of all the teeth except the canine and median premolar, col- lected by Mr. O. A. Peterson from the Lower Santa Cruz beds, ten miles south of Coy Inlet. No. 15,029, an incomplete skull and mandible obtained by Mr. Hatcher from the Lower Santa Cruz beds at the same locality as No. 15,148. No. 15,154, a crushed skull, atlas, third cervical and portions of the right fore and left hind limbs, collected by Mr. Peterson from the Lower Santa Cruz beds, ten miles south of Coy Inlet. The collection of the American Museum of Natural History contains a remarkably perfect skull associated with a part of the mandible (No. 9254, PI. LX, figs. \ — \c] from the vicinity of Felton's estancia on the Rio Gallegos. The accompanying table of measurements shows considerable individual variation in size. This may possibly be a sexual character. MEASUREMENTS. No. Skull, length on median basal line " " premaxillae to condyles Skull, greatest width across zygo- matic arches .... Skull, width across postorbital proc- esses ...... Cranium, length, condyles to anterior border of orbit .... Cranium, least width of brain case Face, length to anterior border of orbit Nasals, length .... " width at anterior extremity . " " posterior Occiput, height .... " width at base Mandible, length, including condyle . Mandible, length, M? to outer end of condyle ..... Mandible, height of condyle above angle ...... * Width increased by crushing. No. 15,029. .056 .0204 No. .070* No. Ant. Mus. .108 .112 .062 .OI9 •0755 .010 .010 .010 .041 .037 •039 .040 .038 .039 .0088 .008 .008 .020 .020 .018 .023 .036 .090 .036 .0175 4O2 PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. No. No. No. No. 9254. 15,14.8. 15,029. 15,154.. Am. Mus. Mandible, transverse diameter of con- dyle ...... .012 Mandible, depth of horizontal ramus at M? ...... .0125 .013 Mandible, depth of horizontal ramus at posterior premolar ... .010 Mandible, depth of horizontal ramus at anterior premolar . . . -°°95 Upper dentition, length, median in- cisor to M± .... -0585 .055 -0565 Upper dentition, length of space occu- pied by premolars . . . .018 .017 -°I75 Upper dentition, length of space occu- pied by molars .... .0245 .023 .020 .021 Lower dentition, length, anterior border of canine to M? . . .0525 Lower dentition, length of space occu- pied by premolars ... .018 Lower dentition, length of space occu- pied by molars .... .026 Median upper incisor, width of crown .001 Lateral " " " " " .002 .002 Upper canine, antero-posterior diam- eter at alveolar border . . . .0058 .006 .0065 Upper canine, transverse diameter at alveolar border .... .004 .0042 .004 Anterior superior premolar, antero- posterior diameter . . . .004 .004 .0038 .004 Anterior superior premolar, trans- verse diameter .... .0015 .0015 .0015 .0015 Median superior premolar, antero- posterior diameter . . . -0055 .005 .005 Median superior premolar, transverse diameter .002 .0018 .0018 Posterior superior premolar, antero- posterior diameter . . . .0063 .006 .005 .0056 Posterior superior premolar, trans- verse diameter .... .003 .003 .0028 .0028 Mi, antero-posterior diameter . . .007 .007 .006 .007 " transverse diameter . . . .004 .004 .004 .004 M£, antero-posterior diameter . . .0077 -0074 .006 .007 " transverse diameter . . . .006 -0055 .005 -0055 , antero-posterior diameter . . .007 .007 .006 .0064 transverse diameter . . . .0078 .0068 -0055 .0058 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 403 No. No. No. No. 9254. 15,14.8. 15,029. 15,154. Am. Mus. MA, antero-posterior diameter . . .0028 .0028 .0028 .0023 " transverse diameter . . . .0072 .007 .0065 .006 Lower lateral incisor, width of crown .0018 Lower canine, antero-posterior diam- eter at alveolar border . . . .0046 Lower canine, transverse diameter at alveolar border .... -0035 Anterior inferior premolar, antero-pos- terior diameter .... .0045 Anterior inferior premolar, transverse diameter ..... .002 Median inferior premolar, antero-pos- terior diameter .... .006 Median inferior premolar, transverse diameter ..... .0022 Posterior inferior premolar, antero- posterior diameter . . . .0065 Posterior inferior premolar, transverse diameter ..... .0026 M-j, antero-posterior diameter . . .006 " transverse diameter . . . .0027 Mj, antero-posterior diameter . . .0068 " transverse diameter . . . -0035 Mff, antero-posterior diameter . . -0075 .0066 " transverse diameter . . . .0038 -0033 M?, antero-posterior diameter . . .007 .006 " transverse diameter . . . .004 .003 Atlas, transverse breadth . .043 " width of neural arch . . -Oio " " intercentrum . .0056 Third cervical, length of centrum . -O'SS " " width across trans- verse processes .... -028 The following measurements of the manus and pes are from No. 15, 154 : Metacarpal II, length ... -°1SS " " width of proximal end . . .0045 " distal " . -0055 Metacarpal III, length " " width of proximal end . . .004 " " " " distal end ... . -O°5 Metacarpal IV, length ... -Ol8 " " width of proximal end . . . .0036 404 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Metacarpal IV, width of distal end ........ .005 Metacarpal V, length . . . . . . . . . . .012 " " width of proximal end ....... .004 " distal " . .0055 Pollex, first phalanx, length ......... .009 " ungual phalanx, length. ........ .0097 " " " width of hood 005 Second digit, first phalanx, length ........ .0085 " " second " " . . ... .0065 " " ungual " "...'. ... .0098 " " " width of hood ... ... .005 Third digit, first phalanx, length ........ .0085 Fourth " " 0085 " " second " " 0063 " " ungual " width of hood . . . . . . . .0036 Fifth digit, first phalanx, length. ........ .0075 " " second " " x . .0055 " " ungual " " 0085 width of hood .0038 Astragalus, length . . . . . . . . . . . .0135 " width of trochlea ......... .0085 " " neck .0046 Calcaneum, length ........... .021 " " of tuber ......... .0097 Metatarsal I, width of proximal end ........ .005 Metatarsal II, length . . . . . . . . . . .022 " width of proximal end. ....... .0032 " " " " distal "........ .005 Metatarsal III, length .......... .023 " " width of proximal end ....... .004 " " distal " . .005 Metatarsal IV, length . . . . . . . . . . .025 " width of proximal end ....... .004 " " distal " . . . . . . . .005 AMPHIPROVIVERRA MINUTA Ameghino. (Plates LIX, Figs. 3, 30; LX, Figs. 3, 3«.) Amphipromveyra minuta Amegh.; Enum. Syn., pp. 134-135, 1894; Bol. Acad. Cordoba, p. 390, 1894. This species may be distinguished from A. inanzaniana by its smaller size and less robust build. It is represented in the Princeton collection by an incomplete skull and mandible (No. 15,373) associated with a frag- ment of the left ulna, a number of phalanges, the distal end of a meta- SINCLAIR I MARSUPIALIA OF THE SANTA CRUZ BEDS. 405 podial and part of the patella, collected by Mr. Hatcher, from the Lower Santa Cruz beds at Killik Aike. Although smaller than A. manzaniana, the length of the upper pre- molar-molar series is almost the same as in some individuals of that species (cf. PI. LIX, figs, i, 3^). In A. minuta, the nasals are less rounded posteriorly than in A. manzaniana and receive a somewhat longer tongue of the frontals between them. This, however, may be an individual rather than a specific character. In the mandible, the chin is more strongly marked than in A. manzaniana. The dental pattern is identical in the two species. MEASUREMENTS. Skull, length from ant. surface of canine to post, border of preglenoid process. .069 " least width of brain case . . . . . . . . . .010 " width across postorbital processes ....... .015 " width across posterior expansion of nasals. . . . . . .015 Palate, greatest length from anterior surface of canine to palato-narial border. .043 " width between anterior premolars ....... .009 last molars 020 Mandible, length, base of median incisor to condyle ..... .076 " " M^ to outer end of condyle ..... .0295 " height of condyle above angle . . . . . . -O'35 " depth of horizontal ramus below MT . . . . .0105 " " " " , " " posterior premolar . . . .008 " " " " " " anterior premolar . . .0078 " transverse diameter of condyle .... .009 Upper dentition, anterior border of canine to My inclusive . .044 " " length of space occupied by premolars . . .017 " " " " " " " molars ..... .020 Lower " base of median incisor to M j inclusive . . .048 " " length of space occupied by premolars . . .017 « " " " " " " molars 0233 Upper canine, antero-posterior diameter at alveolar border . .0045 " transverse " " " " -°O3 Anterior superior premolar, antero-posterior diameter .004 " " " transverse "... .0018 Median " " antero-posterior " " " transverse . -°°2 Posterior " " antero-posterior " " " transverse . • .0025 Mi, antero-posterior diameter . " transverse " , antero-posterior transverse " °°5 406 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. M&, antero-posterior diameter ......... .006 " transverse " 0057 M-t, antero-posterior " ........ .002 " transverse " 0055 Median inferior incisor, width of crown ....... .0015 Lateral " " " " " 002 Lower canine, antero-posterior diameter at alveolar border . .004 transverse " " " " 003 Anterior inferior premolar, antero-posterior diameter ..... .0045 " " " transverse " ..... .0016 Median " " antero-posterior " . . . . . -0055 " " " transverse " ..... .002 Posterior " " antero-posterior " ..... -OO55 " " " transverse " . . . • . . .0023 MT, antero-posterior diameter . . . . . . . . . .0058 " transverse " ......... .0025 M^-, antero-posterior " ......... .0062 " transverse " .......... .003 My, antero-posterior diameter . . . . . . . . . ' .0064 " transverse " ......... .003 My, antero-posterior " ......... .0065 " transverse " 0035 Patella, width 007 Terminal phalanges, average length ........ .0065 " " " width of hoods ...... .003 RELATIONSHIPS OF THE THYLACYNID^E. Although there is sufficient similarity in structure to warrant placing the Patagonian and Tasmanian thylacynes in the same family, it must not be inferred that the existing genus is the direct descendant of its extinct South American forerunners. The study of the group has failed to show a closer relationship than probable descent from a common pre-Santa Cruz ancestor. While retaining the fundamental family characters, both lines have diverged and in some respects the Santa Cruz forms are more advanced than the existing genus. i. Without exception, the Santa Cruz forms, so far as known, show great reduction of the external styloid cusps in the upper molars, the antero-external style alone remaining, while in Tliylacyuns, "a small ele- ment probably equivalent to style c2 is apparently always present in the first molar, variable in the second, and scarcely distinguishable in the SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 407 third." (Bensley, 1903, p. 108.) This style is well shown in figure la of Plate LXV. 2. The last upper molar in the Santa Cruz genera is more reduced than in Thylacynus, reaching an extreme in Borhycena, where but two cusps remain, the paracone and antero-external style. In neither Amphipro- mverra, Frothy lacynus, nor Cladosictis, is the metacone on MA as strongly developed as in Thylacynus. 3. In the lower dentition of all the Patagonian thylacynes the hypo- conulid is undifferentiated from the entoconid, and the heel of the fourth molar is not only smaller proportionately than in T/iy lacy tins, but has undergone greater reduction, except in Amphiproviverra, in which all the lower molar heels are bifid. 4. In cranial characters Thylacynus is decidedly progressive, while the Santa Cruz forms are conservative. The elongation of the face and pos- terior shifting of the orbit, the great increase in brain capacity, the acqui- sition of palatal vacuities and the prenatal shedding of the milk teeth in the recent genus are all progressive characters. 5. The peculiarities in podial structure observable in l^hylacynus are readily understood, if interpreted as adaptive modifications. The foot structure of the common ancestor of the family was probably not unlike that in Amphiproviverra. Adaptation to a cursorial mode of progression resulted in a reduction of the hallux, as in Frothy lacy nus. With increase in speed and the assumption of a digitigrade gait the complete loss of the hallux and the curious shifting of the tarsal elements noticed in Thyla- cynus have been produced. In connection with the question of the descent of the Patagonian and Tasrnanian thylacynes from a common ancestor, it may be interesting to notice that certain large carnivorous marsupials from the Pyrotherium beds (Amegh., 1897, PP- 97~ioo) named by Ameghino Proborhycena and Pharsophortis retain the metaconid in the lower molars, while the pre- molar formula is unreduced. The loss of the metaconid in the Thylacyn- idae separates them sharply from all other carnivorous marsupials. It is possible that the two genera mentioned, in which this cusp is retained, will be found to occupy an intermediate position between the Thylacynidae and Dasyuridae, but until they are better known it is unsafe to attempt generalizations of so broad a character. Confining the discussion to the mutual relationships of the Santa Cruz 408 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. thylacynes, it is surprising to notice the extent to which they have responded to adaptive specialization. No one of them is ancestral to the others, but Ampliiproviverm is perhaps nearest to the ancestral form in foot structure and shows least reduction in the heel of MT. At the other extreme is Borhycsna which, so far as dentition goes, is a decidedly spe- cialized animal. All the Santa Cruz genera are, apparently, divergent branches of a common pre-Santa Cruz ancestral stock, from which Amphiproviverra appears to have departed less in podial and dental structure than any of the others. DIDELPHYID^E. MICROBIOTHERIUM Ameghino. (Plate LXII, Text-fig. 6.) Microbiotheriiim Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, pp. 6-7, 1887. Hadrorkynckus Amegh.; Nuevos Restos Mamif. Fos. Patagonia Austral, p. 25, Aug., 1891 ; Revista Argentina Hist. Nat, I, entr. $a, p. 311, Oct., 1891. Minute polyprotodonts, comparable in size to some of the smaller South American opossums. Although placed by Ameghino in a separate family, the Microbiotheridae, this genus possesses so many important characters in common with the Didelphyidae that the propriety of its ref- erence to the latter family seems beyond question. Dentition (PI. LXII, figs. 1-6). — The dental formula is I, i, f, I, as in Didelphys. The upper incisors are unspaced and the lateral tooth is separated from the canine by a long diastema. The skull of Microbio- therium tortor in the Princeton collection (No. 15,698, PI. LXII, fig. i) retains in place the third incisor only. This tooth resembles the corre- sponding element in Didelpliys. As in that genus, the median incisors were probably procumbent and approximated at the tips, judging from the inclination of the alveoli, but this portion of the premaxillae has been somewhat crushed and the inference cannot be fully verified. The upper canine is a robust tooth, rather blunt, with the crown but little recurved. The premolars are three in number and closely crowded. The anterior premolar is single-rooted, the median and posterior double-rooted. The latter is the largest of the series. A photograph of a specimen in the La SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 409 FIG. 6. Plata Museum (text-fig. 6) shows that this tooth supports a prominent central cusp, but the minuter details can not be ascertained from the photograph. The first two molars are of the same size, the third is a little narrower transversely and the fourth greatly reduced. All, except the fourth, are triangular in outline and tricuspidate, with conic cusps, a nar- row external cingulum and reduced metacone spur. An antero-external style, designated by Bensley, style ab (Bensley, 1903, pp. 89, 183 et seq.}, is present on all the molars as a distinct tubercle. Style c is well developed on the second and third molars of Microbiotherium tehuelchum, but is only slightly differentiated from the cingu- lum in M. tortor. On the anterior molars the metacone is decidedly larger than the paracone. The metacone spur is greatly reduced compared . Mi^tf«™>» SP-- Palatal ... r ° * view of the skull, x I. The With Its Condition in Dldelpliys or DasyuntS, representation of the molar Owing to the decreased width of the Cingulum. patterns is slightly diagram- M- has the metacone vestigial. The cusps rep- matic- Drawn from an en- resented on this tooth are the protocone, para- larged Phot°graPh of a sPed- . 7 , . ... _ . men in the La Plata Museum. cone, style ab and the vestigial metacone. The protocone is supported on a separate root, as in Didelphys. On the three anterior molars the protocone encloses a basin-shaped depression, on the margins of which two minute cuspules are developed. In shape and pattern, these teeth resemble closely the molars of some of the subspe- cies of Caluromys. All the upper molars are triple-rooted. The lower incisors are spatulate in shape, resembling the incisors of Dasyttnts rather than Didelphys. Unlike these genera, the root of the second tooth in the series is not displaced posteriorly with reference to the roots of the first and third. The first and second incisors only are preserved in the mandible associated with skull No. 15,698. The canine is either short, blunt, and directed anteriorly (M. tortor, PI. LXII, fig. 2), or disproportionately long and pointed (M. teJmelcluiin, PI. LXII, fig. 4). The three lower premolars are double-rooted and are either closely ap- proximated (M. tortor, M. patagonicitm], or spaced (M. tehtielchnm}. The anterior premolar is a small, simple-crowned tooth, situated close to the canine. The median and posterior premolars are provided with prom- 4IO PATAGONIAN EXPEDITIONS I PALEONTOLOGY. inent heels. The latter tooth is the largest of the inferior premolar series and exceeded the molars in degree of elevation of the crown. In all the specimens retaining this tooth, the tip of the crown is abraded to the gen- eral level of the molars. The anterior molars are an almost exact dupli- cation of the corresponding teeth in Didelphys. The first and second are of about the same size, the third is a little narrower than the preceding teeth, and the fourth is considerably reduced. On all, the three cusps of the trigonid are well developed. The talonid is identical in pattern with that of the lower molars of Didelphys in all except MT, in which it is narrower transversely than in the anterior teeth. Unfortunately, the heel of MT is somewhat broken in the only specimen retaining this tooth (No. 15,698) and its pattern cannot be fully determined. It appears to have been similar to the teeth preceding it, but with the cuspules less distinct. All the lower molars are double-rooted. A narrow antero-external cingulum is present on the first and second molars of M. tehuelchum, but is wanting in the other species. Skull (PI. LXII, fig. i and text-fig. 6). — The skull is remarkable for the great length of the premaxillae and the extreme posterior position of the canine. Anterior to the alveoli of the median incisors the premaxillae develop a shelf-like extension. The orbits are large, with elevated super- ciliary borders and prominent postorbital processes, from which the tem- poral ridges converge to the sagittal crest. In No. 15,698 the greater part of the brain-case is wanting. Its narrowest part lies immediately back of the postorbital processes. The auditory bulla and the glenoid portion of the squamosal are asso- ciated with a mandible of M. tehuelchum (No. 15,038). This specimen indicates that the inferior bar of the jugal extended to the anterior border of the glenoid fossa. The bulla is large, elliptical in outline, with the alisphenoid and petrous portions equally inflated and articulating in open suture, as in Dasyimis viverrinus (PI. LXII, fig. 7). The palate is well preserved in the La Plata Museum specimen (text- fig. 6). It is perforated posteriorly by two large vacuities. In No. 15,698, it is so badly crushed that the nature of the perforations cannot be ascertained. Both specimens show the thickening of the posterior palatal border and its extension beyond the last molar, as in Didelphys. The infraorbital foramen is situated above the posterior half of the last premolar and outer anterior root of the first molar. SINCLAIR : MARSUPIALIA OF THE SANTA CRUZ BEDS. 4 1 I The mandible either has the angle moderately inflected and the inferior margin of the masseteric fossa widely separated from the lower mandibu- lar border (M. tortor, PI. LXII, fig. 2 ; M. gallegosense, PI. LXII, fig. 3), or the angle is strongly inflected, with the masseteric fossa extending to the inferior border of the jaw (M. tehuelchum, PI. LXII, fig. 4). The chin is either heavy, with prominent tubercle (M. tortor] or shallow (M. tehuel- chum]. The inferior mandibular border is much less convex than in Didelphys. The rami are unfused, as in that genus. The mental foram- ina are fairly constant in number and position. A large foramen is sit- uated either beneath the anterior premolar, or beneath adjacent roots of the anterior and median teeth in species with unspaced premolars. A second foramen is placed below the anterior root of the first molar. Some specimens show two foramina beneath this tooth (No. 9595 Amer- ican Museum). A minute foramen piercing the anterior portion of the masseteric fossa is present in M. gallegosense, but does not occur in the other species. Skeleton (PI. LXII, figs. 8-12).-- Parts of the right scapula and ulna, an incomplete right humerus, and the first to the sixth cervicals, lacking the neural arches, are associated with the left half of a lower jaw and two upper molars of M. tehuelchum. The body of the scapula (PI. LXII, fig. 11) has been almost entirely destroyed. The spine is prominent, but has been broken in the region of the acromion. The neck is short. The coracoid process is large, with inflected anterior margin. The humeral head (PI. LXII, fig. 12) has been somewhat damaged. In general, it resembles that of Prothylacynus and Cladosictis, but does not overhang posteriorly to so great an extent. The greater tuberosity has been broken off; the lesser tuberosity is prominent and is separated from the head by a wide, shallow groove. The distal end is broad, with powerful supinator ridge and enormously developed inner epicondyle. The margin of the supinator ridge has been fractured and the character of its proximal end cannot be determined. A large internal epicondylar foramen is present. The posterior border of the ulna (PI. LXII, fig. 10) is strongly convex. The shaft is flattened laterally and deeply excavated on either side of the sigmoid cavity. The atlas and axis (PI. LXII, fig. 9) are imperfectly preserved, but the 412 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. former shows distinctly the unfused intercentrum, as in Borhycena and Amphiproviverra. The third, fourth and fifth cervicals (PI. LXII, fig. 8) are closely applied. The tops of the neural arches have been broken, but their bases are almost in contact, obliterating the intervertebral spaces. This may indicate the beginning of a fusion comparable to that in Di- delphys. The transverse processes resemble those of Didelphys, except that the anterior lamina of the transverse process of the third cervical is wanting. In this respect, Microbiotherimn resembles Dasyurus. The bases of the transverse processes are pierced by a canal for the vertebral artery. These processes are imperfectly preserved on the side repre- sented in the figure. The articular surfaces of the centra are flat. A median ventral keel is indicated on the axis. A prominent tubercle is de- veloped on either side of the middle line on the posterior ventral border of the centrum of the fourth cervical. Corresponding swellings are faintly indicated on the lower borders of the centra of the third and fifth cer- vicals also. According to Ameghino (1894, p. 105), the feet were plantigrade and probably pentadactyl. Systematic Position and Affinities. - - The affinities of Microbiotherium are unquestionably didelphid. Among living forms it approaches most closely some of the subspecies of Caluromys laniger. The genus cannot be regarded as transitional to any of the living opossums, as the degree of reduction of the outer cingulum, styloid cusps and metacone spur in the upper molars is greater. It has been suggested (Bensley, 1903, p. 208) that Microbiotherium, or some allied genus, is ancestral to the Cas- nolestidae and the Santa Cruz thylacynes. So far as the latter are con- cerned, the relationship suggested is still problematic, but it is extremely probable, as will be shown later, that the Caenolestidae have been derived from a didelphid ancestor. MICROBIOTHERIUM TORTOR (Ameghino). (Plate LXII, Figs, i, 2, 20.) Hadrorhynchus tortor Amegh. ; Nuevos Restos Mamif. Fos. Patagonia Austral, p. 25, Aug., 1891 ; Revista Argentina Hist. Nat. I, entr. 5^, p. 311, Oct., 1891. Hadrorhynchus torvus Amegh. ; Ibid. This species is represented by an imperfect cranium and mandible (No. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 413 15,698) from the Lower Santa Cruz beds at Killik Aike. The principal cranial and dental characters have already been noted in the discussion of the genus and need not be repeated. Microbiotherimn tortor is somewhat larger than M. tehuelchum and may be readily distinguished by the strong mandibular symphysis, which terminates inferiorly in a prominent tubercle, by the blunt anteriorly di- rected lower canine, the absence of spacing between the lower premolars, the less strongly inflected angle, and the wide separation of the lower borders of the masseteric fossa and angle. MEASUREMENTS. Length, II- MA on alveolar border .021 " anterior border of C-M± . . . . . . . . .012 " superior premolar series ........ -0035 MI-MA 007 " IT~ MI on alveolar border . . . . . . . . -0173 C-MT .0135 " inferior premolar series ........ .004 " MT-MT on alveolar border ........ .0078 Depth of mandible below anterior premolar ...... .004 « " " " MT 0045 " " " M5 0055 Antero-posterior diameters of Mi, M^ and M* each . . . .002 Transverse diameters of Mi and M3. each ..... .0022 diameter " M-S. .002 Antero-posterior diameter of MA . . . .0015 " diameters of MT, M7 and M8- each . .002 Transverse diameters of MT, Mj and Mg- each . .0015 Antero-posterior diameter of Mj . . .00175 Transverse diameter of M? ... . -OO i 2 MlCROBIOTHERIUM GALLEGOSENSE Sp. nOV. (Plate LXII, Figs. 3, 3*.) This is the largest known species of Microbiotherium. The type speci- men (American Museum No. 9591), collected by Mr. Barnum Brown, on the Rio Gallegos, is the right ramus of a mandible, broken at both ends. The posterior premolar and the first two molars are preserved. The remaining premolars and molars are missing. The alveoli of the anterior and median premolars are closely crowded and the former tooth was directed obliquely to the axis of the jaw. Both 414 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. are double rooted. The median and posterior premolars are slightly spaced. No trace of an external cingulum can be detected on the molars. Like M. tortor, the lower margin of the masseteric fossa is separated from the inferior mandibular border by a broad convex surface. The mandible is moderately convex below and increases regularly in depth posteriorly. A minute foramen, communicating presumably with the inferior dental canal, pierces the masseteric fossa anteriorly. It is not present in M. tortor and M. tehuelchum. MEASUREMENTS. Length, anterior premolar to M T on alveolar border . . . . . .016 " of premolar series on alveolar border ...... .006 " " molar " " " " oio Antero-posterior diameters of My and Mj- each ..... .0025 Transverse " «««<<« OO2 Depth of mandible below median premolar ...... .005 " " " " MT 006 " " " " My . . . .007 MlCROBIOTHERIUM TEHUELCHUM AmeghinO. (Plate LXII, Figs. 4, 401, 6-12.) Microbiotherium tehuelchum Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 7, 1887. The left half of a lower jaw associated with two upper molars and parts of the skull and skeleton (No. 15,038) have been identified with this spe- cies. The principal features of the skull and skeleton have already been noticed in the generic diagnosis and do not call for further comment. The species may be recognized by its size, which is smaller than in M. tortor, by the less prominent chin, the long vertically-directed lower canine, the well-spaced premolars, the strongly inflected angle and the close approximation of the lower margin of the masseteric fossa to the inferior border of the jaw. The mandible differs considerably in shape from that of M. tortor and M. gallegosense, having the angle inclined obliquely inward and upward. An antero-external cingulum is devel- oped on the lower molars. The upper teeth have the external styles rather more prominent than in M. tortor. MEASUREMENTS. Length, anterior border of canine to My inclusive .... .014 inferior premolar series ........ .005 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 415 Length, MT-MT on alveolar border 007 Antero-posterior diameters of MT, M^ and Mj each 002 Transverse " " Mf and Mj each ooi 5 " M, . .001 Depth of mandible below anterior premolar ...... .0035 . " " " MT 0044 " " " MT 0048 Antero-posterior diameters of M^ and M* each 002 Transverse " "Ml OO22 " Mi 002 Antero-posterior diameter of auditory bulla oil Transverse " " " " ...... .0048 Width of humerus at distal end " " . . . . . . .0075 Greatest width of radius proximally ........ .0023 Diameter of radial shaft .......... .0013 Greatest width of ulna at lower margin of sigmoid cavity .... .004 MlCROBIOTHERIUM PATAGONICUM AmeghinO. (Plate LXII, Figs. 5, 5*.) Microbiotherium patagonicum Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 6, 1887. A minute species, represented by the anterior portion of a right man- dibular ramus, with the posterior premolar-and two molars in place, col- lected by Mr. Barnum Brown on the Rio Gallegos (No. 9121 American Museum of Natural History). The remaining premolars, the canine and the incisors are represented by alveoles. ,The premolars are closely crowded and the anterior tooth is placed obliquely to the axis of the jaw. There is no external cingulum on the molars. The remaining dental characters, so far as determinable, are of generic importance and have already been noticed. The species may be recognized by its small size. MEASUREMENTS. Length from anterior portion of canine alveolus to Mj inclusive . .007 Length of space occupied by the premolars .... . .003 " " " " " " first and second molars 0028 Antero-posterior diameter of M-j- ..... .0016 Transverse " " ... .001 Antero-posterior " " Mj . .0015 Transverse " " " -OOi Depth of mandible below posterior premolar . . .002 416 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. DIPROTODONTIA. CsENOLESTIDsE. (Plates LXIII, LXIV ; Text Figs. 7-9.) All the Santa Cruz diprotodonts may be included in one family, of which a single survivor remains in Ccznolestes (PI. LXIII, figs. 14-14^). The species are small, possibly owing to competition with the placental herbivores acting as a check on adaptive radiation. Before proceeding to a discussion of the group, it is desirable to offer some explanation of the classification adopted. Ameghino in his latest publication on the Diprotodontia (1903, p. 159) recognizes three families among the Santa Cruz representatives of the suborder : the Abderitidae, Epanorthidae and Garzonidae. These have been referred to in a general way by other writers (Thomas, 1895; Bensley, 1903) as the Epanorthidae. The genus Epanorthus was proposed by Ameghino in 1889 (1889, p. 271) as a substitute for Pakeoihenies (Moreno) Ameghino (1887, p. 5), on the assumption that the latter conflicted with Palceoteuthis, a genus of ceph- alopods. Although Moreno's Palceothentes (also spelled by him Palce- otenthes] is a nomen nudnm, the description published by Ameghino in 1887 gave this term a priority in nomenclature from which he was not at liberty to depart by the substitution of Epanorthiis. The latter name can no longer be retained either for a genus or to designate a family. The writer prefers to group all the Santa Cruz diprotodont marsupials in a single family, which may be called the Ccenolestidce (Trouessart, 1898, p. 1205) from its only surviving and best known representative Ccenolestes. Family: C^ENOLESTIDAi. — Pes, so far as known (Canolestes), non-syndactylous. Sec- torials, when present, restricted to the posterior premolar above and the first molar below. Superior premolars three in number (Palceothentes, C&nolestes), the anterior and median small, the posterior large and trenchant. Functional lower premolars 2-none. Vestigial teeth always present in the lower jaw. Molars rooted, brachyodont, tuberculo- sectorial or buno-lophodont, undergoing progressive complication in the superior series by the addition of a hypocone. Hypertrophied lower incisors lanceolate with cutting edges enamel layer confined to the outer face. First Subfamily : C&NOLESTINsE.— Dental formula Aj^, i (Ccenolestes). Sectorials not developed. First and second superior molars fully quadritubercular, third and fourth tri- SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 417 tubercular (Ccenolestes). Lower molars tuberculo-sectorial, approaching lophodont when worn. Median and posterior lower premolars double-rooted and functional. Genera : Canolestes, Halmarhiphus, Garzonia. Second Subfamily: PALAEOTHENTINAE.— Dental formula ££,, \. Posterior superior premolar and first lower molar sectorial in function. Sectorials unstriated. First upper molar fully quadritubercular, second with rudimentary hypocone, third and fourth tritu- bercular (Palaothentes). Lower molars lophodont. MT with prominent metaconid. Pos- terior lower premolar double-rooted and functional or single-rooted and reduced. Genera : Palceolhentes, Callomenus, Decastis. Third Subfamily : ABDERITIN^E. — Dental formula ££, \. First lower molar with proto- conoid-paraconoid blade developed into a striated sectorial shear with serrate margin, greatly elevated above the general level of the tooth row. Metaconoid absent on M-J-. Second, third and fourth lower molars buno-lophodont. Functional lower premolars wanting in known Santa Cruz forms, the posterior tooth being single-rooted and vestigial. Genus : Abderites. As many of the genera are known only from the lower jaw, it has seemed advisable to insert parenthetically in the preceding descriptions the names of those forms on which important observations regarding the upper dentition and feet are based. In writing the dental formulas, the number of lower vestigial antemolars of questionable homology is indi- cated in italics, while figures in roman type express the number of teeth which can be definitely homologized. In the Caenolestinae (PL LXIII, figs. 8, 9, 14) the full complement of lower premolars is retained. The antemolar formula in Halmarhiphus and occasionally in Ccenolestes (Bensley, 1903, p. 124, PL 5, fig. 38) is the same as in Didelphys. In a specimen of Garzonia in the collection the antemolar formula is nine, but this may be tentatively regarded as an individual peculiarity, since the constancy of its occurrence has not been confirmed. In the case of the Palaeothentinae and Abderitinae, it cannot be determined at present whether the reduction in the number of vestigial teeth is to be accounted for by the loss of incisors, canine or pre- molars. The elaboration of the sectorial lower molar in the larger members of the Palaeothentinae (Callomenus and Decastis] from a tooth of the Hal- marhiphtts type is plainly indicated by the intermediate condition in Palaothentes. In this genus the anterior lobe of Mr is proportionately longer and higher than in Halmarhiphus or Ccenolestes, but the paraconid is decidedly lower than the protoconid, from which it is separated by a distinct notch, as in the latter genus. In Callomenus and Decastis the 41 8 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. notch has disappeared and the protoconid and paraconid are of the same height, forming an elevated trenchant blade. Owing to the absence of transitional forms in the Santa Cruz fauna, so far as known, it is less easy to trace the development of the peculiar notched and fluted sectorials of Abderites, which are to be regarded as highly specialized structures adapted to a piercing habit, suggesting that the animal fed on the eggs of birds. The loss of the metaconid is a further adaptation toward the perfection of the piercing function. The derivation of the sectorial teeth in Abderites from the tuberculo-sectorial type of molar characteristic of the Caenolestinae is indicated by the broad heel, and by the additional fact that the remaining molars, although some- what less lophodont than in the Palaeothentinae, retain both the paraconid and metaconid as distinct cusps. A satisfactory discussion of the derivation of the upper molar patterns in the Caenolestidae is at present impossible, owing to a lack of material illustrating the upper dentition in many of the genera, especially in the more primitive forms. Less uncertainty exists regarding the lower molars. In the Palaeo- thentinae, lophodont molars have been developed from teeth of the primi- tive tuberculo-sectorial type, shown in Halmarhiphtis, by the formation of cross crests uniting the cusps of the talonid and heel. In the buno-loph- odont molars of Abderites, all the cusps of the original tuberculo-sec- torial crown have been retained, except in My. The loss of cusps in this tooth has already been discussed. So little is known of the skull in the majority of the Caenolestidae that any attempt at a discussion would resolve itself into a repetition of Thomas's excellent description of the skull of Ccenolestes (1895). A de- scription of an incomplete skull of Palceothentes will be found on a later page, to which, and to the accompanying illustrations of the skull of Ccenolestes (PI. LXIII, figs. 14-14^) the reader is referred. Little is known of the podial structure of the Caenolestidae. Ameghino (1894, pp. 80, 81) describes the feet of the Santa Cruz representatives of his suborder Paucituberculata (= the Caenolestidae) as follows: "The four limbs were almost equal in length, but the hind feet were longer than the fore. They were plantigrade, with five toes on the hind feet and probably also on the fore feet, with all the toes well developed and without the least trace of syndactyly." SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 419 Thomas (1895, p. 872) states that in Ccenolestes both feet are penta- dactyl. On the fore foot the pollex and fifth toe are provided with dis- tinct nails and the remaining toes with well developed curved claws. The third digit is the longest ; the second and fifth subequal and shorter. The hind foot is non-syndactylous and not modified into a hand, as it is in the opossums. The hallux is short, clawless and not properly opposablc, developed much as in Phascogale wallacei. The remaining digits of the hind foot are subequal, the fourth slightly the longest, and all provided with claws. CsENOLESTINsE. HALMARHIPHUS Ameghino. (Plate LXIII, Figs. 9, ga ; Text Fig. 7.) Halmarhiphus Amegh. ; Revista Argent., Hist. Nat, T. I, p. 308, 1891. This genus is peculiar in combining characters of both marsupial sub- orders. The tuberculo-sectorial 'molars are structurally the same as in Microbiotherium or Didelphys and the antemolar formula is that of the Polyprotodontia, while in the anterior portion of the mandible Diproto- dont features are apparent in the enlargement of the median incisor and the vestigial character of the remaining incisors, canine and anterior pre- molar. Halmarhiphus is of exceptional interest, not only as the direct ancestor of Ccenolestes, but as a constructive stage in the evolution of the Diprotodontia. This will be treated at greater length in the discussion of the relationships of the Caenolestidae. ' Ameghino (1894, pp. 96-103; 1903, p. 159) places Halmarhiphus md. Garzonia in the family Garzonidae. It appears preferable, however, to group them with Ccenolestes as a subfamily of the Caenolestidae, to which they unquestionably belong. This subfamily has been named the Caeno- lestinae after its best known representative. Halmarhiphus is represented in the collection of the American Museum of Natural History by the right ramus of a lower jaw (No. 9593 American Museum) agreeing in size with H. nanus Ameghino. The tip of the median incisor has been broken off and the first two vestigial teeth shed from their alveoli. Otherwise the dentition is complete and unworn. Nothing is known of the upper teeth. The lower dental formula may be written T, T, ^, T, if definite homologies are assigned to the vestigial 420 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. antemolars, instead of explaining their exceptional number as due to reduplication. This is the same dental formula as in the Didelphyidae. It is occasionally observable in Ccenolestes, as noted by Bensley (1903, p. 124, PI. 5, ng. 38). The proximal half of the median incisor is preserved. As in Ccenolestes, the enamel layer is confined to the outer side of the crown. So far as can be judged from the part preserved, this tooth was of much the same shape as in Ccenolestes. Following the enlarged incisor are five minute teeth closely crowded and more or less pronate, which are interpreted as three incisors, a canine and the anterior premolar. The anterior two are represented by alveoli. The first alveolus is displaced toward the inner side, lying beside instead of in front of the third incisor. The lateral in- cisor, the canine and the anterior premolar are identical in shape. The median and posterior premolars are double-rooted functional teeth. Both carry large heels, that on the median premolar being much larger than FIG. 7. Halmarhiphtis nanus, right ramus, crown view, x f (No. 9593 American Museum of Natural History). that on the posterior tooth. The crowns are laterally compressed, with the principal cusp high and recurved. An anterior accessory basal cuspule of microscopic proportions is observable on the inner side of the crown. The first molar is not differentiated as a sectorial and is slightly smaller than the second. The third is narrower than the second, and the fourth quite small. All display the tuberculo-sectorial pattern (text fig. 7). The trigonid is narrow, with the cusps separated by sharp notches. The talonid is broad, with the hypoconid and entoconid enlarged and the hypoconulid small, but distinct. The cusps of the trigonid and talonid are elevated to the same general level in the three anterior molars. In the fourth, the heel is depressed. The protoconid is slightly higher than the metaconid in the first molar, but of approximately the same elevation in the second, third and fourth. A prominent external cingulum is present on all the molars, as in Ccenolestes and Didelphys. SINCLAIR : MARSUPIALIA OF THE SANTA CRUZ BEDS. 42 1 The mandible is of the same depth as in Canolestes, with the lower border approximately horizontal. Two conspicuous mental foramina are present, situated respectively beneath the posterior premolar and the anterior portion of the second molar. Halmarhiphus may be distinguished from Ccenolestes by the sharper separation of the molar cusps, which in the latter are tending toward the crescentic pattern of the Palaeothentinae. The molar crowns are less quadrangular in outline and the metaconid on Mr is less reduced than in Ccenolestes. More important differences would probably appear in the upper dentition. Garzonia is readily separated from Halmarhiphus and Ctznolestes by the single-rooted condition of Mi. HALMARHIPHUS NANUS Ameghino. (Plate LXIII, Figs. 9, ga ; Text Fig. 7.) Halmarhiphus nanus Amegh. ; Revista Argent, T. I, p. 308, 1891; fenum. Syn., p. 101, 1894; Bol. Ac. Cord., p. 357, 1894; Segundo Censo, etc., p. 187, 1898. The right half of a lower jaw of this little animal (No. 9593 American Museum of Natural History) was collected by Mr. Barnum Brown on the Rio Gallegos. Until a larger amount of material has been secured, the generic and specific characters cannot be separately stated. The princi- pal measurements are as follows : MEASUREMENTS. Length of lower dental series from posterior border of alveolus of IT to MT inclusive ......... .008 Length of antemolar series exclusive of IT. . . . . .003 " " molar series . . . . . . . . -005 Width of base of median incisor ..... .0008 Depth ••••••" " .001 Median premolar, antero-posterior diameter . . . .0012 " " transverse "... .0005 Posterior " antero-posterior " . . . -OOi transverse "... .0006 MT, antero-posterior diameter . . . .0014 " transverse "... -001 MJ-, antero-posterior " " transverse " -0012 M^, antero-posterior diameter . . . .0014 " transverse " °°I 422 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. Mj, antero-posterior diameter ......... .001 " transverse " 0008 Depth of mandible below median premolar ...... .0022 " MT .0023 " " " " MT 0022 GARZONIA Ameghino. (Plate LXIII, Figs. 8, Sa, 10-13.) Garzonia Amegh. ; Nuevos Restos Mamif. F6s. Patagonia Austral, pp. 21-22, Aug., 1891 ; Revista Argentina, I, entr. 5^, pp. 307-308, Oct., 1891. Phonocdromus Amegh. ; 6num. Syn., pp. 99-100, 1894. The genus Garzonia is represented in the Princeton collection by the left half of a lower jaw (No. 15,238), associated with parts of both fore limbs, referred provisionally to G. patagonica. Dentition (PI. LXIII, figs. 8, Sa). — The tip of the median incisor has been broken off, but enough remains to show that the enamel is confined almost entirely to the outer side of the crown. Six single-rooted, more or less pronate, vestigial teeth follow the enlarged incisor. Of these the first, fourth, fifth and sixth are preserved. The second and third are represented by roots retained in the alveoli. The antemolar formula of this individual is therefore nine, the highest on record among the Dipro- todontia. The constancy of this character may well be doubted. The median premolar and parts of both walls of its alveolus have been broken away, but enough remains to show the double-rooted character of the tooth. The posterior premolar is a large tooth supported on heavy roots, with the crown laterally compressed and, in some species (Garzonia typicd], elevated considerably above the molar series. In No. 15,238, the enamel has been broken from the tip of the crown, which probably had a slightly greater degree of elevation than is indicated in the figure. The rather prominent heel of the posterior premolar is overhung by the anterior portion of MT. The molars are greatly worn, and the crown pattern, which appears to have been similar to that in Ccznolestes, almost obliterated (cf. PI. LXIII, figs. Sa, I4<5). The last molar is single-rooted, a character which distinguishes Garzonia from Hahnarhipus and Ccznolestes. The crown has been broken off. In MT the trigonid is somewhat higher than the talonid, as in Cczno- lestes. This was probably true also for M^ and M^, but has been obliter- SINCLAIR: MARSUPIALIA OF THE SANTA 'CRUZ BEDS. 423 ated by the wear to which the teeth have been subjected. An external cingulum is present, as in Halmarhiphus and Ccenolestes. Mandible (PI. LXIII, figs. 8, 8a). — In shape the mandible resembles that of Ccenolestes, but is deeper in proportion to its length. The anterior border of the coronoid is straight, without the convex curvature seen in the latter genus (cf. PI. LXIII, fig. 14), and slopes obliquely backward. The horizontal ramus is deepest beneath MT and MY, becoming much shal- lower anterior to the masseteric fossa, which is perforated, as in C&tio/esfes, by a small foramen situated near its inferior border. The angle is strongly inflected. The condyle, which is placed far above the level of the dental series, is flat transversely, convex antero-posteriorly, and con- siderably wider internally than externally. The symphysial union of the jaws was ligamentous, as in all known members of the Caenolestidae, the symphysial impression extending as far back as the posterior premolar. Two mental foramina are present, a large one beneath the posterior pre- molar and a smaller foramen beneath the first molar. Appendicular Skeleton (PI. LXIII, figs. 10-13). — Compared with the size of the mandible, the bones of the fore limb are remarkably short and slender, none of them exceeding the jaw in length. The neck of the scapula (PI. LXIII, fig. 10) is short and moderately constricted. The glenoid cavity is slightly oval in outline and rather shallow. The coracoid process is prominent, but incomplete at the tip. The humeral shaft is strongly curved antero-posteriorly and greatly ex- panded distally. The head is strongly convex in all diameters and pro- jects considerably beyond the shaft posteriorly. The greater tuberosity is low, not extending above the level of the head. The lesser tuberosity has been broken off. The deltoid crest is very prominent, forming a broad flattened area which extends half way down the shaft. The distal end of the humerus (PI. LXIII, fig. 11) is broad, owing to the great develop- ment of the inner epicondyle and supinator ridge. The proximal end of the latter is without hook-shaped termination. The inner epicondyle has been broken off and is restored in outline in the figure. An entepi- condylar foramen is present. The radius (PI. LXIII, fig. 12) is exceedingly slender. The head is oval and capable of some degree of pronation and supination. Distally, the shaft is slightly expanded and triangular in cross-section. The distal epiphysis has been lost. 424 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. GARZONIA PATAGONICA (Ameghino). (Plate LXIII, Figs. 8, 8a, 10-13.) Phonocdromus patagonicus Amegh. ; fenum. Syn., etc., p. 100, 1894; Bol. Ac. Cordoba, pp. 355-356, 1894; Segundo Censo, etc., p. 186, 1898. The left half of a lower jaw (No. 15,238) associated with the glenoid portion of the right scapula, the greater part of both humeri and the left radius and ulna, collected by Mr. Peterson from the Lower Santa Cruz beds, five miles south of Coy Inlet, is referred provisionally to this species. The correctness of the identification depends principally on the measure- ments, which agree fairly well with those given by Ameghino (1894, p. i oo) for Garzonia patagonica. This species differs from G. typica and G. minuta in the less elevated condition of the posterior inferior premolar, and from G. captiva in size. MEASUREMENTS. Length of mandible from posterior border of median incisor alveolus to condyle 0195 " " lower dentition exclusive of median incisor. .... .0105 " " space occupied by antemolars exclusive of median incisor . . .0055 " " " " " molars 005 Median incisor, transverse diameter at base ...... .0008 " " depth at base 0012 Posterior premolar, antero-posterior diameter at alveolar border . . . .0014 " " transverse " 000$ MT, antero-posterior diameter ......... .002 " transverse "......... .001 1 Mj, antero-posterior "......... .0018 " transverse " oon My, antero-posterior "......... .0013 " transverse "......... .001 Depth of mandible below MT . . . . . . . . . .0028 " " " at constriction posterior to M7 ..... .0023 Humerus, length (approximate) . . . . . . . . .014 Radius, length, exclusive of distal epiphysis . . . . .014 " width of head ooi 5 " " " distal end 002 Ulna, length, exclusive of distal epiphysis . ...... -0175 " greatest width above sigmoid cavity ....... .0028 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 425 PAL^.O THENTIN^E. PAL^EOTHENTES (Moreno) Ameghino. (Plates LXIII, Figs. 1-7; LXIV, Figs. 1-2 ; Text Fig. 8.) Palceotenthes Moreno; Patagonia, Resto de un Continente hoy sub- mergido, p. 22, 1882 (nomen nudum). Palteotkenfes (Moreno) Amegh.; Enum. Sist. Especies Mamif. F6s. Pat. Aust, p. 5, 1887. Palceotheutes Lydekker; Zool. Record for 1887, XXIV, Mamm., 54, 1888. Epanorthm Amegh.; Contrib. al Conoc. Mamif. F6s. Rep. Argent., pp. 271-272, 1889. Metaepanorthus Amegh.; fenum. Syn., p. 92, 1894. Paraepanorthiis Amegh.; 6num. Syn., pp. 93-94, 1894. This is the most abundant and best known of the Santa Cruz diproto- donts, at least four species being represented in the collections at Prince- ton University and the American Museum of Natural History. Dentition (Pis. LXIII, figs. 1-7; LXIV, figs. 1-2). --The dental for- mula in Palceothentes is ££], j. Three upper incisors are figured by Ame- ghino (1895, P- 96> %• 76; 1903. P- MI. fig- 62, p. 170, fig. 95) for P. minntus. The premaxillary region has been broken from the only skull in the Princeton collection (No. 15,225) and the incisor formula cannot be verified. The canine has also been broken in this specimen. From the cfoss-section of the root it appears to have been considerably flattened laterally. The premolar-molar series forms a crescent, tapering in width at both ends, with the convexity directed outward. The three upper pre- molars may be either closely crowded (P. intermeditts, PI. LXIV, fig. i) or moderately spaced (P. aratce, PI. LXIII, fig. 2a). The anterior pre- molar is single-rooted in the latter species, double-rooted in P. intenne- ditts. The tooth has been shed in the only specimen of P. aratce in the collection (No. 9549, Am. Museum) and in the skull of P. intermedia* has been considerably damaged. The median premolar is laterally com- pressed, with a central cusp and well marked anterior and posterior acces- sory cuspules. The posterior premolar is a smooth trenchant blade, greatly widened posteriorly and tapering to an edge in front, where a 426 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. minute accessory cuspule is developed in some species. The crown ter- minates in a thick blunt point. The anterior root is very oblique and narrower transversely than the posterior root. The principal wear is on the postero-internal face of the crown, where the tooth shears against the anterior blade of the lower sectorial. The molars decrease rapidly in size posteriorly. Each is triple-rooted, with two roots on the buccal and one on the lingual side. The first is fully quadritubercular ; the second has an incipient hypocone ; the third and fourth are tritubercular. The crowns are bunodont. The protocone and hypocone are unite.d respec- tively with the paracone and metacone by transverse ridges (cf. PI. LXIII, fig. 7), and the latter cusps with each other by a sharp trenchant ridge, which passes over the external cusps, as in Petaurus (PI. LXV, fig. 4). An external cingulum is faintly indicated in some specimens (PI. LXIII, fig. 7), wanting in others (PI. LXIV, fig. i). A rather broad anterior cingulum is developed on the first molar. The median lower incisors are procumbent and lanceolate, with the outer edge of the enamel attenuated (PI. LXIII, figs. 4^, 50), and occasion- ally notched by accidental fractures received during life. The enamel layer is restricted to the anterior face of the crown. Internally, it is re- enforced by a thick rib of dentine. The enamel does not grow persist- ently but covers a limited area, which decreases in size as the tooth wears down (cf. PI. LXIII, figs. 4 an$ 5). In shape the incisors resemble those of the Macropodidae. Following the incisor are four single-rooted vestigial teeth. The an- terior two are remarkably procumbent in P. minutus (PI. LXIII, fig. 46), possibly less so in P. lepidus. The third is not retained in any of the specimens in the collection. The crown of the fourth is antero-posteri- orly elongated and overhangs in front. The posterior premolar is a large double-rooted tooth elevated to about the same extent as the molars. The crown supports a prominent central cusp and more or less elevated anterior and posterior accessory cuspules. These have been used by Ameghino in defining the genera Metaepanorthus and Paraepanorthus. The former he characterizes by the presence of well-defined anterior and posterior accessory cuspules on the posterior premolar, and the latter by the occurrence of the anterior cuspule only. The presence or absence of these structures can hardly be a matter of generic importance, as they vary in size and prominence within the limits of a species (cf. PI. LXIII, figs. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 427 4 and 5). The lower molars are lophodont, the cusps of the trigonid and heel forming crescentic ridges. All the molars are double-rooted and de- crease rapidly in size posteriorly. The first is modified as a sectorial by the elongation of the protoconid- paraconid blade. Vertical ridges are not developed on the sectorial, but the enamel on the outer side is irregularly crenulated. The paraconid is lower than the protoconid, from which it is separated by a distinct notch. In this respect, Palceothentes is intermediate between the Caenolestinae and the "more specialized Palaeothentinae, Callomenus and Decastis. The second and third molars are similar in pattern to the first, but the para- conid is reduced and the horn of the posterior crescent, instead of uniting with the metaconid, as in the sectorial molar, is shifted farther toward the outer side, uniting with the anterior crescent between the protoconid and metaconid, as in Callomenus and Decastis (cf. PI. LXIII, fig. 6a ; PI. LXIV, figs. 5«, 6a). The fourth molar is a small tooth similar to those preceding it, but losing early by wear the details of the crown pattern. External cingula are faintly indicated on the anterior lobes of the first and second molars. Skull '(PI. LXIII, fig. 3; PI. LXIV, figs, i, i*). — A skull of P. inter- medius in the Princeton collection (No. 15,225), lacking the region anterior to the canine and posterior to the glenoid fossae, forms the basis for the following description. In side view (PI. LXIII, fig. 3), the skull is seen to be gently arched, the highest point lying at the junction of the temporal ridges. Back of the orbits, the brain case contracts, becoming widely expanded posteriorly. Postorbital frontal processes are wanting, but well marked temporal ridges give rise to a low sagittal crest. Between the orbits, the frontal is plane. Anteriorly, it sends a broad bar forward between the nasal and lachrymal to join the maxillary, differing in this respect from the Santa Cruz marsupial carnivores. The nasals are very broad behind, rapidly decreasing in width anteriorly. Unlike Ccenolestes, there is no trace of an antorbital vacuity. The pre- maxillae have been largely broken away, but resemble those of Gzno/estes in sending a narrow tongue between the maxillary and nasal (cf. PI. LXIII, fig. 14; PI. LXIV, fig. ia). The lachrymal has but little facial extension. The lachrymal duct opens within the orbital rim, which sup- ports a small but distinct lachrymal tubercle. The anterior margin of the orbit is sharply defined. The jugal arches 428 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. have been broken, but enough of the left squamosal bar remains to show that the malar extended to the glenoid fossa. The squamosal portion of the arch is not inflated, as it is in Petaunis, Trichosurns and others of the higher phalangers. The palate is deeply concave both antero-posteriorly and transversely. The incisive foramina extend beyond the premaxillary suture to a point opposite the anterior premolars. Two antero-posteriorly elongated pala- tal vacuities are present, extending from a point opposite the anterior extremity of M1 to M-. Neuro-vascular canals perforating the posterior margin of the palate are present, as in the majority of marsupials. The palato-narial border is greatly thickened and elevated in a manner resem- bling Didelphys. The infra-orbital foramen is large, opening above the anterior root of the upper sectorial. Mandible (PI. LXIII, figs. 4-60; text fig. 8).- -The mandible is long, slender and shallow in the smaller species, deep and probably less elon- gated in the larger forms. The rami are unfused. The symphysial im- pression extends as far back as the anterior half of the sectorial. The coronoid and angle are not preserved in any of the specimens studied. Two mental foramina are usually present, one beneath the sectorial, the other beneath or slightly anterior to the last vestigial tooth. Occasionally there are two posterior foramina somewhat variable in position. PALEOTHENTES ARATE (Moreno) Ameghino. (Plate LXIII, Figs. 2, 2a ; Text Fig. 8.) Palceotenthes aratce Moreno ; Patagonia, Resto de un Continente hoy submergido, p. 22, 1882 (nonten nudum). Palceothentes aratce (Moreno) Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 5, 1887. Epanorthus aratce (Moreno) Amegh. ; Contrib. al Conocimiento Mamif. F6s. Rep. Argent., pp. 272-273, PI. I, Figs. 10-12^, 1889. Epanorthus aratce (Moreno) Trouessart; Catalogus Mammalium, p. 1202. Represented in the American Museum collection by a fragment of the right maxilla (No. 9549 American Museum) from Santa Cruz, retaining the median and posterior premolars and the second, third and fourth SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 429 molars, as well as the alveoli of the canine, the anterior premolar and the first molar. The species may be recognized by its size, by the presence of dias- temata between the premolars, by the spacing of the canine and anterior premolar, and the single-rooted condition of the latter tooth. FIG. 8. Palaothentes aratce, right ramus, x -|. From an enlarged photograph of a specimen in the Ameghino collection. The mandible, shown in the accompanying figure (text fig. 8), drawn from a photograph of a jaw in the Ameghino collection, is very heavy and deep, with convex lower border. The horizontal ramus increases slightly in depth posteriorly. MEASUREMENTS. Amegh. Coll. No. 9549. Length, posterior border of canine alveolus to MA, inclusive . -0235 " of space occupied by premolars ..... .0125 " " " " " molars .0115 Median premolar, antero-posterior diameter .... .0025 " " transverse " .0015 Posterior premolar, antero-posterior "..... .006 " " transverse " .0035 M^, antero-posterior diameter ....... -OO35 " transverse " -005 Mi, antero-posterior " ...... .0025 " transverse " .0035 MA, antero-posterior diameter ...... .002 " transverse " .0025 Posterior lower premolar, antero-posterior diameter . . . -0035 Length of inferior molar series on alveolar border . . . .015 MT, antero-posterior diameter . . . .0065 M, " "...... .004 43O PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Amegh, Coll. My, antero-posterior diameter ........ .003 MT, " " 002 Depth of mandible below M^ . . . . . . . . .0075 " " " MT 008 PAL/EOTHENTES INTERMEDIUS Ameghino. (Plates LXIII, Figs. 3, 7 ; LXIV, Figs, I, ia.) Palceothentes intermedius Amegh. ; Enum. Sist. Especies Mamif. Fos. Pata- gonia Austral, p. 6, 1887. Epanorthus intermedius Amegh.; Contrib. al Conocimiento Mamif. Fos. Rep. Argentina, p. 274, PI. I, Figs. 15-15^, 1889. Metaepanorthus intermedius Amegh.; fenum. Syn., etc., p. 92, 1894; Bol. Acad. Cordoba, p. 348, 1894. An incomplete skull (No. 15,225) collected by Mr. Peterson from the Lower Santa Cruz beds five miles south of Coy Inlet, and two maxillary fragments (No. 15,952; American Museum, No. 9550), from Killik Aike and Santa Cruz respectively, apparently belong to this species, agreeing closely with the measurement given by Ameghino (1894, p. 92) for the premolar-molar series. The important features of the skull and teeth have already been mentioned in characterizing the genus. So far as can be judged from the small amount of material available, the size of mature individuals is fairly constant. The upper premolars are all double-rooted, without intervening diastemata. The anterior pre- molar is in contact with the canine. An exceedingly small anterior ac- cessory basal tubercle is present on the upper sectorial in unworn teeth. The species is considerably larger than P. lepidus, occupying in point of size a position intermediate between that species and P. aratce. MEASUREMENTS. No. 15,225. No. 15,952. No. 9550. Length of upper premolar-molar series meas- ured as chord of arc . . . .015 .015 Length of space occupied by upper premolars -0075 .0074 " " " " " " molars . .0085 .0085 Anterior premolar, antero-posterior diameter .0018 " " transverse " . .0008 Median " antero-posterior " . .002 .002 " " transverse " .001 .001 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 431 No. 15,225. No. 15,925. No. 9550. Posterior premolar, an tero- posterior diameter. .0043 .004 .004 transverse diameter . .0023 .0022 .0022 M-L, antero-posterior diameter . . . .004 .0038 .004 " transverse " ... .0032 .003 -0033 M^, antero-posterior " ... .0025 .0025 .0028 " transverse " ... .0032 .003 .0033 Mi, antero-posterior " ... .0017 .0018 " transverse " ... .002 .002 MA( antero-posterior " ... .001 .001 " transverse " ... .0013 .0012 Greatest interorbital width .... .0105 Least width of brain-case .... .007 Greatest breadth of nasals .... .0087 Depth of skull to alveolar border at front of orbit 012 Width of palate between anterior premolars . .007 " " " at Ml oio " " " " M± oio PAL^EOTHENTES LEPIDUS Ameghino. (Plate LXII, Figs. 6, 6a.) EpanorthTis lepidtis Amegh. ; Revista Argentina, I, p. 305, 1891. This species, is known only from the mandible, of which the American Museum collection contains four incomplete specimens (Nos. 9596-9598, 9600) found by Mr. Brown on the Rio Gallegos. In size P. lepidus is intermediate between P. minutus and P. interme- dius. The vestigial teeth occupy a shorter space than in P. minutus and are more closely crowded. The alveoli of the first and second are larger than those of the third and fourth, are closely adjacent and strongly inclined forward. The alveoli of the third and fourth vestigial teeth are approximately vertical. The mental foramina vary considerably in size and relative position and are of little diagnostic value. The principal measurements agree closely with those of a natural-size photograph of the type in the Ameghino col- lection, and have been depended on largely for the correctness of the specific identification. 432 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. MEASUREMENTS. No. No. No. No. Type. 9596- 9597- 9598- 96o°- Length, posterior border of median incisor alveolus to My inclusive . . . .014 .014 Length of space occupied by vestigial teeth and posterior premolar .... .005 .005 Length of inferior molar series on alveolar border . . .... .009 .009 .009 Posterior premolar, antero-posterior diameter .0014 .0015 .0015 .0015 " " transverse " .001 .001 .001 MJ-, antero-posterior diameter . . . .0035 -0035 -0035 .0035 -°°3S " transverse " .002 .002 .002 Mj, antero-posterior " .0025 .0025 .0025 .0025 .0025 " transverse " .002 .002 .002 .002 My, antero-posterior " .002 .002 .002 .002 " transverse " .0015 .0015 .0015 My, antero-posterior " .0012 Depth of mandible below posterior premolar . .0035 -0035 " " " " Mj . . . .0035 .0035 .004 .0035 .004 " " " " MT .004 PAL^EOTHENTES MINUTUS Ameghino. (Plates LXIII, Figs, i, 4-5*; LXIV, Fig. 2.) Palceothentes minutus Amegh. ; Enum. Sist. Especies Mamif. F6s. Pata- gonia Austral., p. 6, 1887. Epanorthus minutus Amegh.; Contrib. al Conocimiento Mamif. F6s. Rep. Argent, p. 274, PI. i, Figs. i6-i6a, 1889. Paraepanorthus mimitus Amegh. ; Enum. Syn., etc., pp. 94-95, Fig. 40, 1894; Bol. Acad. Cordoba, p. 350, Fig. 40, 1894. Epanorthiis simplex Amegh. ; Enum. Syn., pp. 91-92, 1894 ; Bol. Acad. Cordoba, p. 347, 1894. Incomplete lower jaws of six individuals from Killik Aike (Nos. 15,706- 15,709, 15,068, 15,624) have been identified as belonging to this little di- protodont. A maxillary fragment retaining the posterior premolar and the molars is associated with one of the mandibles (No. 15,709). Two upper molars (No. 15,999) collected by Mr. Hatcher on the south side of the Santa Cruz River, sixty miles below Lake Argentina, and part of a left mandibular ramus (No. 9122 American Museum) from a locality on the Rio Gallegos, are also referable to this species. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 433 As characters of specific value may be mentioned the small size, shal- low, slender jaws, the presence of a well-marked anterior accessory cuspule on the last upper premolar, and the great length of the space occupied by the vestigial antemolars, which are less closely crowded than in some of the larger species (P. lepidiis]. These teeth are of an exceedingly peculiar shape. The first and second of the series are widely separated from each other at the roots, but the cylindrical crowns curve forward abruptly, lying prone on the alveolar border in such a manner that the tip of the second rests on the root of the first, while the tip of the latter rests on the base of the median incisor (PI. LXIII, fig. \b\ The last vestigial tooth has a small, button-shaped crown overhanging anteriorly. Minute notches in the edges of the median incisors are due to accidental fracturing of the attenuated edges of the enamel layer investing the outer surface of the crown. Two large mental foramina are usually present, situated respectively beneath the last vestigial tooth and the first molar. A photograph of the type specimen of Palceothentes simplex in the Ame- ghino collection, for which the writer is indebted to Professor Scott, shows no differences warranting a separation of this species from P. minutus. MEASUREMENTS. No. No. No. No. No. No. No. No. 15,706. 15,707. 15,708. 15,700. 15,068. 15,624. 15,000. 9122. Length of mandible from tip of median incisor to Mj inclusive .0175 .017 Length of mandible from posterior border of median incisor alveolus to M7 inclusive . . ..013 .013 Length of space occupied by vesti- gial teeth and posterior pre- molar 0064 -0057 Width of base of median incisor .0008 .0008 .0008 .001 Depth " " " " " .0014 .0014 .0013 .0015 Length of superior molar series on alveolar border . . . .006 Ml, antero-posterior diameter . .0024 " transverse " . .002 M^, antero-posterior " . .002 .002 " transverse " . .0018 .002 M-3-, antero-posterior " . .0015 .0015 434 PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. No. No. No. No. No. No. No. No. 15,706. 15,707. 15,708. 15,700, 15,068. 15,624. 15,000. 0,122. M^, transverse diameter . . .0015 .0018 MA, antero-posterior diameter . .0012 " transverse diameter . . .00 1 Posterior lower premolar, antero- posterior diameter . . ..0015 .0016 .0014 .0015 .0015 .0015 .0014 Posterior lower premolar, trans- verse diameter . . . .0008 .001 .0008 .0008 .001 .001 .0008 Length of inferior molar series on alveolar border . . . .007 .007 .008 MT, antero-posterior diameter .0024 .0025 .0025 .0025 " transverse " ..0015 .0015 .0015 .0015 My, antero-posterior " . .002 .002 .002 .002 .0022 .002 " transverse " . .0015 .0015 .0015 .0015 .0015 .0015 My, antero-posterior " . .0017 .0016 .0018 .0018 " transverse " . .0012 .0012 .0014 .0014 M-f, antero-posterior " . .0012 .001 .001 " transverse " . .0009 .0008 .0008 Depth of mandible below posterior premolar .... .0028 .003 .0025 .003 .0025 Depth of mandible below My ..0032 .0036 .003 .0027 " " " " Mj . .0032 .0035 CALLOMENUS Ameghino. (Plate LXIV, Figs. 5, 5^.) Callomenus Amegh. ; Nuevos Restos Mamif. F6s. Patagonia Austral, p. 20, Aug., 1891 ; Revista Argentina, I, entr. 5^, p. 306, Oct., 1891. Although known only from the mandible and inferior dentition, this genus is probably valid, having progressed farther than Palceothentes in the elimination of the vestigial antemolars, as the result of a progressive shortening of the lower jaw. The posterior premolar is double-rooted, but no longer reaches the general level of the molar series, as it does in Palceothentes. It is overlapped and partly concealed by the sectorial blade of the first molar. In view of the individual variation in the number of vestigial teeth shown by some of the phalangers, Callomenus may be re- garded as rather doubtfully separable from Acdestis, which retains four of these teeth (Ameghino, 1903, p. 171, fig. 98) in addition to the double rooted premolar. Callomenus :is represented in the collection by the right half of a mandible SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 435 (No. 15,066) which retains the base of the incisor and all the inferior den- tition except the vestigial antemolars. The coronoid and part of the angle have been broken off. The median incisor is elliptical in cross-section, with the enamel con- fined to the outer side. Between this tooth and the posterior premolar are three closely crowded alveoles for the vestigial, single-rooted ante- molars. The posterior premolar is double-rooted, compressed laterally, with an anterior principal cusp and prominent heel partly overlapped by the trenchant blade of the first molar. The molars are double-rooted and decrease in size posteriorly. The first is greatly enlarged, as in Palceo-y thentes. The ridge uniting the protoconid and paraconid is functional as a sectorial blade, but is not notched as in Palceothentes. The prominent metaconid is united by ridges with the protoconid and cusps of the heel. These ridges form a pair of crescents, the concavity of which is directed internally. The outer side of the tooth is channelled by a deep groove which extends forward and inward toward the metaconid. The second and third molars resemble the first in pattern. In these teeth, the proto- conid-paraconid blade is considerably reduced and the tooth crown more quadrangular than MT. The posterior crescent is deflected outwardly, joining the anterior crescent much nearer to the protoconid than the metaconid. The channel on the outer face of the tooth crown is much shallower than in My. The fourth molar is a minute tooth, with button- shaped crown, which, owing to its worn condition, does not show well the lophodont pattern of the anterior molars. It is evident from the shape of the crown that this tooth was of the same general pattern as the third molar. A large mental foramen is present beneath the last vestigial tooth, and a smaller foramen beneath the sectorial molar. The masseteric fossa is imperforate. The angle is prominent and strongly inflected, and the horizontal ramus evenly convex and of about the same depth through- out beneath the molars. CALLOMENUS LIGATUS Ameghino. (Plate LXIV, figs. 5, $«.) Callomenus ligatus Amegh.; Enum. Syn., etc., p. 88, 1894; Bol. Acad. Cordoba, p. 344, 1894. Three species are recognized by Ameghino (1891, p. 306 ; 1894, p. "88), of which but one is represented in the Princeton collection by the right 436 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. half of a mandible (No. 15,066), collected by Mr. Peterson from the Upper Santa Cruz beds at Killik Aike. Owing to the small amount of material available, it has not been possible to distinguish between generic and specific characters. The principal dependence for the correctness of the identification has, consequently, been upon size. The dimensions given below agree closely with Ameghino's figures for Callomenus ligatus. MEASUREMENTS. Length, posterior border of median incisor to MT inclusive . . . .0155 " of space occupied by vestigial teeth and posterior premolar . . -0055 " " molar series on alveolar border. . . . . . . .010 Width of base of incisor .......... .0015 Depth « « « « 0025 Posterior premolar, antero-posterior diameter ...... .0015 " " transverse " ...... .001 MT, antero-posterior diameter ......... .0045 " transverse " ......... .002 M^-, antero-posterior " ......... .003 " transverse " ......... .002 MJ-, antero-posterior " ......... .002 " transverse " ......... .0015 M-J-, antero-posterior " ......... .0013 " transverse " ......... .001 Depth of mandible below last vestigial tooth ...... .0043 " " " " posterior premolar ...... .0045 " " MT 005 " " MT 0055 DECASTIS Ameghino. (Plate LXIV, Figs. 4, 4«, 6, 6a.) Decastis Amegh. ; Nuevos Restos Mamif. F6s. Patagonia Austral, p. 19, Aug., 1891 ; Revista Argentina, I, entr. 5*7, p. 305, Oct., 1891. In Decastis, premolar reduction is carried one step farther than in Callomenus. The heel of the posterior premolar is still retained, but the whole tooth is smaller and no longer double-rooted. Between the last premolar and the median incisor are four vestigial single-rooted teeth. The base of the incisor is retained in one specimen (No. 9594 American Museum, PI. LXIV, figs. 4, 40), showing a large pulp canal, but broken off too far below the crown to indicate the distribution of the enamel. The tooth is larger than in Callomenus, but similar in shape at the base. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 437 The last vestigial tooth (PI. LXIV, fig. 6), probably representing the median premolar, has a cylindrical, blunt-pointed crown. The crown of the third intermediate (PI. LXIV, fig. 4), overhangs anteriorly. The first two rudimentary teeth have been shed, but from the obliquity of their alveoli, it is probable that they resemble the homologous teeth in Palceo- thentes. The molars are an exact duplication of those of Callomenus, but are slightly more worn in both the mandibles in the collection. The last tooth is double-rooted, but has, unfortunately, been broken off. The number and relative position of the mental foramina is much the same as in Callomenus. The apparent shifting forward of the anterior foramen is due to the reduction in length of the premolar series. The masseteric fossa is imperforate and the coronoid broad and high with prominent anterior margin. The symphysial impression is wider than in Callomenus. Difference in size is the only character available for the separation of the two species recognized by Ameghino (1891, p. 305), of which the larger, D. columnaris, is represented in the collection by two specimens. DECASTIS COLUMNARIS Ameghino. (Plate LXIV, Figs. 4, 40, 6, 6a.) Decastis columnaris Amegh. ; Nuevos Restos Mamif. F6s. Patagonia Aus- tral, p. 19, Aug., 1891 ; Revista Argentina, I, entr. 5«, p. 305, Oct., 1891. Like Callomenus ligatus, Decastis columnaris is known only from lower jaws, of which parts of two are in the collections at Princeton University and the American Museum. No. 15,710, collected by Mr. Peterson at Coy Inlet, is a portion of the left half of a mandible, with two premolars and three molars in place (PI. LXIV, figs. 6, 6a). The second specimen (No. 9594 American Museum) from the Santa Cruz beds, on the Rio Gallegos, is the anterior half of a right mandibular ramus retaining in place two molars, the third vestigial tooth and the base of the incisor (PL LXIV, figs. 4, 4^)- MEASUREMENTS. No. 15,710. No. Approximate length, posterior border of median incisor to MT inclusive. . . . . . . . . . .016 Length of space occupied by vestigial teeth and posterior premolar. .0055 438 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. No. 15,710. No. 9594. Length of molar series on alveolar border . . . . .010 Width of base of incisor ........ .002 Depth "•«•••< .003 Posterior premolar, antero-posterior diameter .... .0012 " " transverse "..... .001 MT, antero-posterior diameter . . . . . . -00475 •°°S " transverse " 0024 .0025 My, antero-posterior " . . ' . . . . . . .003 .003 " transverse " . . . . . . . . .002 .002 MJ-, antero-posterior "........ .002 " transverse " 0015 Depth of mandible below first molar ...... .005 .005 ABDERITINA1. ABDERITES Ameghino. (Plate LXIV, Figs. 3, 30 ; Text Fig 9, a, b.) Abderites Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 5, 1887. Diprotodont marsupials in which the first lower molar is developed as a striated sectorial blade resembling superficially the sectorials of the Plagi- aulacidse. Dentition (PI. LXIV, figs. 3, 3^ ; text figs. 9, a, b}. --The formula for the lower dentition is ^-^> -v The median incisors are not preserved in either specimen of Abderites crassignathus in the Princeton collection (Nos. 15,079, 15,425), but, judging from the size of their alveoli, they were large teeth. The posterior premolar is a very small, single-rooted, cylin- drical tooth closely applied to the anterior root of MT. Between the enlarged incisor and the posterior premolar are alveoli for four single- rooted, vestigial teeth. The molar series is placed very obliquely to the long axis of the jaw, more so than in any other member of the Caenoles- tidae, the anterior blade of the first molar projecting beyond the plane of the outer surface of the mandible. The molars are double-rooted and decrease in size posteriorly, as in all the Caenolestidae. The first has been converted into a very perfect sectorial by the complete obliteration of the metaconid, the great elevation and lateral compression of the protoconid- paraconid blade, and the development of parallel vertical ridges on both the outer and inner faces of the crown. The ridges on opposite sides SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 439 meet along the edge of the " sectorial blade in a series of serrations, the number of which varies with the species. The broad talon is bicuspidate. The second and third molars are quadrangular in outline and bunolopho- dont, with the paraconid present but reduced. The fourth molar is retained in one specimen (No. 15,425); but the crown has been abraded to a flat surface without trace of cusps or ridges. In the fragment of the right maxillary referred by Ameghino (1898, p. 184, fig. 49, II; 1903, p. 142, fig. 64, p. 178, fig. 107) to A. meridion- alis, the second molar is fully quadritubercular, in contrast with the in- completely quadritubercular M- in Palcepthentes. Mandible. — The mandible is much deeper than in Callomenus and Decastis. The coronoid is high, with the anterior margin sloping back- ward. Its base is perforated by a branch of the alveolar canal opening externally behind the last molar (PI. LXIV, fig. 30). The masseteric fossa is broad and occasionally pierced by a small foramen situated some distance below the large irregular opening shown in the figure (PI. LXIV, fig. 3), which is due to fracture. The symphysis is broad and heavy, ex- tending posteriorly beneath the first molar. ABDERITES CRASSIGNATHUS Ameghino. (Plate LXIV, Figs. 3, 30; Text Fig. 9.) Abderites crasignathus (sic] Amegh. ; Revista Argentina Hist. Nat., I, p. 248, 1891. Abderites crassiramis Amegh. ; Rev. Gen. des Sciences, p. 80, fig. 4, 1893. The beautiful little specimen figured on PI. LXIV (No. 15,079) was collected by Mr. Hatcher on the Rio Chalia, and has been mentioned by him in the Narrative of the expeditions (this series, Vol. I, p. 113). A second specimen from Killik Aike (No. 15,425) agreeing with the first in proportions, but lacking the crown of the sectorial, is referred to the same species and is represented in Fig. 9. A. crassignathus is readily identified by the presence of five or six • prominent ridges on the anterior half of MT. The ridges are developed on both the outer and inner sides of the tooth, producing a series of ser- rations on the cutting edge of the sectorial blade. A photograph of the type in the Ameghino collection obtained by Professor Scott shows five prominent ridges on the outer side of Mr extending to the cutting edge 440 PATAGONIAN EXPEDITIONS I PALEONTOLOGY. of the crown. In the Princeton specimen there are six ridges. On the inner side, the fourth ridge extends inferiorly but a short distance, as is well shown in the figure (PI. LXIV, fig. 30). In addition to the main anterior ridges, five short ribs are observable on the outer side of the sec- torial. These probably extended farther toward the cutting edge, but have been partly obliterated by the abrasion of this part of the tooth crown from contact with the opposing tooth in the upper jaw. The pho- tograph referred to above shows two short faint ridges back of the last prominent outer rib. These, however, do not give rise to serrations on the margin. The crowns of the vestigial antemolars have not been pre- served, but in the photograph of the type specimen are seen to be antero- FIG. 9. a Abderites crassignathm, right ramus of mandible, x f (No. 15,425). a, side view ; b, crown view. posteriorly elongated and to overhang in front, as in Palceothentes minutus (PI. LXIII, fig. 46). The fourth vestigial tooth and the posterior pre- molar are in contact. The third and fourth vestigial teeth are separated by a considerable diastema. The alveoli of the first and second are elon- gated antero-posteriorly and inclined obliquely forward, as in Palceothentes. A large mental foramen pierces the mandible below the sectorial, and another is present, below or slightly posterior to the third vestigial ante- molar. MEASUREMENTS. No. 15,425. No. 75,079. Length, posterior border of incisor alveolus to M^ . . . .022 " of space occupied by vestigial antemolars . . . .0095 " " molar series on alveolar border .... .013 .013 Posterior premolar, antero-posterior diameter .... .0007 SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 441 No. 15,425. No. 1 Posterior premolar, transverse diameter .0008 My, antero-posterior diameter ...... .005 " transverse " ...... .0023 height of middle of crown, above alveolar border . . .0045 Mj, antero-posterior diameter ...... .0033 -0033 " transverse ' OO22 .0025 Mj, antero-posterior ' 003 .003 " transverse ' 002 .002 M-f, antero-posterior ' ...... .002 " transverse ' ...... .0017 Depth of mandible below middle of sectorial .... .0065 .0065 " " " M¥ ...... .006 .006 RELATIONSHIPS OF THE OENOLESTIDyE. The most primitive Santa Cruz representative of the Caenolestidae is, undoubtedly, the genus HalmarhipJnts, which represents, with little or no modification, a type which is not only ancestral to the Palaeothentinse, but agrees perfectly with the "minute insectivorous forms which, apart from the diprotodont modification of the antemolar teeth, possessed a full ante- molar formula," indicated by Bensley's studies as the ancestors of the Phalangerinae. In this interesting genus the dental formula and molar patterns are didelphid, affording striking evidences in favor of the theory of the didelphid origin of the Diprotodontia (see Bensley, 1903). It has already been shown (p. 417) that the lower molar patterns in the Palaeo- thentinae are readily derivable from the Halmarhiplms type by a reduc- tion in height of the cusps and the formation of cross crests. So far as our knowledge of Halmarhiphus warrants an inference, there can be little objection to deriving the Palaeothentinse from a similar ancestral form. The chief objection to regarding Halmarhiphus as directly ancestral to the Palaeothentinae arises from the fact that the latter are represented in formations older than the Santa Cruz (P. chubtitensis from the Pyro- therium beds, Amegh., 1897, P- 9^)- That Halmarhiphus is in the direct line of descent culminating in Ccenolestes will, it is believed, hardly be questioned after an examination of the accompanying plates. So far as the lower dentition is concerned, no argument can be advanced to the contrary. The recent genus shows a slight tendency toward the develop- ment of crescents, while, in its Santa Cruz predecessor, the cusps are high 442 PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. and pointed, as in the Didelphyidae. Unfortunately, nothing is known of the upper dentition. Garzonia is more specialized than the other mem- bers of the Caenolestinae, having the last lower molar single-rooted. Some account has already been given of the lower molars in the Palaeo- thentinae and an attempt has been made to show that Palceothentes, Callo- menus and Decastis are members of a closely related series characterized by a progressive reduction in the number of antemolar teeth, reduction in the size of the posterior premolar and increasing perfection in the adaptation of the first molar to a sectorial function (see pp. 427-436). It remains to point out the striking similarity in upper molar patterns exhibited by Palceothentes and certain of the phalangers. Reference to the accompany- ing plates will at once make this clear (see Pis. LXIII, fig. 7 ; LXIV, fig. i ; LXV, fig, 4). Indeed, it is possible to trace in the Palaeothentinae the constructive stages in the evolution of the bunodont type of molar characteristic of the more primitive of the existing phalangers. The devel- opment of the hypocone in the upper molars of Palceothentes is less com- plete than in these phalangers, only the first molar being quadrangular and fully quadritubercular, the second retaining a triangular outline, with incipient hypocone. In the higher phalangers (Phalanger and Tricho- surus, PI. LXV, figs. 3, 3*2), all the molars are quadritubercular, inter- mediate stages in molar complication occurring in Petaums, with three quadritubercular molars, and Dromicia with two. In Ccenolestes there exists the apparent anomaly that a genus more primitive than Palceothentes should have the second upper molar more complicated. This complica- tion, however, may be a measure of the extent of dental evolution in the Caenolestinae in post-Santa Cruz time. The condition in the Santa Cruz representatives of the family is not known, but presumably the upper molars were less advanced than in Palceothentes. In the highly special- ized Abderitinae, judging from Ameghino's figure of A. meridionalis (Amegh., 1898, p. 184, fig. 49, II ; 1903, p. 142, fig. 64, p. 178, fig. 107), the second upper molar is fully quadritubercular. Palceothentes, then, represents an early constructive stage of a progressive complication of the upper teeth, which began with M1 and proceeded backward. Abderites is near the end, if not the terminal member, of a highly specialized line, the intermediate stages of which have not been found in the Santa Cruz beds. Affinities with the Plagiaulacidae have been com- monly assumed from the striking resemblance between the first lower SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 443 molar of Abderites and the notched and fluted sectorial teeth of the plagiaulacid Multituberculata. This resemblance is confined to the an- terior half of the sectorial, the posterior portion in Abderites supporting a large bicuspidate heel. The posterior molars show no indication of cusp reduplication and still retain the three cusps of the trigonid, indicating their tuberculo-sectorial origin. In Hypsiprymnodon and Bettongia the posterior premolars closely resemble the sectorials of the Multitubercu- lata. The Bettongiinae and Abderitinae illustrate a case of convergence, where much the same form has been assumed by totally different teeth. The homologies of the sectorials of the Plagiaulacidae are uncertain, but they have been interpreted as posterior premolars, while in the Abderitinae the sectorial is unquestionably the first molar. In a former paper the writer (1905, p. 81) stated that: "the Caenoles- tidae resemble the primitive phalangers in so many respects that it is im- possible to escape the conclusion, that the two famlies are related and not merely convergent groups. With the exception of Halmarhiphtts, a persistent ancestral type, the Santa Cruz diprotodonts possess specializa- tions in dental structure which prevent their being regarded as direct ancestors 'of the phalangers, but favor the idea that both groups are descended from a common ancestry." While substantially the same con- clusions are still held, it is proper to point out the evidence in favor of the view that the striking similarity in dental structure displayed by the two families may be explained by convergence. So far as the foot struc- ture of the Caenolestidae is known (p. 418), the pes shows no trace of syndactyly, while in all the phalangers it is syndactylous. A further dif- ference appears in the development of sectorial teeth, which, when present in the Caenolestidae, are confined to the posterior superior premolar and first lower molar, while in the phalangers the sectorial function may be transferred to the posterior inferior premolar in some of the higher forms ( Trichosums, Phalanger}. Until the upper dentition, skull and feet of the Caenolestidae, and especially of the primitive members of the family, are fully known, this must remain an unsettled question. At present the ar- guments in favor of the alternatives expressed are about equally balanced. The Caenolestidae lend no support to the latest classification proposed for them by Ameghino (1903, pp. 153-159) in which they are grouped as the suborder Paucituberculata, order Plagiaulacoidea, superorder Dipro- todonta, the latter including also the orders Hypsiprimnoidea and Ro- 444 PATAGONIAN EXPEDITIONS: PALEONTOLOGY. dentia. The characters which they possess are in no respect transitional to the Multituberculata. The Paucituberculata, like the Sparassodonta, are a group founded on a misconception of relationships and should be abandoned. BEARING OF THE SANTA CRUZ MARSUPIALS ON ZOOGEOGRAPHY. The reality of a former land connection between the Australian region and South America is plainly indicated by several lines of evidence based on the distribution of fishes, land shells, decapod crustaceans, plants, and Tertiary marine molluscs (Ortmann, 1902). This land connection is be- lieved to have existed not later than the close of the Cretaceous or be- ginning of the Tertiary, and it is only by such a connection that the distribution of the Thylacynidae can be explained. The direction, continuity or discontinuity of this land bridge need not enter into the present discussion. So far as the Thylacynidae are concerned, there can be little doubt of their South American origin, judging from the marked adaptive radiation which they attained during the Santa Cruz epoch, but whether the same can be said of marsupials in general is still a matter of question. It is believed, however, that the order may be properly regarded as of southern origin and that the occurrence of opossums in North America and Europe may be explained as the result of migration from the southern hemisphere. MARSUPIALIA INCERT^ SEDIS. A large number of Santa Cruz marsupials have been named by Ame- ghino and Mercerat, which are not represented in the collections at Princeton University and the American Museum of Natural History. Many of these have never been figured and are very imperfectly known. At present, the writer is not prepared to add to what has been published regarding them. The names and full references to the literature are here given. BORHY^NA Ameghino. BORHYENA FERA Ameghino. Dynamictis fera Amegh. ; Revista Argentina, pp. 148-149, fig. 53, 1891. Dinamyctis fera (sic] Amegh.; Ibid., p. 314. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 445 Borhycena fera Amegh.; Enum. Syn. des Especes de Mamm. Foss. des Formations Eocenes de Patagonie, pp. 117-119, figs. 45, 46, 1894; Bol. Acad. Cordoba, p. 373, figs. 45, 46, 1894. Borhycena fera Amegh.; Segundo Censo de la Republica Argentina, p. !89, %• 55. 6, c, 1898. BORHY/ENA SANGUINARIA AmeghinO. Borhycena sanguinaria Amegh. ; Enum. Syn., etc., p. 120, 1894; Bol. Acad. Cordoba, p. 376, 1894. ACROCYON Ameghino. ACROCYON SECTORIUS Ameghino. Acrocyon sectorius Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 8, 1887; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 289-290, PI. i, figs. 19, 19^, 19$, 1889; Enum. Syn., etc., p. 121, 1894; Bol. Acad. Cordoba, p. 377, 1894; Segundo Censo, etc., p. 189, 1898. Acrocyon sectorius Amegh., Mercerat ; Re vista del Museo de La Plata, II, p. 55, 1891. ACROCYON EQUIANUS Mercerat. Acrocyon equianus Merc.; Re vista del Museo de La Plata, II, p. 55, 1891. ACROCYON PATAGONENSIS Mercerat. Acrocyo n patagonensis Merc.; Revista del Museo de La Plata, II, p. 55, 1891. CONODONICTIS Ameghino. CONODONICTIS S^EVUS Ameghino. Conodonictis scevus Amegh.; Revista Argent, p. 314, 1891 ; Enum. Syn. etc., p. 121, 1894; Bol. Acad. Cordoba, p. 377, 1894; Segundo Censo, etc., p. 189, 1898. CONODONICTIS EXTERMINATOR Ameghino. Conodonictis exterminator Amegh.; Revista Argent., pp. 314-315, 1891 ; Enum. Syn., etc., p. 121, 1894; Segundo Censo, etc., p. 189, 1898. 446 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. ARCTODICTIS Mercerat. ARCTODICTIS MUNIZI Mercerat. Arctodictis muftizi Merc.; Revista del Museo de La Plata, II, pp. 51-52, 1891. ARCTODICTIS AUSTRALIS Mercerat. Arctodictis australis Merc.; Revista del Museo de La Plata, II, p. 52, 1891. PROTHYLACYNUS Ameghino. PROTHYLACYNUS BRACHYRHYNCHUS Ameghino. Prothylacynus brae hyrhync hits Amegh.; Enum. Syn., etc., p. 124, 1894; Bol. Acad. Cordoba, p. 380, 1894; Segundo Censo, etc., p. 189, 1898. NAPODONICTIS Ameghino. NAPODONICTIS THYLACYNOIDES Ameghino. Napodonictis thylacynoides Amegh.; Enum. Syn., pp. 125-126, 1894; Bol. Acad. Cordoba, p. 381,. 1894; Segundo Censo, etc., p. 189, 1898. AGUSTYLUS Ameghino. AGUSTYLUS CYNOIDES Ameghino. Agustylus cynoides Amegh. ; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, pp. 7-8, 1887; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 289-290, 1889; Revista Argent., p. 315, 1891 ; Enum. Syn., pp. 135-136, fig. 53, 1894; Bol. Acad. Cordoba, pp. 391-392, % 53. 1894; Segundo Censo, etc., p. 191, fig. 58.:, p. 193, 1898. Agustylus cynoides Amegh., Mercerat; Revista del Museo de La Plata, II, p. 54, 1891. AGUSTYLUS CARNIFEX Mercerat. Agustylus carnifex Merc.; Revista del Museo de La Plata, II, p. 54, 1891. AGUSTYLUS PRIM^EVUS Mercerat. Agustylus primcevus Merc.; Revista del Museo de La Plata, II, p. 54, 1891. AGUSTYLUS BARDUS Ameghino. Acyonf bardus Amegh.; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, p. 292, PI. i, figs. 18, i8#, 1889. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 447 Agustylus. bardiis Amegh.; Enum. Syn., etc., p. 136, 1894; Bol. Acad. Cordoba, p. 392, 1894; Segundo Censo, etc., p. 191, i! CLADOSICTIS Ameghino. CLADOSICTIS PATAGONICA Ameghino. Cladosictis patagonica Amegh.; Enum. Sist. Especies Mamif. F6s. Pata- gonia Austral, p. 7, 1887; Contrib. al Conocirniento Mamif. F6s. Rep. Argentina, p. 286, 1889; Enum. Syn., etc., p. 131, 1894; Bol. Acad. Cordoba, p. 387, 1894; Segundo Censo, etc., p. 190, 1898. CLADOSICTIS OXYRHYNCHUS Ameghino. Anatkeriumf oxyrhynchus Amegh.; Enum. Syn., etc., pp. 128-129, 1894; Segundo Censo, etc., p. 190, 1898. CLADOSICTIS DISSIMILIS Mercerat. Cladosictis dissimilis Merc.; Revista del Museo de La Plata, II, p. 51, 1891. HATHLIACYNUS (Amegh.) Mercerat. HATHLIACYNUS FISCHERI Mercerat. Hathliacynus fischeri Merc.; Revista del Museo de La Plata, II, p. 52, 1891. HATHLIACYNUS CULTRIDENS Mercerat. Hathliacynus cultridens Merc.; Revista del Museo de La Plata, II, p. 53, 1891. HATHLIACYNUS ROLLIERI Mercerat. Hathliacynus rollieri Merc.; Revista del Museo de La Plata, II, p. 53, 1891. HATHLIACYNUS LYNCHI Mercerat. Hathliacymis lynchi Merc.; Revista del Museo de La Plata, II, p. 53, 1891. HATHLIACYNUS KOBYI Mercerat. Hathliacynus kobyi Merc.; Revista del Museo de La Plata, II, pp. 53-54, 1891. 448 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. THYLACODICTIS Mercerat. THYLACODICTIS EXILIS Mercerat. Thylacodictis exilis Merc.; Revista del Museo de La Plata, II, pp. 54-55, 1891. AMPHIPROVIVERRA Ameghino. AMPHIPROVIVERRA ENSIDENS Ameghino. Protoproviverra ensidens Amegh.; Revista Argentina, p. 313, 1891. Amphiproviverra ensidens Amegh.; Enum. Syn., etc., p. 133, 1894; Segundo Censo, etc., p. 190, i! AMPHIPROVIVERRA OBUSTA Ameghino. Protoproviverra obusta Amegh. ; Revista Argentina, p. 313, 1891. Amphiproviverra obusta Amegh.; Enum. Syn., etc., p. 133, 1894; Se- gundo Censo, etc., p. 190, i! AMPHIPROVIVERRA CRASSA Ameghino. Amphipromverra crassa Amegh.; Enum. Syn., etc., p. 135, 1894; Bol. Acad. Cordoba, p. 391, 1894; Segundo Censo, etc., p. 190, 1898. PERATHEREUTES Ameghino. PERATHEREUTES PUNGENS Ameghino. Perathereutes pungens Amegh. ; Revista Argentina, p. 313, 1891. Perathereuthes pungens Amegh.; Enum. Syn., p. 136, fig. 54, p. 137, 1894; Bol. Acad. Cordoba, pp. 392-393, fig. 54, 1894; Segundo Censo, etc., p. 191, fig. 58^, p. 193, 1898. PERATHEREUTES OBTUSUS Ameghino. Perathereutes obtusus Amegh.; Revista Argentina, p. 313, 1891. Perathereutkes obtusus Amegh.; Enum. Syn., etc., pp. 136-137, 1894; Segundo Censo, etc., p. 191, 1898. PERATHEREUTES AMPUTANS Ameghino. Perathereutes amputans Amegh.; Revista Argentina, pp. 313-314, 1891. Perathereuthes amp^itans Amegh.; Enum. Syn., p. 137, 1894; Segundo Censo, etc., p. 191, 1898. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 449 SIPALOCYON Ameghino. SIPALOCYON GRACILIS Ameghino. Sipalocyon gmcilis Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, pp. 8-9, 1897; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 292-293, 1889; Revista Argentina, p. 315, 1891; Enum. Syn., etc., p. 137, fig. 55, p. 138, 1894; Bol. Acad. Cordoba, P- 393. %• 55. P- 349. l894! Segundo Censo, etc., p. 191, fig. 58*. p. 193, 1898. SIPALOCYON PUSILLUS Ameghino. Sipalocyon ptisillus Amegh.; Revista Argentina, p. 315, 1891 ; Enum. Syn., etc., p. 137, 1894; Bol. Acad. Cordoba, p. 393, 1894: Segundo Censo, etc., p. 191, 1898. SIPALOCYON CURTUS Ameghino. Sipalocyon curtus Amegh.; Enum. Syn., etc., p. 138, 1894; Bol. Acad. Cordoba, p. 394, 1894; Segundo Censo, etc., p. 191, li SIPALOCYON MIXTUS Ameghino. Sipalocyon mixtus Amegh.; Enum. Syn., etc., pp. 138—139, 1894; Bol. Acad. Cordoba, p. 394, 1894; Segundo Censo, etc., p. 191, il SIPALOCYON ALTIRAMIS Ameghino. Sipalocyon altiramis Amegh.; Enum. Syn., etc., p. 139, 1894; Bol. Acad. Cordoba, p. 395, 1894; Segundo Censo, etc., p. 191, 1898. SIPALOCYON LONGUS Ameghino. Sipalocyon longus Amegh.; Enum. Syn., etc., p. 139, 1894; Bol. Acad. Cordoba, p. 395, 1894; Segundo Censo, etc., p. 191, i! ACYON Ameghino. ACYON TRICUSPIDATUS Ameghino. Acyon triciispidatus Amegh.; Enum. Sist. Especies Mamif. F6s. Patago- nia Austral, p. 8, 1887; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 290-292, 1889; Enum. Syn., etc., p. 141, 1894; Bol. Acad. Cordoba, p. 397, 1894; Segundo Censo, etc., p. 191, li 450 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. Hathliacynus tricitspidatus (Amegh.) Mercerat; Revista del Museo de La Plata, II, pp. 52, 55, 1891. ICTIOBORUS Ameghino. ICTIOBORUS FENESTRATUS Ameghino. Ictioborus fenestmtus Amegh.; Revista Argentina, p. 315, 1891; Enum. Syn., etc., p. 140, fig. 56, 1894; Bol. Acad. Cordoba, p. 396, fig. 56, 1894; Segundo Censo, etc., p. 191, fig. 58 / p. 193, 1898. ICTIOBORUS DESTRUCTOR Ameghino. Ictioborus destructor Amegh. ; Enum. Syn., etc., pp. 140-141, 1894; Bol. Acad. Cordoba, p. 396, 1894. MICROBIOTHERIUM Ameghino. MICROBIOTHERIUM FORTicuLUM Ameghino. Microbiotherium forticulum Amegh.; Revista Argentina, pp. 309-310, 1891 ; Eaum. Syn., etc., p. 105, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 187, 1898. MICROBIOTHERIUM CONSPICUUM (Ameghino). Hadrorhynchus conspicitus Amegh.; Revista Argentina, p. 311, 1891; Enum. Syn., etc., p. 106, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 188, 1898. STYLOGNATHUS Ameghino. STYLOGNATHUS DIPROTODONTOIDES Ameghino. Stylognathus diprotodontoides Amegh.; Revista Argentina, p. 309, 1891 ; Enum. Syn., etc., p. 105, 1891 ; Segundo Censo, etc., p. 187, i! EODIDELPHYS Ameghino. EODIDELPHYS FORTis Ameghino. Eodidetyhys fortis Amegh.; Revista Argentina, p. 310, 1891 ; Enum. Syn. p. 105, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 187, 1898. SINCLAIR : MARSUPIALIA OF THE SANTA CRUZ -BEDS. 45 1 EODIDELPHYS FAMULA Ameghino. Eodidelphys famula Amegh. ; Revista Argentina, p. 310, 1891 ; Enum. Syn., p. 105, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 187, 1898. PRODIDELPHYS Ameghino. PRODIDELPHYS ACICULA Ameghino. Prodidelphys acicula Amegh.; Revista Argentina, p. 310, 1891 ; Enum. Syn., p. 105, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 188, 1898. PRODIDELPHYS PAVITA Ameghino. Prodidelphys pamta Amegh.; Revista Argentina, p. 310, 1891 ; Enum. Syn. p. 105, 1894; Bol. Acad. Cordoba, p. 361, 1894; Segundo Censo, etc., p. 188, 1898. PRODIDELPHYS OBTUSA Ameghino. Prodidelphys obtusa Amegh.; Revista Argentina, p. 311, 1891 ; Enum. Syn., p. 1 06, 1894; Bol. Acad. Cordoba, p. 362, 1894; Segundo Censo, etc., p. 188, 1898. ABDERITES Ameghino. ABDERITES MERIDIONALIS Ameghino. Abderites tneridionalis Amegh.; Enum. Sist. Especies Mamif. F6s. Pata- gonia Austral, p. 5, 1887; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 269-270, pi. I, figs. 6-Se, 1889; Bol. Inst. Geog. Arg., p. 150, figs. 1-3, 1890; Enum. Syn., etc., p. 83, fig. 31, p. 84, 1894; Bol. Acad. Cordoba, XIII, p. 337, fig. 31, p. 340, 1894; Segundo Censo, etc., p. 186, fig. 49, I, II, p. 184, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 3*7, t. II), p. 142, fig. 64, P- 155. fig- 78- P- !76. fig- I05. P- i?8. fig- I07- J903- ABDERITES SERRATUS Ameghino. Abderites terrains Amegh.; Revista Argentina, p. 248, 1891 ; Enum. Syn., etc., p. 84; Segundo Censo, etc., p. 186, li 452 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. ABDERITES TENUISSIMUS Ameghino. Abderites temiissimus Amegh.; Revista Argentina, p. 304, 1891 ; Enum. Syn., etc., p. 84, 1894; Bol. Acad. Cordoba, p. 337, 1894; Segundo Censo, etc., p. 186, 1898. ABDERITES ALTIRAMIS Ameghino. Abderites altiramis Amegh.; Enum. Syn., etc., p. 84, 1894; Bol. Acad. Cordoba, p. 340, 1894; Segundo Censo, etc., p. 186, 1898. MANNODON Ameghino. MANNODON TRISULCATUS Ameghino. Tideus trisulcatus Amegh.; Bol. Inst. Geog. Argentine, XI, cuad. VII- IX, pp. 157, 175, 187, July-Sept., 1890. Mannodon trisulcatus Amegh.; Enum. Syn., etc., p. 84, 1894; Bol. Acad. Cordoba, p. 340, 1894; Segundo Censo, etc., p. 185, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 3^7, II), p. no, fig. 28, 1903. DECASTIS Ameghino. DECASTIS RURIGERUS Ameghino. Decastis rurigerus Amegh.; Revista Argentina, p. 305, 1891 ; Enum. Syn., etc., p. 86, 1894; Bol. Acad. Cordoba, p. 342, 1894; Segundo Censo, etc., p. 186, 1898. ACDESTIS Ameghino. ACDESTIS OWENI Ameghino. Acdestis oweni Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 5, 1887; Contrib. al Conocimiento Mamif. F6s. Rep. Ar- gentina, pp. 270-271, pi. I, figs. 9-9^, 1889; Bol. Ins. Geog. Arg., p. 151, fig. 4, 1890; Enum. Syn., etc., p. 86, fig. 33, 1894; Bol. Acad. Cordoba, p. 342, fig. 33, 1894; Segundo Censo, etc., p. 186, fig. 50^, p. 185, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 30, II), p. 171, fig. 98, 1903. ACDESTIS ELATUS Ameghino. Acdestis elatus Amegh.; Revista Argentina, p. 304, 1891 ; Enum. Syn., p. 86, 1894; Bol. Acad. Cordoba, p. 342, 1894; Segundo Censo, etc., p. 1 86, 1898. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 453 ACDESTIS PARVUS Ameghino. Acdestis parvus Amegh. ; Revista Argentina, p. 305, 1891 ; Enum. Syn., p. 86, 1894; Bol. Acad. Cordoba, p. 342, 1894; Segundo Censo, etc., p. 1 86, 1898. DIPILUS Ameghino. DIPILUS SPEGAZZINII Ameghino. Dipilus spegazzinii Amegh.; Bol. Inst. Geogr. Argent., XI, p. 152, figs. 5, 6, 1890; Enum. Syn., p. 86, figs. 34, 35, p. 87, 1894; Bol. Acad. Cordoba, p. 342, figs. 34, 35, 1894; Segundo Censo, etc., p. 186, fig. 500, bt p. 185, 1898. Dipilus spegazzinianus Amegh.; Anales del Museo Nacional de Buenos Aires, IX (Ser. yi, II), p. 157, fig. 79, p. 172, fig. 99, 1903. DIPILUS BERGI Ameghino. Dipilus bergi Amegh.; Bol. Inst. Geogr. Argent., XI, p. 153, 1890; Enum. Syn., etc., p. 86, 1894; Bol. Acad. Cordoba, p. 342, 1894; Segundo Censo, etc., p. 186, 1898: METRIODROMUS Ameghino. METRIODROMUS ARENARUS Ameghino. Metriodromus arenarus Amegh.; Enum. Syn., p. 87, 1894; Bol. Acad. Cordoba, p. 343, 1894. METRIODROMUS SPECTANS Ameghino. Metriodromus spectans Amegh.; Enum. Syn., pp. 87-88, 1894; Bol. Acad. Cordoba, p. 343, 1894; Segundo Censo, etc., p. 186, 1898. METRIODROMUS CRASSUS Ameghino. Metriodromus crassus Amegh.; Segundo Censo, etc., p. 186, 1898. METRIODROMUS CRASSIDENS Ameghino. Metriodromus crassidens Amegh.; Segundo Censo, etc., p. 186, 1898. HALMADROMUS Ameghino. HALMADROMUS VAGUS Ameghino. Halmadromtis vagus Amegh.; Revista Argentina, p. 306, 1891 ; Enum. Syn., p. 88, 1894 ; Bol. Acad. Cordoba, p. 344, 1894; Segundo Censo, etc., p. 1 86, 1898. 454 PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. CALLOMENUS Ameghino. CALLOMENUS INTERVALATUS Ameghino. Callomenus intervalatus Amegh.; Revista Argentina, p. 306, 1891 ; Enum. Syn., p. 88, 1894; Bol. Acad. Cordoba, p. 344, 1894 ; Segundo Censo, etc., p. 1 86, 1898. CALLOMENUS ROBUSTUS Ameghino. Callomenus robustus Amegh.; Enum. Syn., etc., p. 88, 1894; Bol. Acad. Cordoba, p. 344, 1894; Segundo Censo, etc., p. 186, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 3*7, II), p. 1 16, fig. 34, p. 120, fig. 38, 1903. PALJEOTHENTES (Moreno) Ameghino. PAL^EOTHENTES LEMOINEI Ameghino. Palceothentes lemoinei Amegh.; Enum. Sist. Especies Mamif. F6s. Pata- gonia Austral, p. 6, 1887. Epanorthus lemoinei Amegh.; Contrib. al Conocimiento Mamif: F6s. Rep. Argentina, p. 273, pi. I, figs. 13-14^, 1889; Enum. Syn., etc., p. 91, figs- 36-38> J894; Bol. Acad. Cordoba, p. 346, 347, figs. 36-38, 1894; Segundo Censo, etc., p. 186, fig. 50, e, f,g, p. 185, i! PAL^OTHENTES AMBIGUUS (Ameghino). Epanorthus ambigm-ts Amegh.; Revista Argentina, p. 305, 1891 ; Enum. Syn., etc., p. 91, 1894; Bol. Acad. Cordoba, p. 347, 1894; Segundo Censo, etc., p. 186, i! PAL/EOTHENTES PACHYGNATHUS Ameghino. Palceothentes pachygnathus Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 6, 1887. Epanorthus pachygnatkus Amegh.; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 273-274, 1889; Enum. Syn., etc., p. 91, 1894; Bol. Acad. Cordoba, p. 347, 1894 ; Segundo Censo, etc., p. 186, 1898. PAL^EOTHENTES PRESSIFORATUS Ameghino. Palceothentes pressiforatiis Amegh.; Enum. Sist. Especies Mamif. F6s. Patagonia Austral, p. 6, 1887. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 455 Epanorthiis pressiforatus Amegh. ; Contrib. al Conocimiento Mamif. F6s. Rep. Argentina, pp. 274-275, 1889; Enum. Syn., etc., p. 91, 1894; Bol. Acad. Cordoba, p. 347, 1894, Segundo Censo, etc., p. 186, 1898. PAL/EOTHENTES IN^EQUALIS (Ameghino). Epanorthtts incequalis Amegh.; Revista Argent., p. 305, 1891 ; Enum. Syn., etc., p. 92, 1894; Segundo Censo, etc., p. 186, i! PAL/EOTHENTES COMPLICATUS (Ameghino). Metaepanorthus complicattis Amegh.; Enum. Syn., etc., pp. 92-93, 1894; Bol. Acad. Cordoba, p. 348, 1894; Segundo Censo, etc., p. 186, 1898. PAL^EOTHENTES HOLMBERGI (Ameghino). Epanorthiis holmbergi Amegh.; Los Plagiaulacideos Argentines, etc., Bol. Inst. Geog. Argentine, XI, reprint, p. 17, fig. 8, 1890. Metaepanorthus holmbergi Amegh. ; Enum. Syn., etc., p. 93, fig. 39, 1894; Bol. Acad. Cordoba, p. 349, fig. 39, 1894; Segundo Censo, etc., p. 1 86, 1898; Anales del Museo Nacional de Buenos Aires, IX (Sen 3^, II), p. 172, fig. 100, 1903. PREPANORTHUS Ameghino. PREPANORTHUS LANIUS Ameghino. Prepanorthus lanius Amegh.; Enum. Syn., etc., p. 95, 1894; Bol. Acad. Cordoba, p. 351, 1894; Segundo Censo, etc., p. 186, i! HALMASELUS Ameghino. HALMASELUS VALENS Ameghino. Halmaselus valens Amegh.; Revista Argent., p. 306, 1891 ; Enum. Syn., etc., p. 95, 1894; Bol. Acad. Cordoba, p. 351, 1894; Segundo Censo, etc., p. 1 86, 1898. ESSOPRION Ameghino. ESSOPRION CORUSCUS Ameghino. Essoprion coruscus Amegh.; Revista Argent., p. 306, 1891 ; Enum. Syn., etc., p. 95, 1894; Bol. Acad. Cordoba, p. 351, 1894; Segundo Censo, etc., p. 1 86, 1898. 456 PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. ESSOPRION CONSUMPTUS Ameghino. Essoprion consumptus Amegh.; Revista Argent., pp. 306-307, 1891 ; Enum. Syn., p. 95, 1894; Segundo Censo, etc., p. 186, 1898. PICHIPILUS Ameghino. PICHIPILUS OSBORNI Ameghino. Pichipilus osborni Amegh. ; Bol. Inst. Geog. Argent, p. 153, 1890; Enum. Syn., etc., p. 95, 1894; Bol. Acad. Cordoba, p. 351, 1894; Segundo Censo, etc., p. 186, 1898. PICHIPILUS EXILIS Ameghino. Pichipilus exilis Amegh.; Revista Argentina, p. 307, 1891 ; Enum. Syn. etc., p. 95, 1894; Segundo Censo, etc., p. 186, 1898. GARZONIA Ameghino. GARZONIA TYPICA Ameghino. Garzonia typica Amegh.; Nuevos Restos Mamif. F6s. Patagonia Austral, p. 21, Aug., 1891 ; Revista Argentina, I, entr. 5*7, p. 307, 1891 ; Enum. Syn., etc., pp. 98-99, fig. 41, 1894; Segundo Censo, etc., p. 1 86, 1898. Garzonia tipica Amegh.; Anales del Museo Nacional de Buenos Aires, IX (Ser. 3«, II), fig. 91, p. 167, 1903. (Corrected figure, compare Enum. Syn., etc., 1894, fig. 41.) Garzonia typica Amegh.; Anales, etc., fig. 93, p. 168, 1903. GARZONIA CAPTIVA Ameghino. Garzonia capti'va Amegh.; Revista Argentina, p. 308, 1891 ; Enum. Syn., etc., p. 99, 1894; Bol. Acad. Cordoba, p. 355, 1894; Segundo Censo, etc., p. 1 86, 1898. GARZONIA MINIMA Ameghino. Garzonia minima Amegh.; Revista Argentina, p. 308, 1891 ; Enum. Syn., etc., p. 99, 1894; Bol. Acad. Cordoba, p. 355, 1894; Segundo Censo, etc., p. 196, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 30, II), p. 157, fig. 81, p. 186, fig. 121, 1903. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 457 GARZONIA GRACILIS (Ameghino). Phonocdromus gracilis Amegh.; Enum. Syn., etc., p. 100, 1894; Bol. Acad. Cordoba, p. 356, 1894; Segundo Censo, etc., p. 186, i! PARHALMARHIPHUS Ameghino. PARHALMARHIPHUS ANNECTENS Ameghino. Garzonia annectens Amegh.; Revista Argentina, p. 307, 1891. Parlwlmarhiphus annectens Amegh.; Enum. Syn., pp. 100-101, 1894; Bol. Acad. Cordoba, p. 357, 1894; Segundo Censo, etc., pp. 186- 187, 1898. HALMARHIPHUS Ameghino. HALMARHIPHUS DIDELPHOIDES Ameghino. Halmarhiphus didelphoides Amegh.; Revista Argentina, p. 308, 1891 ; Enum. Syn., p. 101, 1894; Bol. Acad. Cordoba, p. 357, 1894; Se- gundo Censo, p. 187, 1898. Hahnariphus didelphoides Amegh.; Anales del Museo Nacional de Buenos Aires, IX (Ser. 30, II), p. 157, -fig. 80, 1903. Parhalmarhiphus didelphoides Amegh.; Ibid., p. 163, fig. 86, 1903. STILOTHERIUM Ameghino. STILOTHERIUM GRANDE Ameghino. Stilotherium grande Amegh. ; Enum. Syn., p. 102, 1894; Bol. Acad. Cor- doba, p. 358, 1894; Segundo Censo, etc., p. 187, i! STILOTHERIUM DISSIMILE Ameghino. Stilotherium dissimile Amegh.; Enum. Sist. Especies Mamif. F6s. Pata- gonia Austral, p. 7, 1887; Enum. Syn., etc., p. 102, 1894; Segundo Censo, etc., p. 187, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 3^, II), fig. 33, p. 115, fig. 92, p. 167, 1903. CLADOCLINUS Ameghino. CLADOCLINUS COPEI Ameghino. Cladoclimts copei Amegh.; Enum. Syn., p. 103, 1894; Bol. Acad. Cor- doba, p. 359, 1894; Segundo Censo, etc., p. 187, 1898; Anales del Museo Nacional de Buenos Aires, IX (Ser. 30, II), p. 117, fig. 35, 1903. 458 PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. BIBLIOGRAPHY. Ameghino, Florentine. 1887 Enumcracion sistematica de las especies de mamiferos fosiles coleccionados por Carlos Ameghino en los terrenes eocenos de la Patagonia austral. La Plata, 1887. 1889 Contribucion al conocimiento de los mamiferos fosiles de la Republica Argentina (A etas de la Academia Nacional de Ciencias en Cordoba, T. V). 1890 Los Plagiaulacideos Argentines y sus relaciones zoologicas, geologicas y geograficas, pp. I -6 1. Buenos Aires, 1890. Ibid. Boletin del Institute Geografico Argentine, T. XI, pp. 143-201, 1890. 1891 Nuevos restos de mamiferos fosiles descubiertos por Carlos Ameghino en el eoceno inferior de la Patagonia austral — Especies nuevas, adiciones y correcciones, pp. 1-44, Aug., 1891. Ibid. Revista Argentina de Historia Natural, T. I, fas. IV, pp! 289-328, Oct., 1891. 1893 Les premiers mammiferes. — Relations entre les mammiferes diprotodontes eocenes de 1'Amerique du Nord et ceux de la Republique Argentine. Revue General des Sciences pures et appliquees, 4' annee, no. 3, p. 77. 1894 Enumeration synoptique des especes de mammiferes fossiles des formations eocenes de Patagonie (Boletin de la Academia Nacional de Ciencias en Cordoba, T. XIII, p. 259). In the citations from this paper given in the text the paging is that of the separate edition. 1897 Mammiferes cretaces de 1'Argentine. Deuxieme contribution a la connaissance de la faune mammalogique des couches a Pyrotherium. Boletin del Instituto Geografico Argentine, T. XVIII, cuad. 4—9. Separate edition, pp. 1-117. 1898 Sinopsis geologico-paleontologica (Segundo Censo de la Republica Argentina, T. I. Buenos Aires, 1898, p. 113). 1903 Los diprotodontes del orden de los Plagiaulacideos y el origen de los Roedores y de los Polimastodontes. Anales del Museo Nacional de Buenos Aires, T. IX (ser. 3", T. II), pp. 81-192. Bensley, B. Arthur. 1903 On the evolution of the Australian Marsupialia ; with remarks on the relationships of the marsupials in general. Transactions of the Linnxan Society, ser. 2, Zool., Vol. IX, pp. 83-217, Pis. 5-7. Bonaparte, C. L. 1838 Syn. Vert. Syst., in Nuovi Ann. Sci. Nat, Bologna, II, p. 112, 1838 (see p. 8) ; Revue Zool., I, p. 217, Sept., 1838. Cunningham, D. J. 1882 Report on some points in the anatomy of the Thylacyne (Tliylacynus cynocephalus), Cuscus (JPhalangista maculatci) and Phascogale (Phascogale calusa), collected during the voyage of H. M. S. Challenger, 1873-1876. Report of the scientific results of the exploring voyage of H. M. S. Challenger, Zoology, Vol. V, 1882. Dollo, L. 1899 Les ancetres des Marsupiaux, etaient-ils arboricoles ? Miscellanees Biologiques, pp. 188-203. Paris. SINCLAIR: MARSUPIALIA OF THE SANTA CRUZ BEDS. 459 Mercerat, Alcides. 1891 Caracteres diagnosticos dc algunas especies de Creodonta conscrvadas en el Museo de La Plata. Revista del Museo de La Plata, T. II, pp. 51-56. La Plata. Moreno, Francisco P. 1882 Patagonia, resto de un antiguo continente hoy submergido (Conferencias de la Sociedad Cientifica Argentina, Conferencia del 15 del Junio de 1882). Buenos Aires. Ortmann, A. E. 1902 Tertiary Invertebrates. Reports of the Princeton University Expeditions to Patagonia, 1896-1899. Vol. IV, Pt. II. Sinclair, W. J. 1905 The marsupial fauna of the Santa Cruz beds. Proceedings of the American Philosophi- cal Society, Vol. XLIX.pp. 73-81, Pis. I and II. Thomas, Oldfield. 1888 Catalogue of the Marsupialia and Monotremata in the collection of the British Museum (Natural History). London. 1895 On C&nolestes, a still-existing survivor of the Epanorthidae of Ameghino, and the rep- resentative of a new family of recent marsupials. Proceedings of the Zoological Society of London, pp. 870-878. 1895. Trouessart, E. L. 1898 Catalogus Mammalium. ERRATA. Pp. 3-43. Throughout Part I, for Lake Pueyrrydon, Puerrydon Series, etc., read Lake Pueyrredon, Pueyrredon Series, etc. P. iov 1. 2. For relatep, read related. P. 54, 1. 17. For Oven Poin, read Oven Point. P. 79, 1. 14. For T. patagonia, read T. patagonica. P. 84, 1. 19. For escutheon, read escutcheon. P. 97, 11. 21, 23, 27. For P. ibari, read G. ibari. P. 113, 1. 20. For 0. vesicularis, read G. vesicularis. P. 134, 1. 33. For Panopaa, read Panopea. P. 135, heading of species 61. For Venus dijficillis, read V. difficilis. P. 153, 1. 10. For especially, read especially. P. 157, heading of species 88. For Dentalum, read Dentalium. P. 162, " " " 93. For Leptothyra philippi, read L. philippii. Omit of (last word of line). For var. inoratus, read inornatus. »te. For Volulolithes, read Volutilithes. For collosity, read callosity. For R. varians, read B. varians. For R. unguiformis, read B. unguiformis. For Callistoma philippi, read Calliostoma philippii. For Fusus spiralis, read F. subspiralis. For D. octocostellum, read D. octocostellatum. For enota Gcuevensis, read Genota cuevensis. For Pinna semicostalis, read P. setnicostata. For Calliostoma pararatum, read C. peraratum. After No. 26, insert Modiola ameghinoi and change the subsequent numbers 27-117 to 28-118. P. 270, 1. 8. For 117, read 118. P. 280, 1. 23. For Glycimeris ibaria, read G. ibari. P. 284, Footnote. For Tcrrcbratella, read Terebratella. For Siphonalis read Siphonalia. For Crassatella kokeni, read Crassatellites kokeni. For Calliostoma pararatum, read C". peraratum. For P! sarissa, read Fi serissa. For Panopea ibaria, read .P. z$«>7. For This, ma*/ The. For deposits, m*l Sanla Cm/ Kivcr. I fll V.llvi- Iii-I.ill;.;ill;; Infix '..1 me- |ix li\ it lll.ll .r, I ll.it 1 1;; nr< •< I (.n pl.il.- XXI. i/'. Ki;;lii valvr <>l .1 yoiiii;; imliviiliial. i. I .I'll \ ,ilvr of thr same. Ml ihr I (YOU iv) PATAGONIAN EXI'KMTIONS PLATE xxii. i' F v Iterion del •: MI. TERTIARY INVERTEBRATES. PATAGONIAN EXPEDITIONS : PALEONTOLOGY. EXPLANATION OF PLATE XXIII. PAGE Fig. i . PECTEN cf. CENTRALIS Sow. . . . . . . . . 1 1 6 Patagonian beds ; Port Desire. i a. Right valve. i b. Left valve of the same individual. Note. Both valves in fragments; in fig. \b the fragments have not been imbedded quite correctly in the plaster. Fig. 2. PECTEN GEMINATUS Sow. . . . . . . . . 117 Patagonian beds. 2a. Small right valve, agreeing with Sowerby's type ; Oven Point, San Julian. zb. Small left valve of another individual ; Oven Point, San Julian. zc. Larger right valve ; Oven Point, San Julian. zd. Right valve of var. quemadensis v. Ih. ; Darwin Station, San Julian. ze. Lateral view of both valves, the same individual as fig. zd. All the figures are natural size. (VOL. iv) PATAGONIAN EXPEDITIONS VOL.IV. PLATE xxm. F v. Iterson del. Werner i Winter, FrankfortSM . lirh TERTIARY INVERTEBRATES. PATAGONIAN EXPEDITIONS I PALEONTOLOGY. EXPLANATION OF PLATE XXIV. PAGE Fig. i . PECTEN ACTINODES Sow. . . . . . . ' . . . 119 Cape Fairweather beds ; Cape Fairweather. la. Right valve. ib. Left valve of another individual. Fig. 2. PERN A QUADRISULCATA v. Ih. . . . . . . . . 97 Patagonian beds. 2a. Hinge of right valve. Mouth of Santa Cruz River. 2b. Cast of an almost complete specimen. Lake Pueyrredon, base. Fig. 3. MYTILUS MAGELLANICUS Chemn. . . . . . . . "121 Patagonian beds ; Oven Point, San Julian. Inner and outer casts, imbedded in matrix. Fig. 4. MODIOLA ANDINA Ortm. . . . . . . . . 122 Patagonian beds ; Lake Pueyrredon, 400' above base. Cast. All the figures are natural size. (VOL. iv) PATAGONIAN EXPEDITIONS VOL.IV. PLATE xxiv. F. v. kerson del. Werner i Winter. TERTIARY INVERTEBRATES. PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. EXPLANATION OF PLATE XXV. PAGE Fig. i. MYTILUS cf. CHORUS Mol. . . . . . . . . 120 la. Fragment of cast from Patagonian beds of Upper Rio Chalia. ib. Not quite complete cast from Cape Fairweather beds, Cape Fairweather. Fig. 2. MODIOLA AMEGHINQI V. Ih. . . . . . . . . 12 1 Patagonian beds ; Mount of Observation, upper horizon. Left valve. Fig. 3. ARCA PATAGONICA v. Ih. • • • • 93 Patagonian beds ; mouth of Santa Cruz River. 30. Outer view of right valve of a specimen of small size. 3. The same, 2/1. Fig. 7. PATELLA PYGM^A Ortm. . . . . . . . . 161 Upper Magellanian beds ; Punta Arenas. 7#. Side view, natural size. 7artim Werner & Winter. Frankfort °M , Irth PROTHYLACYNUS PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. EXPLANATION OF PLATE LII. PAGE. Fig. i . BORHY^ENA TUBERATA : Cervical vertebrae from the left side (No. 15,701). The fourth cervical is missing . 351 Fig. 2. " EXCAVATA.: First to fifth cervical vertebrae from the left side (No. 15,120) ..... 351 Fig. 3. PROTHYLACYNUS PATAGONICUS : First to fifth cervicals from the left side (No. 15,700). . . . 365 Fig. 4. CI.ADOSICTIS LUSTRATUS : Cervical series from the right side (No. Fig. 5. PROTHYLACYNUS PATAGONICUS: Axis from below (No. 15,700) . 365 Fig. 6. BORHY^ENA EXCAVATA: Axis from below (No. 15,120) . . . 350 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LII. Werner s Winter, Frankfort ' BORHYTENA, PROTHYLACYNUS, CLADOSICTIS PATAGONIAN EXPEDITIONS! PAL/EONTOLOGY. EXPLANATION OF PLATE LIII. PAGE. Fig. i. AMPHIPROVIVERRA MANZANIANA : Atlas from above (No. 15,154) . 398 Fig. i a. " " Atlas from below, showing the in- tercentrum (No. 15,154). Fig. 2. BORYH/ENA TUBERATA : Atlas from above, showing the notches for the passage of the vertebral artery and spinal nerves (No. 15,701). 350 Fig. 2a. " " Atlas from below, showing the rugose sur- faces for articulation with the intercen- trum (No. 15,701). Fig. 3. CLADOSICTIS PETERSONI, type : Atlas from above (No. 15,702) . 380 Fig. 3«. " " " Atlas from below, showing the fusion of the intercentrum with the neu- ral arch (No. 15,702). Fig. 4. BORHY^NA EXCAVATA : Atlas from above (No. 15,120). . . 350 Fig. 4«. Atlas from below (No. 15,120). Fig. ^b. " Atlanteal intercentrum from below (No. 15,120). Fig. 5. PROTHYLACYNUS PATAGONICUS : Atlas from above (No. 15,700) . 365 Fig. 50. " Atlas from below (No. 15,700) showing the fusion of the inter- centrum with the neural arch. Fig. 6. Eighth ? caudal vertebra from above (No. 15,700) . . . 366 F'g- 7- " Tenth? dorsal vertebra from the right side (No. 15,700) . . 366 Fig. 8. Third caudal from above (No. 15,700) . . . . . 366 Fig. 9. BORHY^NA TUBERATA : Fourth ? caudal from above (No. 15,701). 351 Fig. ga. " " Fourth ? caudal from the right side (No. 15,701). Fig. 10. CLADOSICTIS LUSTRATUS : Lumbar series from the left side (No. i5-'7o) 381 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LIII. i F v Iferson del. parnm Werner & Winter, Frankfort °M , Itth. SANTA CRUZ THYLACYNES PATAGONIAN EXPEDITIONS I PALEONTOLOGY. EXPLANATION OF PLATE LIV. PAGE. Fig. i. CLADOSICTIS PETERSONI, type : Skull from the left side (No. 15,702). 391 Fig. 2. PROTIIYLACYNUS PATAGONICUS : Left astragalus, inner side (No. i5»7oo) 370 Fig. 2a. " " Left astragalus, plantar aspect (No. 15,700). Fig. 3. CLADOSICTIS LUSTRATUS : Right pes, dorsum (No. 15,046), preserving arrangement of elements as in matrix . 385 Fig. 4. " " Right manus, dorsum (No. 15,046), show- ing the opposable thumb. Elements arranged as in matrix .... 383 Fig. 5. AMPHIPROVIVERRA MANZANIANA : Right manus, dorsum (No. 15,154). Arrangement of elements as in matrix, with the exception of the ungual of the second digit . 398 Fig. 6. " Left pes, dorsum (No. 15,154), show- ing the opposable hallux . 399 Fig. 7. BORHY^ENA TUBERATA: Left manus, dorsum (No. 15,701). The scaphoid and trapezium are supplied from the opposite side. Association of phalanges conjectural. ..... 353 Fig. 8. PROTHYLACYNUS PATAGONICUS: Left pes, dorsum (No. 15,700), showing the reduced hallux . 370 Fig. 9. " " Phalanges of the fore foot (No. 15,700). Association conjectural. 368 Fig. 10. CLADOSICTIS PETERSONI, type : Left astragalus, dorsal aspect (No. '5-702) . 385 Fig. ii. AMPHIPROVIVERRA MANZANIANA: Claw of right pollex, inner side ^(No. 15,154) . . .... 399 Fig. 12. CLADOSICTIS LUSTRATUS: Claw of right pollex, inner side (No. 15,046) . . 383 Fig. 1 3. BORHY/ENA TUBERATA : Claw, presumably of the fourth digit of the left manus, outer side (No. 15,701) ...... 355. Fig. 14. PROTHYLACYNUS PATAGONICUS: Claw of fore foot, side view (No. 15.700) . 368 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LIV. 6 8 13 F v llorson del. parMm Werner 4 Winter. Frankfort ° W, lith SANTA CRUZ THYLACYNES. PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. EXPLANATION OF PLATE LV. PAGE. Fig-, i. CLADOSICTIS LUSTRATUS : Skull, palatal view (No. 15,170), showing the auditory region . . . . 378 Fig. 2. PETERSONI, type : Right humerus from in front (No. 15.702) 382 Fig. 2a. " " Right humerus from behind (No. 15.702). Fig. 3. " Skull from below (No. 15,702) . 391 Fig. 3«. " " " above (No. 15,702). All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LV F v Iterson del parrim Werner* Winter. Frankfort °M . liln CLADOSICTIS. PATAGONIAN EXPEDITIONS : PALAEONTOLOGY. EXPLANATION OF PLATE LVI. PAGE. Fig. i. CLADOSICTIS LUSTRATUS : Skull and mandible, left side (No. 15,046). The occipital condyles have been re- stored from another specimen . . 378 Fig. 2. " " Skull from above (No. 15,046). Fig. 3. " " " " below " All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LVI. F v Iterson del Werner » Winter, Fra n kfori ° M , Irth. GLADOSICTIS PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. EXPLANATION OF PLATE LVII. Fig. i. CLADOSICTIS LUSTRATUS : Pelvis and sacrum from above (No. 384 Fig. la. " Pelvis and sacrum from below (No. Fig. 2. PETERSONI, type : Right scapula (No. 15,702) . . 382 Fig. 2a. " " Glenoid cavity of right scapula (No. 15.702). Fig. 3. LUSTRATUS: Left scapula (No. 15,046). The posterior deflection of the spine is due to crush- ing . . 382 Fig. 4. PETERSONI, type: Left femur from behind (No. 15,702). 384 Fig. 4«. . " " " " " in front (No. 15,702). Fig. 5. LUSTRATUS: Third? to tenth? caudals from above (No. 15,170) ..... 382 Fig. 6. " Lumbar vertebra from above (No. 15,170). 381 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV PLATE LVII. F. v. Iterson del. parhm Werneri Winter. Frankfort °/M,lrth. GLADOSICTIS PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. EXPLANATION OF PLATE LVIII. PAGE. Fig. i. CI.ADOSICTIS LUSTRATUS : Pelvis and sacrum, right side (No. 15, 1 70). 384 Fig. 2. " " Right tibia from in front (No. 15,046) . 384 Fig. 3. " " Fifth? rib, outer side (No. 15,170) . . 382 Fig. 4. " " Right radius and ulna from the outer side (No. 15,046) 383 Fig. 40. " " Right radius and ulna from the inner side (No. 15,046). Fig. 5. " PETERSONI, type : Right radius and ulna from the outer side (No. 15,702) . . 383 Fig. 6. " LUSTRATUS: Left femur from in front (No. 15,170) . 384 Fig. 7 " PETERSONI, type : Pelvis, sacrum and sixth lumbar from below (No. 15,702). The second sacral is missing .... 384 Fig. "ja. " " " Pelvis, sacrum and sixth lumbar from above (No. 15,702), slightly restored. Fig. 8. " " " Left tibia from in front (No. 15,702). 384 Fig. 9. " " " Left fibula, inner side " " 385 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL iv. PLATE LVIII F v Uerson de! partirn Wen • CLADOSICTIS. PATAGONIAN EXPEDITIONS I PAL/EONTOLOGY. EXPLANATION OF PLATE LIX. PAGE. Fig. i. AMPHIPROVIVERRA MANZANIANA : Skull from below (No. 15,154) . 394 Fig. la. Anterior portion of face from the right side, showing cicatrice of wound (No. 15,154) . . . 400 Fig. ib. Skull from the left side (No. 15, 1 54) 394 Fig. 2. Right tympanic bone, outer side, X f (No. 15,154) • 397 Fig. 3. MINUTA: Skull from above (No. 15,373) • • 4°5 Fig. 3«. " " " " below " " Fig. 4. MANZANIANA : Anterior portion of palate show- ing the procumbent median in- cisor (No. 15,029) . . . 394 Fig. 5. Third cervical vertebra, right side (No. 15,154) .... 398 Fig. 6. CLADOSICTIS sp.: Left half of mandible showing the deciduous tooth and the germ of the posterior premolar below it (No. 15,079). The region occupied by the an- terior premolar has been destroyed by fracture, and in repairing this break, the canine has been too closely approximated to the molars . . 378 Fig. 7. CLADOSICTIS LUSTRATUS : Left half of the mandible from the outer side (No. 15,170) . . . . 376 Fig. -ja. " Left half of the mandible from the inner side (No. 15,170). Fig. "]b. " Left half of the mandible, crown view (No. 1 5.170). All the figures, except Fig. 2, are natural size. (voi,. iv.) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LIX. F. v llerr.on del partim Werner i Winter, Frankfort °M . lith CLADOSICTIS, AMPHIPROVIVERPJA. \ PATAGONIAN EXPEDITIONS! PALAEONTOLOGY. EXPLANATION OF PLATE LX. PAGE- Fig. i. AMPHIPROVIVERRA MANZANIANA : Skull from the left side (No. 9254 American Museum of Natural History) 396 Fig. \a. " " Skull from below (No. 9254 Am- erican Museum of Natural His- tory). The right jugal arch has been restored in plaster. Fig. \b. " " Skull from above (No. 9254 Am- erican Museum of Natural His- tory). The right jugal arch and the right margin of the occiput have been restored in plaster. Fig. ic. " " Skull from behind (No. 9254 Am- erican Museum of Natural His- tory). The extent o f the plaster restoration is indicated by the neutral tint. Fig. 2. " " Mandible, crown view (No. 1 5. 029) • 397 Fig. za. " " Mandible from the left side (No. 15,029). Fig. 3. " MINUTA : Skull from the right side (No. 15,373) 405 Fig. 30. " " Mandible, crown view Fig. 4. " MANZANIANA: Left humerus from in front (No. 15,154) • 398 All the figures are natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV PLATE LX. 2a 3a F v. llerson del parfim Werner A Winter. Frankfort °M., hth. AMPHIPROVIVERRA. PATAGONIAN EXPEDITIONS : PAL/F.ONTOLOGY. EXPLANATION OF PLATE LXI. PAGE. Fig. i . CLADOSICTIS LUSTRATUS : Restoration of the skeleton based on two specimens in the Princeton collection. The head, neck, scapula, radius, ulna, pelvis, femur, tibia, fibula, ribs, lumbar and caudal vertebrae are from No. 15,170. The dorsal vertebrae and feet are enlarged to scale from No. 1 5,046. The humerus is restored from Cladosictis petersoni (No. 15,702) . . . . Fig. 2. PROTHYLACYNUS PATAGONICUS : Restoration of the skeleton from re- mains of a single individual (No. 15,700). The facial portion of the skull, the sixth and seventh cervicals and the fore foot are supplied from Borhyczna tuberata. The calcaneum is restored after Cladosictis . . . . . . . . . . 37 J Fig. 3. BORHY^ENA TUBERATA: Restoration of the skeleton (No. 15,701). The humerus, posterior lumbars, pelvis, patella and hind foot are sup- plied from Frothy 'lacy mus. The fabella, tibia and fibula are from Thylacynus. The fourth cervical has been enlarged to scale after Borhytena excavata (No. 15,120). ...... 355 All the figures are about one fourth the natural size. Conjectural parts are un- shaded ; those supplied from related forms are indicated by oblique hachure. (VOL. iv.) PATAGONIAN EXPEDITIONS: PAL/EONTOLOGY. EXPLANATION OF PLATE LXII. PAGE. Fig. i. MICROBIOTHERIUM TORTOR : Palatal view of a crushed skull (No. 15,698) . 408 Fig. 2. " " Left ramus from the outer side (No. 15,698) . 409 Fig. za. " " Left ramus, crown view (No. 15,698). Fig. 3. " GALLEGOSENSE, type : Right ramus from the outer side (No. 9591 American Museum of Natural His- tory) 413 Fig. 3«. " Right ramus, crown view (No. 9591 American Museum of Natural His- tory). Fig. 4. " TEHUE^CHUM : Left ramus, outer side (No. 15,038). 414 Fig. 4«. " " " crown view " " Fig. 5. " PATAGONICUM: Anterior portion of the right ramus from the outer side (No. 9121 American Museum of Natural History) . . . . . 415 Fig. 5«. Anterior portion of the right ramus, crown view (No. 9121 American Museum of Natural History). Fig. 6. TEHUELCHUM : Second and third superior molars of the left side, crown view (No. 15,038) . . 414 Fig. 7. " " Left auditory bulla from below (No. 15,038). a, alisphenoid ; g, glen- oid cavity ; /, petrous ; pg, post- glenoid process . . . . 410 Fig. 8. Third, fourth and fifth cervicals from the left side (No. 15,038) . . 412 Fig. 9. " " Atlas and axis from below (No. 1 5,038). i, atlanteal intercentrum ; n, neural arch of atlas ; c, centrum of axis . . . . . 411 Fig. 10. " " Right ulna, proximal end (No. 15.038) . 411 Fig. ii. " Incomplete right scapula (No. 15,038) X 2.7 . . . 411 PATAGONIAN EXPEDITIONS VOL.IV. PLATE LXII. 4 >* V v. m 8 s 9 V 10 3a 11 12 F v Iterson dei. partim 'Werner 4 Winter. Fra1 MlCROBIOTHERIUM. PATAGONIAN EXPEDITIONS! PAL/EONTOLOGY. PACE. Fig. 12. MICROBIOTHERIUM TEHUELCHUM : Right humerus, anterior aspect (No. 15,038). The part restored in plaster is indicated by the oblique hachure . . . 411 All the figures, except Fig. 1 1 , are three times the natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS I PALAEONTOLOGY. EXPLANATION OF PLATE LXIII. PAGE. Fig. i. PAL^EOTHENTES MiNUTUS : Left maxilla, outer side (No. 1 5,709) . 425 Fig. 2. " ARAT^E : Right maxilla, outer side (No. 9549 American Museum of Natural History). 425 Fig. 2a. " " Right maxilla, palatal view (No. 9549 American Museum of Natural History). Fig. 3. " INTERMEDIUS: Skull, left side (No. 15,225) . . 427 Fig. 4. " MINUTUS : Left half of mandible, outer side (No. i5,7°8) • 426 Fig. 4«. Left half of mandible, crown view (No. ^ 15, 708). Fig. 4$. Right half of mandible, anterior extremity from the outer side, showing the pro- cumbent vestigial teeth (No. 15,708). Fig. 5. " " Left half of mandible, outer side (No. 15,706). A more robust individual . 426 Fig. 5«. " Left half of mandible, crown view (No. 15.706). Fig. 6. " LEPIDUS : Right half of mandible, outer side (No. 9597 American Museum of Natural History) . . . . . . 431 Fig. 6a. " " Right half of mandible, crown view (No. 9597 American Museum of Natural History). Fig. 7. " INTERMEDIUS: Left maxilla, palatal view (No. 9550 American Museum of Natural History) ..... 430 Fig. 8. GARZONIA PATAGONICA : Left half of mandible, outer side (No. 15.238) ... 422 Fig. 8a. " . " Left half of mandible, crown view (No. 15,238). Fig. 9. HALMARHIPHUS NANUS : Right half of mandible, outer side (No. 9593 American Museum of Natural History) 419 Fig. ga. " " Right half of mandible, crown view (No. 9593 American Museum 6f Natural History). Fig. 10. GARZONIA PATAGONICA : Right scapula, distal end (No. 15,238) . 423 Fig. ii. " " Left humerus, front view (No. 15,238) PATAGONIAN EXPEDITIONS VOL.IV. PLATE LXIII. 'iel parfim Werner & Winter, Frankfort W, li Fig. 12. Fig- 13- Fig. 14. Fig. Fig. PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. The proximal end is restored from the right humerus ..... GARZONIA PATAGONICA : Left radius, anterior aspect (No. 15,238) . " " Left ulna, outer side (No. 15,238) . CENOLESTES OBSCURUS : Skull and mandible, right side (No. 10,559 American Museum of Natural History). " " Skull, palatal view (No. 10,559 American Museum of Natural History). " " Left half of mandible, crown view (No. 10,559 American Museum of Natural History). All the figures are three times the natural size. PAGE. 423 423 424 (VOL. iv.) PATAGONIAN EXPEDITIONS;. PAL/EONTOLOGY. EXPLANATION OF PLATE LXIV. PAGE. Fig. i. PAL^EOTHENTES INTERMEDIUS : Skull, palatal view (No. 15,225) . 430 Fig. la. " " " from above " Fig. 2. " MINUTUS : Left maxilla, palatal view (No. 15,709) . 432 Fig. 3. ABDERITES CRASSIGNATHUS : Left half of mandible, outer side (No. 15.079) • 439 Fig. 3«. " " Left half of mandible, crown view (No. 15.079)- Fig. 4. DECASTIS COLUMNARIS : Anterior portion of the right half of the mandible, outer side (No. 9594 American Museum of Natural History). The pit- ting on the anterior portion of the jaw is pathological ...... 436 Fig. 40:. " " Right ramus, crown view (No. 9594 Ameri- can Museum of Natural History). Fig. 5. CALLOMENUS LIGATUS : Right half of the mandible, outer side (No. 15.066) . 435 Fig. 5«. " " Right half of the mandible, crown view (No. 15,066). Fig. 6. DECASTIS COLUMNARIS : Left half of mandible from the outer side (No. 15,710) . 437 Fig. 6a. " " Left half of mandible, crown view (No. All the figures are three times the natural size. (VOL. iv.) PATAGONIAN EXPEDITIONS VOL.IV PLATE LXIV. F. v llerson del. partim Werner 1 Winter, Frankfor C/ENOLESTID/E. PATAGONIAN EXPEDITIONS: PALAEONTOLOGY. EXPLANATION OF PLATE LXV. Fig. I. Fig. I a. Fig. \b. Fig. 2. THYLACYNUS CYNOCEPHALUS : Skull and mandible from the right side, about two thirds the natural size (i. e., .68). Specimen in the collection of the British Museum (Nat. Hist.) 342 " " Skull, palatal view. British Museum (Nat. Hist.). About two thirds the natural size (i. e., .68). " " Skull, posterior view. British Mu- seum (Nat. Hist.). About three fourths the natural size (i. e., .74). " " Mandible, crown view, about three- fourths the natural size (i. e., .74). Osteological collection, Princeton University. Fig. 3. TRICHOSURUS VULPECULA : Skull of a young individual, palatal view, X -^. No. 413, osteological collection, Princeton University. Young indi- vidual lacking M- .... 443 Mandible of the same individual, crown view, X •£. Skull, palatal view, X \. No. 510, osteo- logical collection, Princeton University. This individual is anomalous in lacking the last upper molar . . . . 426 Mandible of the same individual, crown view, X \. Fig- 3«. Fig. 4. Fig. 4«. PETAURUS AUSTRALIS (?) (VOL. iv.) PATAGONIAN EXPEDI INDEX TO VOLUME IV. Page numbers italicized indicate the most important entry under the heading. The numbers with asterisks refer to text illustrations. Synonyms are in italics. ABDERITES, 417, 418, 438, 442, A 45i. altiramis, 452 crasignathns, 439 crassignathus, 438, 439, *44O crassiramis, 439 meridionalis, 439, 442, 451 serratus, 45 1 tenuissimus, 452 Abderitidce, 416 Abderitinae, 417, 438, 442, 443 Acdestis, 434, 452 elatus, 452 oweni, 452 parvus, 453 Acmsea, 161 heroldi, forma pygmaea, 161 Actaeon, 34, 243 chilensis, 243, 257, 304 semilaevis, 244, 258, 273, 294, 295 semistriatus, 245, 294 subscalatus, 244 tornatilis, 244 turgidus, 244, 304 Actaeonidae, 243 Acrobates pygmaeus, 336 Acrocyon, 445 equianus, 445 patagonensis, 445 sectorius, 445 Acyon, 449 bardus, 446 tricuspidatus, 449 Acyonidce, 339 Africa, connection with Antarctica, 316 443, Africa, South, 12 Agassiz, A., 54, 56, 60 L., 59, 60 Agustylus, 446 bardus, 446, 447 carnifex, 446 cynoides, 446 primaevus, 446 Allen, J. A., 334 Amathusia, 134, 135 angulata, .134, 257, 262, 266, 268, 270, 271, 282, 283, 298 Ameghino, F., 7, 79, 103, 104, 262, 265, 266, 267, 268, 269, 270, 271, 273, 274, 275, 276, 279, 280, 282, 284, 285, 287, 288, 303, 308, 309, 333, 334, 335, 339, 343, 348, 352, 378, 407, 408, 412, 416, 418, 419, 426, 429, 430, 431, 434, 435, 436, 437, 439. 442, 443> 444 America, North, 333, 444 South, ii, 333, 339 connection of, with Antarctica, 314 with Australia, 444 Cretaceous of, 6 Ammonite beds, sec Belgrano beds Ammonites cleon, 36 clypeiformis, 37, 38 leopoldinus, 37 sp., 8 telinga, 37 Ammonites, 7, 9, 10 Patagonian, 38 Amphiproviverra, 340, 343, 344, 345, 346, 356, 363, 365, 370, 376, 378, 385, 394, 407, 408, 412 461 462 INDEX. Amphiproviverra crassa, 448 ensidens, 448 manzaniana, 340, 394, 395, 398, 400, 404 minuta, 340, 394, 400, 401, 404. obusta, 448 A mphiprm n \>crrid(E, 339 Ananchytes, 7 Anatttcriiim, 376 dcfossiim, 386 oxyrhynchus, 447 Anchura, 34 Andes, 9 Annelids, 30 Antarctica, 311, 313, 315, 316, 317, 318, 319 theory, 310 Aporrhais, 34, 200, 293 araucana, 200, 258, 267, 293, 295, 298 occidentalis, 33 patagonica, 5 pes pelecani, 200 protuberatus, 32 sp., 4, 33 speciosa, 200 Aporrhaidae, 33, 200 Aptian, 8, 30 Araucana, 306 beds, 306 Area, pj, 307 bonplandiana, 265 darwini, 90, 261 imbricata, 94 navicularis, 94 nose, 94, 291 occidentalis, 94 oxytropis, 93, 298 paratina, 94 patagonica, pj, 257, 262, 266, 270, 271, 273, 277, 283, 291, 295, 298, 301 pseudonavicularis, 94, 301 tetragona, 94, 291 Archhelenis, 319, 321 Archiamazonas, 321 Archibrazil, 321 Archiguyana, 321 Archinotis, 317 Archiplata, 318, 319, 321, 324 A rchiplata- Archhelenis theory, 320 Arcidae, 93 Arctodictis, 446 australis, 446 munizi, 446 Argentine Republic, 1 i, 19, 30, 42 Arrhoges, 33 Arrialoor group, 12,61 Arroyo Gio, 49, 277, 285, 286 Pequenco, 8, 1 1, 19 Tringuico, 9, 26, 35 Artemis Icviuscula, 147 ponder osa, 146 semilavis, 144 Aspidostoma, 67 crassum, 67 giganteum, 67, 257, 266, 274, 289, 323 Astarte, 9, 10, 22 damesi, 22 peralta, 5,22, 23 postsulcata, 5, 22, 23 Astrapotherium, 287 Atlantis, 321 Australia, 301, 303, 333 Australian region, 444 Tertiary, 301, 302 Avicula, 15 nebrascana, 1 5 (Oxytoma) tardensis, 4, 14 DACULITES, 287 Balanidae, 249 Balanus, 249 apertus, 254 coquimbensis, 254 laevis, 252, 253, 254, 259, 307, 308 var. nitidus, 254 psittacus, 241), 258, 259, 278, 289, 298, 307, 308, 323 var. minor, 249 spongicola, 253 trigonus, 252, 259, 307, 308 unguiformis, 252, 294, 460 varians, 249, 250, 258, 260, 273, 274, 275, 276, 277, 278, 282, 294, 295, 298, 460 INDEX. 463 Basalt canons, see Arroyo Gio Bayle, 1 1 Bazin, 59 Behrendsen, 7, 8, 9, 11, 17, 19, 26, 30, 35, 42 Belgrano beds, 4, 5, 6, 12, 13, 14, 15, 1 6, 17, 1 8, 20, 21, 23, 24, 25, 26, 27, 30, 31, 32, 33, 34, 35, 39- 4", 42, 43 Bensley, B. A., 334, 385, 407, 409,412,416, 417, 420,441 Bettongia, 443 Bettongiina;, 443 Bibliography, 325, 458 Bicego, Sr., 269 Bivalves, 277 Blanford, H. T., 61 Bodenbender, 7, 19 Boettger, O., 152 Bonaparte, C. L., 339 Bonarelli, 36 Borchert, 310 Borhyaena, 334,^/0, 342, 345, 346,^7, 362, 363, 365, 366, 367, 368, 369, 372; 377, 38o, 381, 382, 383, 384, 396, 398,400, 408,412,444 excavata, 340, 347, 349, 357, 360 fera, 444, 445 sanguinaria, 445 tuberata, 340, 347, 349, 351, 356, 360, 365 zitteli, 356 Borhyanidce, 339 Borsonia, 24.2, 322 delucii, 243, 294 patagonica, 242, 258, 267, 294, 295 Bouchardia, 79 tulipa, 80 zitteli, 7p, 257, 262, 268, 278 Brachiopoda, 70 Brazil, Cretaceous of, 6 Brown, B., 386, 413, 415,421, 43 i Brunswick Peninsula, 7 Bryozoa, 64 Buccinidae, 208 Buccinum, 208, 324 (Cominella) annae, 208, 258, 267, 293, 295 Buccinum (Cominella) obesum, var. minor ; 209, 258, 298 obesum, 267 obesnnt, 209 veneris, 208, 293 Bulla, 245 chilensis, 245 conoidea, 246 glaphyra, 247, 294 patagonica, 245,^, 258, 262,267,277, 283, 294, 295, 305 remondi, 24.5, 246, 257, 304, 305 striatissima, 246, 304 triticum, 245 Bullidae, 245 Burtinella, 31 Busk, G., 66, 67, 70 313 Caenolestes, 334, 340, 416, 417, 418, 419, 420,421, 422, 423, 427, 441, 442 obscurus, 334 Caenolestidae, 340, 412, 4.16, 419, 423, 438, 441,443 Caenolestinse, 4.16, 417, 418, 427, 442 Calliostoma, 164 audebardi, 166, 292 biangulatum, 292 cossmanni, 166, 258, 267 cyclus, 169, 292 ditropis, 292 fricki, 165, 304 garetti, 168, 258, 267, 292, 295 iheringi, 169, 258, 267, 292, 296 metrium, 169, 292 observationis, 164, 165, 258, 273, 304 peraratum, i6j, 258, 263, 267, 292, 295, 460 philanthropus, 168, 292 philippii, 164, 165, 257, 304, 460 podolicum, 168, 292 pseudoturricula, 168 santacruzense, 167, 258, 263, 267, 292, 295 Callomenus, 417, 427, 434, 436, 437, 439, 442, 454 464 INDEX. Callomenus i ntervalatus, 454 ligatus, 4.35, 437 robustus, 454 Caluromys, 409 laniger, 412 Calyptraea, 178 (Infundibulum) radians, 179 maculata, 183 mamillaris, 182 Campiche, 16 Camptonectes, 13 Cancellaria, 235 ameghinoi, 262, 264 gracilis, 235, 236, 258, 262, 267, 273 274, 282 var. major, 235 medinae, 236, 258, 267, 273, 298 vidali, 236, 261, 264, 267 Cancellariidae, 235 Cancer patagonicus, 255, 261 Canon de las Vacas, 387 Cape Fairweather, 49 beds, 307, 310 fossils, 258 Capulidae, 177 Carcinides, 256 maenas, 256 Carcinoplacidae, 255 Carcinus, 256 peruvianus, 255 Cardiidae, 132, 135 Cardita, 125, 307 acuticostata, 129 caumotiensis, 129 dunkeri, 129, 291 elegans, 126 elegantoides, 125, 127, 256, 257, 266, 273, 303 granulata, 129 inaequalis, 126, 127, 129, 257, 261, 266, 269, 270, 271, 272, 273, 282, 283 intermedia, 300 patagonica, 128, 257, 260, 266, 282, 291, 295, 300 patagonica, 127, 269 flseudopatagonica, 128 Cardita volckmanni, 126, 257, 279, 298 Carditidae, 125 Cardium, 132 comatulum, 133, 291 difficile, 133 discrepans, 135 fragile, 133, 292 laqueatum, 135 multiradiatum, 132, 298 multiradiatum, 132, 261 philippii, 132, 257, 261, 266, 277, 278, 279, 281, 298 var. pauciradiata, 132 pisum, 133, 257, 261, 266, 277, 278, 298 platense, 265 puelchum, 133, 134, 257, 260, 266, 269, 270, 271, 272, 273, 276, 277, 281, 283, 291, 295 sphaeridium, 134, 298 sulcatum, 135 Caricella, 231 nucleus, 227 Caribbean province, 320 Carnivora, 335, 337, 338, 351, 386 Cassidulidae, Cassidulus, 61 Cassis, 322 Cellaria, 64, 65 angustiloba, 66 fistulosa, 64, 257, 277, 289, 291, 295, 299. 323 Cellariidae, 64 Centetes, 335 Cephalopoda, 35 Cerithium, 322 Chaetopoda, 63 Chenopus araucanus, 200 Chili, 9, 10, ii, 21, 298, 303 Chilostomata, 64 Chione vellicata, 138 Chrysodomus, 209 cancellatus, 209, 258, 267, 293, 295 glomus, 293 glyptus, 210 pilsbryi, 210, 258, 267 INDEX. 465 Chthalamus, 251 antiquus, 250, 251 Chubut, 275 Chun, C., 316, 318 Cidaridae, 5 1 Cidaris, 51 antarctica, 57, 257, 266, 274, 276, 280, 290, 295 avenionensis, 51, 290 dissimilis, 52 perornata, 52 sceptrifera, 52 Cinulia, 10, 35 australis, 5, 34. Cirripedia, 247 Cladoclinus, 457 copei, 457 Cladosictis, 334, 340, 343, 344, 345, 356, 363, 376, 398, 399. 407, 41 1 dissimilis, 447 lateralis, 386 lustratus, 340, 341, 376, 378, 382, 384, 386, 391, 392 oxyrhynchus, 447 patagonica, 447 petersoni, 340, 376, 379, 387, 391 trouessarti, 386 Clark, W. B., 89 Cleoniceras, 36 Clypeastridae, 55 Clypeola corrugata, 179 var. leevis, 180 magellanica, 180 Columbella, 322 Colombia, Cretaceous of, 6 Colorado, 17 Cominella, 208, 209, 324 Conglomerates, lower, 4, 28 upper, 4 Conodonictis, 445 exterminator, 445 saevus, 445 Conrad, T. A., 34, 102, 146 Conus, 322 Cope, E. D., 338 Coquand, 11,12 Coquimbo beds, 308 Corbula, 151 birostris, 152 bodenbenderi, 26 crassatelloides, 4, 5, 26 hatcheri,/5/, 257, 267, 273, 274, 292, 295 278, subaequivalvis, 152, 292 Corbulidse, 151 Cordillera, 6, 7, 285 Cossmann, M., 34, 163, 166, 167, 168, 173, 174, 175, 213, 219, 220/244, 263, 282 Cottonwood Creek, Calif., 28 Coy Inlet, 360, 386, 391, 401, 430, 437 Coy River, 356, 386 Crassatella, 26 kokeni, 123, 263, 295 longior, 125, 262 346, lyeltt, 123, 260 447 plicatilis, 124 ponderosa, 125 quarta, 124 , 383, sulcata, 124 Crassatellites, 123 kokeni, 123, 257, 263, 266, 291, 460 longior, 125, 257, 262, 278, 282, 291, 295 lyelli, 123, 124, 257, 260, 266 melina, 125, 291 quartus, 124, 257, 266, 279 sulcatus, 124, 291 Crassatellitidae, 123 Crenatula aviculiformis, 98 Crenella, 16 Creodonta, 335, 336, 337 Crepidula, 184 dilatata, 185, 259, 307, 308 fornicata, 185 grandis, 185, 289 gregaria, 184, 185, 258, 260, 267, 275, 277, 280, 282, 289, 293, 296, 298; 299, 305 incurua, 184 praerupta, 185, 293 princeps, 185 uncinata, 184 Cretaceous, 3, 4, 5, 13, 14, 19, 27, 294, 296, 300, 444 466 INDEX. Cretaceous, American, 2 1 European, 8, 9 fossils, 9 Lower, 7, 9, 10, 42 of California, 14, 28 of India, 12, 19, 28, 61 of North America, 9 of South Africa, 9, 16, 19, 21 Upper, 7, 8, 10, 61, 287, 288 Crucibulum, 777, 293 dubium, 777, 258, 277, 293, 29$ Crustacea, 247 decapod, 444 Crypta grandis, 185 incurva, 184 Cucullaea, 86, 300, 319, 322 aldrichi, 91, 291 alta, 86, 257, 260, 262, 266, 268, 269, 270, 271, 273, 284, 285, 291, 295, 298, 299 chilensis, 89, 298 crassatina, 89 (Cucullaria) darwini, yo, 257 dalli, 86, 89, 262 darwini, 261, 266, 270, 271, 272, 283, 291, 295 decussata, 89, 291 gigantea, 89, 291 multicostata, 86, 262 taeniata, 91, 291 tridentata, 90, 263, 270 Cunningham, D. J., 372 Cyclostomata, 68 Cypraea, 322 Cyrtoma, 60, 61, 290 posthumum, 60, 257, 278, 290, 294 Cytherea, 319 iheringi, 143 pseudocrassa, 142 rostrata, 143 splendida, 142 splendida, 142 "QALL, W. H., 13, 26, 50, 150, 174, 185, 190, 191, 195, 196, 210, 219, 227, 229, 288, 297, 320 Darwin, C., 7, 101, no, 1 16, 197, 217, 247, 248, 249, 250, 252, 253, 254, 255, 260, 261, 286, 298, 312 Darwin Station, 48, 274., 309 Dasyures, 386, 394 Dasyuridae, 342, 407 Dasyurus, 338, 343, 347, 376, 394, 395, 396, 397. 409. 412 maculatus, 385, 396, 399 viverrinus, 3/6, 395, 410 Davidson, T., 71, 72, 73, 74, 75 Decapoda, 255 Decastis, 417, 427, 4j6, 439, 442, 452 columnaris, 437 Turigerus, 452 Delphinulidae, 162 Dendraster, 56 Dentaliidse, 157 Dentalium, 157 gabbi, 1 60, 292 gayi, 157, 159 giganteum, 159 kickxi, 1 60, 292 (Lsevidentalium) limatum, 5, 28 lebuense, 1 59 majus, 157 mantelli, 159, 160, 299, 301, 302 octocostatum, 160 octocostatum, 160, 262, 268 octocostellatum, 160, 257, 262, 268, 273, 277, 460 patagoiiicurn, 157, 159, 261 solidutii, 159 sulcosum, 757, 1 60, 257, 261, 267, 274, 278, 292, 295, 298, 299, 301, 460 siilcosiun, 1 59 Deseado, Port, 49 Desmoceras, 36, 37 Desor, E., 53, 59 Deshayes, G. P., 90, 149, 175, 244, 246 Didelphyidae, 339, 408 Didelphys, 338, 342, 347, 376, 408, 409, 410, 411, 412, 417, 420, 428 Dipilus, 453 bergi, 453 spegazzinianus, 453 INDEX. 467 Dipilus spegazzinii, 453 Diplodonta, 307 Diprotodonta, 443 Diprotodontia, 333, 334, 416, 422, 441 Diprotodonts, Santa Cruz, 340, 416, 443 Dirck Gherritz Archipelago, 3 1 9 Doering, D. A., 286, 287 Doliidae, 204, 205 Dolium, 204, 293, 322 ovulum, 204, 258, 267, 293, 295 Dollo, L., 385 Donald, C. W., 319 Dosinia, 144, 323 acetabulum, 146, 292 complanata, 144 complanata, \ 44 denselineata, 146, 301 laeviuscula, 7/7, 257, 261, 267, 269, 270, 271, 284 magellanica, 130, 144, 145, 256, 304 matthewsoni, 146, 292 meridionalis, 144, 145, 147, 257, 258, 262, 267, 270, 271, 275, 277, 283, 292, 296, 301, 304, 307, 308 ponderosa, 146, 292 semilaevis, 144, 304 Drillia, 241, 322 koeneni, 242 limatula, 242, 294 perpolita, 242 , santacruzensis, 241, 258, 267, 294, 295 sigmoidea, 242 Dromicia, 442 Dromocyon, 335 Dusen, P., 306 Dynamic tis, 444 /era, 444 T7CHINANTHUS, 61 Echinarachnius, 56, 57, 60 excentricus, 57 julicnsis, 55, 261, 268 mirabilis, 60, 290 parma, 57 Echinidae, 52 Echinobrissus, 61 Echinodermata, 5 1 Echinoidea, 51 Echinus, 53 patagonensis, 52 Ecuadorian province, 320 Entolium, 14 Entrerios beds, 308 Eocene, 287, 295, 296, 302, 306, 317 lower, 287, 288 upper, 286, 287, 288, 305 Eodidelphys, 450 famula, 45 1 fortis, 450 Eogsea, 313 Eogene, 295, 296 Epanorthidce , 416 Epanorthus, 416, 425 ambigitus, 454 aratce, 428 Jwlmbergi, 455 incequalis, 45 5 intermedius, 430 Icinoinei, 454 lepidus, 431 minutus, 432 pachygnathus, 454 pressiforatus, 455 simplex, 432 Eschara acaste, 66 ac/iates, 66 acts, 66 acmon, 66 actcea, 66 gigantea, 67 Escharidae, 67 Essoprion, 455 consumptus, 456 coruscus, 455 Eulima subventricosa, 262, 264 Europe, 444 Cretaceous of, 5 Euthria, 209 Eutrochus, 169 Exogyra, 1 1 couloni, 8, 1 1 , 20 468 INDEX. Exogyra imbricate, 1 2 Extratropical fauna, 322 "T^ACIES of Patagonian beds, 285 Fagus, 306 beds, 306 subferruginea, 306 Felton's estancia, 401 Ficida, 205, 322 Carolina, 205, 261 concinna, 205 Fie us pyrifo rmis, 205 Fimbria patagonica, 261, 264 Fissurella, 161 eurytreta, 161, 257, 263, 275, 301 sp., 262 Fissurellidae, 161 Fissurelloidea malleata, 301 Forbes, H. O., 311, 313, 314, 317 Fort Pierre beds, 15, 17 Fossarus pillula, 263, 264 Fricker, K., 316, 319 Fuchs, E., 306 Fusidae, 221 Fusus, 209, 221 archimedis, 221, 222, 258, 274, 294, 295 burdigalensis, 224, 294 cancellatus, 209 dilatatus, 212 dilatatus, 2 1 1 discors, 239 domeykoanus, 2 1 1 glomus, 210 hector, 222, 294 laciniatus, 2 1 7 nexilis, 210 noachinus, 213, 260 nodosus, 264 obesus, 209 var. minor, 209 encodes, 212 ortmanni, 209 oxytropis, 221, 222, 304 patagonicus, 215, 260 pilsbryi, 210 pyruliformis, 223, 224, 298 Fusus sowerbyanus, 2 1 9 steinmanni, 212 subreflexus, 212 subspiralis, 221, 257, 304, 460 torosus, 223, 258, 267, 294, 295, 298 virgineus, 212 QABB, W. M., 33, 131, 185 Galaxias capensis, 3 1 8 Galerus, 181 araucanus, / 15°. 17°. !97- 203, 204, 212, 216, 223, 224, 227, 237, 238, 239, 251, 252, 260, 261, 264 Sparassodonta, 335, 444 Sparassodontidce, 339 Spatangidse, 62 Speyer, O., 151, 200, 205, 206, 243 Spisula, 26 Steinmann, G., 7, 8, 21, 299. Stigmatopygus, 6 1 Stilotherium, 457 dissimile, 457 grande, 457 Stoliczka, F., 12, 30, 31, 61, 64, 65, 66 Straits of Magellan, Cretaceous of, 6, 7 Strombidas, 33, 201 Strombus, 322 Struthiolaria, 201, 204, 299, 300, 320, 323 ameghinoi, 201, 258, 262, 267, 270, 271, 274, 275, 276, 277, 278, 279, 281, 283, 298, 305 var. multinodosa, 202, 204 chilensis, 202, 298 chilensis, 201 hatcheri, 201, 257, 304, 305 ornata, 202, 203, 258, 260, 267, 268, 269,270,271,272,283, 285,305 var. dcnsestriata, 202, 203 Stylognathus, 450 diprotodontoides, 450 Suess, E., 71, 74, 75 Subpatagonian beds, see Piso Subpatagonico Suprapatagonian beds, 265, 271, 284 See also Piso Suprapatagonico TRAPES patagonica, 5, 23 sp., 5, 24- INDEX. 479 Tarsipes, 336 Tasmania, 301, 333 Tate, R., 73, 82, 85, 113, 235, 301, 302 Tehuelche beds, 307, 308, 309, 310 Tellina, 147 capillifera, 148, 292 jeguaensis, 148, 257, 262, 267, 277, 292, 295, 298 patagonica, 262, 264 perplana, 262, 264 promaucana, 148, 298 • santacruzensis, 263, 264, 267 sp., 5, 24 subfalcata, 149 tehuelcha, 147, 257, 263, 267, 277, 282 Tellinidse, 147 Tennysonia, 69 stellata, 69, 289 subcylindrica, 69, 257, 266, 289, 316 Terebella, 63, 64 magna, 6j, 257, 274 Terebra, 236, 322 acuminata, 237 costellata, 236, 258, 261, 263, 264, 267, 279, 294, 295, 298 var. quemadensis, 237 var. santacruzensis, 237 costellata, 237 undulifera, 237, 261, 263, 264 Terebratella, 74 'dorsata, 74, 257, 266, 277, 278, 279, 281, 323, 324 gigantea, 76, 78, 253, 258, 307 patagonica, 75, 79, 257, 260, 266, 268, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 28 1 , 283, 284, 299, 307, 460 Terebratula macrostoma, 79 patagonica, 75, 260, 268 Terebratulidae, 73 Terebridae, 236 Teredo, 28 torulosa, 28 Tertiary, 48, 297, 444 Thomas, O., 334, 387, 416, 418, 419 Thylacodictis, 448 exilis, 448 Thylacyne, 333 Thylacynes, Patagonian, 406, 407 pre-Santacruzian, 406 Santacruzian, 385, 406, 407, 412 Tasmanian, 406, 407 Thylacynidae, 339, 341, 406, 407, 444 Thylacynus, 334, 336, 338, 339, 341, 342, 343, 344, 345, 346, 348, 349, 350, 351, 352, 353, 354, 355, 356, 362, 363, 364, 365, 366, 367, 368, 369, 370, 371, 372, 379, 38o, 381, 382, 383, 384, 385, 387, 395, 396, 397, 398, 399, 406, 407 cynocephalus, 339, '341, *343, *344, *345. *346 spelaeus, 339 Tideiis trisulcatus, 452 Tithonian, 8 Tornatellaea, 10 patagonica, 5, 34 Toxopneustes, 55, 290 pileolus, 54 precursor, 53, 257, 274, 277, 290, 295 Trichinopoly beds, 19 Trichosurus, 428, 442, 443 vulpecula, 335 Trichotropis patagonica, 263, 264 Trigonia, 8, 9, 10 aliformis, 8, 19 cf. aliformis, 19 hanetiana, 21 heterosculpta, 5, 10, 20 robinaldina, 21 sp., 5, 21 subventricosa, 5, 7, 8, 10, 18 tuberculifera, 19 vau, 9, 21 ventricosa, 9, 10, 19 Trigonoccelia insolita, 9 1 , 260 Triton dautzenbergi, 262 leucoslomoid.es, 219, 220 obliteratus, 263 Tritonidae, 206 Tritonium, 206 bicegoi, 206, 258, 263, 267 dautzenbergi, 264 480 INDEX. Tritoniummorgani,.?o7,258,267,293, 295, 298 obliteratum, 264 tarbellianum, 207, 293 verruculosum, 207, 298 Trochidas, 162 Trochita, 178 araucana, 181 clypeolum, 180 colchaguensis, 183 corrugata, 179, 263, 288 costellata, 178 dilatata, 183 filosa, 179 maculata, 183 magellanica, 180, 262, 288 mamillaris, 182 merriami, 178 parvula, 180 radians, 179 j/>., 261 Trochus biangulatns, 169 collaris, 170, 260 ditropis, 169 fricki, 164, 165 Icevis, 170 macsporrani, 168 magus, 171 philippii, 164 philipii, 164 stoliczkai, 163 vencficus, 170 Trophon, 2/5, 323 geversianus, 216, 217, 218, 289 var. calva, 218 var. varians, 218 inornatus, 218, 309 laciniatus, 216, .2/7, 259, 289, 307 var. gradata, 218 var. inornatus,2/(?, 307, 308, 309,460 var. santacruzensis , 215, 262 laciniatus, 288 leucostomoides, 264 leucostomoides, 262 patagonicus, 2/5, 258, 260, 267, 274, 276, 278, 279, 282, 288, 289, 307 Trophon pyriformis, 220, 262 varians, 218, 262, 307, 308, 309 var. gradata, 2 1 8 Tryon, G. W., 178, 182, 183, 185, 232, 233, 234, 324, 460 Tubicolae, 63 Tubulostium, 7 callosum, 30 pupoides, 4, 8, jo Turbonilla, 174 cuevensis, 77^, 258, 262, 273 iheringi, 174, 263 Turnus, 10, 28 dubius, 4, 6, .27 plenus, 28 Turritella, 192, 323 affinis, 193 affinis, 193, 195 aldingae, 195, 301 ambulacrum, 192, 196, 197, 249, 258, 260, 262, 267, 268, 271, 272, 273, 275, 276, 277, 278, 279, 281, 283, 293. 295, 298, 299, 301,305 apicalis, 195, 293 argentina, 193, 194, 262, 268 bicarinata, 194 breantiana, 195, 258, 261, 267, 272, 273, 282, 293, 295, 298 var. indecussata, 195 chipolana, 197, 293 cingulata, 194 cingulatiformis, 194, 195, 198, 307, 308 couteaudi, 195, 261 darwini, 1 96, 197 elachista, 261, 264a exigua, 192, 257, 304, 305, 306 granulosa, 192, 304 iheringi, 195 innotabilis, 198, 259, 307, 308 parvula, 192 patagonica, 194, 196, 198, 258, 260, 272, 274, 276, 281, 293, 295, 298, 305, 307 perattenuata, 196, 293 sowerbyana, 193 steinmanni, \ 94, 262 INDEX. 481 Turritella suturalis, 192, 193 terebriformis, 196 tricincta, 195 tricincta, 195, 196, 263 trilirata, 192 turns, 192 Turritellidae, 192. T TITENHAGE beds, 9, 10, 12, 16, 19, 21 Undulatae, 20, 21 United States, western, i 5 Urosalpinx, 219 cossmanni, 219, 258, 263, 267 elegans, 219, 258, 267, 298 leucostomoides, 219, 262" 298 laicostomoides, 219, 263, pyriformis, 214, 220, 258, 262, 278, 282 WALDIVIA, 7 V Valentin, 7 Vanikoro sp., 5, J/ Variegated Sandstones, 4 Venericardia intermedia, 129 Veneridae, 135 Venezuela, Cretaceous of, 6 Venus, 135, 323 alta, 142 antiqua, 140 arenosa, 136, 141, 256, 304 burdigalensis, 140, 292 'cancellata, 292 chiloensis 137, 138, 257, 280, 292, 295, 298, 305 clathrata, 138 crassa, 142 darwini, 140, 257, 261, 267, 292, 295, 304 difficilis, 135, 140, 256, 304, 460 hormophora, 138, 301 lezviuscula, 261 landbecki, 136, 304 meridionalis, 137, 257, 260, 267, 272, 273, 277, 279, 282, 292, 295, 298, 300, 301 muensteri, 265 inultila mellata ,138 Venus multittniata, 138, 301 navidadis, 141, 257, 262, 267, 275, 298 patagonica, 261, 264, 269, 270, 271 striatolamellata, 141, 262 subsulcata, 136, 304 uncinata, 263, 264 vellicata, 300 volckmanni, /jp, 140, 257, 262, 267, 277, 278, 279, 281, 298 var. argentina, \ 39 Vermes, 63 Vermetidas, 30, 198 Vermetus, 198 incertus, /pp, 258, 274 intortus, 198, 258, 277, 279, 289, 293,295 Vermicularia sowerbii, 30 Verruca, 248 lasvigata, 248, 258, 275, 289, 323 strcemia, 248 Verrucidae, 248 Villa Beltran, 1 1 Voluta, 226, 289, 323 alta, 227, 231, 232, 260, 261, 264, 269 ameghinoi, 233, 234, 252, 258, 262, 268, 270, 273, 279, 299, 301, 323 ancilla, 227, 232, 233, 234, 323 atkinsoni, 234, 301 brasiliana, 234, 323 colocynthis, 234 domeykoana, 232, 234, 258, 263, 267, 298, 323 dorbignyana, 227, 230, 233, 234, 258, 261, 267, 275, 282, 323 dorbignyana, 226 gracilior, 227, 230, 258, 261, 267, 274, 275, 276, 277, 282 gracilis, 227 gracilis, 227, 229, 261 halli, 232 magellanica, 227, 233, 234, 323 orbignyana, 230 pacifica, 234, 299 patagonica, 262, 264 petersoni, 229, 258, 267 philippiana, 228, 263 pilsbryi, 232, 263 482 INDEX. Voluta quemadensis, 227, 228, 263 serissa, 235, 301, 460 tateana, 235, 301 triplicata, 226, 228, 229, 230, 234, 235, 258, 263, 267, 298, 301 tuberculata, 233 Volutididae, 224 Volutilithes, 230, 460 philippiana, 229 Volutoid series, 227 rt/ALDHEIMIA lenticularis , 73 patagonica, 76 Wallace, A. R., 311, 312, 313, 315 Wanganui beds, 299 Waters, A. W., 66, 68, 70 Weber, M., 318, 336 Weltner, W., 250, 252, 253, 255 White, C. A., 6 White Lake, 3 Whitfield, R. P., 102, 113, 125, 129, 138, 146, 148 Wood, S. V., 94, 95, 124, 156, 199, 241, 460 Woodward, L. P., 156 Wolf, pouched, 333 YLOPHAGA, 28 319 , K. VON, 50, 61, 64, 68, 83, 92, 105, 1 10, 113, 176, 177, 178,489, 190, 199, 217, 234, 252, 299 Zoogeography, 444 Zwartkop River, 19 Q 115 P85 1903 Princeton University Expeditions to Patagonia Reports i^ASci. PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY