ay ely ¢ teen i My MN a Phy Lup oy i Mone Le Madd ip. tte Pesan HUB on PE tih sync eat i ot ¢ My, fi es Mie se a rate. KD Ura \ee Pate /) a lees Ne ANNALS OF THE Peo NEGIE Wee sk UM VoLuME IV 1906-1908 W. J. HOLLAND, Eaitor PUBLISHED BY THE AUTHORITY OF THE BoaRD OF TRUSTEES OF THE CARNEGIE INSTITUTE Z es i ‘ erat ee | 7 be ea,s y q ¢ a i % <9) : 4 ers 7 PRESS oF : é e : THE NEW ERA PRINTING COMPANY 7 ‘ : ~“ LANCASTER. PA. — fas 4 , 7 , i ay "| ch . * mal P é. c ‘ : \ Shy ye me: Uny q pie ae << ' ts Papi OF CONTEN ES: Title-page and Table of Contents List of Plates List of Figures in Text Errata and Corrigenda ; ; : Editorial Notes : : é 3-7, 81-83, I, Plastron of the Pioigsiatne Ee G. R. Wieland . II. Descriptions of New Species of Turtles of the Genus Tes- tudo, collected from the Miocene by the Carnegie Museum ; together with a Description of the Skull of Stylemys nebrascensis. By Oliver P. Hay . III. The Miocene Beds of Western Nebraska and Eastern Wyoming and their Vertebrate Faune. By O. A. Peterson ‘ , , ; IV. A New Species of Roars fr om peniesiianin. By Otto E Jennings : ‘ : V. Merycochcerus and a New Gents of Merrcoidedalste with some Notes on Other Agriochceride. By Earl Douglass : : ; VI. Some New Meeyonidodonts a Earl Douglass VII. On Further Collections of Fishes from Paraguay. By Carl H. Eigenmann assisted by Waldo Lee McAtee and David Perkins Ward ‘ VIII. An Undetermined Element in the Osteology Be the ee sauride. By W. J. Holland. : : IX. The Gastropoda of the Chazy Formation. By Percy E. Raymond : ; X. A Further Occurrence of Grades Americana in Pennsyl- vania. By Otto E. Jennings. XI. A Preliminary Account of the Pleistocene Fanaa Ditov : ered in a Cave Opened at Frankstown, Pennsylvania, in April and May, 1907. By W. J. Holland. XII. Description of Vertebrate Fossils from the Vicinity of Pittsburgh, Pennsylvania. By E. C. Case XIII. Notes on Ordovician Trilobites : Ilenidze from the Black River Limestone near Ottawa, Canada. By Percy E. Raymond and J. E. Narraway . ill PAGES Liv v-Vii iX—X1 xil 159-161 8-14 15-20 21~72 ae 84-98 meh . IIO-157 162—167 » 168=22 . 226-227 . 228-233 » 234-241 - Some Oligocene Lizards. By Earl Douglass - Description of the Type Specimen of Stenomylus Gracilis ‘ : . 286-300 . Brief Descriptions of Some New Species of Birds from TABLE OF CONTENTS. - Rhinoceroses from the Oligocene and Miocene Deposits of North Dakota and Montana. By Earl Douglass. . Fossil Horses from North Dakota and Montana. By . 267-277 . 278-285 Earl Douglass Peterson. By O. A. Peterson Costa Rica and a Record of Some Species not Hith- erto Reported from that Country. By M. A. Car- ricer, jr, : ; : 256-266 . 301-302 . Chelone midas. . Superior view of shell of Zestudo inusitata Hay. EIST OF (Pires. Thalassochelys caretta. See explanation, p. 14. See explanation, p. 14. Inferior view of shell of Zestudo inusttata Hay. Superior view of shell of Zestudo hollandi Hay. . Inferior view of shell of Zestudo hollandi Hay. . Superior view of shell of Zestudo ed@ Hay. . Inferior view of shell of Zestudo ed@ Hay. . Group of skeletons representing three specimens of Promery- cocherus carrtkert Peterson. . Restoration of Promerycochaerus carritkeri Peterson. . Cranium of Promerycocherus vantasselensts Peterson. . Skull of Stenomylus gracilis Peterson. . Side view of skull of Diceratherium niobrarense Peterson. . Palate view of Diceratherium niobrarense Peterson. . Side view of skull and lower jaws and crown view of inferior dentition of Dicerathertum cookt Peterson. . Side view of skull and lower jaws of Dinohyus hollandt. . Palatal view of skull of Dinohyus hollandi Peterson. . Skull of Amphicyon superbus Peterson. . Larahippus nebrascensts Peterson. . Lonicera altissima Jennings. XXI. XXII. XXIII. XXIV. XXV. XXVI. XXVII. XXVIII. XXIX. XXX. XXXII. XXXII. Skull of Pronomotherium laticeps Douglass. Eucrotaphus dickinsonensts Douglass. Lucrotaphus montanus Douglass. Skull of Merycotdes cursor Douglass. Mesoreodon (?) latidens Douglass. Promerycocherus hatchert Douglass. Promerycocherus grandis Douglass. Promerycocherus hollandi Douglass. Ticholeptus breviceps Douglass. Ticholeptus bannackensts Douglass. Dysichthys australe Eigenmann and Ward. Figs. 1-4. Pimelodella mucosa, Pimelodella gracilis, [hering- ichthys megalops, See explanation, p. 156. Vv vl XXXII. XXXIV. XXV. KRAVE. XXXVII. XX XVII. KI sch, soa. 06 LLY. XLV. XLVL XLV. ALVIN, XLIX. . Gastropods from the Chazy Formation. . Gastropods from the Chazy Formation. . Gastropods from the Chazy Formation. . Gastropods from the Chazy Formation. . Gastropods from the Chazy Formation. . Gastropods from the Chazy Formation. . Trilobites from the Black River Formation. List or PLATES) Figs. 1-2. Sheringichthys labrosus. Dentition of Serrasalmo humeralts. See explanation, p. 156. Figs. 1-3. Hemidoras paraguayensts. Homodietus antsitst. See explanation, p. 15¢. , Figs. 1-3. Hemtodontichthys acipenserinus. Loricaria typus. See explanation, p. 156. Figs. 1-3. Sturisoma robusta. Figs. 1-4. planation, p. 156. See explanation, p. 156. Loricarta carinata. Loricaria labialis, See ex- Figs. 1-4. Otocinclus vittatus. Corydoras microps. Cory- doras aurofrenatus. See explanation, p. 157. Figs. 1-3. Parodon paraguayensis. Schizodon borelli. Aphy- ocharax dentatus.‘ See explanation, p. 157. Figs. 1-3. Zetragonopterus argenicus. Tetragonopterus allent. Astyanax pelegrint. See explanation, p. 157. Figs. 1-3. Menkhausta dichrourus. Menkhausia agassizt. Deuterodon 1guape. See explanation, p. 157. Figs. 1-2. Metynnis mola. Myleus levis. See explanation, Pp. 157: Figs. 1-2. Chetobranchopsis australe. dquidens paraguay- ensis. See explanation, p. 157. Figs. 1-3. Mesonauta festivus. Heterogramma Corumbe. See explanation, p. 157. Gastropods from the Chazy Formation. See explanation, p. 221. Gastropods from the Chazy Formation. See explanation, p. 221. Gastropods from the Chazy Formation. See explanation, pp. 221-222. Gastropods from the Chazy Formation. 222-223, See explanation, pp. See explanation, p. 223. See explanation, p. 223. See explanation, p. 223. See explanation, p. 223. See explanation, p. 224. See explanation, pp. 224-225. Wynnea Americana Thaxter. . View of Cave at Frankstown, Pa. . View of Cave at Frankstown, Pa. . Reptilian Remains Found Near Pittsburgh, Pennsylvania. See explanation, p. 240. See explanation, PP. 254-255. LXI. LXITI, LXIII. LXIV. LXV. EXVI, LAVII. LXVIII. LIst OF PLATES, Vil Trilobites from the Black River Formation. See explanation, Pp. 255- Trilobites from the Black River and Trenton Formation. See explanation, p. 255. Aphelops montanus Douglass. Aphelops ceratorhinus Douglass. Teeth of Fossil Horses from North Dakota and Montana. See explanation, p. 276. Skull of Merychippus mtssouriensts Douglass. See explanation pa i276. Miocene Horses from Montana. See explanation, p. 276. Miocene Horses from Montana. See explanation, p. 277. as Publications of the Carnegie Museum _ Serial No. 46 nes ac aS ae “RLIFC IR Be ANNALS \4 "7 : < : Mi Seah, y OF SG aVEWE F THE ay ' ite 1% CARNEGIE MUSEUM Vero ive. No: December, 1906 _ For sale by Messrs. Wm. Wesley & Sons, 28 Essex St. Strand, London, England Messrs. R. Friedlaender u. Sohn, 11-Carlstrasse, Berlin, N. W. 6., Germany, and at the Carnegie Museum, Schenley Park, Pittsburgh, Pa., U.S, A. ; ; 4 19? Fk en kat F} Nye) Bi at ian ay Pa c Nt PUBLICATIONS OF THE CARNEGIE MUSEUM SERIAL No. 46 ANNALS OF THE PAKN EGE are SEU iv Voroly sae: W. J. HOLLAND, Zastor PUBLISHED BY THE AUTHORITY OF THE BoARD OF TRUSTEES OF THE CARNEGIE INSTITUTE DECEMBER, 1906 ANNALS OF THE CARNEGIE MUSEUM VOLUME: EVa “NO? EDITORIAL NOTES. THE task of completing the great edifice of the new Carnegie Library and Institute at the entrance of Schenley Park is gradually nearing the end. The building is the largest structure devoted to similar uses in North America and covers a larger area than any other build- ing of like character which has as yet been erected in the new world. The decoration of the walls and the permanent installation of appa- ratus for lighting and heating are rapidly going forward. In August it began to be possible for the management of the Museum to begin to transfer its contents to the various rooms which they are hereafter to occupy. A great deal had been in storage. The moving -of the cases, their renovation, the cleaning and rearrangement of their con- tents, the orderly arrangement and classification of collections which had long been inaccessible, the mounting by the taxidermists of specimens which hitherto it had been impossible to mount for lack of room, all of these things involve an expenditure of effort which is taxing the time and strength of the working force of the Museum to the uttermost. It has been determined to formally dedicate the building upon the 1rith of April, 1907. To bring the great halls and the accumulated treasures of the Museum into something like the shape which it is intended they shall ultimately have, is a task which in the limited time at the command of the working force is little less than herculean. ‘The director of one of the leading Museums of America, upon the occasion of a recent visit to the institution ex- 3 4 ANNALS OF THE CARNEGIE MUSEUM. pressed astonishment that such an undertaking should even be con- templated in view of the shortness of the time at command. ‘In order to do the work before you, it seems to me,’’ he said, ‘‘ that you ought to take at least two years.’’ But of course that is impos- sible. It no doubt will be two or three years before the Museum will have assumed something like that appearance which it is designed to ultimately have. Nevertheless it is anticipated that when the formal opening occurs a very attractive and interesting beginning will have been made. The Director, whose plans are being faithfully ex- ecuted by his colleagues, cannot refrain at this time from uttering a premonitory note of warning, so that the expectations of the public may not be unduly raised. When it is recalled that there are already in the custody of the Museum no less than a million and a quarter of objects each one of which has its significance, each one of which re- quires to be accurately determined and labeled and catalogued, and. each one of which has to be assigned to a place where it may be readily found and referred to by those who may desire to consult the collections, it will be seen that the mere clerical labor involved in the undertaking is huge. In addition to this is the work of installa- tion and display in the case of large portions of the collections, which calls for the exercise both of scientific knowledge and of artistic sense. When these facts are borne in mind it will be understood that the work assigned to the comparatively small force in the employ- ment of the museum is such as to tax their ability to the last degree. However, as the dropping of water wears away the stone, so the per- sistent and faithful efforts of the gentlemen of the staff of the Museum will result in overcoming the great undertaking which is before them. DuRING the past summer it was not deemed expedient, in view of the necessities existing at the Museum, to send many members of the staff into the field for the purpose of adding to our collections. The only exception which was made was in the case of the section of pale- ontology, for the maintenance of which and for the special carrying on of explorations by which Mr. Carnegie himself makes an annual appropriation. Dr. Percy E. Raymond made a visit to northern New York and to Canada in quest of invertebrate fossils, in which he was entirely successful. Mr. W. H. Utterback during the greater part of the summer continued the work of uncovering and digging up fossils EDITORIAL. 5 at the Agate Spring quarry, permission to explore which has been so kindly accorded to us by Mr. James H. Cook. He was very success- ful in his efforts, and one of the results has been the acquisition of a large amount of most excellent material illustrating the osteology of the genus A/oropus and its allies. A great deal of material belonging to various genera of the extinct Rhinocerotidz was recovered. During the month of September Mr. Utterback undertook to explore a locality known to him in the Laramie beds, which yielded some very fine material illustrative of the osteology of the Ceratopsia. He was un- able to complete his work owing to the advent of winter at the high altitude at which he was laboring, but found enough to show that it is highly probable that we shall succeed in recovering an almost entire skeleton, including the skull, of Zorosaurus, one of the huge horned reptiles of the Laramie. Messrs. Roy L. Moodie and J. W. Bartho- low, acting under the instructions of Dr. S. W. Williston, succeeded in recovering for the Museum from the Hailey Shales of Wyoming a large quantity of interesting material, all of which is believed to be new to science. Upon the whole the work of the paleontological section of the Museum has been as successful as that of any previous year and will be found in the end.to have resulted in a very considerable enlargement of our knowledge of the life of the past. Mr. W. E. C. Topp during the month of July made a brief excur- sion to Canada to the country south of Lake Abitibi, and succeeded in making some interesting observations upon the avifauna of that region and in collecting a number of desirable specimens. THE editor of the ANNALS with sincere sorrow recalls that on July 19, 1906, Sir Walter L. Buller, the distinguished ornithologist, whose works on the birds of New Zealand are classic, ended his earthly labors. The great collection upon which he based his ‘‘ Supplement to the Birds of New Zealand,’’ which is in fact a revision of his earlier work on ‘‘ The Birds of New Zealand,’’ is the permanent property of the Carnegie Museum. A letter from his son states that on several occasions before his death Sir Walter expressed great pleasure at the thought that the collection was lodged in the Carnegie Museum, which he declared to be in his opinion ‘‘one of the finest institutions of its kind in existence.”’ 6 ANNALS OF THE CARNEGIE MUSEUM. THE skeleton of Diplodocus carnegiet has been mounted in the Hall of Paleontology upon bases resembling those upon which the replica was mounted at the British Museum in the spring of 1905. The attitude given the skeleton is in most respects identical with that given the replica on the other side of the Atlantic, with this differ- ence, that the neck has been curved upward and is raised considerably higher than is the case with the facsimile in the British Museum. The mounting of the bones, which are many of them immensely heavy, involved far more mechanical difficulties than the mounting of the replica. Through the ingenuity of Mr. Arthur S. Coggeshall, the chief preparator in the Section of Paleontology, a system of cast steel supports was devised which reduces the amount of metal work ex- posed to view to a minimum and gives the skeleton so far as possible a very graceful appearance, while yet securing absolute rigidity and safety. The editor thinks that the mounting of this skeleton repre- sents the most successful attempt which has yet been made anywhere in setting up so large and cumbrous as well as fragile a specimen. Mr. Remi H. Santrens, who for eighteen years was connected with Ward’s Natural Science Establishment at Rochester, New York, asa taxidermist, has entered into the service of the Carnegie Museum and is doing excellent work, for which his long experience and great ability abundantly prepare him. Mr. FREDERIC WEBSTER has mounted in very lifelike attitude a grizzly bear, two mountain sheep ( Ovis canadensis), and a magnifi- cent Rocky Mountain goat, recently presented to the Museum by Mr. John M. Phillips of Pittsburgh, whose adventures, in company with Mr. William T. Hornaday, are delightfully described in a volume coming from the pen of the latter gentleman and entitled ‘‘ Camp Fires in the Canadian Rockies,’’ which has just been issued by the Scribners. It is to be regretted that more of the hunters of big game in this country do not realize how much they might do to interest and instruct the general public if they would take the pains which has been taken by Mr. Phillips to preserve specimens obtained in the_ chase. Mr. Phillips’ great kindness to the Museum is deeply appre- ciated, and we hope that his public-spirited example may be followed by many others of the sportsmen of the ‘‘ Iron City.”’ EDITORIAL. 7 THE paper upon ‘‘ Early Chinese Writing’’ by Mr. Frank H. Chal- fant, which has been published as Memoir No. 1 of Volume IV. of the memoirs, has been distributed and has elicited from those who are able to judge of its merits comments of a most favorable character. Dr. Friedrich Hirth, of the Department of Chinese in Columbia Univer- sity, says ina letter to the editor: ‘It fills a long felt gap in literature on Chinese philological subjects, and I congratulate you for having brought out a book which will be much appreciated by the scientific world both in America and in Europe.’’ A request has been pre- ferred to the authorities of the Museum to supply copies of the work to a number of colleges in China. This request comes from one of the leading scholars of New York City. The editor cannot refrain from expressing some satisfaction in view of the fact that he was able in the city of Pittsburgh to make the Chinese types which were re- quired in the composition of the work. In this task he was diligently aided by the author, who, no doubt, will in coming years, recall the long days he spent in the office of the Director of the Museum, en- gaged in finishing with a file the bits of metal which were used. A few years ago the production of such a work as this in Pittsburgh would have been impossible. AN extensive and well-illustrated Memoir upon his researches on the Pacific coast of Costa Rica, by Mr. C. V. Hartman, is going through the press and will shortly appear as Memoir No. 1 of the third volume of the Memoirs. A number of plates illustrating ob- jects of jade from Costa Rica and Mexico will be included in this work, the objects figured having been obtained by Mr. Hartman him- self or having been acquired as parts of the Velasco collections pur- chased several years ago by the Carnegie Museum. I. PLASTRON OF THE PROTOSTEGIN. By G: R. WiEvanm In all the earlier discovered skeletons of the huge Cretaceous tur- tles included in the Protosteginze, the hyoplastra were found in normal position in contact with the peculiar T-shaped entoplastron characterizing this subfamily ; while elements definitely referable to the epiplastra were singularly absent. This condition having repeated itself at widely separated localities, and in two genera as represented by fully six specimens approaching completeness, I was led to sup- — pose after the discovery of a median nuchal-like bone in Archelon that the T-shaped entoplastron might represent a fusion of the epiplastra with the entoplastron. ‘This idea of fusion was beset by certain doubts at the time it was discussed in my descrip- tion of the plastron of the type specimen of Archelon, and has proven incorrect. Dr. E. C. Case-and Dr. O: Ps aymalsca held unpublished odinions on the subject, the one being inclined to Accept, “the other disagreeing with the idea of a sup- posable fusion of the anterior plastral elements ; although both had mistakenly identified the T-shaped entoplastron as ee rs we: the nuchal of Protostega. However, as Pre. to ate asia ar Presents ue more example a the exi- Wieland. Left epiplastron. 8€NCies attending the uncovering of the Inner (superior) view. >< 1. fossil record, in spite of the various speci- ¢, anterior limit of entoplastral mens known and the fact that the question overlap. Compare with epi- of plastral structure in the Protosteginze plastra of Aspidonectes, etc., in i Biotin anil: ‘ had thus become an open one, no direct evidence was obtained until two years ago. Then I secured on the west bank of the Cheyenne River where it breaks through the Oligocene Bad Lands of South Dakota, the greater. part of a large Arche/on skeleton including along with the hyo-, hypo-, and xiphiplastra a single epiplastron. This proves of quite unexpected interest, because it is of the out-turned type seen in the Trionychids and 8 SESE a =w¥ 7 R » —s ee WIELAND: PLASTRON OF THE PROTOSTEGIN~. 9 Dermochelys amongst existing, and amongst extinct forms, only in the several genera of the Thalassemydidz of the European Jurassic and Cretaceous, together with Protosphargis of the scaly clays of Italy. The epiplastron of Avchelon as represented in superior view in Fig. 1 is of subcrescentic outline with the anterior limb heavy, and the posterior broadened, flattened, and digitate. The thickness of the heavy anterior end is 4.5 centimeters, and an accompanying humerus is exactly 2 feet in length. As in the Trionychyde there is no true sutural union with the ento- plastron, the contour showing that the superior face of the epiplastron was overlain by the antero-external border of the entoplastron. Be- yond this border the anterior limb of the epiplastron projected about 7 cm. like a broad, short, heavy horn, with its convex sideental. Four Fic. 2. Aspidonectes spinifer. Nether (ectal) view of plastron, xt. oD, epiplastron ; ez, entoplastron; 7, f/, anterior and mesial foramina, (Cf epiplastron of Archelon and Dermochelys, also of Thalassemydide. ) broad, shallow furrows increasing in depth from the inner to the outer side, mark the contact of as many overlying digitations or ridges which may all have been entoplastral, rather than in part hyoplastral. This lack of sutural union and the boomerang-like shape of the epi- plastron show how it must have been the very first bone to be torn out 10 ANNALS OF THE CARNEGIE MUSEUM. by predaceous fishes and sharks, or by wave action, after the turtle went down, and thus explain the rarity of its recovery and why the explorations of thirty years have hitherto failed to reveal so interesting a skeletal part. Inasmuch as the adequate mounting and preparation for description of the original type of Avchelon ischyros has now been begun at the Yale Museum, there is no pres- ent need» to attempt ‘further plastral restoration of this largest of seaturtles. Meanwhile, how- ever, a very good idea of the plastral form may be had by comparison, in combination, of the writer’s figure of the ento-, hyo-, hypo-, and xiphiplastra of Archelon with the figure of the Jurassic Thalassemyd /ydro- pelta Meyert given on page 530 of Vol. Ul. of Zittels= Tiamae buch. ‘The manner in which the epiplastra of Archelon pro- jected anteriorly is quite closely Fic. 3. Dermochelyscoriacea. Plastron paralleled in Hydropelta, except with (@) ectal and (2) ental view of nuchal. that in the former it appears that After Gervais (cf out-turned epiplastra with those of Archelon, etc., in subjoined figures ). there was no epiplastral abutting on the median line, and that the entoplastron is relatively larger. The present determination for the first time of the true type of plastron in the Protosteginze is of far more than casual interest be- cause of the obvious bearing on the most vexed of zodpaleontological problems, the origin of Dermochelys, as well as on the highly inter- esting question of the mono- and polyphyly of the other existing and the various extinct genera of marine turtles. The testudinate plastron while undergoing characteristic variations of form within closed groups is fully as conservative a structure as the carapace. Also, since in Dermochelys the plastron and nuchal are the only parts left for comparison with the normally developed carapace and plastron of other Testudinates, the paleontologic record has been scanned year after year for true marine turtles with a more or less reduced WIELAND: PLASTRON OF THE PROTOSTEGINE. 11 carapace and similar plastral type, which might stand in an ancestral or approximately ancestral relationship. Nevertheless, Dermochelys has retained its isolated position ; fossil evidence bearing directly on its origin has been singularly lacking, in fact going scarcely further than to indicate that Psephophorus of the Belgian Pliocene may have marked the culmination in size of the: Dermochelydide. For a time, however, after Cope’s description of Protos- tega in 1875 this genus, as very imperfectly known, was supposed to largely bridge the gap between Dermochelys and the other marine _ turties, mainly on the ground of its considerable carapacial and its doubtless complete horn- shield reduction. ‘The dis- Fic. 4. Ental view of nuchal, of (2) Asfr- covery, in much better preser- “mectes spinifer (X 4), and (4) the entoplas- Pen off the closely related, "On of Azeeeem tines ee ee nether tubercular process articulating with neural Archelon, aswell as the study of better specimens of /ro/0- muscular attachment. Entoplastral and nuchal stega, however, developed the similarity are correlated in these forms. presence of far closer relation- ship to the Cheloniide than was at first suspected, the writer finally being led to include these forms in a Chelonidan subfamily, this doubtless being their correct morphologic rather than their exact phyletic position. Nevertheless it now becomes possible to codrdinate several hitherto isolated facts. If we regard Dermochelys as the most specialized Testudinate, and the osteodermal mosaic as a secondary structure, the plastron has been more persistent than the carapace, only the ento- plastron having been lost by reduction, whereas the nuchal is the sole remaining carapacial element. ‘The Protosteginz also, though struc- turally speaking members of the Cheloniidz, are now seen to have with their much reduced carapace the same highly characteristic epi- arch of cervical vertebra; 7, a lateral ridge for plastral type as Dermochelys, as well as other minor resemblances which need not now be enumerated. The same is true of the Thalas- semydide of the European Jurassic and Cretaceous, as represented by 12 ANNALS OF THE CARNEGIE MUSEUM. various genera, for the greater part but imperfectly known. More- over these Thalassemyds compose the group, which has been considered to stand most nearly in an ancestral relationship to the existing marine turtles, previous to my demonstration of the structure of the Pro- pleurinz of the New Jersey Cretaceous, and proof that this primitive littoral subfamily includes the forms which virtually bridge the gap between primitive land tortoises and the existing genera of the Chelo- nine. On the other hand it is to be emphasized that several other Fic. 5. Osteopygis gibbi Wieland, < s Plastral view. Primitive semi-marine turtle from the New Jersey Cretaceous, showing the elongate and in-turned epiplastra characteristic of the Cheloniide. The horn shields are not indicated, but are in approximate agreement with Thalassochelys. Cretaceous subfamilies besides the Protostegine are so different from the New Jersey forms, that their ancestry must still be sought for amongst the ‘Thalassemyds. Such are in particular the Desmato- chelydine. It is hence more and more strongly suggested, as the > aa — Neamt Mee St | - i +, ve" “pee WIELAND: PLASTRON OF THE PROTOSTEGIN. 13 facts accumulate, that the flippered turtles represent a great complex of forms which have arisen through repeated invasion of the sea in Mesozoic time, it being indeed not improbable that most of the groups most conveniently grouped as marine subfamilies have thus indepen- dently arisen from more or less nearly related genera of land tortoises. The tracing of such independent lines is, however, doubtless rendered difficult, as much by subsequent homoplastic adaptations, as by the imperfections of the record as known. But while we are not yet ina position to absolutely prove such a polyphyly of the Cheloniidz, the general facts in the case of the Protostegine, their various ear marks suggesting a certain relationship to Dermochelys by way of the Thalas- semyds, together with culmination in the Cretaceous, assuredly suggest independent origin from forms other than the New Jersey Propleurine as so closely related to the Chelonine. The hypothesis is therefore advanced, in conclusion, that: (a) The marine turtles are distinctly polyphyletic ; that is, various more or less distantly related tortoises have from the Jurassic on repeatedly assumed littoral habits, and developed flippers. (4) Five of these distinct lines of marine turtles are exemplified by (1) Dermochelys, (2) the Protosteginz, (3) the Desmatochelydinz, (4) the Chelonine, (5) Carettochelys insculpta, the Fly River Turtle of New Guinea, a flippered pleurodiran with complete reduction of the horn shields. (c) The Ancestry of Dermochelys and the Protostegine falls within the Thalassemyds, or Acichelydide, and the plastron and nuchal also suggest certain affinities between the latter and some ancient form near to the original Trionychid line. As correlative to this hypothesis I may add, though somewhat in repetition, that however one may split hairs about the meager evidence as to the nature of the mutations which have resulted in the osteodermal mosaic of Dermochels, the safe and simple working view is to my mind that his plastron is a turtle plastron, his nuchal a true nuchal, all his other organization likewise testudinate and impossible of homo- plastic origin, and that his ancestors were simply more ancient than those of the Cheloninz, but withal typical tortoises, quite probably falling, as above suggested, within the Thalassemydide, and prob- ably without an osteodermal mosaic. The epineural ossicles of Zoxo- chelys, and the epimarginals of Zy¢o/oma, show well that an osteoder- mal series corresponding to the hornshield system was once far more conspicuous in the turtles than now ; and the keels of Dermochelys are in exact correspondence to such a series. 14 ANNALS OF THE CARNEGIE MUSEUM. EXPLANATION OF PLATES. Plate I. Zhalassochelys caretta, Delaware Bay. Ectal view of plastron. 4. Plastral type of the Cheloniide, except the Protosteginz, with narrow in-turned epi- plastra. The ventral horn-shield, or that imbricating over the hyo-hypoplastral junc- tion, is the only one likely to be clearly indicated in fossil plastra of this type. Plate Il. Chelone mydas (var. or sp. nov.). Southern Atlantic coast of the United States. Plastron with marginals (less nuchal) placed in natural position very nearly. More reduced than the preceding. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate |. Ectal View of Plastron of Thalassochelys caretta, } a ~ _ *) = - < et eT ere > + = 7 ole = a a e € e _ ' F « 7. * aeons “2! :. w . : Lae ~ " 5 ay 1 oe Se b x <2 “ss 7 : a 1) gem ra o-< _ 7 , ay ow v+y ret 2sa* [8 Gra: — 7 a a)". oe s nn ‘ Ms 2 ; : ra sae & - Oe ey rh ; > = fay } ya 8 e . = r af =a i) } ‘ i 1 we oot 1, < ? H * . a® ae : . a ¢# ~ eed » =a : - 4 . — 4 ws . ; . y ad é ? * r ; 7 7 4 ‘ ‘ . ri * > “aa ae, » ~@ ANNALS CARNEGIE MUSEUM, Vol. IV. Plate Il. Chelone Mydas, \ess than 1. ie DESCRIPTIONS OF NEW. SPECIES OF TURTLES OF THE GENUS TESTUDO, COLLECTED FROM THE MIO- CENE BY THE CARNEGIE MUSEUM; TOGETHER WITH: A DESCRIPTIONSOF. THE: SKULL OF STYLEMYS NEBRASCENSIS. By OLIvER P. Hay. (PLatres III-VIII.) Through the kindness of the Director of the Carnegie Museum, Dr. W. J. Holland, the writer has been permitted to study and describe in these pages six turtles, which have been collected by parties sent out from the Museum in recent years. The drawings and the photo- graphs illustrating these species have been furnished by the Museum. TESTUDO PERAGRANS, sp. Nov. This species has as its type No. r1o1 of the Carnegie Museum. It was collected in 1903, by Mr. Earl Douglass, at a point south of McCarty’s mountain and Big Hole River, and north of Dillon, Mon- tana. The deposits are supposed to belong to the Lower Miocene, but they may be a part of the Upper Oligocene. | 4 /, f ") ou y. “Wf 8\\\ WO Z\\" | / \ if! WZ y i eeoerna ek Fic. 1, Superior view of skull of Zestudo peragrans. 1. Fic, 2, Inferior view of the skull of Zistudo peragrans. 1, 15 16 ANNALS OF THE CARNEGIE MUSEUM. The specimen furnishes the carapace and plastron, both somewhat damaged, and the nearly complete skull. On account of the close application of the lower jaw to the skull, the character of the tritur- ating surfaces of the upper jaw cannot be determined. There is, however, no reason for supposing that the species is not a Zestudo. SS a ANI SSS EE) sii USSSA H \\ Sir ey Fic. 3. View of the left side of the skull of Zestudo peragrans. 4. The length of the skull (Figs. 1-3) from the snout to the occip- ital condyle, was very close to 53mm. ‘The width across the auditory chambers is 40 mm. ‘The width of the interorbital space is 12 mm. The orbits are nearly circular, with a diameter of 15 mm. The shell (Figs. 4, 5) is damaged so that little can be determined regarding the neurals and the vertebral scutes. The general form of the shell is well preserved. The carapace is 320 mm. long and 280 mm. wide. The plastron has a prominent epiplastral lip. This projects 30 mm. in front of the gulo-humeral sulci. It greatest thickness is 26 mm. TESTUDO ARENIVAGA, Sp. Nov. This species is based on fragments of a large turtle which has the catalogue number 1509 of the Carnegie Museum. ‘The remains were Hav: DeEscRIPTION OF NEW SPECIES OF TURTLES. | 17 found by a collecting party from the Museum in 1905, about two miles north of Agate Springs quarry, in Sioux county, Nebraska. The de- posits belong to the Lower Miocene. The pygal and the eleventh periph- eral are selected for description. The pygal (Figs. 6, 7) has an extreme height of 112 mm. ‘The width of the upper border is 102 mm; that of the free border, 58 mm. The upper border is notched for the supra-pygal. The thickness of the border which articulated with the eleventh peripheral is 30 mm. The upper, or hinder, surface of the bone is convex in all directions. The free border is acute. The eleventh peripheral (Figs. Fic. 5. Inferior view of shell of 7es- 6, 8) is 45 mm. wide above; 87 sudo peragrans. 2. mm. wide below. ‘The height is 98 mm. The outer surface is nearly plane next the pygal, strongly concave next the tenth peripheral. The greatest thickness is 31 mm. A Fic> 6. . FIG. J; Fic. 8. Fic. 6. Upper side of pygal and eleventh peripheral of Zestado arenivaga. 1, Fic. 7. Section along midline of pygal of Zestudo arenivaga. Fic. 8. Section of anterior end of eleventh peripheral of 7estudo arenivaga. 18 ANNALS OF THE CARNEGIE MUSEUM. TESTUDO INUSITATA, sp. nov. (PLATEs ITI-IV. ) This name is applied to a specimen which was collected by Mr. Earl Douglass, near Canyon Ferry, Broadwater County, Montana. The horizon is regarded as being that of the Deep River, a portion of the Middle Miocene. The number of the type is 311 of the Car- negie Museum. The length of the carapace is 265 mm. ; the width, 200 mm. A peculiarity of the species is found in the forms of the neurals. The first is, as usual in Zestudo, four-sided. Usually in species of this genus one or more of the neurals behind the first is four-sided and one or more eight-sided. In the present species all behind the first are six-sided. The vertebral scutes are about one half wider than long. Their sides are not strongly angulated. The epiplastral lip is 25 mm. Ieng and 58 mm. wide at the base. The hinder lobe is notched, as usual in the genus. The gular scutes extend backward on the entoplastron. The pectoral scutes meet along the midline a distance of 15 mm. ‘Their outer ends are greatly expanded. TESTUDO HOLLANDI, sp. nov. (PLATEs V-VI.) The catalogue number of the type of this fine species is 1561 of the Carnegie Museum. The specimen was discovered by Mr. O. A. Peterson in the upper beds of the Loup Fork, near Running Water, Sioux county, Nebraska. The shell alone is preserved, and this is considerably crushed downward. | The carapace has a length of 305 mm. and a width of 280 mm. The first neural is oval; the second and the fourth, octagonal; the third and the fifth, tetragonal. The proximal ends of the second and the fourth costals are narrow, the distal ends very wide; while the proximal ends of the third and the fifth are wide, the distal ends narrow. ‘The plastron is 325 mm, long, being thus longer than the carapace. The epiplastral lip projects 28 mm. beyond the gulo-humeral sulci and is 70 mm. wide at its base. Its greatest thickness is at a point 45 mm. behind the anterior border. The entoplastron is 61 mm. ong and 58 mm. wide. ee Hay: DeEscRIPTION OF NEW SPECIES OF TURTLES. Y9 The first vertebral scute is considerably narrower than the others. The lateral borders of the second and the third are angulated. The pectoral scutes occupy 24 mm. of the midline. | No. 1570 of the Carnegie Museum is likewise referred to this species. It was collected by Mr. Douglass at Cold Spring, Montana, in what are regarded as Loup Fork beds. This species is dedicated to Dr. W. J. Holland, Director of the Carnegie Museum. TESTUDO EDA, sp. nov. (PLates VII-VIII. ) The single specimen which forms the type of this species was col- lected by Mr. O. A. Peterson, in Loup Fork deposits, at Running Water, Sioux county, Nebraska. The specimen consists of the cara- pace and the plastron, from both of which a portion of the right side is missing. The catalogue number of the specimen is 1535. The shell appears to have been originally somewhat depressed, but this depression has been exaggerated during fossilization. The length of the carapace is 450 mm. The width was 380 mm. The areas of the second and the third vertebral scutes are somewhat concave. On the area of the first there isa low and elongated boss. The lower border of the pygal is somewhat drawn in. ‘There are in this indi- vidual only seven neurals. The first and the third are four-sided ; the second and the fourth, eight-sided ; the others, six-sided. Most of the costal plates behind the first are wedge-shaped, the narrow ends of the second, fourth, and sixth being directed upward ; the narrow ends of the third and the fifth, downward. The first and the fourth vertebral scutes are narrower than the others. The plastron lacks but 5 mm. of being as long as the carapace. _ The epiplastral lip is 55 mm. long and go mm. wide at the base. Its inferior surface curves upward like a sleighrunner. The pectoral scutes occupy 25 mm. of the midline. This species is named in honor of Mrs. Eda Peterson, of Pittsburgh. STYLEMYS NEBRASCENSIS Leidy. Notwithstanding the fact that hundreds, perhaps thousands, of shells of Stylemys nebrascensis have been collected in the badlands of South Dakota, Nebraska, and Colorado, no skull of the species has hitherto been found in connection with parts of the shell. In Prince- ton University there is a portion of a skull, and there is said to bea 20 ANNALS OF THE CARNEGIE MUSEUM. skull in the collection made by Dr. Baur for the University of Chicago ; but with neither of these skulls was there any of the shell. ‘The im- portance of the presence of the shell will be realized when it is re- membered that the species was based on the shell and that there are a few species of the closely related genus Zes¢udo in the same formation. Fic. 9.. Superior view of skull of Stylemys nebrascensis Leidy. 1. Fic. 10. Inferior view of skull Stylemys nebrascensis Leidy. +. From the parts of the shell and limb bones the writer has been con- vinced that S¢y/emys belonged, without doubt, near Zestudo and that it was at the same time a distinct genus. What is now known about the skull confirms these conclusions. The length of the skull discovered by Mr. Douglass is 38 mm.; the width at the quadrates, 30 mm. The interorbital space is 10.5 mm. wide. ‘lhe antero-posterior diam- eter of the orbit’ 1s tr timiae ae vertical, g mm. The _ pterygoid Fic. 11. View of right side of skull Teglon where narrowest is To mm. of Stylemys nebrascensis Leidy. 1, wide. The palate 1s deeply exca- vated and a low ridge runs along its middle line. The lower jaw remains in its natural position and hides the crushing surfaces of the upper jaw. The symphysis is short. As seen from the figures (Figs. 9-11), this skull corresponds in all essentials with that of Zestudo. It is idéntical, too, with the skull preserved at Princeton, which displays the crushing surfaces of the upper jaws. These present the same arrangement that we find in the living genus Gopherus (Xerobates, Agass.), there being especially a longitudinal ridge at the symphysis of the premaxille. On examin- ng closely the Pittsburgh specimen this ridge is seen to be present. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate Ill. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate IV. Interior View.of Shell of Zestudo ‘nusittata Hay, 745 an ed $4 Nes ay i sé ANNALS CARNEGIE MUSEUM, Vol. IV. Plate V. Superior View of Shell of Testudo holland? Hay, 2. 9 ; Weta) © d j . : 7 SS + ‘= : j . as, Di ee ate SUN alive id lye ge we aN Vad x jr. . 2 iy ’ ~ PARAS hyn. den, Pring Inferior View of Shell of Zestudo hollandi Hay ANNALS CARNEGIE MUSEUM, Vol. IV. ANNALS CARNEGIE MUSEUM, Vol. IV. Bani ek “C éa=¥ ‘- a Flate VIII ANNALS CARNEGIE MUSEUM, Vol. IV. Inferior View of Shell of Zestudo Ede Hay, , Ill. THE MIOCENE BEDS OF WESTERN NEBRASKA AND EASTERN WYOMING AND THEIR VERTEBRATE FAUNZ4.! By O.. A. PETERSON. During the summers of the years 1900-1906 expeditions from the Carnegie Museum have been regularly employed each year in col- lecting fossils in the Tertiary deposits of northwestern Nebraska and eastern Wyoming. Much material has been accumulated, and the collections made contain many species new to science. In view of the general interest at the present time in regard to the geological position of the sediments which overlie the Oligocene of this region it has occurred to the writer that an illustration representing an ideal section showing the formations in proper succession, accompanied by a list of the vertebrates obtained in each, would be of assistance to the student of paleontology. A short preliminary description of new species and additional notes on some forms hitherto little known will follow each list. Statements and observations made in this paper are based almost exclusively upon the collections belonging to the Car- ‘negie Museum, as they were made either directly by the writer or with complete knowledge by him of the geological horizons in which the individual specimens were found. A much more extended list might have been given by adding genera and species which have been col- lected and described by others from time to time, but as the correct geological position of these might, in some cases, be somewhat in iy doubt, it has been thought best in the present paper to mainly employ data which have been obtained at first hand. Since the early Government surveys under Dr. F. V. Hayden, dif- ferent writers have, from time to time, written upon the lithological and geological characters of this region, and they will in the present ‘I take this opportunity to thank Dr. W. J. Holland, the Director of the Car- negie Museum and Curator of Paleontology, for his revision of the manuscript of this article, and for granting me the privilege of describing the splendid material with which it deals. To Mr. Earl Douglass I am indebted for many kind criticisms and suggestions. The drawings for the illustrations were made by Mr. Sidney Prentice. . O. A. Peterson. 21 ~ ay) ANNALS OF THE CARNEGIE MUSEUM. paper be alluded to only when necessary to do so in the interests of perspicuity. In 1902 Mr. J. B. Hatcher’ subdivided these beds and mentioned a number of species belonging to their faunzee. While the subdivisions. which Hatcher proposed are here followed in the main, it may be stated that the horizons determined by him are difficult to distinguish on account of their lithological similarity. This is especially true of the Gering and Monroe Creek beds. In Fig. 1 each horizon is defined far more clearly than in nature in order to make the subject plain to the reader. The species given by Hatcher (7. ¢., p. 117) 1m one on two cases are incorrectly identified.® Referring to Fig. 1 attention should be called to the fact that the upper fifty or sixty feet of Squaw Butte appear to represent the Harri- son beds. ‘The Monroe Creek beds, however, run insensibly into the latter and the dividing line is difficult to determine except for the fact that Demonelix, which is characteristic of the Harrison beds, occurs in the upper strata. Along the line of the section Niobrara River (locally known as the ‘‘ Running-water’’) does not cut through the Harrison beds, and the thickness of the Monroe Creek beds and the Gering horizons are here only conjectural. The sedimentary mass apparently decreases in thickness southward and eastward, so that if the lower horizons (Monroe Creek beds and Gering beds) are present at this point they are probably quite thin. The sandhill region, south of the Niobrara River, represented in the illustration is a narrow strip extending east and west. The deposit is of late origin and has the usual zeolian character met with in the more extensive sandhili areas further east in the State of Nebrasba. This deposit rests unconformably on the Harrison and the Upper Harrison (Nebraska **) beds. Spoon Butte is located in the southwestern part of Sioux County, Nebraska. It isa long and narrow elevation with deeply eroded sides and a flat top. The long axis of the butte is directed nearly east and 2 Proc. Am, Phil. Society, Vol. XLI., pp. 113-131, 1902. 3 Hatcher wrongly identified Protolabis for Oxydactylus, and Cyclopidius for Mery- chyus. No Cyclopidius has as yet been found in the Upper Harrison beds. 34 From a verbal statement made recently by Professor H. F. Osborn it now ap- pears that the beds which Professor W. B. Scott referred to the Nebraska beds are of later origin than those, which, in former papers, I have referred to that horizon. This horizon, therefore, may be called the Upper Harrison beds and in this paper will be referred to under that name. ee a ee PETERSON: MIOCENE BEDS OF NEBRASKA AND WYOMING. 23 west, the western extremity lying within the State of Wyoming, in Laramie County. The summit of Spoon Butte, which is represented at the top of the section (Fig. 1) is capped by a hard pinkish-gray sandstone, thirty-five to fifty feet in thickness. In this hard cap of sandstone, which is regarded as the top of the Upper Harrison beds, our party found no fossils. The base of this elevation is composed of the Lower Harrison beds. ind Wate , Running later) Niobrara River Sa a SSS SS SSE Page sama 2S SS INC Byee AA, q Fic. 1. A section of the Lower Miocene of western Nebraska and eastern Wyoming. THE GERING BEDs. The Gering horizon forms the contact between the Oligocene and the lower Miocene formations over extensive areas in eastern Wyoming and western Nebraska, and perhaps also parts of eastern Colorado. At Squaw Butte, the northern limit of the lower Miocene, where the sec- tion represented in Fig. 1 was obtained, the Gering sandstones rest directly on the Leptauchenia clays, which compose the uppermost horizon of the Oligocene formation. No fossils were found in this horizon in the neighborhood of Squaw Butte, but farther east, near Sand Creek and near Chadron, Nebraska, a few remains were found. List OF THE FAUNA. ? Mesoreodon. Leptauchenia.* THE MONROE CREEK BEDS. In the vicinity of Squaw Butte it is quite difficult to separate this horizon from the underlying Gering beds by lithological characters alone. While remains of Leftauchenia were found in the lower part 4The material referred to Lepiauchenia' differs but slightly from that referred to L. decora Leidy, from the Oligocene, and should not in my judgment be regarded as belonging to the genus Cyclopidius, 24 ANNALS OF THE CARNEGIE MUSEUM. of this horizon, none were found at the top by the parties from the Carnegie Museum, but it is quite probable that true Cyc/opidius may occur in the upper levels. Other changes of the fauna occur which help to identify the horizon. LIST OF THE FAUNA. ? Diceratherium. )} These genera are found in the Mesoreodon. lower levels of the beds and may Leptauchenta. ( with equal propriety in part be re- Cameloid fragments (Genus ?). | garded as belonging to the fauna j of the Gering horizon. Canid fragments (Genus ?). Euhapsts platyceps Peterson. Promerycocherus carrikert, 0. sp. Phenacocelus typus, gen. et sp. nov. Protomeryx cederensis Mathew. ee These genera and species are from the upper Monroe Creek horizon. =~ Nothocyon lemur Cope. J DESCRIPTION OF NEw MATERIAL. Diceratherium sp. indet. A few fragments of jaws and teeth were found in the lower level of the Monroe Creek beds which evidently belong to the genus Dicera- therium. The incisor is large and apparently occupied the usual pro- cumbent position which is characteristic of the family. The specimen is probably that of an earlier type of the genus. Mesoreodon melagodon sp. nov. In the collection of the Carnegie Museum are five or six individuals upon which this species is founded. The material was found in the middle and lower Monroe Creek beds near Squaw Butte, Sioux County, Nebraska. All the specimens are crushed and the true contour of the skulls is consequently lost. The front of the skull and lower jaws (No. 1325) of a young indi- vidual, but with all the permanent teeth in position (see Fig. 2) are selected as the type. A second specimen of a fully adult animal (No. 1323) is chosen as the paratype, and consists of the back part of the skull, the lower jaws very nearly complete (see Fig. 3), fragments of lumbar vertebrze, fore and hind limbs, and feet. 7 - we PETERSON: MIocENE BEDS OF NEBRASKA AND WYOMING. 29 The principal differences between Mesoreodon chelonyx, the best known species described by Professor Scott,® and AZesoreodon megalodon Y ff, Wy Yh. ‘Hy Yuu fy i, Ht HAE eum \ La \ muy Maer ) . i) ‘ | NB PrN ra AN \ = a = AKS EE _ YZ Hy GE a Hf Me y if Y y { q ‘ Mh 4 (iif ‘i | / Ld eae ee We | / ry rime Dra NY 7 PIG. 02. 1 nat. size. Type. No. 1325. tens hp 7 "diy 1] mw) i } Wan t a iu \ WH! gs VY S ee WSS | ee a of Uh ‘WS agree SS Ee . Ss ee — FEE aS == ff S = E= a ee 4 Fic. 3. External view of left mandible of Mesoreodon megalodon. } nat, size. Paratype, No. 1323. may be stated by saying that in the present specimens the teeth are relatively heavier, the molars having a slightly greater antero-posterior diameter, the dentition being more crowded ; the occiput probably 5 Am. Nat., Vol. XXVIL., p. 661 (1893) ; Trans. Amer. Philos. Society, Vol. XXVIIL., pp. 125-145 (1894). 26 ANNALS OF ‘THE CARNEGIE MUSEUM. lower ; the muzzle heavier; the nasals longer; the infraorbital fora- men placed further forward; the mandible deeper, with the angle more strongly produced downward and not so strongly produced behind the condyle. The limbs are also shorter and comparatively lighter than those in Mesoreodon chelonyx, as is indicated by the fragments of the limbs found with the paratype (No. 1323). MEASUREMENTS. Type. Paratype. (No. 1325.) (No. 1323. ) mim. mm. Distance from anterior border of orbit to the incisors........... 123 Transverse diameter of muzzle at base of canine.................. 60 Antero-posterior diameter of premolars...............000 sees some 60 Antero-posterior diameter of molars + and 2.0... cesses eee 44 Antero-postérior diameter of M2... ....J.saseeaeudees eee acim 30 Vertical.diameter of mandible at P=... Sig eeaees oe eeeee 34 32 Vertical diameter of mandible at anterior part of Mg ........... 42 Length of premolar series’... . 0... ..adseseas tee eee ee ee 59 56 Length of molariseries:).. 4c. 1 acc selencanmes eoacen tine 2 ame ene eee eS Antero-posterior diameter of Me ....4..gecte tannutucnneeaee een 34 Promerycocherus carrikeri,® sp. nov. (PLATE, EX) (Type. No. 1080 Carnegie Museum Catalogue of Vertebrate Fossils. ) This species is found in the upper Monroe Creek beds and is most nearly allied to Promerycocherus chelydra Cope. The skull is brachy- cephalic. The dentition, of the usual merycoidodont type, is 13, Ct, P4, M3. On direct comparison the skull of P. chelydra is seen to differ from the series in the Carnegie Museum collection by a number of important. characters. In P. chelydra the zygomatic arch is less robust and does not extend below the horizontal line of the dentition, the superior border of the premaxillary is more oblique, the frontal region is less convex, the vertical diameter of the cranium is less, the tympanic bulla is smaller and differently shaped, the posterior nares are located further back, the postglenoid process is smaller and more widely separated from the paroccipital process, the dentition is relatively somewhat shorter, and the type specimen is of smaller size than the average sized skulls of Promerycocherus carrikert. Among the more 6In recognition of Mr. M, A. Carriker, Jr., who pointed out the locality of the specimens to the writer. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate IX Group of Skeletons Representing Three Specimens of Promerycocherus carrikeri Peterson, as they were found in the matrix. The type is specimen No. 1080, The Group is installed in the Gallery of Vertebrate Paleontology in the Carnegie Museum, }. 7. ion ee Ss, tly > ae . 6 5 J Ds tol ‘road pes D> Beda cn ; . 5 a ‘ a bi . MOMENT Io wageniosa) sont BnitasesiqeH edtoteleda % ’ \ ee bs 7 PETERSON: MIOCENE Beps OF NEBRASKA AND WyoMING. 27 important characters enumerated above are the greater vertical diameter of the cranium in P. carrikeri, which is due to the greater elevation of the region back of the orbit. The temporal ridges and the sagittal crest are very prominent, and the zygomatic arch extends downward in a manner suggestive of F/otherium. The vertical diameter of the skull is not greater than that in P. chelydra, except as it is increased by the elevation of the occiput and the downward thrust of the zygoma. ‘The group of three skeletons in the Hall of Vertebrate Paleontology is mounted in the original posi- tion in which the animals were found in the field. The specimens are among the most interesting collected by the writer and will be more fully described in a later publication. MEASUREMENTS, mm ae cree MeareE IN PINSON Noreen xf asa sooe wa ache dies scee'sa Fn secve since nedcapnunnses 316 ienethirom anterior border of orbit to tip of nasals. ....0.4,...... o-.se000... E52 Length from anterior border of orbit to the occipital condyles ............ 174 eet CAT OGG oo ale oe ceh os adg vate ndcee ake sesesanseavisssesintedaves L75 Distance from M2 to the occipital condyle ..............cc0ec08 ceeeeesen aes 132 Sateatest transverse wiameter Of the skull)......2....6..c2.c0ss.sececus nets onses 295 wransverce diameterof the. oecipital condyles...) 002-26 .2. se-ceeee sce ceneee 67 Greatest transverse diameter of brain cavity.......5.....0c0sse0sees cba teamiiae 85 Transverse diameter of muzzle at base of canine.................sceeseevees 1. OH PEUerO. DOSEETION Cidm@meten Of TUCOFDIt: ous. {hihi ca cad's cos .-saseeseedetstdene de 35 DPCLHMCAL GIAPICLEE OL EBCIOL IONE . 0. -cuses nnsveesecicvsceeses ep A eee toes st 35 ereatest vertical diameter of the zygoniatio arch ...._.......scsiecseesocennss 165 Vertical diameter of jugal below middle of orbit... ............2:.s00se0000 40 ee eer ett ON IOMINTIOM . 280). Sar0y aalseia yds fame) yan + dada dd». a0igdy sud vos aveoas is ae PAM AONE MRC SONS TO Wn Foie doa wcican dustin a degece os wan caus ssdewaaleavess 105 ere TM Tae ry ot ae Salta te Sax ow ce waginis. ebach $ ceaine’s to.n oir me 75 imkere-postemor diameter Gf canine near base..........¢.+0.0soscsesusssceeee 16 irausverse diameter of canine near base .....5.6........ccencdscccccese ve couces 19 PenO- POsterior CiamMeten OF Pee esse coeds doseokostiwawer there suedeeuse 16 Ppuusere Purtetior ciamibber Of P20... 6. s..0scceswece's aclen oh eiveveawass ex scues ee 18 RamreRO-PRReENO EATIICLEE Ol eo... occa deccw sun edaans vascice sbabeeee-ssaseces 18 mmtero-posterior diameter of P4,...........6.... TREE ie ene ae ee 16 pemrere-posterior Giameter OF Meo... ..2.-cnesecas dennaceceecscss sees Seats aa oe PenenGl asber Ton CHAMICICT OL Ne cs ca cneves sawsesbensesasourucnsteaces 24 PeteLo-postcrcn digmmeter Of MS sae. (ines elec ee sedan car o0t 32 eee ey TidemINCU Mie th t2 se na ale Uc erbiay Sohieyic o6isesh-s ve woe ud. 2-0 adddenveedndtnn 255 menetn ot alveolar border of mandible, (2... 5... 0.sesccsae+.00sncybassvcdabane 183 merical, diameter O1anele at (ie. CONGYIE ..... 2. ...c0.cesececssuns ene anes erent 120 Vertical diameter of angle at the coronoid process...............sseecesceees 118 Petes) Ctamacter or. miamdible at Mae. o. ccs. casas cov ecevecsssncengnenr ane 58 28 ANNALS OF THE CARNEGIE MUSEUM. Veftical diameter of mandible at canine. :..:.. 0.2 eta eee ae 57 Length of inferior dentition, ..<.5....c..i.0ca0..sisceanpe asa teen 180 Distance from incisors to My:.......0..a00s00-.iaeedaamaiae gama eae 87 Length ‘of molar Series. .......... 0002+ 00c=-s -snguap neh ae ites een 85. Antero-posterior diameter of Pi... /...ces>ocnshpen oeeehaemann nes seen 17 Transverse diameter of Py... 0. vv x2.c0nsts«vens os benep ease semen ene amen 13 Antere-posterior diameter of Pig. iv. Lu cece nae sae one cee nee ER ener pan 17 Antero-posterior ‘diameter of Py..2.5. svn: «supa ieee ekeie® a inane: ape 16 Antero ‘posterior diameter of Pci... 2pns-anaecpemeane an one ean eee 20 Antero-posterior diameter.of M7 «2. ~o.mse¢emane-stanseu h-ueaen eee eee 20 Antero-posterior diameter of My .....- -.cscarexanaeusnsenne ee tee eee 22 Antero-posterior diameter of M ....... yu es = aN, \ NE lle a Tiel so ob en — = S 1 fy ‘ nN A efilh ¢ i f' of Aly: aq SS =e i : = Fic. 7. 1, side view of skull of Merychyus harrisonensis. 2, top view of skull. 4 nat. size, Type. . No. 1341; with that of MWerycotdodon culbertsonit and Phenacocelus typus. The tympanic bulla occupies a large area of the base of the cranium and has a peculiarly depressed and flask-like form. There is a conical swelling on the postero-external angle of the bulla, and a deep emar- gination or pit near this eminence for the tympanohyal which constricts the otherwise broad tube of the external ear. The postorbital process is of moderately large size and the orbit was apparently closed posteriorly. The alveolar border of the maxillary is lower even than in JMery- coidodon culbertsont. The infraorbital foramen is double and is located oe = PETERSON: MIOCENE BEDS OF NEBRASKA AND WYOMING. 39 over the back part of P3. P+ is similar to that in Merychyus elegans, and as in that species, the protocone is shifted partly, by wear, to the anterior portion of the crown. The second premolar is very convex externally ; the base is slightly corrugated, but there is no true cingu- lum. Except the usual small excavation at the antero-internal base, the tooth is so much worn that all the characters of the grinding face are entirely lost. The external face of P3 is slightly concave antero- bole nat. size. Type. Fic. 8. Palate view of skull of Merychyus harrisonensts. No. 1341. posteriorly, and there isa decided cingulum ; the tooth is much worn. P+ has a strong cingulum internally, while externally there is no cin- gulum. The external and internal cingula on M2 are only faintly indicated. M2 has a long postero-external heel ; the tooth is longer than broad. ; MEASUREMENTS. mm. Greatest length of the skull ............:-ssssserreserseceonnersrescceserseecerses 196 Distance from premaxillary to postorbital process of frontal. ......sse000s 107 Distance from postorbital process of frontal to the inion.........-+-seeeee 88 Distance from the premaxillary to the occipital condyles .........+++++++++ 176 Length of the alveolar border of the maxillary..........ceeeseereeeeeeeee ees 96 Distance from posterior end of alveolar border to occipital condyle..... $2 Greatest transverse diameter of the brain Cavity..........-seeseeeseereeen scene 54 Transverse diameter of frontals between DUES ooo aoc clas Conn gsaes soe ebs smane 70 Transverse diameter of occipital condyles..... PLM det . eectackatassdovnann Gas 35 Vertical diameter of occipital condyles..........-..ssssrsereererreersseereeeees 13 Antero-posterior diameter of tympanic bulla.........-.seceesseeeeeeeeeeeeneens 23 18 Transverse diameter of tympanic bulla....:.....:sesseeeesseesertmerseseesene ss 40 ANNALS OF THE CARNEGIE MUSEUM. Vertical diameter of tympanic: bulla: sivas .\ lacie meat eokenteree epee eee 13 Total length of the dentition. .2: 5. 156001 sancanes is acostss ting ore aca s «athe ea 95 Length of premolar series......... wipe 8 viasasth\pe'ale's “ete ie Ogee ee cei fete 35 Lepeth of molar series... .<.c.01 doves sche aaaen Maem enieenes ene ee 45 Anterd-posterior diameter of canine at Das; csescms ees aeen eae eee 6 Transverse diametét of canine atibasé J). .Wie.secanee aes knen tee eee 8 Antero-posterior:diameter Of P41... 12.1.2 jswessgneaundpeeuedeveenecahe ay eeeereea 8 Transverse diameter.of PP) 4.7 cAn22cic ans eae ee a 4 Antero-posterior diameter of P2,,...5:.\ceums coos steanneescens Si ecsosetaneenae fe) Transverse diameter of P2.2/.. sscsicdcueaeeseavens coo eee eee eee eee 7 Antero-posterior diameter of P23... c..2:susece: = aera see a eengee etree aay 10 Transverse diameter of P®, «. i... ccsuasaccdiemgues | ease? oe eee ee ae ee 9 Antero-posterior diameter of P42...) 2. castes wstagsa 0 eee eee eee 9 Transverse diameter of P4:.....5..0\ sasesaviaeesteanene ences eee ee 12 Antero-posterior diameter of MA 0. 20. casaetce- a cpcesns «sep ee eee Seana 12 Transvérse diameter of M74, .. s2:sc.dsnGieaeaen tek oaenieae see eee eee 14 Antero-posterior diameter of ME25 2022 Sa osecwess tee cae chet eee eee re 15 Transverse diameter of M2.....22.caccnt0e clita #fo4-c0 aes oe el 15 Antero-posterior diameter of ME, vicsnc ca enten oeeames dite ee sae se eae 19 Transverse diameter. of M2.7x,,scers acc seen cea tea ca tcnstales See ae ees 16 Merychyus, sp. indet. The specimen (No. 1284) described below was found in the Lower Harrison beds on Squaw Butte, Sioux County, Nebraska. The remains consists of a nearly complete tibia with the middle part of the shaft of the fibula adhering to it; the astragulus, cuboid, navicular, cuneiform, and the greater part of the metatarsals. Inasmuch as no part of the cranium or teeth were found with the specimen the writer hesitates to apply a specific name, though the material may represent a new species. It is provisionally referred to the genus Merychyus. The tibia is quite long with a slender shaft ; the cnemial crest is somewhat less developed, and does not reach as low down on the shaft as is the case in Phenacocelus typus. On the posterior face of the shaft is a prominent ridge which extends from the postero-fibular angle of the proximal end, first obliquely downward, then more directly downward, and disappears 40 mm. from the distal end. This tidge is almost entirely wanting on the tibia of Phenacocelus typus, while that in Merychyus minimus™ is fairly well developed, but in a less degree than in the present specimen. ‘The internal malleolus is directed vertically downward and the groove for the internal condyle of ‘To be described later in this paper. ANNALS CARNEGIE MUSEUM, Vol IV. Plate XII. Yi Y, ve 7. . LE if\\ HK) Yh \ Wye Uf UY cit m o\ f) Skull of Stexomylus gracilis Peterson, 1. Rr PETERSON: MIOCENE BEDS OF NEBRASKA AND WyomING. 41 the astralagus has a fore and aft position very like that in AZerychyus. The fibula had a complete shaft, judging from the comparatively large fragment adhering to the shaft of the tibia. The astragalus and the cuboid are high. ‘The navicular and cal- caneal facets of the cuboid are very nearly of equal size. The facet on the cuboid accommodating metatarsal five is located on the tibial side of a small lip on the outer angle, which extends below the artic- ular facet for the fourth metatarsal. The palmar hook of the cuboid is quite robust. The navicular and cuneiform are injured by erosion, but enough of them is preserved to show their close similarity to the corresponding bones in Merychyus. ‘The second and fifth metatarsals are relatively longer than in Werychyus minimus and not nearly so heavy as those in Phenacocelus typus. In comparing the second metatarsal (which is broken off at the distal epiphysis) with the cor- responding bone of Merychyus minimus (No. 1331, Car. Mus. Cat. Vert. Foss.) it is seen that while the bones are of the same length the one here described is much the heavier of the two. It is also notice- able that metatarsals three and four are proportionally smaller than those in Merychyus minimus, and, when complete, no doubt they would be shorter. The proximal articulation of metatarsal five is very small and fits neatly into the overhanging lip on the fibular angle of the cuboid referred to above. MEASUREMENTS, mm. Puen lene onthe fraoment of thé tibia. 2.05 3..6..vecse.ctiscdevecscee xsuree 127 a panisverse diameter Of distal End Of tibia... .cecjcsceccceeccecussceccdeavacens 19 Embero-posterior diameter of distal end of tibia.......s0sesecccerceesersacses 14 eee EOL ASIACAINS 2... .5cccdnescnedenadencasdevagcrodene vanes Stave 25 Greatest transverse diameter of astragalus......2.< sseocccsscscseetivwesessts 14 Wertical diameter of cuboid, anteriorly,.........-...ss00cee A et apiticohtatine nee 15 fixamey crac diametermot cuboid; anteriorly, ...icc.cieccces canceevcssdisendeceee 9 Werueal ciameter Of mavicular, anteriorly .. i... jeceresdarsecdhenciced “edenie 8 Werlical diameter Of Ecto-meso-cuncifOrms ...;..6....0cceveccesccvess cerconecs 6 Transverse diameter of tarsus at head of metatarsals.............0.seceeeees 19 Length of metatarsal II from proximal end to epiphysis of distal end... 44 Stenomylus gracilis, gen. et sp. nov. (PLATE XII.) (Evpe. No. 2610,'Carn..Mus. Cat. Vert. oss. This remarkable cameloid form was found by the writer in the Lower Harrison beds on the Niobrara (Running Water) River, Sioux 42 ANNALS OF THE CARNEGIE MUSEUM. County, Nebraska, in 1901. The type specimen consists of the greater portion of the base of a skull with the lower jaws complete, the limbs, the feet, and several cervical vertebree. Only the skull and lower jaws are free from the matrix, and these are illustrated. The skull and lower jaws display characters which are entirely differ- ent from any cameloid living or extinct known to the writer. Certain cranial characters are quite suggestive of the Oligocene genus //yfz- sodus, but the limb and foot structures are distinctly cameline. PRINCIPAL CHARACTERS. I, Cz, P29, M2. The more striking characters of the skull are the position of the posterior nares with relation to the pterygoids. The latter are located in the usual position, but they are very heavy and are completely united at their origin, their extremities diverg- ing downwardly and outwardly. The posterior nares are large, ovate in outline, and are located well forward with the anterior border op- posite the posterior part of M+. ‘This is well shown in the illustra- tions. The tympanic bulla is of moderately large size ; it is well co- ossified with the paroccipital process as in the camels generally. The basicranial axis is strongly angled. The occipital condyles are quite large and there are large accessory facets on the basioccipital. The Occiput is rather low. The lambdoidal crests are broken off superi- orly, but enough is preserved to show that they are of considerable prominence. ‘There is no sagittal crest. The brain cavity is large and the frontals have a great transverse expansion between the post- orbital processes, which give to the orbit an oblique outward position. The alveolar portion of the maxillary bone is very high for the ac- commodation of the extremely hypsodont teeth, the crowns of which, however, emerge but little below the alveolar border. ‘The infraor- bital foramen is above M+. Anterior to P2 the skull is unknown. | P3 is small; it has a simple crown and two roots. The ectoloph of P* is smooth and nearly straight antero-posteriorly ; there is, on the antero-external angle, a well developed vertical rib; the internal border is evenly rounded from before backward. ‘The molars, espe- cially the second and third, are of great antero-posterior diameter, while transversely they are much compressed. ‘The vertical ribs on the antero-external angle of M2 and M3 are prominent; otherwise the external surfaces of the teeth are smooth. The intercrescentic cavities of the second and third molars are large and deep. ‘The pos- PETERSON: MIOCENE BEDS OF NEBRASKA AND WyOMING. 43 terior crescents of M® are but little worn, indicating a rather young, though fully adult, individual. The principal characteristics of the mandible are seen in the shallow temporal fossa, the delicate and backwardly projecting coronoid pro- cess, and the rapidly decreasing depth of the horizontal ramus of the jaw from the angle to the symphysis. The incisors are quite large ; they are closely crowded, and their crowns are spatulate. The in- cisors, canine, and P; form a continuous closed series. The canine is incisiform and P+ has a single root and a high trenchant crown. P, is isolated from P; and Pz by diastemata; the tooth is small and _two-rooted. Pz is also a small two-rooted tooth which is unworn in the type specimen. Pz is compressed laterally, but has a normal an-- tero-posterior diameter. The molars increase rapidly in length from before backward ; they are much compressed transversely. ‘The limbs are long and slender and the metatarsals are codssified half their length from the proximal end. The navicular and cuboid are separate as in the camels. The metacarpals are codssified in a less degree than the metatarsals and are nearly as long as the latter. } MEASUREMENTS. mm Distance from P® toand including the occipital condyle..................45 152 Distance from M2 to and including the occipital condyle................... 76 Meuuealdiameter of the occipital conGyle 2.4.4, ccccssccseessecssecocceeecese 20 eaasyverse diameter of the occipital condyle..;...... 2.2.0. se: scasesesscouns 32 Transverse diameter of the frontals at postorbital process, approxi- ee ty Ene laiehs, foie umes Asia ones Saka Ga sees sae ure cde okeEC LENT aneae 95 Puree posterior ctameter Of PS. oi: ..vspax ta nseyenies seas densds vocertecdcaree 6 nMSMOnR eT MRCteE OE PT, vo cca ccs ode ovaceeendeenaesateusswbeaas neveeseeeree 3 SaPeRarpbEe Rial Giammeter Of PM. 3. .cycse tsnivnctusnersapede odeaesscens¥ one sot 10 meamametse ciameter of P* posteriorly... occ cecessonsecnarndeateneudveracssene 7 maton ma meriar ciameter OF Me. aie .c cies pa esis dpi sniseeeeudr sendy ede sebast 15 eee oiatpeter OF NTA.. | |... ice icin nesnnavnrdespbentaneees acre t ates a rae fe) PeMreEO-POntenOn diatieter Of (NEA coi iccccse cag eccete vs sex scm cas aan cans ensse nee 25 rwmwotmerciarieter Of Nl2 s/s s.cap avevesescecesncederscscsescnecuesnswongesas fe) mmera-postenor diameter Of Me ir... ....scececccscecseessencesacevccnseceranse 26 Bir eB GISe NAMMCTCR OF IE Din ck cove ener ccncniedamcinacccescaqeneseucavévevenren 8 ieremtest fength Of/the Mandible. oo... J. vesecceaseunendanetengdneansccesens as 184 DSH OH fasy AL-COLON@IG PLGCESS ic. .o4 cin cscne pea ssenseeescccsespanoerscsnnnens 97 Depth of jaw at posterior PON, RES wwncc ap pcccgad Geeusteuhasyeitnadse's 47 Pepin of jaw at diastema in front Of P2..............cccssceensncrecnscnncsees 18 Peer MIRETION COMCMOM 2... coisas ccatcsccensseecseccevevceccccesdencesvause 137 Beeteppsterior diameter Of PE... io. 6. cewensevcosncaneccesccvasessavnpecsuse 5 44 ANNALS OF THE CARNEGIE MUSEUM. Antet o-posterior diameter of Py. i...00. (200s. sientedeens sae eet 5 Antéro-posterior diameter of Py....0......c1vaceseoyecen enemas MOE RR SE 5 Antero-posterior diameter of Pz). .... 51.2... +. same yore ee ane ee 9 Transverse diameter of Pz 2. ..<..sc.sdeenswssa tae anee eee eee ae armas ean 4 Antero-posterior diameter of My li. .c.2 i and 5 .../0.ivee..tenntena uel euaeeaineten ee pave: eaee eee re 22 Antero-posterior diameter of carnassial tooth ................cccsseeseceseeeeees 15 Transverse diameter of carnassial tooth at base..................seceeeceeeeeees 8 Antero-posterior diameter of heel of carnassial tooth.....................0000 4 A PRovISIONAL LIST OF THE FAUNA OF THE AGATE SPRING FossIL QUARRY. _Parahippus. ; Dicerathertum niobrarense Peterson. Diceratherium cooki Peterson. Moropus elatus Marsh. Dinohyus holland Peterson. 2? Merycocherus. Merychyus elegans Leidy. Amphicyon superbus sp. Nov. LVothocyon annectens sp. nov. Diceratherium niobrarense Peterson. (Sczence, Vol. XXIV., No. 609, pp. 281-282, 1906.) (PLATES XIMI. and XIV.) (Type. No. 1271 Carnegie Museum Catalogue of Vertebrate Fossils. ) In comparing this species with the type of Diceratherium nanum Marsh,” and also with the complete and well preserved skull of that species (No. 7324) in the American Museum of Natural History,” certain differences are observed. In the type of Diceratherium nanum, P+ is relatively larger than that tooth in Dicerathertum niobrarense, the nasals are somewhat more pro- duced in front of the nasal horn-cores, and are also apparently heavier than in the latter species. In the specimen in the collection of the 15'The front of askull and Jower jaws in Yale Museum, No. 526. 16 This specimen was referred by Cope to the species Diceratherium nanum, and is apparently identical with the type so far as comparison can be made. £ SYOSIa}JI ISUIADAGOLU ULNIABYJVAIIIGZ JO |[NAS FO MELA IPIS (Wwe YOR Xe il il Oe iy Zp; pater iia Zod <— De ze = , i) Wy NA YG Neh ene Wo of HANS INS i | S SSS en S Q) ene \\ \\ | WSS ST . Ree \\) } {SS Se Nae \ SS —— Wi? 2S = ———— ; " } 1 yh] / TMX. #421 Al CA ‘WNaSNW SISSNYVO STVNNV E SUOSAIJO ISMIADAGOIU ULNLAIYJVAIIZTT JO MOA : ee oo he) é& 4, Y “AIX 81d "AL ISA ‘WAASNW JISANYVOD SIVNNV *Y SUOSTIII 7Y¥OOD MUNIAIY]DAIIIGT JO UOIUAC] AOLMaFUT JO MALIA UMOAD puv sMUf IOMO'T PUP [[NYS JO MIT. OPIS tnt IZ @ i Y Wwe ' Nee Zh, ~~ o MAd t/ \ Wo 3 y = - a Ans é : Ca ] \ \ H "i ie oF y t Soren . N zs } LG / WAY ’ Z q . 5S Lp oy) Whe lhe & | = y SSS35° % " 3 —Y a , ne HED. GXeS LU Ip . SEY’ a ’ S g Th / - (a AT hn S AINA LS <—————g = is SS AVA a, ——== Y A, — "AX SF ld ‘AL “ISA ‘WNASNW SISANYVO STVNNY PETERSON : MIOCENE BEDS oF NEBRASKA AND WyoMING. 47 _ American Museum of Natural History P+ and the nasals show very close similarities to those of the type specimen in the Yale Museum. The specimen in the American Museum is further distinguished from the type of Diceratherium niobrarense by the less convex contour of the skull from side to side, the smaller brain case and occipital condyles, the more complicated grinding surfaces of the teeth, and by a small = Z SF i Fic. 11. Posterior view of skull of Déceratherium niobrarense. i nat. size. ape, No. 1271. tubercle in the median valley on the internal side of M2. The type of D. niobrarense and the specimen in the American Museum referred to D. nanum are very nearly of the same size. The type of D. arma- tum represents an animal of larger size, and there are other differences which were pointed out by Marsh (Am. Jour. Science (3), Vol. IX, Der2A2, 1875 ). Diceratherium cooki Peterson. (Science, Vol. XXIV., No. 609, pp. 281-282, 1906. ) (PLATE XV.) The modification of the teeth and the general construction of the skull of this species at once separates it from other American species. The European species Dicerathertum minimum Cuvier is perhaps the most nearly allied. It is interesting to note that the configuration of the triturating surfaces of the teeth in Déceratherium nanum is more nearly similar to the present species than to Deceratherium niobrarense and Diceratherium urmatum, and that the two latter have the teeth complicated in nearly the same degree. 48 ANNALS OF THE CARNEGIE MUSEUM. In the abundant material of Diceratherium cooki in the Carnegie Museum a great variation in the size of the nasal horn cores is ob- served. This is undoubtedly due to differences inage, sexual charac- Fic. 12. Top view of skull of Diceratherium cooki. 4 nat. size, Type, No. 1572. ters, and individual variations. When the material from the Agate Spring Fossil Quarry is freed from the matrix full descriptions of the various forms will be given in the publications of the Carnegie Museum. Fic. 13. Palate view of skull of Diceratherium cooki. 1 nat, size. | Type, No. Plate XVI ANNALS CARNEGIE MUSEUM, Vol. IV. Wha ] Yj hi) Yj, \ Ys | i Uf 4 wh 4 LY \ | | 1 6 aws of Dinohyus hollandé Peterson, 5- Side View of Skull and Lower J Plate XVII. ANNALS CARNEGIE MUSEUM, Vol. IV. f ( ‘ G \ ((! Ai CEé£ 2 = = ~s ‘ I SJ = SS i ; \ (x 7D)yy \ \) Bai ity) 1 b Palatal View of Skull of Dinohyus hollandi Peterson, ¥. PETERSON: MIOCENE BEDS or NEBRASKA AND WyomInc. 49 Dinohyus hollandi Peterson. mescrence, N. S,, Vol. XXII., No..555, pp. 211-212, 1905; No. 570, P- 719, 1905.) (PLaTEs XVI. and XVII.) The chief differences which distinguish the present genus from Llotherium are found in the modification of the dentition. The first upper premolar is proportionately larger, and P2 is reduced antero-pos- teriorly and enlarged transversely when compared with that tooth in Lilothertum. ‘The deuterocone of P4 is relatively larger than in the Oligocene genus. In the inferior dentition a change corresponding to that in the upper jaw is apparent. Premolars one and two are large while premolar three is of a relatively less antero-posterior diameter than in Z/o- thertum. Yhe crowns of the premolars are more obtusely pointed than in Zlotherium, and the tubercles of the molars are less distinctly separated than in the latter genus. | The zygomatic process of the jugal is much enlarged when com- pared with the corresponding process of the “otheria from the Oligo- cene in the collections of the Carnegie Museum. ‘This process in Dinohyus reaches to the glenoid cavity and assists in forming a strong buttress in front of it; while in the Oligocene genus (especially those forms from the lower levels) the process is rather slender and does not reach the anterior border of this articulation. In the American Museum of Natural History is a well preserved specimen of Elotherium ingens (No. 572) from the Protoceras beds of South Dakota. The zygomatic process of the jugal of this specimen forms an independent downwardly directed process about 38 or 40mm. in front of the glenoid cavity, but furnishes no apparent support for the condyle of the lower jaw. ‘The broad and downwardly extended process of the jugal below the orbit, which is characteristic of £/o- thertum is much reduced in Dinohyus. On the inferior border of the mandible the anterior pair of protuberances is but slightly noticeable in Dinohyus. More striking differences than those above given may be found in the structure of the limbs and feet when these portions of the type specimen shall be extracted from the matrix. It will then also be carefully compared with LElotherium humerosum Cope, the large form 50 ANNALS OF THE CARNEGIE MUSEUM. from the John Day formation of Oregon, which is represented only by limbs and feet in the collection of the American Museum of of Natural History. MEASUREMENTS. (Type specimen No. 1594, Carn. Mus. Cat. Vert. Foss. ) mm Greatest length of skull... i...:02s- ses scehene-¥eeceneehtaeeie geen ae tiaee ae isa 098 Distance from premaxillary to postorbital process of jugal................ 528 Distance from postorbital process of jugal to and including occipital COTA YTS oo. ic ccneeneciseabbeesicnnieioe pag emmante es site tieime eet este aaien es ate mene 305 Length of alveolar border of maxillary and premaxillary.................. 465 Distance from median incisor to ME ....42)2ucs ee seen ence eae 330 Length of molar series ....25..0dsntyevae sens snes ene t ene 47 ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XVIII. “y is ~S ti Ss A N= 4 LN 2, eee oe BE Taw Skull of Amphicyon superbus Peterson, oe PETERSON: MIOCENE BEDS OF NEBRASKA AND WyoOMING. 51 Transverse diameter Of Moy .........:.sssecoccecccesecccestesscercvscrsscasssnvens 36 Antero-posterior diameter Of My..........scsccsecserescecscsereeecessescercesens 49 Transverse diameter Of Mg ............ccccsececeecessescccceccccecscevscncsverenes 37 Amphicyon superbus, sp. Nov. (Type Specimen No. 1589, Carn. Mus. Cat. Vert. Foss. ) (PLaTeE XVIII.) This species is represented by an exceptionally well preserved skull with the lower jaws, and portions of the skeleton. Only the skull and lower jaws are freed from the matrix and are illustrated in connection with this description. The specimen was found near the Agate Spring Fossil Quarry and in the same horizon. The type," No. 1589, is remarkable for its close resemblance to Camis; in fact were it not for the relatively smaller brain case and the presence of M® it might well be taken for a member of that genus. The principal cranial characters of the type are as follows: The skull is mesetocephalic, and the brain case is small. 13, Cj, P4, M2. On comparing it with Camzs occidentalis the following dif- ferences are noticeable. The fossil represents an animal very nearly the size of a fully adult gray wolf, but the skull is shorter and broader, the occipital condyles smaller, the basioccipital and basisphenoid of greater transverse diameter, the paroccipital processes further separated from the tympanic bulla, the mastoid larger, the tympanic bulla smaller, and the postglenoid process heavier. The most important differences from that of Canis occidentalis are perhaps the relatively smaller brain case, the presence of M3, the smaller sectorial tooth, with the larger internal tubercle, the larger M2, and the shorter and broader skull. In size and position all the foramina are nearly identical with those in Canis occidentalis. The mandible, excepting the larger and deeper temporal fossa, is very similar to that of the wolf. On comparing the type with that of Amphicyon americanus Wort- man (American Journal of Science, Vol. XI., pp. 200-204, Igor) it is seen that the skull is smaller, the canines smaller and more rounded, the premolars larger, and the second and third premolars smaller. The internal cusp on the fourth upper premolar of Amphicyon superbus has a greater development, M* has a more internal position in the alveolar border than in Amphicyon americanus. 17 Two individuals of the same species were found together. 52 ANNALS OF THE CARNEGIE MUSEUM. A fragment of a mandible (No. 408, Carn. Mus. Cat. Vert. Foss. ) of Amphicyon ? crucians Filhol from the Quercy formation of Europe, with the molar series in place, shows close resemblance to the specimen under consideration. In 4. crucians Mz is relatively smaller and the anterior portion of the crown is more elevated than in A. superbus. MEASUREMENT. mm. Greatest length of skull....... An ircon tata daaghd ee tsimeen Raceeten ates eg sate io ene Distance from occipital ‘condyle to tmciSOrs.. jgpcen aneseeat tenner 228 Distance from occipital condyle to M=...7 7... :ss.s-a5 Re ores epi Senne AAT 5 112 Distance from M® to.and including ‘the tncisors)..22i2..2,.02. ernest eee 117 Greatest transverse diameter of the slulll...,..\..c.tsssssnpseenee ogee ee 148 Transverse diameter of occipital condyle............... \nienigs oSeeae oe eee ae 46 Greatest transverse diameter of the brain case.................. a Maha es 58 Transverse diameter of the frontals at paroccipital processes.............. 68 Transverse'‘diametet of muzzlé at canine. <....2%¢-.c1s2 dnacnwdeeiee gee eee 51 Length ‘of premolar sertesin i. sac ssancpeas en eee a ddedle aeeeeae see eaeame 55 ength of molar’ serves. :/. svc s7.05 sth sone one ve nash capes be eee ada eee 34 Antero-posterior diameter of canine at base.......... Tadao abelancines anagem eet 13 Transverse diameter of canine at ‘bases, .. 2..0.2.<.0 sl oaxsen seme, cae eee 10 Antero-postetior diametet Of P21. .ccacewayessave aseravnqesne: tenspeew mea ieeee 6 Transverse diameter .of Pi 7 ic cseure ct cutee sone: heen ene bee ee eee 5 Antero-posterior diameter of Pes y15) ps. 2. tlomsntek danlee dace cee eee ee CARE ‘Transverse diameter of P2.).. i siceaag. certs es veut ato eannc serene “ad seth ot 5 Antero-posterior diameter of P3...... ge sia aaa fat Seacea rea a eeaugeuies 13 Transverse diatneter of (P25 vssoucdsadan se can eaeee see eae ee ee 7 Antero-posterior diameter Of LAK gesa ness esseecratecesasreuass ce aeckue en eee 22 Transverse diameter of P4, anteriorly...... Feit ahi macinate et cote eee 14 Antero-postetior dianreter:of DRS... yess G i.e tee cat walodautels a etguts saeaene ee 19 Transverse diameter of NMA..3c..027 nit seis caeent hone bein ee Oat ee eee 24 Antero-posterior diameteriol M2 wy sia ieek tae cae tae eS an Sane sees 13 ‘Fransvetse diameter of M2 \c5. festian vn eencrtenase sub tauceece cue ee 18 Antero-postérior diameter Gf M25. cide. as catetebasaret ack aa dentine ceneemee 5 Transverse diamieter of M@,>.c:d:,cnjsecsee cae cases ae ee 8 Greatest length of mandible ....... ve de digs Coke dnapy th oanaavokeets tohe at Cameo 182 Height of mandible at coronoid process: .ijaj.ceisesceey saeea j vance bartine $2 Length of premolat series. .<.:.si...dsseneeneee 370 Distance from occiput to posterior border of orbit’.....\.../:.sss 274... c:ccavessrdscaconseesizesdevacssharctecrvoseoassetsece | 22 natmeremsce Ciametet OF WE. 5. . ...csiscus eivaretancedccnsw Oleubes ebavcods ciuksbertiGMnalvecctiecs 29 imvera-pesterion tiameter Of M2 ois: cnsszpravinneowbsettaennstaxs thsevbstbsastcaesinns, 23 mmr reren climate Oh 2 sey co 8 ie ck awd abncptenaiisamsandowr aowadearmanodanel> [29 Antero-posterior diameter of Ma acieeutgem tar reaensataaeties Retreat wisi 20 Transverse diameter of M3, iatdtionly! das oee se api adie aici Greatest length of mandible, aputosinndl BORG Se faae wavarvencienenHanen seaMese: teeviee "EO Vertical diameter of mandible at coronoid proces............ssscssecesecsseesseeee 175 Vertical diameter of mandible at Mg3...................04 bd cMicngatvetumubanive sec pS Vertical diameter of mandible at hieaemed in frank ne E. Bele BO ReeMA it OL LOmer PLEMOlAar SCTIES:.......:025 0000 s---2 deeeare meeeee 73 Verticalsdiameter of mandible at anterior part of Moi7.ci<:-..:euee eee 26 Vertical diameter of mandible at anterior part of Pq. .5..-.:0.. erst ee en |) ee 4 | Babe | ; | | tet ie les all ie =H ace n a ie if : ) baths wre ie lal ana ae ag . ‘ A Elke | se, ce bie ase! os 7h oh ee Aes eae i a te | Non k AN Dae a wh ia aegis lara ia | maka Feet hy vices Son aoa ta iat eee “wn : Fey, A. ES - ete hs a 4 OMe 5 os he hegetia 4 | et: okie ‘2 EX aS ¥ es tine oa ea 4 5 ’ ‘GONTENTS, ©). 00 Us eae Editorial Notes 26a ee eo he i Bec Sune 2 I. © Plastron of the Protosteninge: By G. R. WIELAND . 8 II, Descriptions of New Species of Turtles of the. Genus. é ie Testudo, collected from the Miocene by the Carnegie Museum ;. together with a Description of the Skull oe of Stylemys nebrascensis. ‘By Oxiver P. Hay Bi! ie ad 5 | oe III. The Miocene Beds of Western Nebraska and Eastern . . “Wyoming and their Vertebrate Fauna. ‘By 0. Bee a PETERSON. ~ SSR Pa sea SO ore he Relate tro area aie en a IV. A New Species of Lonicera from Pennsylvania. By” ee | SOTTO. Hs JENNINGS. 91S as reo Oita eae ea a ICATIONS OF THE CARNEGIE MusSEUM SERIAL No. 49 ANNALS OF THE CARNEGIE MUSEUM WOOL vi No. iT. W. J. HOLLAND, Editor PUBLISHED BY THE AUTHORITY OF THE BOARD OF TRUSTEES OF THE CARNEGIE INSTITUTE JULY, 1907 ANNALS OF THE PARNEGIE MUSEUM VOLUME. ITV., “NOW He EDITORIAL NOTES. Tuts number of the ANNALS goes to press while the echoes of the voices of those who participated in the dedication of the new and enlarged buildings of the Carnegie Institute are still sounding through its halls. The occasion was most notable from every point of view. A memorial volume which will keep green the memory of the fes- tivities is being prepared by the Secretary of the Board of Trustees. The expressions of pleasure and satisfaction which fell from the lips of the distinguished men who were assembled on the occasion were most gratifying, especially to those who through many years have been laboring to bring about the consummation which was witnessed on April the 11th, 1907. It is not for the editor of the ANNALS of the Museum to speak of what was said in reference to other departments of the institution, but it was to him the cause of profound pleasure to hear from-the lips of those most competent to judge words of unfeigned appreciation as they passed through our still unfilled and unfurnished halls. Our attitude in the Museum has been for some time past more or less apologetic, but our friends who visited us were unanimous in declaring that apologies were not in order, and that out of sincere hearts they were able to congratulate us upon what has already been achieved. It would savor of inordinate vanity to quote what was said by them by word of mouth and by letter, but these pleasant expres- sions have left their impress upon the minds and hearts of all, and we find ourselves, in spite of the fatigues and cares incident to the 81 82 ANNALS OF THE CARNEGIE MUSEUM. occasion cheered and prepared to address ourselves with great courage and assurance to the tasks of coming years. Pittsburgh has a Museum which may well challenge comparison with the other great Museums of America. For this we gratefully thank the generous man whose kind heart made possible what has been accomplished. WE anticipate with pleasure the coming meeting of the American Association of Museums, which will take place in the first week of June. ‘The time was perhaps a little unfortunately chosen, as the en- gagements of many of those, who are members of the Association and who are connected with our colleges and universities, or, who are in- tending to avail themselves of the summer months for collecting ex: peditions, will prevent them from being present. Nevertheless every- thing indicates that there will be a good representation of those who are engaged in the work of Museums in different parts of the country. Pleasant and profitable discussions will no doubt take place and a num- ber of excellent papers will be presented. Every effort will be made by the Committee of the Trustees who are charged with entertaining the Association to make their visit to Pittsburgh agreeable. THROUGH the kindness of Mr. H. H. Jack and Mr. E. H. L. Page the attention of the Director of the Museum was called to the fact that in the latter part of April there had been discovered in the quar- ries of the American Lime & Stone Company, near Frankstown, Pa., the remains of a mastodon and of some other animals lying on the floor of a small cave which was reached in the process of excavation. Mr. O. A. Peterson of the Museum was despatched to the spot. He returned after a few days bringing enough material with him to make it plain to the Director that he was justified in requesting Mr. Peterson to return to the locality and continue further search. The result of his investigation has been the uncovering of a remarkable number of species and individuals of animals, a number of which are now extinct in Pennsylvania, the bones of which were commingled on the floor of the cave. WG Vyas Leek ig Gili Ee Wy Yi Uj ——— aS W WSS a Skull of Pronomotherium laticeps Douglass, 4. ay 7 * a ‘el ea ue ae -) - t evidently does not belong to either of these genera. IV. AFFINITIES OF PRONOMOTHERIUM. I do not know of anything very closely related to Pronomotherium laticebs except Pronomotherium altiramus and ‘* Merycocherus’’ rusticus. It may be from beds later than Merycochwrus proprius but probably not a direct descendant. It is doubtful if it is a descendant of ‘‘Merycocherus’’ rusticus, Dut it was probably more nearly con- temporaneous with the latter, than with the former. I know of nothing in lower horizons which is likely to prove ancestral to any of these. Pronomotherium was an extremely aberrant artiodactyl. The upper lip and snout were certainly greatly modified to correspond with the extreme modification of the bones of the head. The character of the skull could hardly tell in a plainer manner, that the possessor 21+«* New Species of Merycocheerus,”? Am. Jour. Sc:,- Vol. Bebs, 1900, p. 755 Figi?2. 22 [bid., p. 77: 23 /bid., p. 79, Fig. 4. 4 Tbid., p. 73, Fig. 1. 25 «« The Vertebrate Fauna of the Ticholeptus Beds,”’ Amer. at., AX, p. 368. 98 ANNALS OF THE CARNEGIE MUSEUM. had a large lengthened snout or proboscis. How long it was of course is not known but it was probably quite long. ‘The position of the condyles and the extremely heavy mandible indicates that the head was carried with the facial axis in approximately a vertical position or approaching a right angle to the vertebre of the neck. If we may judge by what appears to be its nearest known relatives, and by what few fragments of bone are preserved, this animal was short limbed, like most of the later Merycoidodonts ; but we must await the discovery of more complete skeletons. V. AGE OF THE FLINT CREEK AND MADISON VALLEY BEDs. These beds both belong to the Loup Fork Epoch as it is usually understood. Either they are not quite contemporaneous or else they represent a somewhat different ecological condition, at least there isa different assemblage of fossils. This will be thoroughly discussed later after a revision of the fauna has been made. JZylagaulide, but perhaps of different genera, occur in both formations, Pronomotherium appears in the Flint Creek Beds and a related form Pronomotherium altiramus in the Madison Valley Beds. Pa/eomeryx ? appears in both beds but of smaller size in the latter. Procamelus of large size occurs in both. . On the other hand My/agaulus is found in the Flint Creek and Deep River beds. TZzcholeptus occurs in the latter, and perhaps in the former, as the specimen, Merychyus smitht is much more like Zicholeptus. Paleomeryx ? borealis occurs in the Deep River beds, and what appears to be the same species, in the Flint Creek beds. It appears most probable, from the evidence, that the Flint Creek beds are in some ways intermediate between the Deep River and Madison Valley formations, yet some things in the first seem more modernized or specialized than in the last, yet we cannot judge by this, for there have been very highly specialized mammals all along the line, and some characters that were supposed to be modern are quite ancient. A careful study of the Horses, Camels, and other fossils of these beds may furnish a better basis for correlation. Matthew * thinks that the Pawnee Creek Loup Fork holds a position distinctly lower than that of the Niobrara, Santa Fe, and the Repub- lican River Basin. ‘‘It seems most nearly equivalent to the upper beds of Smith Creek, Montana (Deep River substage).’’ 26 «* Fossil Mammals of Colorado,’’ pp. 373-4. ees a ® \ ~ eS ES Se Qo, f° ql aah Vy Lhd TA Wa RAN ey ea Ns Wy a IKK Sti ‘Al ‘ISA ‘(WN3SNW JIDSSNYVOD STVYNNY Douctass: New MERYCOIDODONTS. 49 VI. SOME NEW MERYCOIDODONTS. By EARL DOUGLASS. During the various expeditions of the Carnegie Museum in Mon- tana and North Dakota many remains of Merycoidodonts have been obtained from various localities and geological horizons. Much of this material was supposed to be new to science, but it was not until the types and other specimens in the museums at Washington, Prince- ton, New York, and New Haven were examined, and the collections in the Carnegie Museum were cleared from the matrix and made accessible for study and comparison, that the new species could be intelligently described. ‘The collection has been placed in the hands of the present writer by the Director of the Museum, Dr. Wm. J. Holland, with the request that preliminary accounts of the new mater- ial be given, pending the preparation of a more complete memoir. The following species are based, it is believed, on sufficiently com- plete material. Other remains which are interesting and are undoubt- edly new, but not complete enough to be satisfactorily treated as types, will be described in a later paper. Eucrotaphus dickinsonensis sp. nov. (PLATE XXII.) (Type No. 1584, Carnegie Museum Catalogue of Vertebrate Fossils. ) The type, which was collected by Earl Douglass in 1905, consists of a nearly complete skull and mandible with the greater portion of the spinal column, and fragments of limb and foot bones. ‘The specimen was found near the top of the thick nodular beds of the Middle White River (‘‘ Oreodon ’’) horizon of the Little Bad Lands near Dickinson in North Dakota. Though not suspected at the time when the specimen was collected, it 1s barely possible that it may have come from the upper beds which contain remains of Hucro- taphus major (?). The remains were not imbedded in their original position and they may have been derived from a higher level, though the specimen is quite different from any species of ELucrotaphus so far found in the upper beds. It is also different from the one specimen of Eucrotaphus bullatus ? which was found in the upper portion of the ‘¢Oreodon’’ Beds. The size is about the same as that of Merycoidodon culbertsont. The upper antero-posterior line of the skull is convex, the brain-case well 100 ANNALS OF THE CARNEGIE MUSEUM. rounded, and the sagittal crest low. The posterior portions of the nasals are attenuated and the lachrymal pits shallow. The face is concave above the premolars, especially anterior to the infraorbital foramen, which is over P3. The tympanic bulle are very large and high and are elliptical in horizontal section. The pit for the tym- panohyal and the stylomastoid foramen are nearly equal in size. The mastoid process is fairly heavy, but does not extend far downward between the exoccipital and the external auditory meatus. The paroc- cipital processes are five-sided at the base, four-sided lower down, and three-sided near the tips. There is a postero-external ridge and the ‘antero-internal side, which is closely pressed against the postero-exter- nal side of the bulla, is concave. ‘There is no foramen rotundum. The crowns of the upper premolars incline backward or the outer crescents have a long convex cutting border and a shorter concave posterior border. ‘The fourth upper premolar is different from that of Merycoidodonts in general, in having two anterior fossettes as in P4, Pee and The chin is slightly concave. The anterior border of the ascending ramus of the mandible is nearly perpendicular and the coronoid process is not deflected backward. ‘There is a small sharp cusp on M2 between the posterior external crescent and the heel, which I have not observed in any other Merycoidodonts. | The specific name refers to the prosperous town of Dickinson, which is not far from the locality where the specimen was found. MEASUREMENTS. mm Length of skull, basal méasutement ,;...,2 schoo. 5. vsaacpn~t else ees 184 Width of ‘skull at posterior orbits 1.0 22s. ccestecenaes op iene saegene ee cgeneeen ie 105 Length..of molar. premolar Series 2.5.2 5< +" f as to : 7 oe : ” : ese e o\~, ae yy 4 ¥ % , ’ a Py, Fea el ee Le Sbect + =e ~ekt = rate ee ; . re le od. ‘ Fs : j , F = & Py ee rs = 2! Ss A i, oo RIX 73% ld “£ ‘ssvpsnoq «o0s1no saprooAuapy JO |[NyS ‘Al “ISA (WNASNW JISINSVYD STVNNY Douc.Lass : NEw MERYCOIDODONTS. 101 men obtained in this locality and horizon, but the beds evidently overlie the Lower White River, which occupies a large area just north of this outcrop. The upper molar teeth have nearly the same antero-posterior diam- eter as the corresponding teeth of the type of Ovedon ( Eucrotaphus ?) major Leidy,* but the transverse diameter is a little greater. The species is a little larger than the specimen figured as Oreodon major in Leidy’s ‘‘Extinct Mammalian Fauna of Dakota and Nebraska’’ ; all the upper premolar and molar teeth are longer than wide ; the incisors are small; the molars increasing in length pos- teriorly ; P; overlapping P; inwardly; the length of the upper and lower molar series is nearly one and one-fourth times the length of the corresponding premolar series ; the tympanic bulle are inflated, and of medium size ; the paroccipital processes are flattened on the inner surface where they press against the posterior outer portion of the tympanic bullz. They are slender and are directed antero-posteriorly below the bullz, but somewhat twisted on themselves near the ends ; the arrangement of the elements at the posterior base of the skull is nearly as in the species of Promerycocherus from the Cafion Ferry Beds ; the orbits are fairly large, the malar moderately deep ; the pos- terior portion of the zygomatic arch is only moderately heavy; the sagittal crest is high and thin; the infraorbital foramen is located over P4; the anterior portion of the mandible is low, but increasing in depth backward to the angle which is very large and rounded ; the coronoid processes are low. MEASUREMENTS. mm, ie emres rere le PP GOLAL s2 foe a. a 5n'90 dines shu F eRe tek nes caanbdessenen Tews tersweass 239 Rees NORM EO Bran whe A eal. step scene amas ata Shh ok dus eX youdecsiwachsaearacenes 72 Pei Or MODE PLEMONAL SELES, coeds cueceseecmseacs Vedueunddee’ coueduessscocceuntees 390 Beret orn MP Per Cental SELES, <5. ...+.0rawansvekvivecubesinssessnacees.sansceses 213 Width of skull just anterior to glenoid articular surface................0065 180 Peete Ga ose Gt POSELIOL Of NLS... soc,cdeecsis ovens ccncsestsonsacvvaress . 100 106 ANNALS OF THE CARNEGIE MUSEUM. Length of molar-premolar serieS.2:3f, srs... Gbinergiee: G. es dete syeseser .. 168 Length. of premolar Series... .0...-ssseunddqtes Pr ee ees eee an pe Bee hoe 82 Length of molar, series. ..:0.02as SS en = mig Mh MULL! = SSS = = a > Mss ~ ° NN » SST Ticholeptus bannackensts Douglass (Type), 3: Douctass: NEw MERYCOIDODONTS. Length of lower molar-premolar S€TiCS......... seeeeereceeeeeeerseeeereers oes 125 Length of lower premolar SerieS.......ssssereeeeeeeesesersrererenr ane naeeer seen es 55 Length of lower molar Seri€s..... ..ssseeseeeeeeessereeeeeees supine adept aed stpaeek 7° Length of radius....3.....-..cceeosesssecsenceareserseeeeerscssoaneesenseeeaenececes 155 Length of metacarpal TIL ...........scsseceeseeeeeeeeneseeceeeeeeee enn eeeteesnerss 71 Width of shaft of metacarpal LIT .............ceceeeee scence eeecer ene reeeereeeees I2 Length of femur, head to distal end............cseeeessseeeeeeeterseereeeceeees 205 Length of tibia.........esceeseee sence eeee tenons Sete phic nae nyeee been ex selones 175 Height of cuboid.............00.cseeeeesccrenseceneersenensesecnensecenaasererceceee 22 MEUELS OL CUDOIG cx cc coc se nov cecau'nccaviennvnncvsosanetcetestssscceesernenvererersnase ih Length of metatarsal L11...............:sseeessseeeeensereeeenessneeesenee eer ans es 73 Width of metatarsal LI1.................ccscccecccccesscccncecccccccccccsesescccoes 13 Named after the old mining towns of Bannack. ° 109 VII. ON FURTHER COLLECTIONS OF FISHES FROM PARAGUAY.!' By Cart H. EIGENMANN ASSISTED BY WALDO LEE MCATEE! AND ~ Davip PERKINS WARD. The fresh-water fish fauna of tropical America is by far the richest in the world. It comprises about one tenth of all known fishes. It is entirely distinct from the North American fauna and from the Pata- gonian fauna. Its center of greatest diversity lies in the Amazons about the mouth of the Rio Negro. From this point it becomes at- tenuated northward until it reaches the vanishing point on the borders of the United States. Southward it extends on the eastern slope to somewhere, no one knows exactly where, south of Buenos Aires. On the western slope it does not extend so far south. The key to the great diversity of the tropical American fauna is to be found in the enormous single water system extending from 10° north to 35° south latitude, and from 50° to 79° west longitude, pro- viding a continuous north and south water-way of more than 3,000 miles and an east and west course of over 2,000 miles. It embraces the basin of the Orinoco, the basin of the Amazons, and the basin of the La Plata, draining over 3,000,000 square miles of territory, or an area about equal to that of the entire United States, exclusive of dependencies. It has long been known that the Orinoco and the Amazons are connected by the Cassiquiare, the waters of which at times flow one way, at times another. The following is from the ‘* United States of Brazil’’ issued by the Bureau of American Republics. ‘« Another remarkable phenomenon of the Paraguay is the mingling of its principal head waters with those of the affluents of the Amazon. An affluent of the Jaurt River is sufficiently near the Guaporé River to be connected with the latter by a canal. The Aguapehy, another tributary of this river, is separated from the Alegre by a narrow isth- mus 5 kilometers wide. In the eighteenth century an attempt was made to open up a canal here, and owing to the abundant rains a large canoe of twelve oars succeeded in passing from the one river to the other. One of the governors of the State also endeavored to open up a canal 1 Contributions from the Zodlogical Laboratory of Indiana University, No. 65. 110 EIGENMANN: COLLECTIONS OF FISHES FROM PARAGUAY. 111 10 kilometers long in another part of the isthmus, but on account of the small amount of trade it was never completed. ‘This would con- nect Montevideo and Para by a continental waterway 8,300 kilometers long. In the near future it is probable that railways will take the place of the canal. There are many places on the edge of the pla- teau, farther to the east, where a simple cut of a few meters would connect the tributaries of the Amazon with those of the Paraguay, transforming eastern Brazil into an island. ‘There is a space of but too meters between the Estivado, a small tributary of the Tapajoz, and the Tombador, which empties into the Cuyaba.’’ In 18917 I called attention to the great similarity of the faunz of the La Plata and the Amazons. ‘The former was at that time known to differ from the latter by negative characters only. At the same time I called attention to the radical difference between the fauna of the La Plata and the Amazons and the fauna of the coastal streams that empty into the Atlantic between these two great rivers. Every additional collection from the Paraguay, and indeed from the entire system of the La Plata, tends to emphasize the similarity of its fauna with that of the Amazons. The interest that centers in the Paraguayan fauna becomes apparent from the above considerations. It is through the basin of the Para- guay that the La Plata has probably received its Amazonian character. For this reason accurate representations of the members of the fauna are greatly desired and an attempt has been made to supply these in the photographs of a number of actual Paraguayan specimens which accompany this paper. The basin of the Paraguay approaches in area the basin of the Ohio plus that portion of the basin of the Mississippi north of the entrance of the Ohio and exclusive of the basin of the Missouri River. The area then is very large. ‘The means of communication for the most part are primitive, and probably but a fraction of the fauna is known. Two hundred and fifty-four species have been recorded from this basin. The regions that will probably yield the richest rewards in the future are the mountain brooks and pools of central Paraguay, the ponds of the Chaco, and the mountain sources of the Pilcomayo. In the Proceedings of the Philadelphia Academy of Sciences for 1903, pp. 497-537, Eigenmann and Kennedy reported on a collection of fishes made by Professor J. Daniel Anisits in the basin of the Paraguay. 2 Proc. U. S. Nat. Mus., Vol. XIV., pp. 1 et seq., 1891. 112 ANNALS OF THE CARNEGIE MUSEUM. Professor Anisits has made further collections of the fresh-water fishes of the basin of the Paraguay which were sent to Indiana Univer- sity for identification. By these collections several new species and new genera are added to the South American fauna, the known distri- bution of many species is extended, and what is, perhaps, of most importance, the South American fauna is rid of a number of nominal species, which have been relegated to synonymy. The collections of Professor Anisits are the most important and extensive which have been made in the basin of the Paraguay, and his energy and enthu- — siasm, together with his great care in the preservation and labelling of his specimens, promise to make the basin of the Paraguay ichthyolog- ically among the best known regions of the neotropical realm. Of particular interest in the present collection are the new species of Dysichthys, the only other species of which came from the Peruvian Amazon, and Homodietus, a new genus of Stegophilini, some of the species of which live in the gill-cavities of the larger Siluroids. The following localities are represented in the collections : 1. Rio Paraguay at Porto or Puerto Murtinho, Tuyuyu, and Cor- umba in Matto Grosso. 2. Rio Otuguis, a western tributary of the Paraguay in Paraguay near the boundary of Bolivia. 3. Bahia Negra on the west bank of the Paraguay in northern Paraguay. 4. Puerto Max in the lime region of Paraguay. 5. Tributary of the Paraguay in the Chaco Paraguayo. 6. Ipané-Tuya on the Paraguay. 7. Rio Paraguay and Laguna Pasito at Ascuncion. 8. Rio Negro, a tributary of the Paraguay opposite Ascuncion. g. Laguna Ipacaray. to. Mountain brooks at Sapucay, Central Paraguay. 11. Villa Rica, and a small brook of Colonia Gonzales. | The types of new species are in the collections of Indiana Univer- sity. A full set of cotypes are contained in the collections of the Carnegie Museum. BUNOCEPHALIDZ. 1. Bunocephalus rugosus Eigenmann and Kennedy. One specimen from Corumba (322).° 3 Professor Anisits’ collector’s number. ANNALS CARNEGIE MUSEUM, Vol IV. Plate XXX\I. Figs. 1-2. Dysichthys australe Eigenmann and Ward. (Type.) Ventral and dorsal views. EIGENMANN:: COLLECTIONS OF FISHES FROM PARAGUAY. 1138 2. Bunocephalus dorie Boulenger. Two specimens from small brooks at Villa Rica (464) ; another from the Laguna Pasito (462). 3. Dysichthys australe Kigenmann and Ward, sp. nov. (Plate XXXI.) Type No. 10123, one specimen, 28 mm. long from Corumba (317). meeynes, No. 10124, ten specimens from Corumba (317). This species may be distinguished from Dys¢chthys coracotdeus, the only other species of the genus, as follows: a. First dorsal ray a little nearer tip of snout than base of caudal; maxillary barbels Samim base of pectorals. DoT bess... liseas wees scscresounernencee ” coracoideus. aa. Distance of first dorsal ray from base of caudal one and one half times as great as its distance from tip of snout; maxillary barbels not reaching to base of a fio Seial i siaas wane Ai hocipae Vs Ue dake aime nsciscinee stab seun’ australe. Body slender, its greatest width at base of pectorals, 3% in the length ; depth at origin of dorsal 6 in the length; head depressed ; snout rounded ; two ridges diverging from near the central portion of the snout, running backward above the eye, meeting again to form the nuchal crest, leaving a diamond-shaped depression between the ridges ; nuchal crest continued back to base of dorsal fin; a crest on each side beginning at the operculum and running parallel with the lateral and nuchal crests; the ridges and knobs of the head well developed ; interorbital space very concave ; the part of crest bounding the orbit especially strong; a knob before and another behind the eye. ‘The eyes placed almost laterally below the ridges ; eye 1% in the snout, 7 in the head, 3 in the interorbital ; maxillary barbels not reaching to base of pectorals by % of their length. Coracoid processes parallel behind, their length 2 in the distance between them. Humeral processes slightly shorter ; skin everywhere covered with very censpicuous papille, those on the sides of the body arranged in about seven rows; distance of dorsal fin from tip of snout 2% times in the length ; pectoral spine armed on both sides with long hooks. Dark brown, speckled with lighter ; fins light brown ; belly speckled with white; head 5; depth 6; width 3%; D.I,4; P.I, Way Vo eeenceg: C, 10. The only other’species of this genus was described by Cope from Nauta on the Marafion about 2,000 miles from the present locality. SILURID. 4. Rhamdia quelen (Quoy and Gaimard). One specimen from Corumba C385): 114 ANNALS OF THE CARNEGIE MUSEUM. 5 Pimelodella gracits (Valenciennes) (Plate XXXII., Fig. 2). One specimen from Corumba (352), and. another from Laguna Ipacarai (450). 6. Pimelodella lateristriga (Miiller and Troschel ). Seven specimens from Villa Rica (469). 7. Pimelodella mucosa Eigenmann and Ward sp. nov. (Plate XXXII., Presa) Type No. 10125. One specimen from Bahia Negra (399). ‘This species is most nearly allied to P. eigenmanni and buckleyt from which it may be distinguished as follows : a. Lower caudal lobe the longer. 6, Maxillary barbel reaching to middle of caudal; postmental barbel to middle of pectoral spine ; pectoral spine curved, with about fifteen straight, feeble spines on its posterior surface, anterior surface with minute denticulations along basal part and feeble recurved hooks near its tip, 14 in the head. mucosa. aa. Upper caudal lobe the longer. c. Maxillary barbel reaching beyond origin of anal ; postmental about to middle of pectorals ; pectoral spine as long as the distance between base of maxil- lary barbel and the opercular border, its inner edge distinctly though feebly serrate ; adipose dorsal 32 to 4 in the length............ ezgenmannt. cc. Maxillary barbel reaching to origin of anal; postmental to tip of pectoral; pectoral spine as long as the distance fromthe anterior border of eye to the opercular margin, practically smooth on its inner edge; adipose fin a littke more than 14 in the total length....2:..0.¢./cssssmetneeeree ee buckleyt. Body compressed posteriorly ; head sub-conical, depressed in front, its width 1% in its length, its-depth at the base of occipital process 14%, its width at the angle of the mouth 224 in the head; occipital process rather slender, its width at base 234 in its length; maxillary barbel reaching to middle of caudal; mental barbel to operculum ; postmental to middle of pectoral. Gill-membranes separate to below anterior margin of eye ; very conspicuous pits along either side of the lower surface of the head. Eye 12-in the snout, 3% in the head, 1 in interorbital, slightly nearer to posterior margin of operculum than to snout. Dorsal spine equidistant from snout and anal; its highest rays equal in height to the head in length. Adipose dorsal 3% in the length, its distance from the dorsal fin being 1% in the length of the latter. Caudal long and deeply forked, the lower lobe wider and longer than the upper, 3% in the length. Anal rounded, its longest ray 14 in the head. Ventrals inserted on a vertical line with fourth dorsal ray, 1% in length of head; pectoral spine 14% in ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXII, . Pimelodella mucosa Eigenmann and Ward. (Type.) . Pimelodella gracilis Valenciennes. 8. 3-4. Lheringichthys megalops Eiger.mann and Ward. (Type. ) oa ) * all * diigo e - EIGENMANN: COLLECTIONS OF FISHES FROM PARAGUAY. b1d the head curved, with about fifteen straight, feeble spines along the middle of the posterior surface, anterior surface with minute denticu- lations along the basal part, and feeble recurved hooks near its tip. Color light brown, a dark lateral band; fins blotched with black. mead 4; depth 5; D.1,6; A. ro. 8. Pimelodus ornata Kner. Cabezudo. One specimen from Corumba (358). 9. Pimelodus clarias (Bloch). One specimen from Porto Murtinho (286), one from Bahia Negra (398), two specimens from Corumba (291). 10. Pimelodus fur Reinhardt. One specimen from Corumba (part of 290). 11. Pimelodus valenciennis (Kroyer). One specimen, Laguna Ipacarai (451). 12. Lheringichthys megalops Eigenmann and Ward sp. nov. (Plate XXXIi., Figs. 3-4). Type No. 10126. One specimen, 175 mm. Bahia Negra, Rio Paraguay (No. 4oo). This species resembles /heringrchthys labrosus, the only other species of the genus. In every character behind the head the two species appear identical. The differences in the head, comparing specimens of same size, are as follows: a. Interorbital concave; width of occipital process at its base equal to its length ; widest part of fontanel about 2 in interorbital. Eye 33 in the head; snout 21 in the head ; postorbital part of head 32; interorbital 5; upper Jip scarcely REE e WOO EVE) TAMING 8515 ctcime's sae e raswopiecait Sdde hou a oivvige whic selu'syeectien’ megalops. aa. Interorbital flat; occipital process distinctly narrower than long; widest part of fontanel about 3 in interorbital. Eye 41 in the head, 2 in the snout; snout 21%; postorbital portion of head 3,5,; interorbital 4 in the head; upper lip Wit an aeep moter in the middle. 20... co. .tenensaacetes-abckesdvacedemedes labrosus. Body not as wide as deep at shoulders, compressed toward caudal ; head sub-conical, slightly depressed, its greatest width 1% in its length, its greatest depth about 2 in its length. Snout conical; en- tire upper portion of head granulose or striate; fontanel scarcely reaching to posterior margin of eye ; occipital process as wide at base as long; distance between the nostrils slightly greater than half the diameter of the eye. Maxillary barbels extending very little, if any, beyond tip of caudal; postmental barbels extending to middle of pectorals ; mentals not reaching pectorals. Eye elongate, very large, 116 ANNALS OF THE CARNEGIE MUSEUM. 32 in the head, 144 in the snout, 7 in the interorbital region. Pre- orbitals prominent; interorbitals distinctly concave ; mouth very nar- row, jaws not equal; lips very thick, upper lip with its free margin. reflexed, scarcely, if at all notched ; upper lip extending beyond the lower by Y diameter of the eye; teeth of the jaws in very narrow bands; no teeth on the vomer ; teeth on the pterygoids inconspicuous. Dorsal spine strong, 114 in length of head, serrate on posterior mar- gin; distance of adipose fin from dorsal slightly less than length of adipose, which is 44% in the length. Caudal forked. Pectoral spine strongly serrate on both margins, serrations on inner margin a little the stronger, the spine a little shorterthan the head. Humeral proc- ess triangular, pointed behind, but not spine-like, scarcely reaching to middle of pectoral spine, its surface granulose-striate. Ventrals reaching to anal, margin of anal concave, some of the anterior rays ex- tending much beyond tip of last. Anterior upper portion of the body spotted, otherwise plain. Head 3% 3 depth44); °D: LroguArar 13. Lheringichthys labrosus (Kroyer) (Plate XXXIII., Fig. 1). One specimen from Corumba (290), and four from Bahia Negra (382°). 14. fHemtsorubim platyrhynchos (Cuvier and Valenciennes) Jiripoca. One specimen from Corumba (359). ~ 15 Doras costatus (Linn. ). } One specimen from Corumba (364); another from Laguna Ipacarai (449). 16, Doras weddelli Castelnau. One specimen colored like the type figured by Castelnau. ‘Tribu- tary of Rio Pilcomayo (445). 17. Oxydoras eigenmanni Boulenger. Eight specimens from Corumba (365). 18. Hemidoras paraguayensts Eigenmann and Ward sp. nov. (Plate AXMTV, Fig. Gee Type No. 10127. One specimen, Corumba (366). This species is closely related to zatfererz, from which it differs in the characters set forth in the following key : a. Depth equal to length of head ; lower jaw sometimes with patches of teeth. Max- illary barbel extending to below the eye; caudal deeply forked...... matterert. aa. Body distinctly deeper than length of head ; lower jaw without teeth. Maxillary barbei extending to gill-opening. Caudal scarcely forked, the lower lobe wider and longer than the upperiisis.csseuccsvencmameete es paraguayensis sp. nov. Saree ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXIIl. . Theringichthys labrosus (Kroyer). . Dentition of Serrasalmo humeralis (Cuvier and Valenciennes ) y ‘i Ff Y 7 i} : ata Ae ri < ’ > i P| eee ‘Se a »! ‘ ¥ ; 4 ae Ys 7 ¢ : w yo r la 4 AS ~ ny « x - ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXIV. Fig. 1. Hemidoras paraguayensis Eigenmann and Ward. (Type.) Figs. 2-3. Homodietus anisitst Eigenmann and Ward. (Type.) nd gh ar , ar | & ie , pte Ret td ’ een i ey i Tk aes > i E re i Ya*s eS = “eras ae Cue =i. * - ) =, we 9 f in 2 — Se « cai oa me So * = ‘ . oe “ ent = ‘es cam 7’ e 2s . z . ¥ : be << Lah Lf *« . A ' ~ ,; eee é 7% ¢ 7 J ‘ ] Ln ? a Tig Adda ia. : i ! oy 98 > - or cP > . fa e ¢ ~ . Lae , aA r Bas inal & _ < : 7 ¥ e a+ ; s a 2 i, ? x a « x A ‘d a ‘ r 2 “ * rs * i 5 x . y ‘ et) 2 = “ . v ¢ ; - * t . te ae js - | i ao ‘ +e P/ r aeub wi ol ‘ . 7 7 , — a “ Hy oe1F : ; 2 = am o - - Fr, 4 = . oe } - i - a A Nv . i yo % ¢ e ’ j i sia o" , 4 7 é a - ! * 7 + * re EIGENMANN : COLLECTIONS. OF .FISHES FROM PARAGUAY. 117 Body short and deep ; ventral outline almost straight to base of anal; profile steep to nostrils, rounded at base of dorsal spine ; head short, slightly deeper than wide, its depth equal to its length. Inter- orbital region convex. Occipital region steep and convex. Fontanel an oval opening somewhat shorter than the eye, situated in a groove 1¥%4 times as long as the eye. Eye large, 14 in the snout, 3 in the head, 1% in the interorbital; snout short, rounded. Barbels all connected by a membrane, those of the maxillary extending to gill- openings, with a few cirri on their outer margins ; gill opening ex- tending forward to about % the distance of the eye. Coracoid proc- ess exposed, striate. Humeral process broad and long and with a strong keel, its surface striate. A few imbedded scutes in front of a large scute which is connected with the dorsal plate; the following lateral scutes much lower, each with a large median hook and a series of fine marginal teeth above and a series of one large and several fine teeth below the median hook. ‘The lateral scutes on the caudal ped- uncle have a more enlarged median hook, and the fine marginal teeth fewer, or none. No plates on dorsal or ventral surface. Distance of the dorsal spine from the snout 24 in the length. Dorsal spine slightly curved, its anterior margin with rather long and fine teeth to near the tip, its posterior margin with short, wide-set teeth. Space between dorsal and adipose fins 334 in the length. Adipose fin as high as long. Caudal scarcely emarginate, the lower lobe much wider and somewhat longer than the upper. Anal fin as high as long, rounded. Ventrals not reaching anal by % their length. Pectoral spine strong, longer than the dorsal spine, reaching to the second third of the dorsal fins; both margins serrate, both surfaces striate. Color light brown; silvery below lateral scutes; anal and pectoral fins peppered with dark. ireadai6-5 depth 22 ; lat.l.29 elem Ares) V..7; P.I, 8. 19. Auchenipterus nigripinnis Boulenger. Bagre Sapo. Two specimens, Corumba and Puerto Max (353). 20. ZLrachycorystes striatulus (Steindachner). ‘Two specimens. ‘Tributary of Rio Pilcomayo (444). PYGIDIIDZ. flomodietus Eigenmann and Ward, gen. nov. Type: Homodietus anisits’ Eigenmann and Ward. Dorsal behind the ventrals ; no teeth on the vomer ; gill-membrane 118 ANNALS OF THE CARNEGIE MUSEUM. united with the sides above and with the isthmus below, leaving the opening a very small slit in front of the pectorals; a longer and a shorter maxillary barbel on each side ; upper jaw and lips with about eight widely separated series of teeth, the teeth narrow, more or less spoon-oar-shaped, those of the inner series slightly larger; teeth of the lip very movable, those of the jaw more firmly attached ; lower lip without teeth, three series of the teeth on the lower jaw, those of the innermost series largest and forming a compact series. All the teeth more or less angularly bent backward near the tips. Opercle with about 4 spines ; subopercle with 6 ; anal short, behind the origin of the dorsal; head depressed, the eye directed upward, mouth inferior ; caudal emarginate; very large glandular swellings behind the pectoral. This genus is most closely allied to Stegophilus and Miuroglanis. It differs from Svegophidus in having two maxillary barbels and from Miuroglanis in having an emarginate instead of a rounded caudal and in having several series of labial teeth which are apparently wanting in Miuroglants. The genus Stegophilus as defined by Eigenmann and Eigenmann includes both species with a rounded and an emarginate caudal. Were the species large we should not hesitate in placing them in sep- arate genera and a different rule should not apply here. The genera of the Stegophiline may be defined and distinguished as follows ; a. Upper lip with several series of numerous, small, movable teeth ; each jaw with several series of minute teeth ; mouth inferior. 6, Gill-membrane broadly united with the isthmus. c. Caudal widely forked, the upper lobe produced in a filament; a single MaARtAary DALE thaccacsxieeind gdceseteeaen ds Peon eeeeaees Pseudostegophilus. cc, Caudal emarginate. d, A’ swoglesmaxallarg: barbel ico. scnc he. dese auhms Vandellia. Jf. Teeth broad, incisor-like in both jaws ; caudal forked ; two maxillary barbels. Paretodon. 21. Homodietus anisits’ Eigenmann and Ward sp. nov. (Plate XXXIV., Figs. 2-3). Type No. 1ors5, one female 43 mm. long; small creek at Villa Rica, Paraguay (466). Head 61% ; depth 534; D. 8; A. 8. Elongate compressed, head much depressed ; head nearly as wide as long; snout broad, its horizontal outline rounded; mouth very large, hemicircular in outline, 24 in the head ; lower surface of head flat, upper arched, the eyes directed upward, sidewise and forward ; eye equals snout, 3% in head, about equal to the interorbital ; no free orbital margin; origin of maxillary barbels below the last quarter of the eye ; outer and longer barbel shorter than eye, inner barbels much shorter, minute but distinct ; a distinct thin, free, lower lip extending for but a short distance from the angle of the mouth ; teeth of upper lip distinctly visible when mouth is closed ; pectorals extending some distance beyond the axillary gland, their length 224 in distance from their base to ventrals, equal in length to their distance from the tip of the mouth ; origin of dorsal equidistant from tip of caudal and poste- 6 Liitken in Videnskabelige Meddelelser for 1891, Kjobenhavn, 1892, describes Acanthopoma as follows : Caput depressum parabolicum, cauda subcompressa: maxillze in tentaculum breve continuz ; os inferum, supra seriebus dentium minutorum plurimis ; rimz branchiales confluentes, membrana branchiostega cum isthmo gulari haud connexa; spinz oper- culares et interoperculares plures ; pinna dorsalis post pinnas ventrales, inter has et pinnam analem posita. ACANTHOPOMA ANNECTENS. Maal : rere de ath eae rk oon coMae aa deed Pie dicsiks oS «i dawide adee Se 100 mm. ROME Cte ROMO Ang a cacuigcenee era wCavcesenk des seves 00s auain ie. ~ BN OS MSY set rtpe Lily aah tg co taal che van ncw'o avin meso Ree Legemets stgrste Hojde........ eee Be toads to ee 10.5 * Fra Snudespidsen til Rygfinnen........ .........00 Ber Pte e- Geass eae eiiespidsei El GAttet Sno le. a dcann canted Saecannes Dar art 120 ANNALS OF THE CARNEGIE MUSEUM. rior margin of eye; caudal with numerous accessory rays, slightly emarginate, the upper lobe longest, longer than pectoral ; anal inserted below the end of the dorsal ; ventrals short, reaching the vent; vent equidistant from tip of mouth and tip of caudal. Straw-color, back with numerous large, conspicuous, stellate, black chromatophores and many more smaller, much less conspicuous, brown ones ; sides with a few small stellate, black, chromatophores, gradually giving rise toa regular series along the middle of the tail; a dusky streak along the sides between the myotomes of the body and the thin covering of the abdominal cavity ; a small, intense black spot at the base of the middle caudal ray; middle caudal rays dark, becoming intensely black toward tip; oblique bars extending from the end of the second ray below median dark one downward and forward to the tip of the lower caudal fulcra and then as a black line forward along the tips of the fulcra; another one like it in all respects from the tip of the second ray above the median dark one upward and forward to the tip of the caudal fulcra and then forward along their tips as a black line ; remaining fins more or less dotted. | Alimentary canal straight, without convolutions or bends, the thin- walled stomach lying lengthwise and giving rise toa short, thin in- testine which merges into the much longer and larger, but thin-walled large intestine which appears to be filled with minute grains of sand. LORICARIIDA. 22. Hemiodontichthys acipenserinus Eigenmann and Ejigenmann. (Plate XX XRVays Mig 1): One specimen. Corumba (No. 333). 23. Sturtsoma robusta (Regan) (Plate XXXVI., Figs. 1-3). One specimen from Corumba (354). 24. Loricarta typus (Bleeker) (Plate XXXV., Figs. 2, 3). Four specimens from Corumba (331), one from Puerto Max (406), and two from Laguna Pasito, Ascuncion (460). — 25. Loricaria apeltogaster Boulenger. One specimen, Corumba (348). This specimen differs from the other specimens described in having the entire surface of the lower lip covered with short cirri. 26. Loricaria carinata Castelnau (Plate XXXVII., Figs. 1-2). One specimen from Puerto Max (405) and two from Corumba (330). Three from a brook at Villa Rica (465). ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXX\ Fig. 1. Hemdodontichthys actpenserinus (Kner.). Figs. 2-3. Loricaria typus (Bleeker). % v 12 ov i Dahan ae _Sial ‘ . - . ~ ¥ ‘ * : ~ ¢ =. .* uy ‘ ov Es ‘ . ’ . * os . . Tt) ‘ « = ¥ v . = ‘ 4 _ = hy a - . ’ ' - . . “Sag ; . r 1G 1 u ‘ ~ i . . if “> F' ANNALS CARNEGIE MUS®SUM, Vol. IV. Plate XXXVI. 'Fig. 1. Stur’soma robusta (Regan) ¢, dorsal view. Fig. 2. Sturtsoma robusta (Regan) 9, dorsal view. Fig. 3. Sturtsoma robusta (Regan) ¢, ventral view. ry a ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXVII. Figs. 1-2. Loricarta carinata Castelnau. Figs. 3-4. Loricaria labialis Boulenger. aie Vie Tiel ie Sa “ls i e wisps Ms PA ‘i g “f\ fi ; / a all’ ign : p> es { (a ‘© : * tee eds nt aR ne eee oe , ’ \¢ x Pye FOS ie Oe ak ft UD eas Leaner A . oi » a 7 ae bc : - a) ie -, > % J ae oh gt Y 4 4 Se ~e 4 = r . a : : we ; Plas af * » ; ‘ - ek, 4 io. we aA a fi i.” Aa W ‘ . = + _ a ¥ . * - 4 ; 7. ar ’ * > ae eee / ed — * a’ . ay - —— 4, oe : \ Beye PLoS Tra hh * % *hy 4 ~ ‘ 1 \ q A erg ad ¢ (“. few » Pay : : ia TA tee A oe ts Bl 7 é a _ = etc eet AS ee | , 2 1 OG 7 hay ena are ~ eo? vn . rhe | . ed ~ ee a he eee | a + bs yeh r / » Ms x ; ‘aa , 4 4 of + < eine Se a eee br a) on -_ ‘ ' * z iy Pye ae ‘ 4 4 ? j oF s A r. ae ' « ‘ * * ' 4 ¥ ’ ve jana ,* “— ‘ . 3 . - ‘ . a 7 -_ ¥ . R S é , wer Ps. * - a Sh a A? y fe : oT be Purl a “T 7 oa a, i. *% ’ i’ : “ a \ see ; a® $ i@ ’ { vn rs my 1 D4 a 6 : \4 ‘ « £ > : “ f nie ®t Vie ~- 1 a P — ' os te ‘. 7 ‘ J ‘ ve 7 a , _ % ts ; F Ha NLS ‘ re 1 aa & s “A os a < ore ee j a 5 hd ee ia ee aS ; ais: ig iL ’ > - F ar te ' , ° ON 6? le ee : ne ie ie sa eee y \ ' * \ r 77 A CAs , : al eo! i : m be 4 sf im ae > 5 * + Lars : - 4 P d ts ti“ * i 4 2 } Yt. 4 £ ’ F* & = i wv Veolia ‘ 7 + . ’ ‘ a Sete . y ais ala Va , ‘ ’ re ie * Fis ~ .} e : 4 y My 7 ' ieee al é \ yl Fi t C) ree: ‘ ‘ i a ob < 7% ie. 1% H - x J a at AY a o f : = ; : i * \ uM EIGENMANN: COLLECTIONS OF FISHES FROM PARAGUAY. 121 27. Loricaria parva Boulenger. Ten specimens, Corumba (339, 349, 300, 301). Four from Laguna Pasito, at Ascuncion (461). 28. Loricaria labiahs Boulenger. (Plate XXXVII., Figs. 3-4). Four specimens from Corumba (319). 29. Ofocinclus vittatus Regan (Plate XXXVIII., Fig. 1). One specimen (part of 304), Corumba; two specimens (446), tributary of the Pilcomayo. Description of the Corumba Specimen. — Body cued in front, compressed toward tail; its greatest width less than depth; profile straight, less steep than in afins. Occipital process terminating in an elevated, triangular process; ventral and dorsal profiles similar ; sides of head vertical, eyes distinctly lateral. Interorbital convex ; a groove running from snout to nares. All the bones of the head hispid ; head nearly covered with spines ; spines minute on occipital, strongest on border of snout. Orbit 2 in the snout, 4 in the head, 2 in the interorbital. Snout pointed. Lower surface of the head naked except a triangular space below the eye. Lateral plates hispid. * Distance of dorsal fin from tip of snout 2% in the length. Dorsal spine as long as head. Caudal forked. Pectoral extending little beyond origin of ventrals. Ventrals reaching nearly to anal. Color light brown above with a broad dusky bar extending from snout to end of caudal; tips of caudal dark ; black bands extending from end of dark band on middle caudal rays first backward, then forward, form- ing with the median band a y-shaped figure. neces. Ul, GA. 6; V. 6; Pie C. 16 3 Lat: 1.23: The two from Pilcomayo are very beautiful, well-preserved speci- mens, which are slightly longer than the Corumba specimen, being 34 mm. and 25 mm. long. They agree with it inall except the color, which is faded. ‘Traces are visible of the dusky lateral band and markings on the caudal fin. ‘The specimens have the appearance of having been preserved in some corrosive sublimate preparation, or having lived in a cave; one of them has D.I, 7. KEY TO THE SPECIES OF OTOCINCLUS. a. Sides with a longitudinal band. b. Middle caudal rays black. D.I, 6 or 7. c. Tips of caudal dark; black bands extending from end of the dark band on middle caudal rays first backward and then forward, forming with the median band a Y-shaped figure; a well-defined band from tip of snout to caudal, widest at the caudal peduncle, blackest on caudal. 122 ANNALS OF THE CARNEGIE MUSEUM. Eye 4 inthe head ; ‘lat. 1:23... ,e7cpebeasseee eee ee villatus. cc. Fins, aside from the basal caudal spot, plain; lateral band a weak line on lateral line's: lat OL 2m eee otk nig Oe 9 Balen ees vestitus. 6b. Middle caudal rays not black; D.I, 7; Dat. 123-25 © coca ence affinis. a@. Sidesispotteds “Di, 73 Acl, § 3" late Las. d¢. Six spots along lateral line; a series of corresponding spots along the back; dorsal and caudal spotted. Eye 4; lower lip thin and entirely naked. Hextlts. dd. ‘Two large spots along the lateral line; a series of corresponding dorsal spots ; caudal with 3 vertical dusky bars, sometimes spots. Eye 4.25; lower lip' coarsely tubeneular;.\304.a. keer tee eeed ene eke meee Jimbsiatus. 30. Plecostomus plecostomus (Linneus). Three specimens, Corumba (340). These specimens with several sent in the first collection which were not recorded from Ascuncion ; Rincomeda on Rio Apa, Arroyo Trementina give us a range of speci- mens from 60 to 390 mm. in length. ‘There is a great modification in the shape of the head. It becomes lower, broader, and very much more rounded with age. The caudal, which is cross banded in the young, becomes spotted. The dorsal which has a single series of spots between the rays comes to have two series. It is evident from the series that P/ecostomus boulengeri is but the young of Plecostomus plecostomus, as Regan has recently stated. 31. LPlecostomus johni Steindachner. Iam inclined to think that both Plecostomus commersoni and ver- miculdaris mentioned in the first paper on Paraguayan fishes and P/e- costomus ternetzt Boulenger should be placed here. ‘The vermicularis is asmall specimen and its identification must be more or less of a guess. | Our largest specimen is about 260 mm. long. In this the region in front of the gill slit, a band across the breast and a large triangular patch on the middle of the belly are granular and there is a series of larger plates along the sides of the belly between the pectorals and ventrals. Otherwise the lower surface is naked. In the smaller specimen there is also a median granular band between the ventrals. In the shape of the head, spines of the lateral plates, lateral line, etc., the specimens agree with Plecostomus ternetzi. Our larger one differs in having the dorsal with large spots, but these spots are quite obscure. In the smaller specimen the dorsal is uniform dark, as in the type of ternetzt. 32. Xenocara gymnorhynchus Kner. One specimen, mountain brook at Sapucay, Central Paraguay (452), 3 ey re A hie ac ae i Ul re ot hed «Si =) 2 mi < sa . - a 1 v, ‘ % 4 yi 7 fa - ¢ Tw , ‘ “« fi ‘ as o we La a ' 2 hal fa ‘ F ¥ § a -é ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXVIII Fig. 1. Ofoctnclus vittatus Regan. Figs. 2-3. Corydoras microps Eigenmann and Kennedy. Fig. 4. Corydoras aurofrenatus Eigenmann and Kennedy. EIGENMANN: COLLECTIONS OF FISHES FROM PARAGUAY. 123 CALLICHTHYID. 33. Callichthys callichthys Linneus. Two specimens from Bahia Negra (412). 34. Hoplosternum pectorals (Boulenger ). ‘ Two specimens from a swamp near Ascuncion (297). Two others, Laguna Pasito, at Ascuncion (463). 35. LHoplosternum littorale (Hancock). One specimen from Bahia Negra (411). 36. Corydoras microps Eigenmann and Kennedy (Plate XXXVIII., Pigs. '2,:\3-).. Four specimens, mountain stream, Sapucay, central Paraguay (453), six from small brooks at Colonia Gonzales, Villa Rica (467). 37. Corydoras aurofrenatus Eigenmann and Kennedy (Plate De XN IF, Fig. 4). Two specimens, small brook at Villa Rica (472). 38. Corydoras australe Eigenmann and Ward, sp. nov. Type No. 10129, one specimen, Corumba (304). Cotypes, 10130, five specimens Corumba (304); two specimens (447), tributary of the Rio Pilcomayo. This species is very closely related to hastatwvs and may be identical with it. The coloration of the caudal in the two species is identical, but the lateral band in Aastatus is jet-black, while in this species it is an indistinct line. Anterior profile strongly rounded and nearly vertical to the nostrils, less steep and almost straight from nares to dorsal. Head slightly longer than deep ; width 1% in its length; occipital process slender and triangular, meeting the dorsal plate ; fontanel quite elongate, ex- tending into the occipital bone ; preorbital small. Eye very large, orbit one in the snout, 3 in the head, 124 in interorbital. Mouth inferior, snout rounded, short ; rictal barbels extending to middle of eye ; lower lip terminating in two barbels. Coracoid process striate, forming a ridge on the side of belly. Distance of the dorsal spine from the snout 2 in the length, height of dorsal spine a little less than length of head. Caudal deeply forked. Pectoral spine longer than dorsal spine, its surface striate, comparatively free from serrations. Straw color. An indistinct dusky line from gills'to base of caudal, terminating in a large arrow-shaped spot which is bordered posteriorly with white, which itself is narrowly margined with blackish, the caudal dusky beyond (as in fastatus). A faint line from behind 124 ANNALS OF THE CARNEGIE MUSEUM. ventrals to behind anal. Head back of eyes dusky, the color extend- ing down and back on the sides posterior to the gill-openings. Head 3%; depth 2%;.D.1, 7; Pla 7 pV eee eee & CHARACIDA. 39. Hoplias malabaricus (Bloch.) ‘Trahira. Three specimens, Corumba (362); one specimen, Bahia Negra (396). 40. L[Hoplerythrinus uniteniatus (Spix). Two specimens, Bahia Negra (402). 41. Pyrrhulina australe Kigenmann and Kennedy. Five specimens, Upper Paraguay at Corumba (309, 316). 42. Psectrogaster curviventris Eigenmann and Kennedy. Blanquillo. One specimen, Bahia Negra (374), and one specimen Puerto Max (408). 43. Curimatella alburnus (Miiller and Troseieiga One specimen, Bahia Negra (375). 44. Curimatus elegans nitens Holmberg. Curimatus elegans paraguayensis E. & K. ‘T'wo specimens, Ascuncion (276). Twospecimens from mountain streams at Sapucay (457). 45. Curimatus bimaculatus Steindachner. One specimen, Corumba (357). 46. Prochilodus scrofa Steindachner. One specimen, Bahia Negra (396). 47. Anisttsia othonops’ (Eigenmann & Kennedy). One specimen, Bahia Negra (403). 48. Parodon paraguayensts Eigenmann (Plate XXXIX., Fig. 1). Nineteen specimens, Ascuncion (275). 49. Wanognathus boreli Boulenger (Plate XXXIX., Fig. 2). Anostomus fasciatus Eigenmann & Kennedy, Pica Phila. Acad. Sci. 1903, 512, 1904 (Rio Paraguay at Ascuncion and Estancia la Armonia). The specimens mentioned by E. & K. and three additional ones from Corumba (338, 357) and one from Puerto Max (409) differ from fasciatus in having no caudal spot. They differ from the typical dissimilis Garman in having but 40-42 scales in the lateral line and the head 4% in the length, instead of 4%. 7 Misprinted orthonops in E. & K.’s article, 7. c., 511. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XXXIX. aa eKGRAKALKS + 2 - + ofan, “ Or Gr fin Bin, i Fig. 1. Parodon paraguayensts Eigenmann. 2. Schizodon borelli Boulenger. . 3. Aphiocharax dentatus Eigenmann and Kennedy. ¥ 4 7 i es pha < Gap wo ee ¢ ‘ 43 a ® \ if a a =v . . ’ ro oi a — EIGENMANN: COLLECTIONS OF FISHES FROM PaRaAaGuay. 125 50. Leporinus frederici (Bloch). Two specimens, Puerto Max (407). 51. Leporinus trifasciatus Steindachner. A single specimen from Bahia Negra, 310 mm. (401), and another one from Ascuncion. Head 324; depth 4; eye 5% in the head, 3 in interorbital. Scales 6—42-5. 52. Leporinus affinis Giinther. One specimen (195), Arroyo Trementina. 53. Leporinus hypselonotus Giinther. Two specimens (413) from Puerto Max, are probably the young of this species. Length 18 mm. Depth 3; head 3. Body with 8 transverse bands, the last on the caudal peduncle. Some of these near the middle of the body have a tendency to split at the top, Y-shaped. Head with a diamond-shaped, dark blotch on top, enclosing an oblong, lighter portion, and with a line from eye, around lower lip, and a dark opercular blotch. Form somewhat different from that of the adult, especially as to the head. Lower jaw more prominent, dorsal profile of head not concave but forming part of an even line from dorsal fin to snout. Teeth strap-shaped, a pair of stronger ones set slightly behind the others in the mandible. 54. Characidium fasciatum Reinhardt. Two specimens, one (471) from Villa Rica and the other (448) from a tributary of the Pilcomayo. ‘The latter specimen is very pale, having a faint lateral band, a black spot at base of middle caudal rays and the lateral bands faint, most distinct on the back. The Villa Rica specimen, on the contrary, is highly colored, the lateral band wide and the vertical bands distinct. The dorsal with horizontal bands and the caudal with cross-bands. Dei; A: 8; lat.1.37. The pectoral in the.Pilcomayo specimen is feeble, not reaching the ventrals ; that in the Villa Rica specimen is strong and reaches a little beyond origin of ventrals. The Villa Rica specimen has the general habit and coloration of a darter ( Z¢heo- stoma). 55- Odontostilbe paraguayensis Eigenmann & Kennedy. Five specimens, Corumba (329). 56. Odontostilbe trementina Eigenmann & Kennedy. One specimen, Puerto Max (part of 388). 126 ANNALS OF THE CARNEGIE MUSEUM. 57. Chetrodon interruptus (Jenyns). Puerto,Max (part of 388). Seven specimens (473), Colonia Gonzales. 58. Aphiocharax dentatus Eigenmann & Kennedy (Plate XXXIX., Fig. 3). Nine specimens from Corumba (272, 282, 274, 367) one from Puerto Max (386), and ten from a small tributary of the Rio Negro, a tributary of the Rio Pilcomayo (443). Back deep*yellow, opercle golden. In the!original account of this species three typographical errors occur. For type No. ‘‘10030’’ read 10032, and under cotypes for “*70030 and 10031’ read Loosqjand 16je35, 59. Hemigrammus luetkent Boulenger. (Part of 388), Puerto Max. Three specimens from a pond at Colonia Gonzales, at Villa Rica (470). 60. Hemigrammus kennedyt Eigenmann. Two specimens from Corumba (327), and several from Puerto Max (389). 61. Hemigrammus ulreyt (Boulenger). Specimens 34—40 mm. long, Corumba (328). 62. Tetragonopterus argenteus Cuvier. Lambari. (Plate XL., Fig. 1.) Zetragonopterus chalceus Kigenmann & Kennedy, 1. c., 523. Eight specimens from Porto Murtinho (287). Five specimens from Bahia Negra (383). . 63. Zetragonopterus allent® EKigenmann & McAtee sp. nov. (Plate XL. Pres 23 Mojarra Lambari. Type No. 10158, a specimen 110 mm., Cor- umba (363). Cotype, No. 10159, a specimen 8s mm. Corumba (363); Cotype, No. 10160, a specimen 95 mm., Rio Otuquis (372); Cotypes, No. Io161, two specimens, 60 and 95 mm., Ascuncion (280). This species agrees very closely with specimens of Astyanax mu/tt- radiatus from Ascuncion, Rio Paraguay, but differs notably in the greater depth, and the sharp up-turn of the nape making the profile distinctly a Zetragonopterus profile. It is very probable that Stein- dachner’s Zet/ragonopterus multiradiatus is a true Tetragonopterus, in- asmuch as his largest specimens were but two inches long and yet had 8 For J. A. Allen, of the American Museum of Natural History. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XL. Fig. 1. Tetragonopterus argenteus Cuvier. Fig. 2. Tetragonopterus allent Eigenmann and McAtee. Fig. 3. Astyanax pelegriné Eigenmann., EIGENMANN: COLLECTIONS OF FISHES FROM PARAGUAY. 127 a depth equal to half the length. It is more than probable that the Astyanax multiradiatus of Eigenmann and Kennedy is a species dis- tinct from Steindachner’s. Head 4; depth 2; D. 11; A. 40-41; scales 11-50-8. Compressed, more elongate than in avgenfeus ; anterior profile con- cave: Snout blunt, interorbital convex. Eye 234 in head, not equal to the interorbital, 14f longer than snout. Maxillary small, not reach- ing to below eye; no teeth on maxillary. Origin of dorsal equidistant from tip of snout and base of middle caudal rays or nearer the former, inserted distinctly behind the ven- trals, not falcate, the longest ray not nearly reaching adipose ; anal not falcate, the rays of nearly the same height its entire length ; ventrals about reaching anal; pectoral a little beyond origin of ventrals. A faint humeral and caudal spot, an indistinct silvery lateral band, fins gray, outer pectoral ray dark. ASTYANAX Baird and Girard. ASTVYANAX Baird and Girard, Proc. Acad. Nat. Sci., Phila., Vol. VII., p. 27, 1854 (argentatus ).. Pecilurichthys Gill, Ann. N. Y. ye Nat. TLISt., 1585, D.. SA, (drevoortit = bimaculatus ). The genus Astyanax is one of the oldest and most widely distributed genera of South American Characins. It is now one of the dominant South American genera, being found from the Rio Negro on the bord- ers of Patagonia to the United States, and on both slopes from Peru north to Mexico. Its counterpart is found in Africa as Pefersius. It has hitherto been associated with a number of other genera under the name Zetragonopterus with the following characters : Premaxillaries with two series of multicuspid teeth, none of them enlarged ; maxillary with or without teeth ; mandible with a single series, those in front very strong, graduated and multicuspid, those of the side abruptly minute and usually conical; gill-rakers setiform ; no predorsal spine ; maxillary short ; the snout and maxillary together less than half the length of the head. The species with the characters just noted represent several genera which may be divided as follows: a. Caudal and anal both naked. 6. T.ateral line interrupted. 128 ANNALS OF THE CARNEGIE MUSEUM. c. Maxillary with usually conical teeth along its entire edge. Hlolopristis Eigenmann. cc. Maxillary with o-5 conical or multicuspid teeth at its anterior upper GAGS hoc cfd rma sn adedenicaye mee eon eRoe nee meee eee Hemigrammus Gill. - 66. Lateral line complete. c. Maxillary with o-I0 teeth. ad. Nape rather abruptly elevated, profile very concave, depth less than half the length, preventral region flat, with lateral keels. Tetragonopterus Cuvier. dd. Profile little if at all concave, depth usually less than half the length, preventral region rounded. Astyanax Baird and Girard. cc. Maxillary with teeth along its entire edge......... Hemibrycon Giinther. aa. Caudal scaled. é. Anal scaled, long, rounded, with over 40 rays......... Markiana Eigenmann. ee. Anal naked, with 23-28 rays. j- Premaxillary teeth, parallel’ 3 /.2.0)...2eens Menkhausia Eigenmann. ff. Anterior series of premaxillary teeth in a wavy line. Bryconamericus Eigenmann, KEY TO THE SPECIES OF ASTYANAX BAIRD AND GIRARD. a. Anal rays 30 to 49. 6, Lateral line 55; anal 45; head 3.6; depth 2.63; lateral line 55. Eye greater than snout by one third of its diameter, greater than interorbitall by £4 of M5 | AiAMNObERs 55240105. c0hs ones coe eee erythropterus (Holmberg) 1. 66. Lateral line with 50 scales or fewer. c. Anal 40-48. ad. Scales 42-50. e. Anal 47-48 ; scales small, 47 in lateral line. J. A few rudimentary teeth on the maxillary ; asilvery lateral band ; ventral profile much arched; snout pointed ; A. 48.............spilurus (Cuvier & Valenciennes) 2. ff. No teeth on the maxillary; a very distinct silvery lateral band ; an indistinct caudal and humeral spot ; maxillary reaching anterior margin of orbit ; depth 2.3; eye 3 in. the head ; scales 9 or 10-47-10 or 11; A.47. hauxwellianus (Cope) 3 ee, Anal 41-45. g. Scales 8-46-8 ; two teeth on maxillary ; lateral band not conspicuous ; humeral spot bordered with silvery in front and behind; caudal spot large, continued to end of middle caudal rays; depth 3; eye 24%; A. 41. rivett Pelegrin 4. gg. Scales 10-45 to 50-8 or 9; a small silvery lateral band ; a dark spot on shoulder and one on base of caudal 9 9 These spots are very obscure in alcoholic specimens, while in specimens first pre- served in formalin they are conspicuous and a black band replaces the silvery band of the alcoholic specimens. ——— EIGENMANN : COLLECTION OF FISHES FROM TPARAGUAY. 129 depth 2%-224; A. 41-45; maxillary with a single, rather large caducous tooth ; eye 2.4 in the head, about equal to interorbital ; head 4.1-4.5. pelegrint Eigenmann 5. ggg. Scales 8-42-7 or 8; A. 42; eye 3.7 in the head, 1.7 in the interorbital; head 3.7 in the length; a caudal SPOt 2.5. hcaptueieeeeensar vedas correntinus (Holmberg) 6. dd. Scales 37-38. hk. Several small teeth on the maxillary; head 32; cepth 3 or somewhat more than 3; snout 4 in the head; a silvery lat- eral band bordered above by blue-green; a round humeral spot; caudal spot, when present, extends to the end of some of the caudal rays. A. 43; scales 6-37 or 38-4. batrvdzt (Steindachner) 7. Ah. Maxillary with a single smali tooth. Scales in lateral line 37 or 38; caudal and humeral spots round, both very dis- tinct; middle of caudal fin thickly dotted with black ; maxillary not reaching orbit; head nearly 4; depth about BY, i} LM. QO-A2 ase iaboetgeseuane tabatinge (Steindachner) 8. cc. Anal 30-41. j. A silvery lateral band ending in a black caudal spot, sometimes two lateral spots in front ; maxillary very short, 2 in the eye; depth 2%-3; head 4-4%; eye 2%-2% in head; D. 11, posterior to ventrals; A. 35-40. Lat. line 44-47. jeste ( Boulenger) 9. jj. A black band extends from base of caudal fin along its middle rays; maxillary extending to eye; eye 223-3 in head; depth 224-234 ; head 4; A.34-40; scales 8 or 9-40 to 50-8 or 9; last dorsal over first anal ray........ brevirostris (Giinther) 10. 27, Scales in the lateral line less than 40 (except in adramis, sometimes in brevoorti and bimaculatus). k. Anal rays 36 or fewer, rarely as many as 39 in dzmaculatus ; maxillary with a few teeth or none. 7. Depth more than three in the length, the form long and slender ; maxillary without teeth. m. No caudal or humeral spots; a silvery lateral band, most distinct posteriorly ; head 3144; eye 2% in the head. A. 30; scales 5—35-3..as¢zctus (Ulrey) 11. 71, Depth less than 3 in the length; a distinct caudal spot. n. No humeral spot; depth 2%4-23; eye 324-4 in the head; maxillary with o-3 teeth, extending consid- erably beyond anterior margin of eye; a conspicu- ous black band on the caudal peduncle, becoming wedge-shaped on the caudal. A. 29-31. Scales 7 or 8-37 or 38-6...... maximus (Steindachner) 12. nn, A distinct humeral spot. (See adramis.) o. Maxillary long, extending to below center of eye ; 130 ANNALS OF THE CARNEGIE MUSEUM. eye 3in the head; maxillary without teeth; a black humeral spot, a silvery lateral band, becoming black on the tail and extending up on the caudal. A.30; scales 8-37 or 38-7; depth 297%, sn seeaeaone moort ( Boulenger) 13. oo. Maxillary extending little, if any, beyond origin of eye. fp. Dorsal plain. g. Body deep; humeral spot usually hori- zontally elongate. r. Scales in 19 or 20 rows, 10-43 to 47-8 org. Anal 28-32. Caudal plain or with an indistinct spot ; humeral spot indistinct or want- ing. Dorsal distinctly behind the ventral, the pectoral reach- ing the ventral. Depth 2%; head less than 4; scales in 18 or 19 rows, 10-43 to 47-8 or 9. abramis (Jenyns) 14. vr. Scales in fewer than 17 rows. s. Moderately compressed. ¢. Teeth of the inner series of the premaxillary with their posterior surface convex, the denticles corresponding to the convexity, arranged in a U-shaped line. v. Scales 6 to 8-30 to 40-5 to 8; max- illary with o-4 teeth, extending somewhat beyond the front margin of the eye. Depth 2-2.1; head 4-43 ; eye 3 in the head. A. 27-39. bimaculatus (Linnzeus) 15. tt. Teeth of the inner series of the premaxillary alike in front and be- hind, the denticles ar- ranged in a_ nearly straight line ; scales 6- ——————————————— ‘ EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. 131 40-6 ; maxillary with a single tooth, extending to below origin of pupil ; depth 2.4; head 4; eye a little more than 3 in head.),;.A..33- orthodus Eigenmann 16. ss. Greatly compressed. Scales 7 or 8-39 or 40-6 or 7; a humeral and a caudal spot; the pectorals extend be- yond origin of ventrals ; maxillary toothless, ex- tending somewhat beyond front margin of orbit ; head 34-333 depth 2%-2Y ; A. 38-41. caucanus (Steindachner), 17. gg. Humeral spot circular or vertically elon- gate. uw. Depth 2-2.2; scales 8 or g—36 to 40-8 or 9; a humeral and a cau- dal spot; maxillary with one tooth; head 3.66; A. 38-39; dorsal and _ ventral outlines equally curved. atratoensis Eigenmann 18, wu. Body more elongate; humeral spot circular. Anal rays 38. A silvery lateral band. Dorsal a little behind the origin of the ventrals, the anterior anal rays elongate. Maxillary extending to near the anterior margin of theeye. Depth2.4; head 3.5; scales 8-39-10. stilbe (Cope) 19. fp. Dorsal plain or with a broad, oblique, dark band across the middle. Scales 8—39-7 ; maxillary extends a little beyond the origin of theeye; D.11; A. 31; depth 234 ; head 3%...dar¢lettii (Giinther) 20. aa. Anal rays 26-29. (Sometimes as many as 36 in cordove. ) v. A series of seven, deep brown, longitudinal bands. Maxillary extending little, if any, beyond anterior margin of eye, with one tooth. Head 32; 132 ANNALS OF THE CARNEGIE MUSEUM. depth a little more than 2 in the length. A. 27; scales 5-31-4. (See Also wider: bimraculatis. )...0;secancss denveneener steindachnert (Eigenmann) 21. vu. A single lateral band or none. z. Caudal without vertical band, x. Scales 28-35. y. A caudal spot. z. Humeral spot horizontally oval. A. Lateral band black. (See also démaculatus.) Scales in 14 rows. Dorsal fin behind the base of the ven. trals. Maxillary with one tooth extending a little beyond the anterior border of the eye; interorbital space much greater than the eye. Head 3%; depth 3; A. 28; scales 7-34-6. wappi (Cuvier & Valenciennes) 22. zz. Shoulder spot if present vertically elongate, sometimes faint. £&. Caudal spot band-like, continued on the middle caudal rays; scales in less than 15 rows. C. A humeral spot. YD. Dorsal behind the ventrals. £, Maxillary with 2-7 teeth; anal 27-32; depth 2%%-3; head about 4; maxil- lary equal to eye; eye 3 in head, equal to interorbital. Scales 7-37-6. rutilus mnicaraguensis Eigenmann 23. EE. Maxillary with o-2 teeth. F. Head 4, or more than 4, in the length. A band-like caudal spot sometimes extending for- ward to the indistinct humeral spot. Maxillary extending dis- tinctly beyond the anterior mar- gin of the eye. Head 4-4%; depth 2.25-3; A. 25-30; scales 6 or 7-30 to 39-4 % to 6. rutilus (Jenyns) 24. FF. Caudal bordered above and below by yellow; a silvery lateral band ; a humeral spot ; head not 4inthelength. A. 27-35. Scales 7-38-7. Maxillary with a single tooth. teniurus (Gill) 25. LFF, A. 24-32; origin of the dorsal behind the base of the ven- trals ; maxillary reaching an- 0 Rutilus jequitinhonhe has depth 3. / EIGENMANN: COLLECTION OF FISHES FROM ParaGcuay. 1338 terior border of the orbit ; head 4; depth 2.25-2.75; scales 7 to 9-35 to 39-6 or 7. rutilus eneus (Giinther) 26. FFFF, Depth 3.33; head 3.66; A. 29; scales 7-37-5; maxil- lary long, equal to eye; eye 2.5 in head; _ interorbital 3.66...cuvieri (Liitken) 27. DD. Dorsal over ventrals ; snout less than 4 in the head ; interorbital flattish. A. 27, beginning behind the dorsal. Depth 3; head 4; lateral line 37. petenensis (Giinther) 28. CC. No humeral spot. G. Dorsal behind ventrals ; snout about 4 in the head ; interorbital convex. A. 26-30, be- ginning under last ray of dorsal; depth 3-314 ; head 324-4. Lateral line 36-40 ; no maxillary teeth ; a silvery lateral band which may become black on caudal. stmus (Boulenger) 29. GG. Head 3%-4% ; depth 23/-3; interorbital much more than diameter of the eye. Maxillary terminates below the anterior margin of the eye; origin of dorsal be- hind the ventrals. A. 26-30; scales 6-35 to 37-6 to 8. peruanus (Miiller & Troschel) 30. £B, Caudal spot not continued on the middle rays. #7, Scales in 15 rows, 7-36-7. Humeral and cau- dal spots present, anal and ventral with broad red margins; eye 324 in the head, much smaller than the convex interorbital. Maxil- lary extending to anterior margin of eye. A. SAFSA3LA tte cower humilis (Giinther) 31. HH. Scales in 10-14 rows ; lateral line 34-38. f. Head 4, or less than 4, in the length. J. A humeral spot. Maxillary reaching anterior border of eye, not to end of first suborbital. Head 33(-4; depth 2.4-2.7; A. 23-28; scales 6% or 714-34 to 37-5% to 6%. Twoor three maxillary teeth. Jischeri (Steindachner) 32. J/. Maxillary ceases in front of the verti- 11 See also eneus. 134 ANNALS OF THE CARNEGIE MUSEUM. cal from the pupil and end of first suborbital. Dorsal considerably be- hind origin of ventrals; pectorals reaching beyond ventrals. Eye about equal to the slightly convex interorbital, 3 in the length of the head. Depth 23; head 3%; A. 26; scales 614-37 or 38-5; a humeral spot.....cavoline (Gill) 33. /f. Head 4 or more in the length ; scales 5 %4- 35-4. Anal, dorsal and caudal fins with red markings; scales in 9% rows. Caudal and humeral spots indistinct. Head 4.2; depth 3.3; A. 26-27 ; scales 5 14-35-4... 0 phenicoplerus (Cope) 34. yy. No caudal spot; a humeral spot. &. Maxillary with 3 small teeth; eye 2.3 in the head ; snout 4.5, the flattish interorbital 3; depth 2.66; head 3.6. A.28; scales 5-35-4; a silvery lateral band, a vertical humeral Spot. 12.600. Se asratauaees megalops Eigenmann 35. KK. Maxillary, with numerous minute teeth, extending be- yond the front of the orbit. Snout shorter than eye ; eye 2.8-3. A. 29-30; scales 5 or 6-32 0r 33-4 to 5. Head 32; depth 24%-2% ; humeral spot elongate WOHIZontAlly... cess vaviccesss bahiensis (Steindachner) 36, KEK. Maxillary without teeth. Silvery band not edged with green above; origin of dorsal between ven- tral and anal; pectorals to origin of ventrals, ven- trals nearly to anal. Head 434 ; depth 31%; eye 2;%, in head; D. 10; A. 27; scales 5—37-3. alburnus (Hensel) 37. xx. Scales 9-40 to 46-10. Longitudinal series of scales 20. Caudal and humeral spots generally absent; origin of dorsal over root of ventrals. Width of interorbital greater than the diameter of the eye. Head 4; depth 3; A. 26-31. Interorbital very convex, 224 in head; eye equals snout, 324 in head; depth of caudal peduncle about \% of the depth...cordove (Giinther) 38. xxx. Scales in 15 or 16 rows, 8-39 to 45-7; A. 28 or 29; head 3.6 or 3.5; depth 2.7-3; eye large, 2,6-2.8; interorbital 3.25 in head; maxillary as long as eye, with two narrow teeth ; dorsal behind origin of ventrals; humeral spot faint, caudal spot distinct, not continued on middle rays. emperador Eigenmann & Ogle 39. aaa. Anal rays 16-25, rarely 26 or 27 in fasciatus and eneus ; 3, to 5 series of scales below the lateral line in all but e@eus (6 or 7) and fasciatus (4% to7%4). ZL. Maxillary without teeth ; two or three scales below the lateral line. M. No caudal spot. EIGENMANN: COLLECTION OF FISHES FROM ParRaGuay. 135 NV. Depth 4.7. Longitudinal series of scales 12. A broad silvery lateral stripe ; no caudal spot. Maxillary toothless, rather wide, extends little beyond anterior border of the orbit. Head 4.2; epi 4. 7's A. 25. sie ate dca sete cst cos ene deo longior (Cope) 40. NN. Depth 2.5-3.5 in the length. O. A silvery lateral band sharply edged above with a dark band. Dorsal fin a little behind the ventrals. The pectorals not entirely reaching the ventrals, the ventrals reaching the anal. Head 34; depth 3-334 ; scales 5-32-3. A. 21-22. copet (Steindachner) 41. OO, The broad silvery lateral band not edged above with dark. Maxillary extending nearly to the front margin of the eye ; anterior dorsal and anal rays elongate. Head 42; depth Se §° Scales: 4=35—9ir Rl kG 231.42. «. diaphanus (Cope) 42. VOO. The narrow silvery band edged with greenish above ; humeral spots distinct. Origin of the dorsal just behind the ventral; the pectoral reaches to the middle of the base of the ventral. Head 32-334 ; depth 24-2§; A. 24-25; scales 5-32 or 33-3...cod/etti (Steindachner) 43. MM. Caudal and humeral spots present. Anal 24; scales 5-28-3. Head 4; depth 2%. Origin of the dorsal fin behind the ventral. Max- illary extending to the front margin of the eye. oligolepis (Giinther) 44. LL. Maxillary with teeth. P. Middle caudal rays black. Q. One to three teeth in the maxillary. Rk. A. 18-25. Lateral line with 37-40 scales; a silvery-gray lateral band; a dark caudal spot fading out forward. Usually a humeral spot. Maxillary with two small teeth. RR. A. 24-27. Origin of the dorsal behind the base of the ven- trals; maxillary toothless, reaching anterior border of the orbit. Head 4; depth 24%-23f; A. 24-27; scales $0 -S5- to JO—G Ole kar veccscesnienes eneus (Giinther) 26. RRR. A. 18-25, rarely as many as 27. A dark caudal spot ex- tending to the end of some of the rays and fading out anteriorly ; a silvery lateral band; an indistinct humeral spot ; ventral and pectoral fins with red. Head 4, longer than its depth at the occiput; depth 214-3; scales 5 or 6-33 to 40-4% to 7................ fasciatus (Cuvier) 46. RRRR. A. 19-21; three or four scales below the lateral line ; head 4-414 ; depth 2%-3; scales 5-35 to 37-3 or 4; eye 3 in head........fasczatus theringi (Boulenger) 47, QQ. Numerous minute teeth on the maxillary, which extends either a little or distinctly behind the front margin of the orbit. Eye small, 3.5-4 in the head. A vertically elongate humeral spot 136 ANNALS OF THE CARNEGIE MUSEUM. and a longitudinal caudal stripe extending to the end of the rays and fading out anteriorly, Head 34-334 ; depth 3-3% ; scales 5-33-4. A. 16-18..............7ezynst (Steindachner) 48. PP. Middle caudal rays not black. S. Scales in lateral line 34-38. T. A. Igor 20." Almost colorless ; caudal rays sometimes dusky, a grayish lateral band. Pectorals reaching 24 to ventrals, ventrals 24 to anal. Maxillary, with 2 teeth, reaching to eye. Head 4, depth 31-334 ; scales 4 or 5-35 to 38-4. mankhaust (Eigenmann & Kennedy) 49. 77. Anal 17-18. Yellowish above, white on the sides; lateral band plumbeous above, silvery below; a plumbeous humeral spot. Head 4% to 4%; eye 2% in head; depth 31-32. Lateral line 34-36. Dorsal band, large part of anal, the body near it and the median spot of the caudal more.or less. red. ./sivahesees rubropictus (Berg) 50. 77T. Anal 18; scales 37 or 38; depth 31; head 44; plumbe- ous, fins dusky; pectorals reaching ventrals; ventrals nearly to anal; caudal Jobes rounded; snout blunt, lower jaw distinctly shorter than upper; a faint humeral spot. mouth very small, maxillary not reaching eye; eye 314 in head; interorbital very convex, less than 3 in head; depth of caudal peduncle little less than half the greatest GEPUM hen, stancnes so eigenmanni Evermann & Kendall 51. SS. Scales 5-31-3. Silvery lateral band; a diffuse caudal spot; no humeral spot. Head 3% ; depth 23f. A. Ig. paucidens (Ulrey) 52. 64. Astyanax pelegrint Eigenmann (Plate XL, Fig. 3). Pecilurichthys multiradiatus Eigenmann, Proc. Phila. Acad. Sci., 1903, 521. (Not Zetragonopterus multiradiatus Steindachner. ) Steindachner describes his mu/tiradiatus as having depth 2; head 3.6; eye 3 in head = interorbital; A. 40-41; scales 10-41 to 42-9. Inasmuch as Steindachner’s specimens were 2 inches long and their depth cannot be ascribed to old age, they should very probably be classed with typical Zedragonopterus. Several specimens from the basin of the Paraguay, mentioned in the paper quoted above, and an additional one (380) from Bahia Negra, have the following formula: Depth 2-2.5; head 4-4.5; A. 41-45; scales 9 to 11-45 to 52-7 to 8. These represent a species certainly distinct from the +wuw/tradiatus of Steindachner. It may be identical with Cope’s hauxewellianus from Peru, from which it differs in the number of anal rays, there being 47 in hauxewellianus. 2 26 in one specimen. EIGENMANN: COLLECTION OF FISHES FROM PaRaGuay. 1387 65. Astyanax bimaculatus lineatus Perugia. Many specimens, Ascuncion (281) ; Corumba (344, 368) ; Bahia Negra (380) ; Puerto Max (388, 389, 393) Sapucay (456). Steindachner states that dimaculatus (maculatus) varies in the shape of the body, with age and sex, and with the habitat in rapid clear waters or in stagnant water. Variation in depth, anal rays and scales of A¢maculdatus recorded by : Length Depth A. Scales I. Giinther 31-34 7-39-7.5. 2. Steindachner, 214-3 inches 2.4-2.5 os} Oe EE ane Bed) ra 7.5 or 8.5-35 or 36-6 to 7.5. $ 4% 6. > 2. 33-24 ¢ 3. Steindachner, Cauca & Magdalena 2.1-2.5 32-39 7.5 or 8-35 to 36-7. 4. Liitken (acustris) 2 24-32 5 to 7-34 to 36-41%. Rio Grande do Sul 27 7-39-5. Sixty-five specimens from the basin of the Paraguay have the fol- lowing rays, counting the rudimentary ones: Two have 27, one has 28, four have 29, thirty-two have 30, eighteen have 31, five have 32 and three have 33. The specimens from the Paraguay system have usually a small accumulation of pigment cells at the tips of the scales which form horizontal series. It is probable that such a specimen has served Perugia for his type of Zeatws. Perugia’s specimen had 28 anal rays with nothing said as to whether or not the rudimentary rays were counted. It is seen above that some specimens have 28, not counting the rudimentary rays, so that his specimens fall within the limits of bimaculatus whether he counted these rays or not. The most impor- tant difference is in the number of scales in the lateral line, which, according to Perugia, is 34, whereas the specimens examined by us from the basin of the Paraguay have the scales 36-40 in the following ratio: twelve have 36, twenty have 37, seventeen have 38, eight have 39 and one has 4o. There are no other differences and I have little doubt that Perugia’s lineatus is identical with dcmaculatus. If it is desirable to distinguish the differences which the Paraguayan specimens show they may be termed Astyanax bimaculatus lineatus. 66. Astyanax fasciatus (Cuvier). P. scabrifinnis Eigenmann & Kennedy, 1. c., 521. 138 ANNALS OF THE CARNEGIE MUSEUM. (With 281), Rio Paraguay, at Ascuncion; four specimens from Villa Rica (part 470). 67. Astyanax theringt (Boulenger). Seventeen specimens (455, 456), from mountain streams of Sapacar 68. Menkhausia dichrourus (Kner). (Plate XLI., Fig. 1). — A single beautiful specimen 57 mm. long (299) from the Rio Para- guay at Tuyuyu, Matto Grosso, probably belongs to this species. The proportion and measurements differ somewhat, and are as follows: D. 9; A. 26; depth 3.5 (the type has depth 2%); head 4.5; scales 5— 35-3. Maxillary without teeth, much curved. Middle caudal rays and the tips of all the rays dusky or black, the tips of the lobes beyond the scales black. A well-defined lateral band extending from caudal to below front of dorsal where it is rapidly contracted and tends to disappear before reaching the faint humeral spot. A low adipose ridge extending from the adipose fin half way to the dorsal. We have other specimens from Corumba and Asuncion. 69. Menkhausta agassizt (Steindachner) (Plate XLI, Fig. 2). Two small specimens Nos. 10164, 40 mm. and 10165, 33 mm. from Corumba (305) agree very closely with this species but differ in having the lateral line incomplete. If this is a constant character and not the condition of the age of the specimens, these specimens must be held as species distinct from agasszz¢. Until further specimens are available this question may be left in abeyance. The great similarity of the specimens with a complete lateral line and an incomplete one is in itself not proof that the two sets of speci- mens belong to the same species. There are remarkable cases of parallelism between other species concerning whose generic difference there cannot be the slightest doubt. Head 3.5; depth 2.5; D. 10; -A. 23; scales 5-24-3342; pores on 7 or 8 scales. Maxillary without teeth; eye 2.5 in the head; snout 4%; body deep, robust but moderately compressed ; dorsal and ventral outlines equal ; caudal peduncle short and deep. Dorsal behind ventrals, the highest ray about equal to the length of the head ; caudal lobes about 114 times as long as the middle rays ; highest anal ray about twice the height of the lowest, the fin not dis= tinctly falcate ; ventrals reaching the anal; pectorals reaching beyond origin of ventrals. Top of head and tip of lower jaw minutely but densely punctate ; large b ANNALS CARNEGIE MUSEUM, Vol IV. Plate XLI. Fig. 1. Maenkhausta dichrourus (Kner). Fig. 2. Menkhausia agassizt Steindachner. Fig. 3. Deulerodon tguape Eigenmann. (Type.) wer Aer 7 . mr er Y a . urd ae, ae e r mi f Ae z yal ¥) i) ALA wa) OL aa ; cb Bae? Oph? Er RO 9 oe Sa WOU Lat ns Napa DE RCETC A hy a ae eevee age ots pene st aa aes “fi : ag : hee. wel aay Yi, 3 ¥e |") ,, sh peak ta ROR," aicigh ai calle AEA AUNTY catty Ne a rn | >» Le ae J eh tty “ue aL TP aed ¢ Pe a! @, Ce ‘ ,* ‘ Lary zi , oghes (hu), Bo seer | A. aL ae Pe Cel © wt ih ay * wy ne Lt er ra Be cre ey | ; Pe? are he, fag : “os SUAS 7 i os 4 ete q ; ey 7 wy tte ee F r ‘ A Tp Cts? cub ohn pe ‘i yer 7 * : a} _— = ne r : ai) as ¥S, a ae i = ¥ . 1 } i ‘crs %. mJ : 4 4 > a, * - fea pen 7 2: sh Pri ie oe Lan a NE RR Sopa Oye fi 7 ; i we tue = i 7" Ay 4 ; 5 - q ho . ae o ihe Papceehel eu %, PIN ant vn ear ce? Maa aoe 4 “9 a . yy ~. ‘A ~ -_ F r iv » rk , 7 tay r Piel 7 ~ -~4 { ol , a) : y ; od <4 by i> ak y i re AS 7 @ 5 ; j rr 06 pea 7 : Ete at it) As ed + é ~ ‘ N : ie ‘J a j i 4 f, 4 4 - it o rn 4 «| soe J o , ae ‘ ’ es @ ve J ~ S a J - ‘4 A J “4 ’ ” “a a ' % eM h } 5 x fs ' ! ‘ bee ‘ wn * ¥ y i « yn os ' ; 7” : ; S , : .f . * ’ ‘ “ a . ’ -~ - Phy * ® sll gt 8 2 a. in , sf EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. 139 color cells on upper part of cheek and opercle ; back and sides above the lateral streak densely punctate, the margins of the cells blackish ; a row of cells along the margin of the scales below the lateral streak ; belly and breast colorless; region above anal peppered; a vertical shoulder spot and a dark lateral streak ; back of caudal peduncle punc- tate, the peduncle otherwise without pigment except a narrow posterior margin, which, together with the basal part of the caudal is jet black, the band being about as wide as the eye and slightly oblique ; caudal immediately behind the band without pigment; tip of caudal and all the other fins punctate. 70. Lryconamericus exodon Kigenmann gen. et sp. nov. Pecilurichthys dichrourus Eigenmann & Kennedy, in part, Proc. Acad. Nat. Sci. Phila., 1903, 522 (Ascuncion). Type: No. 10298a, Indiana University Museum, a specimen 45 mm. long, over all. Puerto Max, J. D. Anisits. Cotypes : 10298, Indiana University Museum, 5 specimens. Puerto Max. Cotypes: 10297, Indiana University Museum, 3 specimens, Ascun- cion. Cotypes: 11264, Indiana University Museum, 2 specimens, Ascun- cion. The specimens differ from all other astyanaciform species I have been able to examine in the position of the anterior teeth of the pre- maxillary. I have both males and females and there are no apparent secondary sexual differences. Head 4.4-4.33; depth 3.66-4; D. 9 or 10; A. 23-25; scales 5-39 or 40-4. Slender, tapering regularly from near the middle of the pectorals to the caudal; head blunt, mouth moderate, terminal ; maxillary extending beyond origin of eye; snout and maxillary not quite 2.5 in head; maxillary not slipping under preorbital; cheeks covered by suborbitals ; interorbitals slightly convex, very little greater than eye ; eye not quite 3 in head. Maxillary with two, three, or more pointed teeth; premaxillary with four, five, or more pointed teeth of the usual sort in the inner series ; outer series consisting sometimes of five teeth in each premaxillary, the second and fourth teeth withdrawn from the line of the first and fifth and alternating with the space between the first and second and third teeth of the inner series; the anterior series thus form two imperfect series. The first, third, and fifth teeth directed forward 140 ANNALS OF THE CARNEGIE MUSEUM. slightly ; sometimes there are but four or three teeth, in which case the first and last are directed forward and the other one or ones are withdrawn from the line connecting them. . Dentary with four large, many-pointed teeth and several small, mostly conical ones on the side. Dorsal considerably behind the ventrals ; adipose well developed ; anal margin oblique, but little concave; ventrals small, not reaching anal; pectorals not ventrals. A small humeral spot; a well-defined silvery lateral band from in front of dorsal to caudal; middle caudal rays, margin and tips of caudal lobes black. DEUTERODON * Eigenmann gen. nov. A genus of Tetragonopterinz distinguished as most of the genera of Caracidee by its dentition. Teeth in the premaxillary in two series, those of the outer series few (2 on each side), and separated from each other, expanded at the tip, each with a large median and two small lateral cusps ; teeth of the second row very broadly expanded at the tip, each with a long, pointed median cusp and three graduated cusps on the sides ; the teeth becom- ing smaller, but not notably so toward the sides, 5 on each side ; maxillary with a few (2 on one side, 3 on the other), similar teeth ; mandible with a single series of teeth, 10 on each side, regularly graduated from in front back; the teeth little expanded at the tip, each with a large and strong median cusp and two or three much smaller lateral cusps ; the teeth and jaws so arranged that their action is shear-like, in contrast to Astyanax and related genera where the lower jaw is apparently thrown forward when opened and its anterior teeth point up and back when it is closed, the arrangement being similar to that in the Myline, the second series of teeth in Astyanax being further back than in the present genus. Gill-rakers setiform as in Astyanax ; no precumbent dorsal spine; nares close together ; gill-membranes free from the isthmus and from each other ; lateral line complete. Eats plants. Its teeth enable it to cut out pieces of aquatic plants with great neatness. 71. Deuterodon iguape Eigenmann sp. nov. (Plate XLI, Fig. 3).. Type No. 9265, a specimen 110 mm. long from Iguape collected by Dr. H. von Ihering. Tetragonopterus fasciatus Eigenmann & Nomis (not Cuvier). Re- vista Museu Paulista, IV, 357. '’ debrepoc repeated, ddobc¢ tooth ; in allusion to the simi «rity of the mandibular eeth. seer © ee ee ee ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XLII. . Metynnts mola (Eigenmann and Kennedy). i - Myleus levis Eigenmann and McAtee. EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. 141 This species is the image of Astyanax fasciatus Cuvier, but differs generically from the species in the character of the teeth. Bread 3.75; depth 2.6:;. D.. 11; A. ‘2,22 3 scales 6—34-5. Dorsal and ventral profiles nearly equally curved in front of the dorsal and ventrals ; head somewhat pointed, the snout rather long ; mouth nearly horizontal, the maxillary oblique, reaching at least to front of pupil; eye large, equal to snout in length, 3% in head, less than interorbital width. Origin of dorsal slightly behind origin of ventrals, a little nearer tip of snout than base of middle caudal rays, its height less than length of head ; anal moderate, its margin slightly concave ; ventrals not reaching anal ; pectorals not to ventrals. Free margins of the scales with a series of non-converging sub-par- allel lines. In alcohol, straw color; a silvery lateral band overlying a dark band ; a well-marked round humeral spot smaller than eye; a well- defined caudal spot continued to the end of the middle rays; no other marks on the fins. 72. Chalcinus angulatus Agassiz. Chalcinus angulatus curtus EK. & K., 1. ¢., 24. Bahia Negra (384), Corumba (360). 73. TLhoracocharax stellatus (Kner). Bahia Negra (385) and Corumba (357). 74. Pygocentrus natterert (Kner) Piranha. Rio Paraguay at Porto Murtinho, and Corumba (285 and 342). 75. Serrasalmo spilopleura Kner. Piray, Pirambé. Rio Otuquis; Rio Paraguay at Ascuncion, and Porto Murtinho (278, 284, 377, 378, 395). 76. Serrasalmo humeralis Cuvier and Valenciennes, Pirambéva. (Plate XXXIII, Fig. 2). Porto Murtinho (384), Bahia Negra (410). 77. Metynnis mola (Eigenmann & Kennedy) (Plate XLII, Fig. 1). A specimen somewhat larger (No. 376) than any of those recorded Perore, may be different. It is fromthe. Rio Otuquis. A.:37; D: 18; ventral serre 29-2, none of them two-pointed; back under dorsal without cross-bands, cross shades behind the dorsal. | Sides be- low the lateral line and along the lateral line spotted. In two small specimens (No. 361) from Puerto Murtinho, Matto Grosso, there are no cross-bands on back. 142 ANNALS OF THE CARNEGIE MUSEUM. 78. Mylossoma albiscopus (Cope) Piranha. Bahia Negra (394). 79. Myleus levis Eigenmann & McAtee, sp. nov. (Plate Xin Fig.-2). Type No. 10156, Bahia Negra, Dec., 1901 (296) 143 mm. | Head 3.5; depth 1.5 ; D. 27; A. 36; abdominal serre 45 ; lat line about 112. Form rather polygonal; nearly straight or slightly concave from snout to end of occipital crest, thence to within 2 cm. of the dorsal fin the anterior margin is strongly arched, thence is straight to the dorsal. From the front of the dorsal to the caudal the back is evenly curved. Caudal peduncle slender, about equal to eye, 2.8 in head. The basis of the anal is straight and meets the ventral profile at a large angle (about 60 degrees). ‘The ventral surface is bounded by a straight, horizontal outline from anal to a point slightly in advance of the ventrals, thence evenly arching to in front of the pectoral, where it is slightly concave. Lateral line complete, decurved from above middle of pectoral to below end of dorsal. Opercle rounded ; maxillary small, strap-shaped, partly sheathedin | first suborbital, not reaching eye. Mandible with 5 strong teeth on each side, the inner 3 large and with slightly wavy edges, the outer two quite small. There are 2 hooked, conical teeth in the middle behind the anterior row ; premaxillaries with two series of teeth, the outer composed of ro teeth similar to those of the mandible and the inner with 4 broad, concave-topped teeth in a straight row across the jaw. Jaws equal; snout .8 in eye. Dorsal fin rounded, highest in front. Caudal broad, subtruncate ; pectorals 1.2 in head ; ventrals narrow, 2 in head, not reached by the pectorals, nor reaching vent. Anal strongly falcate, the third ray very long and heavy, 1.2 times head. Color plain dark brown, lighter below lateral line. Fins clear, except a wide basal portion of the dorsal dusky and a heavy black blotch on the margin of the anal from the ci to the 13th ray, widest on the 4th and 6th rays. 80. Charax cahurus Eigenmann & Kennedy, sp. nov. (Plate XLIII, Hie ry In Proc. Acad. Sci. Phila., Eigenmann & Kennedy describe a specimen, No. 9969 of. the I. U. Collection. It was supposed to be the young of sguamosus. The present collection contains specimens of ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XLIII. Fig. 1. Charax caliurus Eigenmann and Kennedy. Fig. 2. Charax sgquamosus Eigenmann and Kennedy. EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY, 143 sguamosus ranging from 70 to 280 mm. These leave no doubt that the specimen (No. 9969) under consideration represents a distinct and most strikingly marked species of the genus. 81. Charax squamosus (Eigenmann & Kennedy) (Plate XLIII, pis. 2). Of this species we have a specimen from Bahia Negra (404) 230 mm. to base of caudal. The profile becomes more and more concave with age, as the result of the humping of the nape, so that in the largest specimen the depression is 6 mm. below the line joining the tip of the snout and the nape. Formalin specimens all show a dis- tinct humeral spot, a larger caudal spot and a well-defined black band uniting the two. In alcoholic specimens the ‘:umeral spot is not apparent and the black lateral band is hidden b__ ue silvery band over- lying it. 82. Cynopotamus knerit Steindachner. Specimens from Bahia Negra (373) and Corumba (341). 83. Reboides bonariensis Steindachner, Soirt pintada. Corumba (356). It is very probable that this specimen, as well as those mentioned as mcrolepsis by Eigenmann & Kennedy, are 2. bonariensis. 84. Reboides prognathus Boulenger. Rio Otuquis (371). 85. Acestrorhynchus ferox Giinther. Pez de Cachorro. Corumba (332). GYMNOTID&. 86. Ligenmannia virescens (Val. ) Cubiha. Corumba (293, 346). 87. Sternopygus macrurus Bloch & Schneider. Corumba (294). 88. Gymnotus carapus Linneus. Corumba (295). PCCILIIDA. 89. Girardinus caudomaculatus Hensel. Numerous specimens ; exceedingly abundant in a mountain brook, Arroyo Itoroto at Sapucay (458). BELONID-. 90. Potamoraphis guianensis (Schomburgk). Pez d’ Aquelha. Tuyuyu (298). 144 ANNALS OF THE CARNEGIE MUSEUM. SCLENID. gt. Pachyurus bonariensts Steindachner. Corumba (250). CICHLID:. 92: Chetobranchopsis australe Kigenmann & Ward, sp. nov. (Plate SL EV, Hie. ea):. Type No. 10157, one specimen Iro mm. to base of caudal. Bahia Negra. Profile from tip of snout to occipital region straight, thence evenly curved to base of caudal. Depth 124, head 1% ; eye 3% in head, I in snout, 124 in interorbital. Cleft of mouth oblique, not reach- ing to the vertical from the orbit ; cheek scaled ; upper part of oper- culum scaled ; 14 scales in front of dorsal; to scales in front of ven- trals. Dorsal spines of medium strength, increasing in height pos- teriorly, the highest being 14 times orbital diameter ; pectoral long, some of the rays in the upper half much produced, longer than head, reaching to origin of second half of anal; ventrals produced in a filament which reaches to 3d anal ray. Anal and caudal fins densely scaled ; soft dorsal with a very few, indistinct scales at base. Color light brown ; a dark blotch on cheek and a dark spot on sides below lateral line ; ventral and anal fins margined with black. D. XIV, 135 A.V, 16; Jat.1: toe-6; depiiyias.. C. orbicularis from the Amazon has fifteen dorsal and six anal spines. 93. -quidens tetramerus (Heckel). Cara. | Ascuncion ; tributary of Paraguay in Chaco Paraguayo ; Corumba ; Bahia Negra; Puerto Max and Villa Rica (288, 343, 381, 391, 415, 468). An examination of 9 specimens gives D. XV, 10 in all speci- mens; “A. Ill, 9 in seven; Ill;:10 in two ;. lat. 16-— To imeone,, 17 + 8 in three, and 17 + 9 in five specimens. 94. guidens paraguayensis Eigenmann & Kennedy (Plate XLIV, | Pig. 27). Ascuncion ; Corumba; Rio Otuquis; Puerto Max (414, 417, 418, 292, 284, 320,.260,) 370, 200): An examination of 29 specimens gave the formulas; D. XIII, 10 in one specimen; XIV, 9 in 21; XIV, 10 in 3 and XV pees specimens. A. III,‘7 in one, IJ], 8 in 18 and Il], 9 in ge; constant, 16+ 10. Of the specimens with A. III, 9, six came from a mountain brook at Sapucay (454). ANNALS CARNEGIE MUSEUM, Vol. IV. Plate XLIV. Fig. 1. Chetobranchopsis australe Eigenmann and Ward. Fig. 2. guidens paraguayensis Eigenmann and Kennedy. a d é ‘ 4 4 i- = ’ = . , : : & ‘ ‘ *. = ile « ar q vt) ¢ =. f is a a - at i 7 ee 6 ee a i 2 ks = Le & ’ ; go + ° + ‘ > ‘ 4 ~*~ io eS ar a" eter a ~~ a Pom am «iL @ ~ fh = A a” 7 p= ‘ Pt é weak = ae or "ets Sat oe ee “ “ - * 7 A ‘ | , ‘¢ . 5 bey" : ; ite ¥ - £ t eis ~ ‘h 3a Be ewe 4 rf ame ’ { ‘ mes a a Wes)» ‘ ; 7? vl ’ i Ne a toa ® oe 1 eS » oa - mas : Thess hh > Way a bf s > om, : =% ws Yan ot aie *< ee ee ot 2 a J ac i R » . ’ 4 > ‘ Me EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. 147 These specimens differ but slightly from Giinther’s and Steindach- ner’s descriptions of ¢enzatum, hitherto known only from the Amazon. Dorsal XIV-XVI, 5%; A.III, 5% to 6%. Depth 22-(3) ; head 3 (3); head as long as high (longer) ; eye 3 in the head, longer than snout, equals interorbital ; preorbital 2 in eye; highest dorsal spine equals length of eye or a little longer; soft dorsal and anal reaching middle of caudal when laid back ; caudal rounded; pectorals about reaching anal, the threads of the ventrals beyond origin of anal. Lateral line with 4-7 canals, 21-23 scales in the lateral series. A dark streak from eye forward, another downward and backward; a black band broken into spots along the middle of the sides to the end of the caudal; a series of dark spots along the base of the dorsal ; anterior margin of anal and outer pectoral ray sometimes black. ACHIRIDZ. 103. Achirus jenynsi (Giinther). Corumba (347). EIS? OF FISHES SO FAR RECORDED, FROM THE. BASIN OF THE, PARAGUAY. In the following list one star (*) indicates that the species is pecu- liar to the basin of the Paraguay ; two stars (***) that both the genus and species are peculiar to it; A indicates that the sfeczes is also found in the basin of the Amazon; B, that the gezuws is found in the basin of the Amazon; P, that the species is peculiar to the basin of the La Plata; C, that the species is also found in the coast streams of eastern Brazil, and S, that it is found in the Rio San Francisco. TRYGONID. 1. A. Potamotrygon hystrix Miiller & Troschel. 2. A. Potamotrygon dumerili Castlenau. ASPREDINID. B.* Dysichthys australe Eigenmann & Ward. B.* Bunocephalus rugosus Eigenmann & Kennedy. B. P. Bunocephalus theringt Boulenger. B.* Bunocephalus dorite Boulenger. 148 ANNALS OF THE CARNEGIE MUSEUM. SILURIDZ. PIMELODINA. A. Pinarampus pirtnampu (Spix). B. A. Luctopimelodus platanus Giinther. A. S. Pseudopimelodus zungaro (Humboldt). * Pseudopimelodus cottoides Boulenger. A.C. Rhamdia quelen (Quoy & Gaimard). 12. P. Heptapterus mustelinus Valenciennes. A. Rhamda sebe@ knert Steindachner. A. x Ae 5. 1a\e 12. 14. Pimelodus ornata Kner. Iie: Pimelodus albicans (Cuvier & Valenciennes). 16. S. Pimelodus clarias (Bloch). Pimelodus valenciennis (Kroyer). S. Pimelodus fur Reinhardt. 19. ** Sheringichthys labrosus (Liitken). 20. ** Sheringichthys megalops Eigenmann & Ward. 21. A. Pimelodella gracilis (Valenciennes). 22. * Pimelodella teniophorus Regan. 23. A.C. 8S. Pimelodella latertstriga (Miiller & Troschel). 24. * Pimelodella mucosa Eigenmann & Ward. 25. A. Sctades pictus (Miiller & Troschel). 26. A. Hemtsorubim platyrhynchus (Cuvier & Valenciennes). 27. B.S. Pseudoplatystomus coruscans (Agassiz). 28. A. Sorubim lima (Bloch & Schneider). DOoRADINE. 29. A. Doras granulosus Valenciennes. 30. A. S. Doras costatus (Linneus). 31. * Doras maculatus Valenciennes. 32. * Doras nebulosus Eigenmann & Kennedy. 33. A. Doras weddeli Castelnau. 34. B.* Oxydoras kneri (Bloch). 35. * Oxydoras eigenmanni (Boulenger). 36. * Hemidoras paraguayensis EKigenmann & Ward. AUCHENIPTERIN~. 37. B. A. TZracheliopterus coriaceus (Cuvier & Valenciennes). 30.4" mai la ks nigripinnis (Boulenger). 43. PPD Pb eee HR EO EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. Trachycorystes galeatus (Linneus). ae . A.C. Zrachycorystes striatulus (Steindachner). AGENIOSIN-. . A. Agenetosus valenciennes Bleeker. . A. Agenetosus brevifilis (Cuvier & Valenciennes ). HYPOPHTHALMID&. A. Hypophthalmus edentatus Spix. PYGIDIIDA. . * Pyotdium borelli Boulenger. 45: 46. . ** Homodietus anisitst Eigenmann & Ward. * Pygidium brastliense (Reinhardt). P. Pygidium cordovensts (Weyenberg). LORICARIIDA. PLECOSTOMIN &. Plecostomus plecostomus (Linneus). Plecostomus johni Steindachner. Plecostomus commersont (Valenciennes). Plecostomus vaillantt Steindachner. Plecostomus borelli Boulenger. .C. Plecostomus robint Cuvier & Valenciennes. S. Plecostomus wucherert Giinther. Hemiancistrus vittatus (Steindachner). Cochliodon cochhodon Kner. Pterygoplichthys multiradiatus (Bloch). Pterygoplichthys anisitst Eigenmann & Kennedy. Pterygoplichthys juvens Eigenmann & Kennedy. Pterygoplichthys gigas Boulenger. Pseudancistrus barbatus (Cuvier & Valenciennes). Xenocara gymnorhynchus (Kner). A. Ancistrus cirrhosus Valenciennes. Ancistrus cirrhosus dubius Kigenmann & Eigenmann. Ancistrus hoplogenys (Giinther). >*eNa0-F . * Oxyropsis tnexpectatum Holmberg. . *® Ofocinclus vittatus Regan. 149 150 ANNALS OF THE CARNEGIE MUSEUM. LORICARIINE. 68. A. Hemiodontichthys acipenserinus (Kner). 69. B.* Sturtsoma robusta Regan. 70. * Sturisoma barbata Kner. 71. * Loricaria parva Boulenger. 72. P. Loricaria catamarensis Berg. 73. A. Loricaria phoxocephala Eigenmann & Eigenmann. 74. * Loricaria maculata Bloch. 75. A. Loricaria typus Bleeker: 76. * Loricaria labialts Boulenger. 77. P. Loricaria anus Valenciennes. 78. A. Loricaria cataphracta Linneus. 79. A. Loricaria carinata Castelnau. 80. * Loricaria apeltogaster Boulen ger. 81. * Loricaria macrodon Kner. 82. * Loricaria laticeps Regan. 83. * Loricaria platycephala Kner. CALLICHTHYID&. 84. B.* Corydoras microps Figenmann & Kennedy. 85. * Corydoras paleatus (Jenyns). 86. * Corydoras aurofrenatus Figenmann & Kennedy. 87. * Corydoras australe Eigenmann & Ward. “Cae ee Os Callichthys callichthys (Linnzeus). 89. * Callichthys callichthys asper Quoy & Gaimard. 90. * Callichthys callichthys hemiphractus Hensel. gt. * Hoplosternum pectoralis (Boulenger). 92. A. Hoplosternum littorale (Hancock). CHARACID. ERYTHRININE. 93. A. C. 8S. Hoplias malabaricus (Bloch). 94. A. C. Hoplerythrinus uniteniatus (Spix). PyYRRHULININA. 95. * Pyrrhulina australe Kigenmann & Kennedy. 96. A. Pyrrhulina brevis Steindachner. ee: EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. CURIMATINE. 97. B.* Psectrogaster curviventris Eigenmann & Kennedy. 98. A. Curimatella alburnus (Miiller & Troschel). 99. * Curimatella alburnus australe Kigenmann & Kennedy. too. A. Curimatus spilurus Giinther. lor. * Curimatus gilli Eigenmann & Kennedy. to2. A. Curimatus nasus Steindachner. 103. * Curimatus nigrotenia Boulenger. 104. * Curimatus elegans nitens Holmberg. 105. A. Curimatus bimaculatus Steindachner. 106. A. Cwurimatus rutiloides Kner. 107. C. Curimatus gilberti Quoy & Gaimard. 108. B. A. Anodus latior Spix. CHILODINZ. tog. B.S. Prochilodus argenteus Agassiz. 110. C. Prochilodus scrofa Steindachner. 111. P. Prochilodus lineatus (Valenciennes). HEMIODONTIN&. 112. * Anisitsia othonops Eigenmann & Kennedy. 113. A. Hemtodus unimaculatus (Bloch). 114. * Hemtodus semiteniatus Kner. 115. A. Hemtodus microlepis Kner. 116. A. Parodon suborbitalis Cuvier & Valenciennes. 117. * Parodon gestri Boulenger. 118. * Parodon paraguensis Steindachner. 11g. * Parodon tortuosus Eigenmann & Norris. ANOSTOMATINE. 120. ** Schizodon borelli (Boulenger). 121. S. Schizodon isognathus Kner. 122. A. Schizodon dissimilis (Garman). 123. A. Schizodon fasciatus (Spix). 124. A. Lahtlliella nasutus (Kner). 125. A. Leporinus striatus Kner. 126. A.C. Leporinus frederici (Bloch). 127. A. Leforinus obtusidens (Valenciennes). 151 152 Bone 129; 130. {31. es 134, T340 125. 136. 137. 133. 120: 140. 141. TAZ. 144; 144. 145. 146. £47: 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. EOL; 162. 163. 164. 5 ee ae oe ANNALS OF THE CARNEGIE MUSEUM. Leporinus trifasciatus Steindachner. Leporinus eques Steindachner. Leporinus affinis Giinther. Leporinus hypselonotus Giinther. ay oy Leporinus fasciatus (Bloch). re Ok Leporinus controstris Steindachner. Characidium fasciatum Reinhardt. Characidium laterals (Boulenger). TETRAGONOPTERINA. B.« Odontostilbe paraguayensis Eigenmann & Kennedy. * Odontostilbe trementing EKigenmann & Kennedy. * Chetrodon ribetrot Eigenmann. B. * Cheirodon tnterruptus (Jenyns). * Cheirodon caliurus Boulenger. a 7a A. *K Cheirodon insignis Steindachner. Cheirodon natterert Steindachner. floloshesthes pequira (Steindachner). Aphyocharax dentatus Eigenmann & Kennedy. A. Aphyocharax alburnus Giinther. * Aphyocharax stramineus Eigenmann. A. Aphyocharax anisitst Kigenmann & Kennedy. * Aphiocharax rathbuni Figenmann. oe Flemigrammus gracilis Reinhardt. ? Hemigrammus caliistus (Boulenger ). * Hemigrammus anisits’ Eigenmann. C. Hemigrammus litkeni (Boulenger) C. Hemigrammus ulreyt (Boulenger). * Hemigrammus tridens Eigenmann. sk Hemigrammus kennedyt Eigenmann. A. Tetragonopterus argenteus Cuvier. Ae, Tetragonopterus orbicularis Cuvier & Valenciennes. * Tetragonopterus allent Eigenmann & McAtee. * Tetragonopterus ternetzi Boulenger. A. Astyanax fasciatus (Cuvier). P. Astyanax theringt (Boulenger). A. 7 * pp at Se te ANNALS OF THE CARNEGIE MUSEUM. MYLINZ. Metynnis mola Eigenmann & Kennedy. Metynnis hypsauchen (Miiller & Troschel). Myleus astertias (Miller & Troschel). Myleus levis Eigenmann & McAtee. Mylossoma aureus (Agassiz). Mylossoma albiscopus (Cope). Piaractus brachypomus (Cuvier). CHARACINA, Charax gibbosus (Linneus). Charax squamosus Eigenmann & Kennedy. Charax calliurus Figenmann. Restes molossus (Kner). Reboides microlepis (Reinhardt). Rebowdes bonariensts Steindachner. Reboides prognathus (Boulenger). Cynopotamus humeralts (Valenciennes). Cynopotamus knert Steindachner. Cynopotamus magdalene (Steindachner).” Salminus brevidens Cuvier. ACESTRORHAMPHIN. Acestrorhynchus ferox (Giinther). Acestrorhamphus hepsetus (Cuvier). CYNODONTIN. Raphiodon vulpinus Spix. STERNOPYGID. Sternarchus albifrons Linneeus. Rhamphichthys reinhardti Kaup. Rhamphicthys marmoratus Castlenau. L[1ypopomus brevirostris Steindachner. S. Sternopygus macrurus (Bloch and Schneider). S. Ligenmannia virescens (Valenciennes). SS. Gymnotus carapo Linneus. [he record of this species for the Paraguay is on the authority of Perugia. It probably should be eliminated. EIGENMANN: COLLECTION OF FISHES FROM PARAGUAY. . SYNBRANCHID. . B. A. Synbranchus marmoratus Bloch, STOLEPHORID. . A. Stolephorus olidus Giinther. PCCILIID. A. Rivulus punctatus Boulenger. A. Gtrardinus caudomaculatus Hensel. . C. Cnesterodon decemmaculatus (Jenyns). B. * Fundulus paraguayensis Eigenmann & Kennedy. B. * Fundulus balzanii (Perugia). * > Llyodon paraguayense Eigenmann. BELONIDZ. . A. Potamorrhaphis guianensis (Schomburgk). A. Tylosurus amazonicus (Steindachner). SCLENIDA. Pachyurus bonartensis Steindachner. Pachyurus schomburgkit Giinther. * -Plagtoscion ternetst Boulenger. > CICHLIDA. B. * Chetobranchopsts australe Eigenmann & Ward. A. Astronotus ocellatus (Agassiz). A. Atquidens tetramerus (Heckel). _C. 4quidens portalagrensis (Heckel). A. &quidens dorsigera (Heckel). * Equidens paraguayensis Eigenmann & Kennedy. A. Aquidens vittata (Heckel). . A. Cichlasoma bimaculata (Linneus). . B.* Heterogramma corumbe Eigenmann & Ward. . ®E Heterogramma borelli Regan. . *K FLeterogramma trifasciatum Eigenmann & Kennedy. Mesonauta festivus (Heckel). Crenicichla lepidota Heckel. Crenicichla adspersa Heckel. Crenicichla vittata Heckel. vege 155 156 ANNALS OF THE CARNEGIE MUSEUM. 247. A. Crenicichla saxattlis (Linneus). 248. B.* Batrachops ocellata Perugia. 249. P. Batrachops semifasciatus Heckel. 250. * Batrachops ocellatus Perugia. 251. A. Satanoperca pappaterra Heckel. 252. B.* Geophagus balzani Perugia. 253. A. Geophagus jurupart Heckel. PLEURONECTIDZ asa. P. . ix x List oF FIcuURES I~. Texz: Nothocyon annectens. Side view of upper and lower jaws. 2 od Superior and inferior dentition Merychyus minimus Peterson. Right side of skull Alurocyon brevifacies Peterson. View of right side of skull RS es of Palate view of skull “ “ es Fragment of pelvis ; head of femur ; metapodials Tooth of Rodent Bone belonging to Mosasaurus ee Dalle. Three views : convex surface, concave surface and view showing curvature of the bone . : : : : Two views of sections across bone of Mosasaurus lemonntert Bone belonging to CZzdastes tortor Cope Raphistoma stamineum Hall. An enlargement of the upper surface . Scalites angulatus. Top and side views . e a Outline drawing Bucania sulcatina. An enlargement of the nietion od the sintice Maclurites magnus Lesueur. A natural section of the operculum Diagrammatic section of cave at Frankstown, Pennsylvania Eryops. Dorsal vertebra, left side . iS Rib. : : é i : ; : : i Anterior view of dorsal spine of caudal vertebra. Tooth of Desmatodon holland: . Teeth of various Diadectids Chevron bone of a Diadectid ; Naosaurus raymond. Part of the doveal spine. Ilium of an undetermined reptile. External view Aphelops montanus, Femur : i ceratorhinus. Calcaneum of type sf : Outline of back of skull m % Basi-occipital view of skull. a 4 Lateral view of back of skull Aphelops sp.? Humerus Teleoceras? sp.? Lateral and superior views of ‘Eigen of skull Glyptosaurus montanus. Top view of skull + ie Side view of skull Rhineura hatchert. Side view of skull = as Top view of skull vs * Palatal view of skull. FPeltosaurus granulosus. Top view of part of skull = of Side view of part of skull : 3 Portion of a mandible List OF FIGURES IN TEXT. x1 -Stenomylus gracilis. Diagram of skull. ; , ? : : 287 a iy ae Fourth cervical vertebra. Left side : . 288 | ee . Fifth te iS Pe hs , . 288 es Sixth es ke Sar F . 289 be e ee ss ty Anterior view. « 286 i te. ) Re ate ea ek fe (Py Bie eee igh eee Sa Sar wa eae Be mee anys «ai Siete 7 RRR eT oN WY Woke i eoke : } ase tik Rss > Suhtnae GLA nin Bel nde aie +f . \ re Si Bite iene A) Sol he ie ae 4 eine 6M Sy Sat, a a nay We ae Ce st ec SNR Breed £ el ‘of cS La a] : J é ¢ j ~~ i es. Shay / L <4's Vita . * . ‘ Abe ‘ef ir. a + * . Ne ; 4 . A te ® ) a . i Rs Ea, i: ° ¢' m4 2 ' - . @ ) - ? -. 5 , rd 1 a! - * > boo » ¢ i 7. * al a 7 aM LS ay, ce) é ‘ ‘ ; ‘ < \ ’ a > ee tae . eu ; “ i AONE 7 ~! ~) i 1 ; = wo we T f by 1 sere 1 . ’ hoe - Me 4 4 t ¥ } a re} J : 7 y nm: - a Oe , iw - oF . PB ¢ of . i ¢ : 4 y « a a Me ‘ 6 ba Sep Se Cie, Tee ‘ + : a (e \ é q ~ ’ : a < “a . ‘ - ts bs ’ = eS ’ ' } a F ‘ a ] - . ® “ ad > ® - - . i . , La i y ‘ = . i. ‘ ij f - ‘ \ \ 4 ‘ i " ’ ‘ ’ a o ERRATA AND CORRIGENDA. fee 2, ane 1, for ** Dec., 1906,’" read ‘* March 21, 1907."’ Page 23, footnote, for ‘‘ Lepiauchenia,’’ read ‘‘ Leptauchenia.’ Page 24, line 12 from below, for ‘‘melagodon,’’ read ‘‘ megalodon.’ Page 41, line 3 from below, for ‘‘ oss,’’ read ‘‘ Foss.’’ Page 51, line 4 from below, for ‘‘second and third premolars,’’ read ‘«second and third upper premolars.’’ Pare gs, line 13, for ‘‘ Harrison or Upper Monroe Creek beds,’ ‘‘ Harrison beds.”’ Page 96, lines 35 and 36, for ‘‘ Upper Monroe Creek or Harrison beds,’’ read ‘‘ Harrison beds.”’ Page 126, No. 58, read ‘‘ Aphyocharax’’ for ‘‘ Aphiocharax.’’ Ditto p. we 652, No, 148. Page 169, 3d Jine from bottom, for ‘‘ Canaa’ faeerie2, une 12, for ““Te’’ read ‘* The.” Page 177, line 12, for ‘‘ umbiculated ’’ read ‘‘ umbilicated.’’ Page 184, 3d line from bottom, for ‘‘ Scalits’’ read ‘‘ Scalites.’ Page 250, line 17, for ‘‘ Ways’’ read ‘‘ Way.”’ Piate Alt, legend’; for ‘* % *’ read. ‘* %.”’ 3 ’ read ’ read ‘‘ Canada.’’ rf Xili REO rue Wee d nw aes g rr - ‘ ae + %. ai a ft : % = 6 = y ee, tesa *) AG eel Bh cn ~ a ; aay - tere sui iy ; posta Bed ee ‘7 ie ; ; ay ae Paris tt . ni iene a fe pol ae es a src sak) yout ath Pa to ee ee Hs By bai Swe ae ‘ioe Sieg, Lae ae Aca nae LA: ad ial. ry + ae i Wie vie a en4, £ ; " : lee ; , fe eX, A sia Fe < wt er. ‘4 ‘ sae 5 i . Tr % 7 y + a » "sg > ie¥o te Hit ee + Ty. r . ae nt Reo, i {oe eR ete eg. ore (aaa a ax oe : Ai 7 ae ts a ; Rede | Tees ee le 20h) 0 we AL othe a i ; m “a mie ty asd] Sara a: wee Arty } wh abe <3 ci Spt mms tice ia of ta 4 tists ips SA Par eae a ” ‘ ‘ = « s y ea i? A isn a ae: | weg Ce * Sen. GU Cues salle » . i ® < ~ . — ae - r ~ 5 a é a 7 Ade? « A ‘ j rg > ¥ i- > by * PUBLICATIONS OF THE CARNEGIE MUSEUM SERIAL No. 50 OF THE Wor, IV, Nose lil annoy W. J. HOLLAND, Eaitor PUBLISHED BY THE AUTHORITY OF THE BoarRD OF TRUSTEES OF THE CARNEGIE INSTITUTE APRIL, 1908 = i cae S te ee aR : i ‘ Jo er ~ Bee: ee a eee a ie iy? ’ ‘ 6 i : i Teh . : ys * = “gf iv - ada.) ‘ r 2 f rvs we I ; t 4 y ‘ My, ~~ 5 wu f I bs 5 + M J * oA rT i} . ’ ‘ j = - ‘ y ¥ , : ’ mvs ar a . * - ?) ‘ a 3 “ Sit ee eae O01 Al eel _— ee EE EEE Pe N Ae Seen NE GLE MUSEUM VOLUME IV. NOS. Jib anp-1V. EDITORIAL NOTES. THE Second Annual Meeting of the American Association of Mu- seums, which was held in the Carnegie Museum June 4-6, was well attended, and many who were present have since written expressing their appreciation of the hospitality shown them by the Trustees of the Institute and the citizens of Pittsburgh. The papers which were read and the discussions which took place were interesting and instruc- tive. The American Association of Museums may be regarded as having been fairly launched upon a career of usefulness, and takes its place as one of the important associations of scientific men in the western hemisphere. The proceedings of the meeting held in June will shortly be published. Dr. A. E. Ortmann, Dr. Percy E. Raymond, Mr. O. A. Peterson, and the Director of the Carnegie Museum attended the sessions of the Seventh International Zodlogical Congress which were held in Boston at the end of August. Papers were read by all of the gentlemen who represented the Carnegie Institute. A LARGE number of men distingu'shed in the walks of science have recently visited the Carnegie Museum. Among those from abroad whom we have had the pleasure of welcoming were Mr. S. F. Harmar, of King’s College, Cambridge, and Mr. Arthur E. Shipley, of Christ’s College, Cambridge, who are associated in the editorship of the Cam- 159 160 ANNALS OF THE CARNEGIE MUSEUM. bridge Natural History, and hold very high rank among British zodlogists. Other visitors have been Sir John Murray, famous by reason of his connection with the ‘‘ Challenger Expedition,’’ Dr. Charles W. Andrews, of the British Museum, best known by his ex- ploration of Christmas Island and his remarkable paleontological dis- coveries in the Fayam ; Dr. G. Severin, Director of the Royal Museum in Brussels ; Dr. David von Hansemann and Dr. Richard Heymons, of the University of Berlin; Dr. Lithe, of the University of Koenigsberg ; Dr. H. Schauinsland, Director of the Museum in Bremen ; Dr. Ver- sluys, of Amsterdam, well known through his work in connection with the ‘‘Siboga Expedition,’’ and Dr. J. E. Hoyle, Director of the Man- chester Museum. A constant stream of American men of science passing through ‘‘ The Gateway of the West,’’ as Bancroft styles Pitts- burgh, have made it a point to interrupt their journeys going and coming in order to spend a day at the Museum of the Institute. We have also enjoyed the honor of visits from many, both from home and abroad, distinguished in other than the walks of science. One of the most famous of these was General Kuroki, the illustrious Japanese commander. Tue expedition of Mr. W. H. Utterback to the fossil fields of the west was suddenly terminated by the summons which came to him to return to the bedside of his father, who was dying at his home in Franklin, Indiana, and who has since passed away at a ripe age, greatly honored and respected by all who knew him. The results of Mr. Utterback’s labors were somewhat lessened by the bad weather which he encountered, but he sent in twelve large cases containing a great deal of valuable and important material illustrating the osteology of the Ceratopsia. In October Mr. Utterback returned to the Museum. Work upon the replicas of the skeleton of Diplodocus carnegtet intended as a gift, one for His Imperial Majesty, the German Emperor, the other as a gift to the President of the French Republic, has been carried on vigorously during the past summer and fall, and it is ex- pected that these two reproductions will have been put in place by the end of June of the coming year. A rearrangement of the exhibits in the Hall of Paleontology at the National Museum in Paris is taking place in order to accommodate the skeleton of the western monster, and Dr. Brauer, the Director of the Royal Museum of Natural His- i EDITORIAL. 161 tory in Berlin, writes that with sorrowful heart he is taking down the skeletons of some fine whales which have hitherto adorned the light-court (Lichthof) of the great Museum in Berlin in order to make room for the Diplodocus. He adds in his letter the remark that ‘‘when such monsters of the past are resurrected from their tombs the living fauna is compelled to take a back seat.’’ It is certain that the Diplodocus during its long life in Jurassic times did not give as much trouble to thoughtful minds as it is now doing. Wesincerely hope before another twelve months has rolled its course to be able to place alongside of the Diplodocus now standing in the Hall of Paleontology in the Carnegie Museum the skeleton of a Bronto- saurus which we possess, and also if possible the skeleton of a Mosa- saurus, of which we have a fine example. SUPERB work in the way of mounting several groups of recent mam- mals is being accomplished by the Messrs. Santens, and at the next celebration of Founder’s Day, 1908, the Gallery of Mammals will pre- sent aneven more attractive appearance than it does now. THE generosity of Mr. H. J. Heinz in sending to the Museum from London, where he acquired them, some thirty-three beautiful examples of ancient watches made two or three hundred years ago, is most cer- tainly appreciated. Mr. Heinz is laying the foundation fora fine horo- logical collection which will be of great interest not only to the his- torian, but also to the student of an interesting branch of mechanics. Mr. Joun D. HaseMan, who went to Brazil at the beginning of October in order to carry on explorations in some hitherto little known portions of that vast country, was very kindly received by Professor J. C. Branner, upon his arrival in the province of Bahia, and reports himself as having addressed himself to his task with great hope of much success. VIII. AN UNDETERMINED ELEMENT IN THE OSTE- OLOGY OF THE MOSASAURIDZ:. By’ W. ‘J. “Hoeriann. In the collection of Baron Ernest Bayet, purchased in 1903 by Mr. Andrew Carnegie for the Museum of the Institute in Pittsburgh, was a skeleton of Mosasaurus lemonnieri Dollo, discovered at Cuesmes, Bel- gium. It is mounted free, and though lacking the phalanges of the paddles, is otherwise one of the best specimens representing the genus and species in existence. Associated with the skeleton was a bone which had evidently been identified as a portion of the sternum, for it was placed on the floor of the case in which the skeleton was displayed in such a position as to indicate that the preparator supposed that it might be properly assigned to the sternum. Dr. A. Smith Woodward, who reported upon the Bayet Collection to the Trustees of the British Museum with a view to its purchase, and who kindly allowed me to make a copy of his report preserved in the files of the National Mu- ~ seum in London, in speaking of the skeleton of Mosasaurus says that itincludes a ‘‘sternum?’’ ‘The reference of the bone to the sternum was evidently a matter of doubt in the mind of Dr. Woodward. Last summer, when freeing from the matrix and mounting the bones of a fine specimen of C@dastes tortor Cope, now displayed as a slab mount in the Carnegie Museum, a portion of a similar bone was found asso- ciated with the remains and lying very near the lower margin of the inferior maxillary bones. In the summer of 1906 I called the atten- tion of Dr. S. W. Williston, who was visiting me, and who has de- voted more time and attention to the osteology of the Mosasauridze than any other American student, to the bone belonging to the spec- imen found at Cuesmes. I suggested to him that the bone might be possibly regarded as a glossohyal bone, or that it might be the xiphis- ternum, stating that I was inclined to prefer the first hypothesis. At that time the specimen found associated with the remains of Clidastes tortor had not turned up. Subsequently, having found the latter specimen, I wrote to Dr. Williston, enclosing sketches of both bones, and asking him to again give me the benefit of his great knowledge of 162 vet ae HOLLAND: OSTEOLOGY OF THE MOSASAURID. 1638 the osteology of the group and to indicate to me where the bones, in his judgment, ought to be placed. In a letter written by him on Jan- uary 13, 1907, he says: ‘‘Tt is rather a humiliating confession to make, after thirty years’ acquaintance with the Mosasaurs, to say that your bone ‘stumps’ me, but it is a fact. The lingual bone has been figured by Marsh for Tylosaurus, a copy of whose figure you will find in Volume IV of the University Geological Survey of Kansas, Plate XXXI, Figures 1-3, there wrongly ascribed to Platecarpus. Somewhere else I make men- tion of the hyoid bones of P/latecarpus being very much like the fig- ures, having found specimens ata later date. Mosasaurus and Ci- dastes are scarcely distinct generically, and they differ in so many ways from Platecarpus and Zylosaurus that one would expect to find some differences in the hyoids, but scarcely so great differences as your bone would indicate. ‘ Se Tey Fic. 5. An enlarge- panes ; carina is elevated. ment of a portion of the surface of a specimen of One locality has furnished a number of very Bucania sulcatina (Em- small specimens of this species, varying 1.5 to mons). 4mm. in diameter. These young individuals always show more rotund whorls and a raised carina, but otherwise the same surface markings as the adult. (See figure 14, Pl. LV.) The adults are from 50 to 7o mm. (in’sreatess diameter. Locality. — Common in the Chazy at Crown Point, Valcour Island, Plattsburgh, and Chazy, New York; and Isle La Motte, Vermont. Figures 3 and 4, Plate L, and figure 15, Plate XLIX, are from speci- mens in the Carnegie Museum. Figures 16 and 17, Plate XLIX, and figure 13, Plate LV, are from specimens in the United States National Museum. Figure 14, Plate LV, is from a young specimen in the Carnegie Museum. NS — ee ee ee ee ee RAYMOND: GASTROPODA OF THE CHAZY FORMATION. 197 Genus Oxydiscus Koken. Oxydiscus catilloides Raymond. (PLATE LV, FIGURES 9, IO.) Bucania catilloides Raymond, 1906, Annals Carnegie Museum, Vol. mil, p:. 5,76. DESCRIPTION. Shell very small, compressed, with two or three whorls which expand gradually. The carina is narrow, but distinct and elevated. The form is much like that of Oxydiscus acutus (Sowerby), but the umbilicus is wide enough to expose all whorls. On none of the specimens are the surface markings preserved. Locality.— A very rare fossil in the Chazy, found only on the west side of Valcour Island where it is associated with Aucanza sulcatina, Lygospira acuttrostris, and species of Phylloporina. The type is in the Carnegie Museum. Genus Tetranota Ulrich and Scofield. Tetranota bidorsata? (Hall). (PLATE L, FIGURE 5.) Bucania bidorsata ? Raymond, 1906, Annals Carnegie Museum, Vol. III, part 4, p. 514 e¢ seg. (In faunal lists. ) From a layer of decomposed limestone on the south end of Valcour Island, a considerable number of imperfect specimens of a species of Tetranota have been obtained. All are casts of the inner whorls, and are not sufficiently well preserved to permit of accurate identification, but they appear to be closely related to Zetranota bidorsata. DESCRIPTION. Shell small, closely coiled in one plane, umbilicated on both sides. Surface of the cast, which shows only the inner whorls, marked by four keels, two marginal and nearly obsolete, two central and con- nected by a flat or slightly concave band. ‘The shells show no other surface markings and the form of the aperture is not shown. These shells may be compared with either 7e¢ranota bidorsata (Hall) or Zetranota obsoleta Ulrich and Scofield. It differs from the former species in having the revolving keels less strongly developed on the 198 ANNALS OF THE CARNEGIE MUSEUM. inner whorls, while it differs from Zetranofa obsoleta in having the slit band of the cast flat instead of distinctly grooved. Locality.—In the lowest beds of the Maclurites division of .the Chazy on Valcour Island, New York. The figured specimen is in the Carnegie Museum. Family CyrTouitip# Ulrich and Scofield. Genus Trigyra genus nov. Trigyra ulrichi sp. nov. (PLATE L, FIGURE 6.) Trigyra genus nov. Cyrtolitidse with two strongly developed medio-lateral carinz on the body whorl. Type, Z7gyra ulrichi sp. nov. This genus is founded to receive a single species, and is based on a small testiferous specimen in the National Museum. DESCRIPTION. Shell small, closely coiled in one plane, umbilici rather narrow. Whorls angular at the boundaries of the umbilici, the back of the whorl broadly and gently convex. The body whorl is marked by five sharp revolving keels, two at the boundaries of the umbilici, two in the middle of the back of the whorl, placed about half way between the sharp V-shaped central keel and the keels at the umbilici. Ac- cording to Dr. Ulrich, who has recently removed the shell from the matrix, the two medio-lateral keels are not developed on the inner whorls. The surface markings consist of fine lines of growth which sweep gently back from the edges of the umbilici to the central keel. Beside these markings the surface is covered with fine wrinkles trans- verse to the lines of growth, and minute pits, producing the surface characteristic of the Cyrtolitidze. This shell has somewhat the general appearance of a Ze¢ranota, but differs in its reticulate surface, V-shaped central keel which does not carry a slit band, and in having an odd, rather than an even, number of keels. Locality. — From the Chazy Limestone on Isle La Motte, Vermont. The holotype is in the United States National Museum. — ANNALS CARNEGIE MUSEUM, Vol. IV. Sydney Prentice del. Gastropod from the Chazy Formation. Plate: lt es) ba ANNALS CARNEGIE MUSEUM, Vol. IV. Sea eas Sydney Prentice del. Gastropods from the Chazy Formation. Plate LIl. RAYMOND:-. GASTROPODA OF THE CHAZY FORMATION. 199 Family EvoMPHALID& de Koninck. Genus Maclurites Lesueur. Maclurites magnus Lesueur. ChIATE U, FIGURES 1; 25 Prate L].; Prate LIT.) Maclurite magna Lesueur, 1818, Journal of the Academy of Natural eeiences, Philadelphia, Vol. I, p. 312, Pl. 13, figs. 1, 2, 3- Maclurea magna Emmons, 1842, Final Report of the Second District of New York State Survey, p. 276, fig. 1. Maclurea magna Hall, 1847, Paleontology of New York, Vol. I, p. 26, Pl. 5, figs. ra—-1d; Pl. 507s, figs. 1a—10. Straparollus magnus Emmons, 1855, American Geology, p. 156, Pl. ap. 15. Maclurea magna Raymond, 1902, Bulletin of American Paleontology, fee ltp.°305, Pl. 18, fig. ro. Maclurites magnus Clarke and Ruedemann, 1903, Bulletin of the New York State Museum, No. 65, p. 542. This species was one of the first of American Paleozoic fossils to be described. It has always been considered the characteristic fossil of the Chazy, and as such has become widely known. It is not, how- ever, by any means as widely distributed as it has been reported to be, and it has constantly been confused with Maclurites logani Salter, Maclurites bigsbyi Hall, and other species of the younger formations, not so much because of its similarity as because of the familiarity of the name Maclurea magna. From this fossil the Chazy has been called the Maclurea Limestone. Maclurites magnus is a very common fossil in the Chazy from Orwell, Vermont, north to Montreal, but it should be pointed out that the Beekmantown formation contains more species and probably as many individuals of this genus as the Chazy, and that the Stones River and Trenton also afford numerous shells of the same genus, with more or less modification of form. Maclurites magnus is distinguished from JZaclurites logant by the more gradual increase in the size of the whorls and by the operculum, which is horn-shaped in Maclurites magnus and oval in Maclurites logani. ‘Yhe various Beekmantown species described by Billings, have in general more numerous and more rounded whorls, although there are exceptions. The following is Lesueur’s description : 200 ANNALS OF THE CARNEGIE MUSEUM. ‘¢ Genus Maclurite. «Generic Character. Shell discoidal, much depressed, unilocular ; spire not elevated, flat; umbilicus very large, with a groove formed by the projection of the preceding whorls, not crenulated. ‘« Species AZ. magna. Shell obtusely carinated on the exterior upper edge ; whorls rapidly increasing in size ; aperture on the left, irregularly oval, horizontally depressed above, lips not reflected.”’ Lesueur’s original specimens were from Basin Harbor, Vermont, and subsequently other specimens were obtained by him from Kentucky. The specimens actually described as Aaclurite magna were undoubtedly from the Chazy formation, as is shown by the figures accompanying the description (Plate 13 of the article cited above). Hall’s description : ‘« Sinistrorsal, discoidal, depressed turbinate ; breadth more than twice as great as the height; spire flat, a slightly depressed line at the sutures; whorls about six, gradually increasing from the apex, eS y \ 8 : 5 -% % % : RA: = z ih => 1 Bo ices WS 4 PRB hace s 3 SS Fic. 6. A natural section of the operculum of JZaclurites magnus Lesueur. These sections are very frequently seen on the surfaces of the layers in the middle portion of the formation. ventricose, flattened above, obtusely angular on the outer edge; sur- face marked by fine striz, which, upon close examination, are found to be produced by the imbricating edges of lamellz ; striz undulat- RAYMOND: GASTROP@DA OF THE CHAzZY FORMATION. ‘201 ing, bending backwards from the suture and forward in passing over the . edge of the shell ; aperture obtusely trigonal, depressed above, slightly expanded beyond the dimensions of the whorl just behind it; axis hollow, umbilicus broad and deep, extending to the top of the spire.’’ Operculum.— The operculum of Maclurites magnus has been referred to by Billings, Ulrich, and others, but does not appear to have been figured. As shown in the figures on Plate LII, the operculum is large, heavy, horn-shaped, with the nucleus twisted to the right. In the inner right hand corner of the operculum is a long process which pro- jects downward into the shell and forms a place for the attachment of muscles. The specimen figured was silicified and was removed from the limestone matrix by etching. ‘The inner side was exposed first, and the drawing for figure 2, Plate LII, was made before etching was complete, but the muscular process was imperfectly preserved. In an attempt to free the shell entirely from the matrix the inner edge of the shell was destroyed so that when figure 1 was made the specimen was imperfect, but the outline is restored from the figure previously made. Viewing the operculum from the outer side, the lines of growth are very prominent and there is a slight depression extending from the nucleus along the left margin to the lower left hand corner. Figure 1, Plate L, shows a view of the exterior of a calcified specimen from Crown Point. On the right the shell has been removed and the depression shown is the opening left by the removal of the basal por- tion of the muscle-process. Locality. — This species is common in the Chazy Limestone from Orwell, Vermont, to Montreal, Canada. It occurs also in Eastern Tennessee. ‘The figured specimens are in the Carnegie Museum. Maclurites atlanticus Billings. Maclurea atlantca Billings, 1859, Canadian Naturalist and Geologist, Mole IV; p..459: This species was described from specimens obtained at the Mingan Islands and was separated from Maclurites magnus by the difference in form of the operculum. No specimens of this species have been found in the region of Lake Champlain. ‘The following is Billings’ original description : ‘* Maclurea atlantica. ‘* Description. — Whorls about four, flat or gently convex on the lower side, ventricose above, and obtusely angulated at the edge of 202 ANNALS OF THE CARNEGIE MUSEUM. the umbilicus. In a specimen with four whorls the diameter is three inches and seven lines ; the width of the last whorl at the aperture, on _ the flat, lower side, is sixteen lines; at the termination of the third whorl six lines, and of the second whorl three lines. ‘The first whorl occupies about two lines of the diameter in the center. In the same specimens, if on a line drawn from the aperture straight across the shell, the width from the outside of the aperture to the center is two inches and two lines, and from the center to the termination of the line on the posterior side one inch and three lines . ‘‘'The operculum found associated with the specimens is elongated, flat or a little concave on one side, moderately convex on the other, curved like a short Cyrfoceras, but not in the same plane, — the apical half being gradually turned towards the flat side, so as to constitute a sub-spiral curve.’’ Genus Eccyliomphalus Portlock. Eccyliomphalus fredericus Raymond. (PLATE LILI, FIGURES I-3.) Eccyliomphalus fredericus Raymond, iI902, Bulletin of American Paleontology, p. 305, Pl. 18, fig. 4. DESCRIPTION. ‘«Very loosely coiled, making but one volution ; tapering rather abruptly at the apex ; test thin, marked by distant lamellose lines of growth ; cast rounded, smooth ; section nearly circular. ‘¢ Diameter of largest specimen 3.5 centimeters; greatest diameter of outer coil, 1 centimeter.”’ Locality. —'This species was described from Crown Point, where it is not uncommon. It occurs also at Valcour and Valcour Island, New York. The holotype is in the Cornell University Museum. The speci- mens figured on Plate LIII are from Crown Point and were a part of the series of individuals used in drawing up the original description. They are therefore paratypes. These specimens are now in the Carnegie Museum. Eccyliomphalus kalmi Raymond. (PLATE LIII, FIGURE 4.) Eccyliopterus kalmi Raymond, 1906, Annals of the Carnegie Museum, Vol, 1, p75 76. ANNALS CARNEGIE MUSEUM, Vol. IV. Plate LIll. Sydney Prentice del. Gastropods from the Chazy Formation. RayMonpd: GASTROPODA OF THE CHAzY FORMATION. 203 DESCRIPTION. Shell small, loosely coiled, apex acute, incurved, but not making a closed volution. ‘The shell is roughly quadrangular in cross-section, flattened on top, sloping outward and downward on the side and obtusely angulated at the lower angle; rounded below and on the inside. ‘The specimen is a cast and does not show any surface mark- ings. This species may be separated from both Lccylomphalus fred- ericus and £. prochvis by its quadrangular cross-section. ccyliom- phalus fredericus is circular in cross-section and Lccylomphalus pro- clivis triangular. Locality. — A very rare fossil in the Chazy Limestone at Sloop Bay, Valcour Island, New York. Named for the explorer and naturalist, Pehr Kalm, who visited this region in 1749. The type is in the collection at the Yale University Museum. Eccyliomphalus proclivis Raymond. (PLATE LITI, FIGURE 5.) Lccyliopterus proclivis Raymond, 1906, Annals of the Carnegie Museum, Vol. III, p. 576. DESCRIPTION. Shell loosely coiled, apex acute, scarcely incurved. Cross-section of cast triangular. Upper surface flat, the lower side acutely angulated. Both inside and outside are gently convex. The cross-section of this shell easily separates it from either of the preceding species. Locality. — A very rare fossil in the Chazy at Crown Point, New York. The type is in the collection of the Carnegie Museum. Eccyliomphalus sp. ind. (Puiate LIII, Ficure 6.) On the surface of a fragment of sandstone from the basal beds of the Chazy at Valcour Island are three small specimens of a species of Eccyliomphalus. ‘The shell is small, nearly circular in cross-section, regularly and evenly coiled, the whorls almost touching. ‘This form is more like Eccyliomphalus fredericus than any of the other species, but differs from that shell in its more regular and more closely coiled whorls. 204 ANNALS OF THE CARNEGIE MUSEUM. Locality. —In the Chazy Sandstone on the south end of Valcour Island. ‘The figured specimen is in the Carnegie Museum. Genus Eccyliopterus Remele. Eccyliopterus vagrans Raymond. (PLATE XLIX, FIGURES 10, II.) Flelicotoma vagrans Raymond, 1905, American Journal of Science, Series 4, Vol. XX, p. 270. This small but characteristic shell occurs only rarely in the Chazy and may readily be identified by the fact that the spire is depressed below the general surface and there is a sharp ridge on the outer angle of the body volution. DESCRIPTION. Shell small, the spire flat and depressed below the plane of the highest points on the upper surface. Outer edge of the body whorl angular, raised as a high sharp ridge on the body whorl. Lower surface of shell rounded, the umbilicus very wide, disclosing all the whorls. Section of the body whorl quadrilateral, angular above, rounded below. Surface marked by fine lines of growth which turn back on crossing the angle of the upper surface. ‘This species is related to Lecylopterus belottensts Ulrich and Scofield. Locahty.— A rare fossil in the Chazy at Valcour Island, New York. The types are in the Carnegie Museum. Genus Helicotoma Salter. Helicotoma whiteavesiana sp. nov. (PLATE XLVIII, FicuRES 11, 12.) In the Aylmer Limestone at Aylmer and at the Hog’s Back, near Ottawa, Canada, there occur thousands of casts of a small species of Flelicotoma. No specimens have been observed which retain any satisfactory surface markings. The shell consists of three flat-topped whorls which expand gradually from the apex. ‘The spire is low and step-like. The lower portion of the body whorl is rounded, the umbilicus rather wide. The periphery of the last whorl has an elevated rounded ridge, beneath which is a slight concavity. One specimen (Plate XLVIII, figure 12) shows the outline of the outer lip. This species resembles He/icotoma verticals Ulrich but differs RAYMOND: GASTROPODA OF THE CHAZY FORMATION. 205 in being higher in proportion to the width and in having the upper _ surface of the whorls slightly convex instead of concave. Locality.— Common in the Aylmer Limestone at Aylmer and at the Hog’s Back, near Ottawa, Canada. ‘The type is in the Yale Uni- versity Museum. Family TROCHONEMATID# Zittel. Genus Trochonema Salter. Trochonema biangulatum (Hall). (PLATE LIII, FIGURES 9g, I0.) Pleurotomaria biangulata Hall, 1847, Paleontology New York, Vol. ep. at, Fl. 6, figs. 62, 60. The type of this species cannot now be located, but there are speci- mens in the Hall collection in the American Museum of Natural His- tory labeled Pleurotomaria biangulata which agree fairly well with the original description and figure and may, perhaps, be accepted as authentic examples of Hall’s species. ‘The shells so labeled are small, flattened, with few volutions, and, although they show no surface markings, they apparently belong to the genus 77ochonema. At Valcour Island this small gastropod occurs sparingly in two or three localities and may be readily separated from all other species of Trochonema in the Chazy by its small size, few whorls and small umbilicus. Its nearest ally in the Chazy is Z7rochonema dispar, but it is readily distinguished from that shell by its small umbilicus and its square-shouldered whorls. DESCRIPTION. Shell small, with about three whorls which expand gradually. Spire low, raised very little above the body whorl. Body whorl with two carine, which are about equally distant from the axis of the shell, thus forming the flattened side of the whorl usually seen in this genus. Upper surface of the whorl slightly convex, often almost flat. Lower surface gently convex. Umbilicus small. All specimens are casts, and show no traces of surface markings. One specimen is to mm. in greater diameter and 7 mm. high. The body whorl is 3.5 mm. wide at the aperture. Other specimens are of about the same size, one or two being slightly larger. Locality. — This species occurs in the Chazy at Valcour Island, New York. The figured specimen is in the Yale University Museum. It is from the south end of Valcour Island. 206 ANNALS OF THE CARNEGIE MusEuUM, Trochonema dispar Raymond. (PLATE LIIT, FIGURES 7, 8.) Trochonema dispar Raymond, 1905, American Journal of Science, - Series 4, Vol. XX, p. 378. This species is rather rare and is closely related to Zvochonema umbilicatum “all. It differs from that species principally in the greater convexity of the under surface of the whorls and the less pro- nounced angularity of the body whorl. DESCRIPTION. Shell rather large, consisting of three whorls with depressed spire and very large body whorl. ‘The suture is very deep and the whorls are almost free. ‘The body whorl has a flat revolving band on the outer side. The top is flat and sloping and the lower side strongly convex. ‘The surface markings are not shown. ‘The umbilicus is large in the cast but rather small in testiferous specimens. Locality. — Fairly common in the Chazy at the south end of Valcour Island. It is rare elsewhere on Valcour Island and at Chazy, New York. The type is in the collection of the Carnegie Museum. Trochonema rectangulare Raymond. (PLATE LIII, FIGURES II, 12.) Lophospira rectangularis Raymond, t905, American Journal of science, Series 4, Vol! 3205p. 377. Although no surface markings are present in any of the specimens of this species, a fancied resemblance to Lophospira perangulata (Hall) led the writer, in the article referred to above, to describe this species under the generic name Lofhospira. This species somewhat resembles Trochonema robbinst Ulrich and Scofield, and Zrochonema niota Hall, but can readily be distinguished from both those species by its fewer and more rapidly expanding whorls and the smaller umbilicus. The figures show that it is not similar to any of the other species of Trochonema in the Chazy. DESCRIPTION. Shell fairly large, with five volutions. Body whori very large, spire small. ‘The last three whorls have the sides parallel to the axis of the shell. Body whorl with two keels, separated by a flat band, the width “~¢£ — a.) j Mae ~ rs = q 6 a a \ hy = * " . t . i 4 P ‘ ‘ 4 mei 2 he aa I co 4 int S . j i @ j ~ a ® , ' 7 474 ] ’ £ “ ‘. - 7 | hs t i } ’ A ¥ ' ¢ ' . s ° ° . ; P - - al > x, — a F < ’ +] £ : : P ie . 7 A H S ‘ . = ' a ~ f : ; ’ Mi s . © * = Mi = r id . , ia 7 ad ‘ are ; a € Z , > i. * i “ Y p 7 © 2 - f ’ ca f 4 ’ 5 ES " ’ ‘2 ; a ; ; is ; : 4 e, ~ ¥ ah - “4 | o th 4 oa . Sarre 7 > : = ; : i ¥ ® a ? a «t > 3 i - M ei ae ‘ 2 , ¥ * ae ao (= 7: 7. ® = akg ae m t é ; Tk 84 ¢ a) ~~ ¥ : 7 a < 4 - ret ‘ - a 7 ie Ses A+) ron ay, + ms ua 5 . pO an tT ey . “ idee 4 , — ~ y, ; ‘ iwi din =) © ee Aa ‘ A « y) i i] ay; es Fos i - ‘ : a a. Pesta’ Sd, ¥ > » se bet ey y — ; ne - F ~ 6s : . ‘ = ) j. ha ae 2 : ; i / * F ; = * ee rs ae ce = re ‘a rics : £ ¢ : hs ~~ , : Ay « a i = ew 7 ' F r ‘ 5 4 ~ aa Pe 2 a 3, é os ‘ ete 3 vay sj > . 7 ne ‘ v mend é a nag m . . 4 : 7 - J » 4 « 7 F F - es + : a . « c. i i + * b * ; : . y 4 3 : oe \ hers eee A " + - b! a a ~~ = —s r . ars ary N 7 - ¢ ~y . par) , ‘ :

: < % | wer) | tae pe vas | Plate LIV. ANNALS CARNEGIE MUSEUM, Vol. IV. tel ° 10€ ¢ ey Prent Sydn ~ from the Chazy Formation. Ss tropod RAYMOND: GASTROPODA OF THE CHAZY FORMATION. QBO7. of which increases with the expansion of the shell. Aperture large, nearly circular. The section of the body whorl has its upper side nearly straight, meeting the straight outer side at an obtuse angle. The inner and lower sides of the section are rounded. ‘The umbilicus is very small. All the specimens in the collection are casts of the interior and do not show anything more than traces of the surface markings. Locality. — A rare species from the Chazy at Valcour Island, New York. The type is in the collection of the Carnegie Museum. Trochonema hudsoni Raymond. (PLATE LIV, FIGURES I-3. ) Flolopea hudsont Raymond, 1905, American Journal of Science, Series aval, XX p. 373. Specimens of this species are usually casts of the interior and in that condition they show the smooth rounded whorls of a Ho/opea, under which genus they were originally described. Certain revolving lines on one of the casts caused Dr. Ulrich to suspect that the shell might be a Trochonema of the same type as the Zvochonema obsoletum described by him from the Trenton of Kentucky. Further search in my collec- tion brought to light a fragment of the body whorl of a testiferous specimen which shows distinctly the flattened peripheral band. The description must therefore be considerably revised. DESCRIPTION. Shell Ho/opfea-like, with rounded volutions. There are usually four whorls, the body whorl being about three times the height of the spire. A testiferous example shows a wide, flattened peripheral band, above which the whorl is somewhat flattened, and directly above the carina almost concave. Below the band the shell is gently convex. In the cast some specimens show a very slight flattening of the upper side, while in other specimens the cast of the body whorl is regularly rounded, with two or three faint revolving ridges below the center of the whorl. In the casts the whorls of the spire are regularly rounded, and the suture is very deep. ‘The spire of the specimen retaining the shell is broken away. ‘The umbilicus is small, almost closed by the thicken- ing of the inner ip. The aperture is almost circular in outline and the inner lip is free from the body whorl except in the posterior inner 208 ANNALS OF THE CARNEGIE MUSEUM. side. One specimen is 28 mm. high and 22 mm. in greatest diam- eter; the spire is 7 mm. high. In the same specimen the transverse diameter of the aperture is 13 mm. and the antero-posterior diameter is 14mm. A second specimen is 24 mm. in height and 20 mm. in diameter. Locality. — This species is rather common at Crown Point, Valcour Island, and Chazy, New York. ‘The original of figures 1 and 2, Plate LIV, are in the Yale University Museum, and the original of figure 3 in the Carnegie Museum. Genus Eunema Salter. Eunema altisulcatum Hudson. Lunema altisulcatum Hudson, 1905, Report New York State Paleon- tologist. for 1903, p...291, PL 5, fis. 3; This species is not represented in the collection studied by fae writer, and it has been described so recently that the description is not repeated here. The species occurs in the Chazy Limestone at Valcour Island, New York. Eunema epitome Hudson. Eunema epitome Hudson, 1905, Report New York State Paleontologist fer 1903, p: 2090, Pl: 4ngs. 6, 7 This species, which has been very fully described by Hudson, occurs in the Chazy on Valcour Island, New York. Genus Gyronema Ulrich and Scofield. Gyronema historicum (Hudson). (PLATE LIV, FIGURES 5, 6.) Lunema historicum Hudson, 1905, Report New York State Paleontolo- pist for 1903, p-°268,-Pli4; fig. 5. Cyclonema ? normaliana Raymond, 1905, American Journal of Science, Series 4, Vol, 2X 3p. 37%. An examination of Hudson’s figure and description of /Aunema historicum has convinced me that Cyclonema ? normahana Raymond was founded on larger and more complete specimens of the same species, with, however, the detail of surface markings less perfectly preserved. ————— ye ee ie a i ie a ee Mati ei ti RAYMOND: GASTROPODA OF THE CHAZY FORMATION. 209 DESCRIPTION. Shell small, with about four rather flat-sided whorls which expand gradually. Apical whorl smooth, while the succeeding whorls bear four to six revolving ridges. _ Sutures deeply impressed, the side of the previous whorl meeting the top of the next whorl at a right angle. The lower surface of the body whorl is gently convex ; the umbilicus is closed. In addition to the sharp revolving keels, the surface is marked by fine, nearly vertical lines, between which are somewhat deep depressions which do not cross the revolving keels, thus giving the shell a somewhat cancellated appearance. Locality. — This shell is found near the Normal School at Platts- burgh, New York, and on Valcour Island. The holotype is in the New York State Museum. The plesiotypes are in the Yale Museum. Gyronema leptonotum Raymond. (PLATE LV, FIGURE I5.) Eunema leptonotum Raymond, 1905, American Journal of Science, merics 4, VOl. XX, p: 378. DESCRIPTION Shell small, with about four whorls which expand gradually toward the base. The whorls are all convex, the sutures deeply impressed. The first three whorls are smooth and //olofea-like. ‘The fourth, or body whorl, is ornamented by five sharp revolving ridges, equally spaced. ‘These ridges are crossed by fine vertical lines which are close together and give the ridges a flattened appearance. The aperture is not seen. The height of the shell is 5 millimeters; the width at the body whorl 3.5 millimeters. It is evident that this species is closely allied to Gyronema historicum Hudson, but may be readily distinguished from that shell by its some- what slenderer form and the presence of two or three smooth apical whorls instead of one. This shell is not uncommon in the Chazy, but on account of its small size and liability to exfoliation it is often overlooked, or is in too imperfect a condition to be positively indentified. Localhty.— Lower Chazy at Chazy, New York. The type is in the Yale Museum. 210 ANNALS OF THE CARNEGIE MUSEUM. Gyronema ? rotalineum Raymond. (PLaTE LIV, FIGURE 4.) Gyronema ? rotalineum Raymond, 1906, Annals Carnegie Museum, Vols is prays DESCRIPTION. Shell of medium size for the genus. Whorls five in number, spire acute, volutions rounded. The shell expands gradually from the apex, the body whorl occupying about half, or a little less than half, the total length. All the whorls are robust, nearly circular in section. The suture in the cast is deep. ‘The first three whorls are smooth, but the last two show two or three revolving ridges on the lower side of the whorl, the upper side being smooth. In the middle of the body whorl is a narrow concave band. ‘The aperture was apparently rounded, the columellar lip somewhat excavated. There is no um- bilicus. The height of one specimen is 18 mm.; the width of the last whorl, 14 mm. Locality. — This shell occurs very rarely in the Middle Chazy at Sloop Bay, Valcour Island, New York. ‘The type is in the collection | of the Yale University Museum. Gyronema microclathratum (Hudson). Hlolopea microclathrata Hudson, 1905, Report New York State Paleon- —tologist for. 1003, p. 204, Pla, figs: 3, 4. This species, which appears to possess more of the generic charac- ters of Gyronema than of Holopea, has been very fully described by Hudson in the locality mentioned above. ‘The shell is found in the Chazy, at Valcour Island, New York. Order CTENOBRANCHIATA Schweigger. Suborder PLATYPODA. Family SUBULITIDZ Lindstrém. Genus Subulites Conrad. Subulites prolongatus Raymond. (PLATE LIV, FIGURE 13.) Subulites prolongatus Raymond, 1905, American Journal of Science, Series. 4, Vol. XX, p.-394- ial RAYMOND: GASTROPODA OF THE CHAZY FORMATION, BVA ) A single fragmentary specimen from the ‘‘ Trilobite layers ’’ in Sloop Bay represents a very elongate, conical species of Swdulites of the type of Swbulites elongatus Hall, but about the size and shape indicated by the fragments of Sudulites pergractis Ulrich and Scofield, illustrated in the ‘‘ Paleontology of Minnesota,’’ Volume III, part 2, Plate 81. The whorls are, however, longer in our species than in Sudulites fer- gracilis and the shell is even slenderer in proportion to the height. DESCRIPTION. Shell small, elongate, fusiform, with about six (?) whorls (one specimen shows body whorl and three above). ‘The whorls are long and narrow, decreasing slowly and regularly toward the apex. The body whorl is about equal to the length of the two whorls above it, and is contracted below. ‘The aperture is not shown. ‘The height of the fragment is 29 mm. Probably the total height was about 35 mm. Locality. — From the Middle Chazy at Sloop Bay, Valcour Island, New York. The type is in the collection of the Yale University Museum. Genus Cyrtospira Ulrich and Scofield. Cyrtospira raymondi ( Hudson). (PLATE LIV, FIGURES 14, 15.) Subulites raymondi Hudson, 1905, Annual Report of New York State Paleontologist for 1903, p. 293, Pl. 4, figs: 1, 2. This little shell which Hudson described from specimens obtained at Valcour Island occurs also in McCullough’s sugar bush at Chazy in the layers with Spherocoryphe goodnovt. Ulrich and Scofield’s genus Cyrtospira, which was proposed for species near Swbulites but differing in the short curved form and large aperture seems well founded, and the present species falls within the limits of that genus. In its double curvature it resembles Cyr/ospira bicurvata Ulrich and Scofield from the Stones River group, but the spire of the species from the Chazy is higher. Hupson’s DESCRIPTION. ‘« Shell small, fusiform ; apical angle about 44°, length of speci- men, with apical whorl, or a little more, lost, 9.5 mm. Greatest thickness across axis at middle of shell 3.4 mm. Whorls five or six ; a1? ANNALS OF THE CARNEGIE MUSEUM. penultimate whorl showing a rapid elongation, body whorl 6 mm. long or considerably longer than the spire. ‘« Aperture elongate, oblique, narrow, with well carmaul anterior. canal; inner wall of aperture nearly straight; outer lip convex, gradually increasing its distance from the axis for about one-fourth its length, remaining nearly parallel for another fourth and then slightly increasing its convexity to anterior extremity. With the aperture toward the observer, the shell appears slightly angulated at a little above the middle on the right. Turned toward the left through go°, the right hand outline is more uniformly convex. Suture but slightly impressed. Surface smooth.’’ Locality. — A rare species in the Chazy, found so far only on Valcour Island and at Chazy, New York. Family Lirrorinip& Gray. Genus Holopea Hall. Holopea scrutator Raymond. (PLATE LIV, FIGURES 7, 8.) Flolopea scrutator Raymond, 1905, American Journal of Science, Series 4, Vol. XX, p. 379. DESCRIPTION, Shell of medium size, about three whorls, the body whorl consti- tuting by far the larger part of the shell. Spire depressed, sutures not deep. Aperture elongate, oval, entire. Umbilicus small. The specimens usually occur as casts, but on a few the shell is pre- served. It shows no markings except a few growth lines which run diagonally back across the whorl. When the specimens are exfoliated, the suture lines are much more deeply impressed and the spire appears higher. This shell is easily recognized by the low spire, the shallowness of the sutures and the general depressed form of the shell. Locality. — Common in the Chazy at Valcour Island and Chazy, New York. ‘The types are in the Yale Museum. Holopea harpa (Hudson). (PLATE LIII, FIGURE 13 ; PLATE LV, FIGURES 16, 17.) Straparolliina harpa Hudson, 1905, Report New York State Paleon- tologist for 1903, p: 202, El; ties; aoe : RAYMOND: GASTROPODA OF THE CHAZY FORMATION. 2138 Straparollina harpa Raymond, 1906, The Nautilus, Vol. XIX, p. 101, two figures. This is a rare species and has so far been found only in the typical locality on Valcour Island. ‘Two small specimens from that locality are of unusual interest, as they preserve traces of color markings. The specimens are very small, the larger being less than one quarter of aninch in diameter. The body color of the shells isa light yellow, which is the prevailing color of the fossils in the particular stratum from which these specimens were taken. Around the top of the body whorl, adjoining the suture, is a narrow, brownish-gray band. Below it is a band of the yellow body color, and then, about the middle of the whorl, another brownish-gray band, more deeply colored than the one on the top of the whorl. Below this principal band is another light yellow band, and adjoining the umbilicus, the color is orange. The yellow color is undoubtedly due to the iron of the decomposed limestone from which the fossils were obtained, but the brown tints may give some hint of the original coloring of the revolving bands. These are probably the oldest shells on which color markings have been observed, dating, as they do, from Middle Chazy time. The oldest instances of color preservation previously recorded in America are those reported by Professor O. C. Marsh, and Dr. Theodore G. White. Professor Marsh described (Proc. Am. Assoc. Adv. Sci., XVI, p. 326, 1868) certain markings on the shell of a specimen of Lindoceras ( Cameroceras) proteiforme Hall from the Trenton forma- tion in New York. Dr. White mentions (Trans. N. Y. Acad. Sci., Vol. XV, p. 85, 1896) two specimens of Holopea symmetrica Hall from the Black River formation of the Rathbone Brook, N. Y. section, which preserved the original shell material, and one showed the iri- descent luster of pearl. Quite a number of cases of color preservation have been recorded from the Devonian and Carboniferous, but examples from the older formations are exceedingly rare. The specimen shown on Plate LV, figures 16, 17, has been consider- ably crushed, which accounts for many of the differences between it and the specimen figured on Plate LIII, figure 13. For a complete description of this species, see the article by Pro- fessor Hudson cited above. Locality. — The specimens are from the east side of Valcour Island. The original types and the specimen here figured, are all in the Yale University Museum. 214 ANNALS OF THE CARNEGIE MUSEUM. Holopea? plauta Raymond. (PLATE, LIV, FIGURE 9g.) Flolopea plauta Raymond, 1906, Annals Carnegie Museum, Vol. III, P- 577: DESCRIPTION. © Shell small, depressed. Spire scarcely elevated above the level of the body whor]. There are usually two and a half or three volutions, the body whorl expanding very rapidly. Along the outside and at about the middle of the main whorl is a slight ridge, and below this revolving ridge are a series of low broad folds running back into the umbilicus. With this exception the surface is smooth. The aperture is rounded below and rather acute above. ‘The umbilicus is very wide. This shell in some ways resembles /olopea scrutator, but may easily be distinguished from it by the almost total absence of a spire and by the wide umbilicus. Locality. — A rare shell found in the Chazy on Valcour Island and at Chazy, New York. The type is in the Carnegie Museum. Holopea sp. (PLATE LV, FIGURES II, 12.) Another species of //o/ofea is indicated by two imperfect specimens from Valcour Island. ‘These specimens are somewhat irregular in out- line, and the body whorls show corrugations or wrinkles such as are seen in several species of this genus. In neither of the specimens is the spire preserved, but the shell apparently had two or three volutions, which expanded gradually, the body whorl being free for about half a volution. ‘The surface is marked by very fine, wavy, raised lines which have a forward bend on the upper surface of the whorls, turn backward on the periphery and again make a forward-pointing lobe below. The largest specimen is 9 mm. in greatest diameter, and was prob- ably about 7 mm. high when complete. This species differs from FLolopea scrutator and Holopea ? plauta in the vagrant body whorl. Locality. —'The only specimens so far found are from the Chazy Limestone on the west side of Valcour Island. The figured specimens are in the Carnegie Museum. Plate LV. _ TANI, PTT Lane gf, nani? Sydnev Prentice del. } : ; Gastropods from the Chazy Formation. RAYMOND: GASTROPODA OF THE CHAZY FORMATION. 215 Family XENOPHORID# Deshayes. Genus Clisospira Billings. Clisospira bassleri sp. nov. (PLATE LIV, FIGURES 16, 17.) Dr. Ray S. Bassler of the United States National Museum has brought to my attention a small specimen of a species of C/sospira from the collection of Chazy fossils in the National Museum. ‘These fossils were collected on Isle La Motte, Vermont, and associated with the C/isospiva were numerous specimens of Raphistoma stamineum, Scenella montrealensis and Bucania sulcatina. DESCRIPTION. Shell small, sinistrally coiled, expanding gradually. Apical angle 70°. Volutions 314, plump, convex, the last one extending all around, making a large, nearly circular aperture. The shell covers all but the initial whorl, and is thin, smooth, without surface markings. Suture rather deeply impressed. The cast of the initial volution shows that the young shell was loosely coiled, almost in one plane. This shell is very similar to the two shells described as Cusospira curtosa by Billings and to the C@sospira orientalis Whitfield from the Trenton of Wisconsin. From the original specimen of C7/sospira curtosa Billings, our specimen differs in the less expanded aperture, the smaller size and the greater number of whorls. From the second specimen described by Billings under the same specific name, C/sospira bassleri differs in lacking the reticulated surface markings. From Clisospira occidentalts our species differs in the absence of the indis- tinct undulations found on that species, the sharper apical angle, and in possessing one more whorl. ‘The holotype isg mm. in diameter at the aperture and 7.5 mm. high. This species differs only slightly from two of the species previously described, yet there are differences, and they should be pointed out. The writer believes that more harm may be done by uniting unlike forms than by a too fine discrimination between closely related ones. The fossils found in the Chazy are usually poorly preserved, and in studying them, it has been the policy of the writer to give new names whenever it was not possible to prove identity with species already described. Locality. — A rare species in the Chazy on Isle La Motte, Vermont. The holotype is in the United States National Museum. 216 ANNALS OF THE CARNEGIE MUSEUM. Suborder CONULARIDA * Miller and Gurley. Family CONULARIIDZ Walcott. Genus Conularia Miller. Conularia triangulata Raymond, (PLATE LIV, FIGURE 18. ) Conularia triangulata Raymond, 1905, American Journal of Science, Vols S, pis70- A rare fossil in the Chazy is a small, slender Conularia, which is almost triangular in cross section. It is not, however, three sided, as it at first appears, but really six-sided, as each of the angles is trun- cated near the apex so that there are three broad faces and three very narrow ones. DESCRIPTION. Shell small, slender, slightly curved, six sided, but three of the sides are sO narrow as to give the shell an almost triangular cross section. The narrow faces alternate with the wide ones, the narrow faces trun- cating the angles which the wide faces would make if prolonged till they met. Along each of the faces, both wide and narrow, is an ele- vated line which extends longitudinally along the center of the face. The surface markings consist of numerous fine transverse striz which bend backward on crossing the raised line. The best specimen in the writer’s collection is broken at the tip and at the aperture, but, as it stands, is 38 mm. long. The original length was at least 8 mm. more. At the larger end the three wide faces are each 7 mm. wide, and the narrow faces are each 1.5 mm. wide. At the small end the wide faces are 2.5 mm. wide and the narrow faces are reduced to practically nothing, thus showing that in young stages the shell was triangular. Locality. — The type, which is in the collection of the Carnegie Museum, was found in the upper part of the Chazy on the southeast point of Valcour Island (Cystid Point). This species occurs also near Smuggler’s Bay in layers a little lower in the formation. SPECIES UNRECOGNIZABLE OR BELONGING TO OTHER FORMATIONS. Euconia amphitrite (Billings). Pleurotomaria amphitrite Billings, 1865, Paleozoic Fossils of Canada, Yo ly pias *This suborder is left with the Gastropoda as the classification used in Eastman’s Le) translation of Zittel’s ‘‘ Griindzuge ’’ is here followed. —— = RAYMOND : GASTROPODA OF THE CHAZY FORMATION. A Euconia amphitrite Ulrich and Scofield, 1897, Paleontology of Min- mesota, Vol. III, part 2, p. 954. This species was described by Billings from shells obtained at the Mingan Islands from the ‘‘ Chazy or Black River.’’ No similar speci- mens have been found in the Chazy of the Champlain or Ottawa Val- leys, but there are four similar species in the Beekmantown, one from the Mingan Islands, one from Newfoundland, and two from Fort Cassin, Vermont. It is very possible that Auconza amplhitrite is also from the Beekmantown. Murchisonia ? hermione Billings. Murchisonia hermione Billings, 1865, Paleozoic Fossils of Canada, Welt p- 33, figs.-34, 35, 35¢- From the same locality and formation as Huconia amphitrite. Plethospira hyale (Billings). Murchisonia hyale Billings, 1865, Paleozoic Fossils of Canada, Vol. I, Py 33° Plethospira hyale Ulrich and Scofield, 1897, Paleontology of Minne- sota, Vol. III, part 2, p. 1009. This shell was described from one specimen found at Phillipsburgh, Canada, ‘‘ in beds holding fossils approaching in aspect to those of the Chazy or perhaps Black River Limestone.’’ Tryblidium eubule (Billings). Metoptoma eubule Billings, 1865, Paleozoic Fossils of Canada, Vol. I, p2\33. Tryblidium eubule Whiteaves, 1884, Paleozoic Fossils of Canada, Vol. HT, ‘p: 30. This species also is from Phillipsburgh, probably from the same beds as Plethospira hyatle. Platyostoma auriforme (Hall). Capulus auriformis Hall, 1847, Paleontology of New York, Vol. I, p. ae, Pl. 6, figs.-6a, 60. Stomatia auriformis Emmons, 18 55, American Geology, p. 157. This species was described by Hall as coming from the Chazy Lime- stone at Galway, Saratoga County, New York. As the Chazy forma- tion does not extend as far south as Galway, it is probable that this is 218 ANNALS OF THE CARNEGIE MUSEUM. not a Chazy fossil. No specimens of it have been found in the typical region. Emmons refers it to the Trenton Limestone (Joc. ci¢.). Pleurotomaria antiquata Hall. Pleurotomaria antiquata Hall, 1847, Paleontology of New York, Vol. dip. ay Bi ie. ay .No more specimens lke the original have been found. ‘The type is in the New York State Museum at Albany. List OF SPECIES OF GASTROPODA DESCRIBED FROM THE CHAZY FORMATION UP TO THE TIME OF THE PUBLICATION OF THE PRESENT PAPER. SYNONYMS AND DOUBTFUL SPECIES IN ITALICS. Archinacella deformata (Hall). Archinacella propria Raymond. Bellerophon sulcatinus Emmons ( Bucania sulcatina). Bucania catillotdes Raymond (Oxydiscus catilloides). Bucania champlainensis Whitfield (Bucania sulcatina). Bucanta rotundata Hall (Bucania sulcatina). Bucania sulcatina (Emmons). Capulus auriformis Hall (Platyostoma aurtforme). Clisospira bassleri Raymond. Conularia triangulata Raymond. Cyclonema ? normalana Raymond (Gyronema historicum ). Cyrtospira raymondi (Hudson). Eccyliomphalus fredericus Raymond. Eccyliomphalus kalmi Raymond. Eccliomphalus proclivis Raymond. Eccyliopterus kalmt Raymond (Eccyliomphalus kalmi). Eccyliopterus proclivis Raymond (Eccyliomphalus proclivis). Eccylopterus vagrans Raymond. Eotomaria obsoleta Raymond. Euconia amphitrite (Billings). Probably not Chazy. Eunema altisulcatum Hudson. Eunema epitome Hudson. Eunema historicum Hudson (Gyronema historicum). Eunema leptonotum Raymond (Gyronema leptonotum). oa RAYMOND: GASTROPODA OF THE CHAZY FORMATION. Gyronema historicum (Hudson). ~ Gyronema leptonotum Raymond. Gyronema microclathratum (Hudson). Gyronema rotalineum Raymond. Helicotoma vagrans Raymond (Eccyliopterus vagrans). Helicotoma whiteavesiana Raymond. ® Holopea harpa (Hudson). FHlolopea hudsont Raymond (Trochonema hudsoni). Flolopea microclathrata Hudson (Gyronema microclathratum ). Holopea? plauta Raymond. Holopea scrutator Raymond. Hormotoma infrequens (Billings). Liospira docens (Billings). Lophospira aspera (Billings). Lophospira billingsi Raymond. Lophospira perangulata (Hall). Lophospira rectangularis Raymond (Trochonema rectangulare). Lophospira rectistriata Raymond. Lophospira seelyi Raymond. Lophospira subabbreviata (d’Orbigny). Name to be dropped. Maclurea atlantica Billings ( Maclurites atlanticus). Maclurea magna Emmons (Maclurites magnus). Maclurea striata Emmons (Raphistoma striatum). Maclurites atlanticus Billings. Maclurites magnus Lesueur. Metoptoma ? dubia Hall (Archinacella deformata). Metoptoma eubule Billings (Triblidium eubule). Metoptoma montrealensis Billings (Scenella montrealensis). Murchisonta abbreviata Hall (Lophospira subabbreviata). Murchisonia asper Billings (Lophospira aspera). Murchisonia decurta Hall (Lophospira subabbreviata). Murchisonia ? hermione Billings. Probably not Chazy. Murchisonia hyale Billings (Plethospira hyale). Murchisonia infrequens Billings (Hormotoma infrequens). Murchisonia perangulata Hall (Lophospira perangulata). Murchisonia subabbreviata a’ Orbigny (Lophospira subabbreviata). Orbicula ? deformata Hall (Archinacella deformata). Oxydiscus catilloides Raymond. Paleacmea ? irregularis Raymond. 219 220 ANNALS OF THE CARNEGIE MUSEUM. Platyostoma auriforme (Hall). Probably not Chazy. Plethospira hyale (Billings). Probably not Chazy. Pleurotomaria amphytrite Billings (Euconia amphytrite). Pleurotomaria antiquata Hall. Not recognizable. Pleurotomaria biangulata Hall (Trochonema biangulatum). Pleurotomaria calyx Billings (Raphistoma stamineum ). Pleurotomaria creviert Billings (Raphistoma stamineum). Pleurotomarta docens Billings (Liospira docens). Pleurotomaria tmmatura Billings (Raphistoma immaturum). Pleurotomaria pauper Billings (Raphistoma stamineum). Raphistoma immaturum (Billings). Raphistoma planistria Hall (Raphistoma stamineum). Raphistoma planistria parva Hall (Raphistoma stamineum ). Raphistoma stamineum Hall. Raphistoma striatum (Emmons). Raphistomina undulata Raymond. Scalites angulatus Emmons. Scalites planistria Emmons (Raphistoma stamineum ). Scalites staminea Emmons (Raphistoma stamineum ). Scalites striata Emmons (Raphistoma striatum). Scenella montrealensis (Billings). Scenella pretensa Raymond. Scenella robusta Raymond. Stenotheca dubia Whitfield and Hovey (Archinacella deformata). Stomatia aurtformis Emmons (Platyostoma auriforme). Straparollina harpa Hudson (Holopea harpa). Straparollus magnus Emmons (Maclurites magnus). Subulites prolongatus Raymond. Subulites raymondi Hudson (Cyrtospira raymondi). Tetranota bidorsata ? Hall. Trigyra ulrichi Raymond. Trochonema biangulatum (Hall). Trochonema dispar Raymond. Trochonema hudsoni Raymond. Trochonema rectangulare Raymond. Tryblidium eubule (Billings). Probably not Chazy. ws ~~ & °F. or) C= a ae bo pee RAYMOND: GASTROPODA OF THE CHAZY FORMATION. a EXPLANATION OF PLATES. PLATE XLVI. 1. Archinacella deformata (Hall). A specimen from Valcour Island, enlarged four diameters. | 2. Side view of the same specimen, with same magnification. 3. The same species. A specimen from Aylmer, Canada, to compare with figure 2. Enlarged two diameters. 4. The same species. A specimen from Crown Point, N. Y., enlarged two diameters. 5. The same species. A specimen which is partially buried in matrix, four times natural size. Valcour Island, New York. 6. Another view of the same specimen. 7, 8. Archinacella? propria Raymond. Top and side views. Enlarged four diameters. Chazy, New York. 9, 10. Scenella montrealensis (Billings). Side and apical views of a specimen from Chazy, New York. Four times natural size. II, 12, 13. Scenella pretensa Raymond. ‘Three views of a specimen from Lenoirs, Tennessee. Enlarged four diameters. e PLATE XLVIL. I. Scenelila robusta Raymond. ‘Top view of the typical specimen. Twice the natural size. Valcour Island, New York. 2, 3. The same species. Top and side views of a specimen from Lenoirs, Ten- nessee. Twice natural size, 4. Raphistoma striatum (Emmons). Side view, showing thickened inner lip. Aylmer, Canada. Twice natural size. 5. Another view of the same specimen, showing form of outer lip, and surface markings. Twice natural size. 6. Same specimen viewed from below, to show the form of the lip. Twice natural size. ; 7,8. Raphistoma striatum (Emmons). Top and bottom views of a cast of a specimen from King’s Bay, Cooperville, New York. Natural size. 9, 10. The same species. ‘Top and bottom views of a cast collected at Aylmer, Canada, to be compared with figures 7 and 8. Natural size. 11. Raphistoma stamineum Hall. Side view of the inner whorls of a specimen from Chazy, New York, showing the flat strize of a small shell. Twice natural size. 12. The same species. Top view of a specimen from Chazy, New York. Notice the slender whorls, as compared with those of Raphistoma striatum. Natural size. 13. The same species. A small specimen enlarged four diameters, to show the way in which the strize are sometimes gathered into fascicles. Valcour Island, New York. PLATE XLVIII. 1. Raphistoma stamineum Hall. View of the spire of a small specimen from Valcour Island. Enlarged four diameters. 2. The same species. Bottom view of a specimen from Chazy, New York, to show absence of umbilicus in a testiferous specimen. Twice natural size. 292 ANNALS OF THE CARNEGIE MUSEUM. 3. The same species. Bottom view of an exfoliated specimen, showing umbilicus. Natural size. 4. The same species. Enlargement of a part of the surface of an adult specimen, to show the interruption of the strize on the top of the whorl. Four times natural size. - 5, 6. Top and side view of a specimen which appears to be the same species, Twice natural size. Lenoirs, Tennessee. For further illustrations of Raphistoma stamineum, see plate LV. 7. Raphistomina undulata Raymond. View of the spire of a specimen from Val- cour Island, New York. Twice natural size. 8, 9, 10. The same species. Top, side, and bottom views of a specimen from the same locality as the last. Enlarged two diameters. It, 12. Helicotoma whiteavesiana Raymond. ‘Top and side of a specimen from the ‘‘ Hog Back,’’ near Ottawa, Canada. ‘Twice natural size. 13. Scalites angulatus Emmons. A fragment showing the depression just beneath the shoulder of the whorl. Natural size. 14. The same species. A specimen from Plattsburgh, New York, which shows the inner lip well. The outer lip is somewhat crushed and deformed. Natural size. 15, 16. The same species. The natural sections on the rock surface. They show the high spire and the shoulder of these shells. Natural size. Plattsburgh, New York. PLATE SLX: 1. Lophospira billingst Raymond. Side view of a specimen from Aylmer, Canada. Natural size. 2. The same species. Another specimen lacking the spire, but with a less distorted body whorl. Natural size. 3, 4. Lophospira rectistriata Raymond. Side views of two specimens from Chazy, New York. Twice natural size. 5. The same species. Side view of the body whorl of a specimen from Chazy, New York, showing course of the surface markings. Enlarged two diameters. 6. The same species, Side view of the body whorl of another specimen. Twice natural size. 7, 8. Lophospira perangulata (Hall). Two casts, each lacking the apical whorl. Enlarged two diameters, Valcour Island, New York. 9. Hormotoma species ind. A fragmentary specimen on a piece of the matrix. Natural size. From Maclurea Point, New York. 10. Eccyliopterus vagrans Raymond. Lower side of a specimen from Valcour Island, New York. ‘Twice natural size. I1. The same species A smaller specimen, viewed from above. Twice natural size. Valcour Island, New York. 12. Eotomaria obsoleta Raymond. Internal cast of a specimen from Crown Point, New York, ‘Twice natural size. 13, 14. The same species. ‘Top and side views of the paratype, from Valcour Island, New York. Twice natural size. 15. Lucania sulcatina (Emmons). A partially exfoliated specimen, showing the lines of growth. Natural size. Crown Point, New York. 16. The same species. A specimen from Isle La Motte, Vermont, showing the form of the aperture. Natural size. bE NI nn rg =“, aw] /— —_———— a RAYMOND: GASTROPODA OF THE CHAZY FORMATION. oe 17. The same species. Side view of a specimen from Isle La Motte, Vermont. Natural size. For specimens of this species showing surface markings, see plate LV. PEATERE. 1. Maclurites magnus Lesueur. The outer view of a partially exfoliated oper- culum. Crown Point, New York. Natural size. 2. The same species. Cast of a specimen from Lenoirs, Tennessee. Natural size. 3. Bucania sulcatina (Emmons). Side view of a partially exfoliated specimen, showing the lines of growth only. Natural size. 4. The same species. Aperture of another specimen from Crown Point, New York. Natural size. 5. Zetranota bidorsata ? (Hall). Internal cast of a small specimen from Valcour Island, New York. Twice natural size. 6. Trigyra ulricht Raymond. A testiferous specimen from Isle La Motte, Vermont. The very minute pits on the surface could not be brought out by this process. Twice natural size. 7. Operculum of an unknown gastropod. Valcour Island, New York. Natural size. PLATE LI. 1, 2. Maclurites magnus Lesueur. Top and bottom views of a specimen from Valcour Island, New York. Natural size. Prare (if, I, 2, 3. Maclurites magnus Lesueur. Outer, inner, and side views of the oper- culum. Natural size. Valcour Island, New York. 4. The same species. Side view of the specimen figured on the previous plate. Pirate LI. 1, 2. Eccyliomphalus fredericus Raymond. ‘Two views of a specimen from Crown Point, New York. Twice natural size. 3. The same species. A somewhat distorted specimen from Crown Point, New York. Twice natural size. 4. Eccyliomphalus kalmit Raymond. Upper side of a specimen from Valcour Island, New York. Enlarged four diameters. 5. Eccyliomphalus proclivis Raymond. Lower side of a specimen from Crown Point, New York. Twice natural size. 6. Eccyliomphalus sp. ind. Enlarged four diameters. Valcour Island, New York, 7, 8. Trochonema dispar Raymond. Bottom and side views of a specimen lack- ing the spire. Twice natural size. Valcour Island, New York. 9, 10. Trochonema biangulatum (Hall). Side and top views of a specimen from Valcour Island, New York. Twice natural size. II, 12. Zrochonema rectangulare Raymond. ‘Two views of a specimen from Valcour Island, New York. Twice natural size. 13. Holopea harpa (Hudson). Four times natural size. WValcour Island, New York. (After Hudson. ) iy 7 224 ANNALS OF THE CARNEGIE MUSEUM. PrIATesLiv; 1. Zrochonema hudsoni Raymond. A cast, partially buried in the matrix. Twice natural size. Valcour Island, New York. 2. The same species. The body whorl of a testiferous specimen from Chazy, New York. Twice natural size. 3. The same species. Another cast from Valcour Island. Twice natural size. 4. Gyronema rotalineum Raymond. A cast from Valcour Island, New York. Twice natural size. 5. Gyronema historicum (Hudson). A specimen from Plattsburgh, New York. Type of Cyclonema ? normalianum Raymond. ‘Twice natural size. 6. The same species. A figure of the type, after Hudson. Four times natural size. 7. Holopea scrutator Raymond. A cast from Valcour Island, New York. The inner lip is obscured by matrix. Twice natural size. 8. The same species. Top view of a specimen retaining a part ofthe shell. Val- cour Island, New York. ‘Twice natural size. 9. Holopea ? plauta Raymond. A specimen from Chazy, New York. Twice natural size. 10. Paleacmeairregularis Raymond. A small specimen from Chazy, New York. Four times natural size. 11. The same species. A larger specimen, showing a trace of the suture. Twice natural size. Chazy, New York. 12, The same species. A specimen with broken apex. Natural size. Chazy, New York. 13. Subulites prolonga‘us Raymond. Valcour Island, New York. Twice natural size. 14, 15. Cyrtospira raymondi (Hudson). Two views of the type. After Hudson. Four times natural size. ; 16. Clisospira bassleri Raymond. Side view of a specimen from Isle La Motte, Vermont. Twice natural size. 17. The same species. Apical view of the same specimen shown in figure 16. Four times natural size. 18. Conularia triangularis Raymond. ‘The type, and diagram of section. Val- cour Island, New York. Twice natural size. PLATE LV. I. Scenella montrealensis (Billings). A specimen from Isle La Motte, Vermont, showing surface markings. Twice natural size. 2. Lophospira aspera (Billings). One of the types, from the Mingan Islands, Canada. Natural size. 3. Lophospira seelyi Raymond. A specimen from Isle La Motte, Vermont. Twice natural size. 4. Hormotoma infrequens (Billings). The typical specimen, on a piece of the matrix. Natural size. 5, 6, 7. Raphistoma stamineum Hall. Side and bottom views of a specimen from Isle La Motte, to show form of lips and surface markings of an adult. Natural size. RayMOND: GASTROPODA OF THE CHAzyY FORMATION. 225 8. The same species. Side view, to show excavated columella. Natural size. Isle La Motte, Vermont. 9,10. Oxydiscus catilloides Raymond. Back and side views of the type. Enlarged six diameters. Valcour Island, New York. 11. Holofeasp.ind. The side of a specimen, to show surface markings. Enlarged six diameters. 12. The same species. A larger specimen, viewed from above. The spire is broken off. Enlarged six diameters. Valcour Island, New York. 13. Buccania sulcatina (Emmons). A specimen from Isle La Motte, Vermont. Twice natural size. 14. The same species. A very young individual from Valcour Island, New York. Six times natural size. 15. Gyronema leptonotum Raymond. A specimen from Chazy, New York. Six times natural size. 16, 17. Holopea harpa (Hudson). A somewhat distorted specimen, showing color markings. Six times natural size. Valcour Island, New York. xX. A FURTHER OCCURRENCE OF WYNNEA AMERI- CANA IN PENNSYLVANIA. By Otto E. JENNINGS. CPiLAgE evi) The rare fungus, Wynnea americana, was first described by Thaxter in 1905’ from specimens collected by himself, in 1888, at Burbank, Tennessee, and at the same place and on the North Carolina moun- tains at Cranberry, in 1896. A specimen is also mentioned collected by Mr. E. Wilkinson at Mansfield, Ohio. In September, 1906, Mrs. Jennings and myself found a few speci- mens of this species at Ohio Pyle, Fayette County, in southern Penn- sylvania, as noted and described by Sumstine in the Journal of Mycol- ogy.” In early September, 1907, Mrs. Jennings and myself found the fungus growing quite abundantly along the lower slopes of the deep wooded valley of Meadow Run, four miles south of Ohio Pyle, and on September 8, 1907, a few specimens were found by Dr. T. D. Davis, Prof. D. R. Sumstine, and the writer on the estate of Mr. James R. Mellon, on the Laurel Hill Mountains of northeastern Westmore- land County, near New Florence. The Ohio Pyle and New Florence specimens were both found in shaded soil derived from sandstone formations rich in humus, and in both places were near the bottom of the slope where the soil is con- stantly moist. The aérial portion of the fungus, resembling a rabbit’s ear, varied from a purplish red, when young, to a dark velvety brown, when old. ‘The underground portion consisted of a very irregularly lobed, tuber-like body, always comparatively large and much heavier than the aérial portion. This tuber-like sclerotium was tough and cartilaginous ard usually presented the appearance of a perennial 1Thaxter, Roland. A New Species of Wynnea. Contr. Crypt. Lab. Harvard Univ. LX. Bot. Gaz. 39: 241-247.. April, 1905. 2Sumstine, D. R. Note on Wynnea americana. Jour. Mycol. 12: 59. March, 1906. 226 ‘ssuluual ‘a '°G Aq poudeisoj0t “OZIS TBanywu Jyey-ou I Gi | IS | J] “IOJXVY YT, VUDITAAUD pauud Af 1 i y ‘ y sige etal ‘Al “ISA ‘WAASAW JIDSSNYVO STVNNY JENNINGS: OCCURRENCE OF WYNNEA AMERICANA IN PENNA. 227 growth, many parts of the tuber showing old shells or husks, which were evidently the remains of former protuberances. The accompanying photograph, taken by the writer at Ohio Pyle, Pa., Sept. 1, 1907, shows the appearance of the species when just dug out of the earth. The photograph shows the specimens one-half natural size. CARNEGIE MUSEvuM, November 20, 1907. XI. A PRELIMINARY ACCOUNT OF THE PLEISTOCENE FAUNA DISCOVERED IN A CAVE OPENED AT FRANKSTOWN, PENNSYLVANIA, IN APRIL AND MAY, 1907.' By W. J. HoLianp. (Plates LVII-LVIII.) In the latter part of April of the present year the writer received messages from Mr. H. H. Jack and Mr. E. H. L. Page of Hollidays- burg, Pa., calling attention to the fact that there had been discovered in the quarries of the American Lime and Stone Company near Franks- town, Pa., the fossil remains of a number of large animals, and sug- gesting that a careful examination of the locality should be made. Mr. O. A. Peterson of the Section of Paleontology in the Carnegie Museum was promptly dispatched to the spot with instructions to report upon the character of the deposits. He returned after a few days bringing with him some material of sufficient interest to cause the writer to feel justified in requesting him to return to the locality and to carefully continue the work of recovering whatever might be found. He spent nearly three weeks in the work, being visited during that time by the writer, who made a careful inspection of the site and helped a little in the task of recovering specimens. A great deal of assistance was accorded to Mr. Peterson by Mr. James King Henry, the Superintendent of the American Lime and Stone Company, who not only instructed his men while at work to help Mr. Peterson so far as was possible, but himself with great kindness at times personally aided Mr. Peterson in the laborious task of taking up the deposits which were encountered. The quarries of the American Lime and Stone Company are situated on the top of a hill rising about four hundred feet in height above the banks of the Juniata. The kilns of the Lime and Stone Company are located in the historic hamlet of Frankstown, in Blair County, on the line of the Petersburgh branch of the Pennsylvania Railroad. ‘The limestone in which work is being carried on, is a fine compact blue ' Paper read at the Seventh International Zodlogical Congress, held in Boston, August, 1907. 228 ANNALS CARNEGIE MUSEUM, Vol. IV. Plate LVII. View of the eastern end of the cave after its nerth and south walls had been removed in great part together with the top. Mr. Peterson is seen stand- ing by the remnant of the north wall. Above him may be seen stalagmites pendant from the rocks. HOLLAND: PLEISTOCENE FAUNA AT FRANKSTOWN, Pa. 229 limestone, the purity of which has caused the product of the quarries to be much sought after by manufacturers of paper. The rock has been extensively faulted, and the fissures, which vary in thickness from a few millimeters to as much asa foot in diameter, have been filled with calcite. The rock is lower Devonian, locally known as Lewistown limestone. The hill, in which the cavern is located, con- Fic. 1. Diagrammatic section of the cave before its walls were removed in the process of blasting and excavation. A. Open chamber. B. Space filled with earth and stones in which bones were found. C. Fissure closed with rocks. D. Lime- stone strata dipping to the south. E. Soil bearing forest growths. tains several small caves, or grottoes, the floors of two of which were -explored by Mr. Peterson, but neither of them showed any evidence of extinct animal life. The cave in the limestone quarry was appa- rently originally a cleft or fissure, which, owing to falling in of the upper strata, had gradually become roofed over by large blocks, which had been in part cemented together by stalactitic deposits with which earthy matter washed down from the surface had become mingled. 230 ANNALS OF THE CARNEGIE MUSEUM. No evidence existed on the top of the hill of an approach to this cave, and it was only as the process of blasting was being carried forward that its existence was revealed. The cave itself, as nearly as can be determined, was about forty feet in length, averaging from six to eight feet in width, and at the highest point was not more than ten or twelve feet high. Its floor was about thirty feet below the top of the hill. The bottom of the cave was filled in at most places to a depth of ap- proximately two feet with red soil traversed everywhere by bands and layers of dark material rich in Organic matter and somewhat spongy in texture. A deeper crack at one point descends to a depth of about fourteen feet. This it was impossible to explore at the time, but borings made by Mr. Henry after our visit, yielded no bones. Mingled with the finer deposits were fragments of stalactites and pieces of stone, which had fallen from the roof and sides of the cave, varying from large blocks several feet in diameter to mere chips. These angular fragments were so commingled with the finer material as to make the work of recovering the bones which lay at the bottom of the cave a task of much difficulty. The fall of rock in past ages had led to the fracture of many of the bones, which had been crushed down into the looser material in which they were imbedded. As the work of quarrying went forward from time to time portions of the floor of the cave were brought down. Mr. Peterson was always on hand to investigate, and from the stuff, which was dislodged, he carefully extracted everything which could be preserved, and he also went forward into the cavern and endeavored to explore the floor before the blasting and quarrying were advanced. Rains were con- stent, and Mr. Peterson worked most of the time in mud. As the result of his labors, upon which he deserves to be congratulated, there were brought back to the Carnegie Museum a great number of speci- mens, mostly in a broken and shattered condition, due to no fault of the collector. A superficial examination of the collections taken from the cave shows the remains of a number of species of mammals, birds, and reptiles. The genus A/fega/onyx is represented by a number of teeth; to what species they should be referred has not yet been determined. It is possible that another genus of Adenfata is represented, but this can only be decided aftcr a more careful investigation than the writer has had time to bestow upon the subject. The genus Zafzrus is repre- ANNALS CARNEGIE MUSEUM, Vol. IV. Plate LVIII. View of the eastern end of the cave after its sides and top had been almost entirely removed, taken from the southern edge of the quarry. The two men are standing with their backs to the north wall of the cave. = ky au . A555 Nj HOLLAND: PLEISTOCENE FAUNA AT FRANKSTOWN, Pa. 231 sented by the third and fourth left lower premolars of a small species about the size of Zapirus americanus Brisson. ‘The Suzde@ are repre- sented by the remains of a number of peccaries. A remarkably well-preserved mandible I determine to be that of Ducotyles penn- sylvanicus Leidy, the type of which is the fragment of a mandible of an immature specimen, which was found in Hartman’s Cave near Stroudsburg, Pa. Our specimen is the lower mandible of an adult, accompanied by other portions of the skeleton, including well-pre- served metapodials and vertebree. The animal was, judging from what we possess, considerably larger than either of the existing peccaries. The lower jaw is longer than in either Décotyles labiatus or torquatus, and the metapodials are longer. ‘The canine teeth, which are wanting in Professor Leidy’s type, are also much longer and more formidable weapons than in either of the existing species of the genus. Another peculiarity is the fact that the jaw at its inner angle flares outwardly, whereas the existing species of Dvzcotyles are characterized by the recession of the lower jaw at the angle. ‘The species might well be referred to the genus P/atygonus, in which, however, it has not been placed by Hay in his Catalogue. Lydekker regards Platygonus as practically synonymous with Decotvles or Zayassu, but the distinction between the two genera the writer believes to be valid. Portions of a skeleton referable to ABzson were uncovered. The Cervide@ are repre- sented by three species, one of them of very large size, possibly repre- senting Cervalces, the other two smaller. ‘There has not been time to carefully decide the generic location of these species, which are prin- cipally represented by the teeth of the lower jaws. A fragment of an antler, possibly belonging to Carzacus virginianus, was dis- covered. The remains of mastodons were exceedingly plentiful. Mastodon americanus is represented by portions of one mature and five or six immature specimens. ‘The remains of rodents are very numer- ous. Numerous jaws of Lrythizon, Lepus, Scturus, Fiber, and other smaller genera are recognizable. Chiroptera are present in the remains of one or more species. The Ursid@ are represented by two species. Fragments which are no doubt correctly referable to Ursus americanus were found, but far more interesting are the remains of several individuals of a huge bear which J identify without any hesita- tion as Arctodus haplodon Cope. The presence of the entepicondylar foramen in the humerus, the two masseteric fossee of the mandible and the peculiar shape of the first inferior true molar are characters 222 ANNALS OF THE CARNEGIE MUSEUM. which serve to reveal at once the identity of the animal with the species described by Cope. ‘The material in our possession is better than that at the command of Professor Cope. It is, unfortunately, fragmentary, but very well preserved so far as it goes, and it is quite abundant, though not sufficient in quantity to enable us to restore the skeleton. ‘The genus JZephitis is represented by at least one species. The remains of the genus Cazzs are quite numerous. Several well- preserved jaws show an animal not differing greatly in size from Canis occidentalis, but differing somewhat in the dentition, which more nearly resembles that of the Esquimaux dog of to-day, of which it may have been the ancestor. ‘The Felidz are represented by fragments of an animal comparable in size with the jaguar. Unfortunately there is not much of the skeleton preserved, except the bones of the foot. In addition to the remains which I have mentioned there are the remains of several other smaller carnivores represented by teeth and portions of jaws which it has been impossible as yet for lack of time to identify. The remains of birds are not infrequent, and among them the bones of a large species undoubtedly belonging to the genus A/e/eagris are rather numerous. The collection also contains the remains of Ophidia and Batrachia. These are scattered and so completely disarticulated as to make resto- ration apparently hopeless, and even determination of the genera problematical. It has been of course impossible, on account of the brief time at the command of the writer since the material was received, to carefully work it out and to study it with sufficient care to enable a correct reference to be made in all cases. Mr. Peterson, who took up the material, is inclined to the view that from thirty to forty species and several hundred individuals are represented in the material which has been secured. A comparison of the fauna found in this cave with that discovered in a similar cave at Port Kennedy, in the valley of the Schuylkill, and which was made the basis of several papers by the late Professor E. D. Cope, shows many points of similarity and some marked differences. The scarcity of the remains of AMegalonyx, which were very abundant in the Port Kennedy deposits, is noticeable. Some of the neo-tropical genera reported by Cope are not found among the remains taken from the Frankstown cave. But the existence of the peccary, the tapir, and HOLLAND: PLEISTOCENE FAUNA AT FRANKSTOWN, Pa. 283 of Ursus haplodon, which, as has been pointed out by Cope, shows affinities to the fossil Ursidz of South America, rather than to existing North American species, shows that at the time when these deposits were made there was a commingling on the soil of Pennsylvania of nearctic and neotropical genera, though the nearctic forms predomi- nated. The deposits in the Frankstown cave were probably a little later in their origin than those at Port Kennedy. Probably the most interesting of all the numerous specimens ob- tained in the cave at Frankstown is the mandible of a very young mas- todon, sufficiently well preserved to enable a study to be made of the early stages of the dentition. It is, I think, the most perfect speci- men of its kind in existence in any collection. It recalls at once the publication by Dr. Godman in 1830 of his genus 7e¢/vacaulodon for the reception of certain specimens which Dr. Isaac Hays in 1834 figured in the Transactions of the American Philosophical Society for that year. Dr. Hays with great chivalry defended the position of Dr. Godman, which had been assailed, and maintained the valid character of the genus Zetracaulodon, which was based principally upon the exist- ence of four incisors, or tusks, in the lower jaw. No better evidence of _ the correctness of the position of Dr. Godman’s critics could be found than that given by the remains discovered in the Frankstown cave. The presence of the remains of numerous infant mastodons and of various species of artiodactyles in the cave, associated with the remains of the huge Arctodus haplodon, suggests that the latter, at a time when what later became a sealed cavern was still an open cleft in the rocks, preyed upon young mastodons, which they may have separated from their mothers and chased over the edge of the cliff-like wall. Falling to the bottom they became an easy prey to the great bears, just as calves to-day fall a prey both to the black bear and to the grizzly. In like manner these bears dragged into this place, which was their lair, in order to feed their young, the carcasses of deer and other hoofed animals. Or they may have driven them over the rocks, as it is pos- sible to imagine that the young mastodons were driven. —_ y \ —= } » ain Ly * iy — SS (ae y if : | — ) ces aM Min E a. ~ - fh ty \ = > i a — a NIN \ i" { \ iH ) Wien —-, Y, 4 N \ Wy . ny : i ay = : Ws Wl} A Vee i I = Aphelops montanus Douglass. | i a j ne 250 | yy \\\\ Sy UY 9 % Z forward trend of the posttympanic, +. DOUGLASS : RHINOCEROSES FROM NORTH DAKOTA AND MONTANA 257 moderately broad and converge backward until they are 2 cm. apart ; then they begin to diverge 8 cm. anterior to the occipital crest. The upper posterior portion of the skull resembles that of specimen No. 840, which is described in this paper, but in the latter specimen the supratemporal ridges diverge more rapidly toward the occiput. There are two infraorbital foramina in the maxillary above the third and fourth premolars. The lower of these foramina is large and round. ‘The other foramen is just above and a little posterior to the first and is oblong-oval in form, with the apex of the oval antero-inferior. The malar is rather shallow beneath the orbit, but is deeper farther back. The posterior upper angle of the zygo- matic portion of the squamosal is low. The external auditory open- ing is entirely surrounded by the temporal bone, as the postglenoid and posttympanic portions are in contact by the enlargement and which is large, rounded on the outer surface, and somewhat roughened by shallow depressiosn. The paroccipital processes, as seen from the side, are quite broad. They are near the occipital con- dyles. The occiput appears to have been nearly perpendicular, not much inclined either forward or backward. It is low in propor- tion to the length of the skull and is broad, not narrowing rapidly Fig. 1. Femur of Aphelops montanus. i, upward as in specimen No. 854 (Aphelops ceratorhinus ?). The middle of the occiput immediately above the foramen magnum is very convex transversely and projects backward overhanging that opening. 258 ANNALS OF THE CARNEGIE MUSEUM. The depth of the mandible is greater than that of the type of Aphe- lops ceratorhinus, it is much thinner transversely, and the angle is not so large. The femur is long and very slender for so large a rhinoceros. Only portions of other bones are preserved. MEASUREMENTS. 7 nm Length of skull from tips of nasals to crests of OCCiput............sceceee 567 Width of skull including zygomatic arches, greatest .............2.eeeeeeees 330 Width of skull at: post-tympanies 7.2, Jessen tceesecees proms Sree oie 200 Height of -skulh at Orbit. -20se. seu ccesnieen conece sve vsegeanaue canes tenant ae eee 205 Fieipht-of,-skeulll) af occiptt .:.1 see 3-ars cc ap eolnn apeeade eee eee me Oa 200 Length of nasals, approximately: ..2.niccssteee ta teeten os coos nee ieee 170 Height of paroccipital processes, 2. ..<...2te./acees tresses eer ee A. 51 Width of sparoceipital processes at base:.2.\\. dig: ac: sete 51 Height of mandible under P(. \..~irscotneana che est esmbaeetson ty: kane een 85 Height of ascending ramus of mandible, estimated................0.-ceeeee 255 Width of ascending ramus of mandible from M, backward................. 150 Length of upper molar-premolar Series....:. s: cos. «rsaiike se=ceqtieneas sane 265 Lengthsof lower molar-premiolar ‘Series ...7.. J... dees eaveshbceee eee eee 268 Length-of ‘femur abOut.:. ...%..asavaiedantegudetvous faqeebs dedet oats eee eee 520 Proportion of height of skull at occiput to length of top of skull == 35. 20hoo. The Flint Creek beds, from which this rhinoceros was obtained, are closely related in time to the Deep River beds of the Smith River Valley in Montana, though Paleomeryx ? borealis is the only species that is now known to be common to the two formations. COMPARISONS OF APHELOPS MONTANUS WITH OTHER SPECIES OF RHINOCEROSES. Comparison with A. ceratorhinus. Aphelops montanus much resembles A. ceratorhinus in the size and proportions of the skull, in having a complete upper premolar series, and in having teeth which are not strongly hypsodont. Some of the differences are: the lack of horn-rugosities on the nasals of Aphelops montanus, the absence of an inward-turning scroll-like border on the posterior outer portions of the nasals, the different form of the poste- rior basal elements of the skull, and of the postero-superior portion. There are some differences in the proportions of the teeth, especially the molars, as is shown in the following measurements which are arranged for comparison : 7 a 3 Douc ass : RHINOCEROSES FROM NortH DAKOTA AND MONTANA 259 A, montanus. A. ceratorhinus. mm. mm, Paget OR Pc a vcis ba dat'veby cen teaaelseowantos es 27 26.5 PLE ihe fk wid teac o's dc'ap osencenimonewapeyence’s 20 22 Meme tle OF Fo. Seana eas sande ceentasnesereesssens 36 35 ANGLO IICH Gh 9S ca ptieaTacins rays csnonedee@awcemiemeayee 40 43 Det rOt Po. ac cccec es. 6. cee cseeenn ane sadawensos 41 42 EN Oo oo 5 vintine daisooins Sea aaiae dean cnen oe 56 BSS GE OE PS. 3, ciscst oan ceceacnasannes varaaaie annten 40 46 petite aso or oan Jonuswsand eonwe@anaemat ose 58 58 Length of Mi... 6 5csc0sccscsasnavearensenensons a 50 DiC CIE Ch Serie oe dees occa ne ynceinietnle sea nmeemaer 63 55 rem Ot OF WT 65). faciecesns-cseaeosnineadeateseelvsane 60 52 nea SE a ot aia sian v sasneneng ta memaeeasher 62 58.5 Die ety OE MOS cies ss. ccse cna ctenesies -on ven amare 44 45 Widtiiok § ..5....: Licadsvecaaibuduetma enna mates 56 53 Comparison with Aphelops Megalodis. A. montanus agrees in having the smooth elongated nasals, which are devoid of horn-rugosities, and in the comparatively brachyodont teeth, but it is a more dolichocephalic type, has a lower occiput, more slender zygomatic arches, and a broader roof to the brain-case. Our knowledge of Miocene rhinosceroses is still too incomplete to allow us to arrange them all under their proper generic names ; but as A. megalodus is the type of Aphelops, the present series which resembles it in so many particulars, should, for the present, be assigned to that genus, though the resemblance may not be due to a very close relationship. This species, but for the fact that it has no horn-rugosities on the nasals, would, with Aphelops ceratorhinus, come under Professor Os- born’s definition of the third phylum of Miocene rhinoceroses. This phylum, according to Professor Osborn," is distinguished by decidedly long limbs and feet, long skull, brachyodont teeth, and flattened pointed nasals with small terminal horn-rugosities. It is barely possible that the type of Aphelops montanus is a female and the male possessed horns, though I do not think it probable. 1<« New Miocene Rhinosceroses,’? Bulletin of the American Museum of Natural Fiistory, Vol. XX, 1904, p. 324. 260 ANNALS OF THE CARNEGIE MUSEUM. Aphelops ceratorhinus Douglass.’ (PLATE LAITY.) The type of Aphelops ceratorhinus is No. 857, Carnegie Museum | Catalogue of Vertebrate Fossils. This specimen was not all acces- sible or fully prepared for study when the first description was made. Fragments of the skull were originally found in a cattle-path on a steep bluff on the east side of the Lower Madison Valley in Montana. By digging into the sand, portions of a much broken skull and a nearly complete mandible were found. A good portion of one side of the skull has been put together, but the upper posterior portion is still wanting. Fragments of vertebree and limb-bones were found weathered out just below where the skull and mandible were obtained. From the same beds as the type are other portions of the skulls, which, judging by the forms of the nasals and the basal portions of the skulls belong to the same or nearly related species. These enable us to get the approximate proportions of the missing parts of the cranium of Aphelops ceratorhinus and to make the restoration of the skull which is given in Plate LXIV. Principal Distinguishing Characters. — Size large ; skull long (dolichocephalic); nasals long and slender, with small terminal horn-rugosities ; exter- nal auricular opening closed below ; post-tympanic expanding outward in wing-like processes ; parocci- pitals concave behind and in front, separated from occipital condyles by a concave area; cheek teeth brachyodont with cingula on interior portions of premolars ; P4 to M3 with crotchets ; mandible long and comparatively shallow, but thick and heavy; ascending ramus only moderately high, angle rounded, alveoli for canines large. Judging from the posterior portion of the skull of another individual (No. 854, see Fig. 3), which is smaller, the occiput was low and nar- rowed upward. ‘The nasals turned downward slightly at the tips just anterior to the nasal rugosities. The borders of the nasals, beginning at the rugosities, expand and have a decidedly downward trend for more than one half of the distance backward. On the posterior half Fic. 2. Calca- neum of type of Aphelops cerator- hinus. i, 1 «New Vertebrates from the Montana Tertiary,’’ Annals of the Carnegie Mu- seum, Vol. II, No. 2, 1903, p. 195. Plate LXIV. IV. ANNALS CARNEGIE MUSEUM, Vol. Aphelops ceratorhinus Douglass. DouGLass : RHINOCEROSES FROM NortH DAKOTA AND MonrTana 261 the borders are folded or turned inward apparently for the purpose of strengthening the nasals for the support of the anterior weapon of de- fence. The anterior portion of the frontal plane is slightly convex transversely, but nearly flat. Fic. 3. Outline of back of skull of Aphelops ceratorhinus? {- (No. 854.) The paroccipital process is different from that of any other species which I have observed. It is situated at some distance anterior to the occipital condyles and there is a quite large concave space between the yk * Z ey r SS SPQ We =~ WS 3 Fic. 4. Basi-occipital view of skull of Aphelops ceratorhinus? 4. (No. 854.) two. The process is concave on the anterior and posterior surfaces. The anterior surface is trough-shaped, but the posterior has a large depression at the base of the process, and farther down a concavity 262 ANNALS OF THE CARNEGIE MUSEUM. which is smaller and deeper. The post-tympanic is very different from that of Aphelops montanus. It is not rounded antero-posteriorly, but is wing-like, expanding antero-externally. It is quite thick antero- ~ posteriorly. : Measurements will be found in the original description of 4. ceratorhinus. Aphelops ceratorhinus ? Carnegie Museum Catalogue of Vertebrate Fossils, No. 854. This specimen consists of the posterior portion of a skull. The posterior basal portions of the crania are preserved both in this speci- men and in the type of Aphelops ceratorhinus and the corresponding parts agree in all essential particulars, though, as would be expected, there are some minor differences. The occiput is moderately high for the width and narrows upward as in Aphelops megalodus. ‘The supraorbital ridges unite for a short distance forming one low narrow a ridge which represents the sagittal crest. The paroccipital processes and the post-tympanics are essentially like those of the type of the species. The former have deep posterior concavities in the bases, and a prominent ridge extends, external to the cavities, from the posterior portion of the processes to the base of the occipital condyles. The anterior faces of the paroccipital processes are trough-shaped. The post-tympanics are not so large and rugose as in the type. The external auditory opening is long, comma- Figg, Latetel San cee shaped, and below this the post-tym- skull of Aphelops ceratorhinus? 1. panic and post-glenoid processes are (No. 854.) nearly in contact for some distance. The median basal portion of the skull has a nearly flat area between the occipital condyles and anterior to them. Anterior to this is a high, narrow, median ridge in front of which there is a rounded knob, or protuberance, just posterior to the pterygoid fossa and between the glenoid articular surfaces. SS ee ee Ae ws q DoucG.Lass : RHINOCEROSES FROM NORTH DAKOTA AND MONTANA 263 MEASUREMENTS. mm. PPCIEMT OL OCCIPUT, jj dnsinnjem sins imabe consesia see pdaviranhsledeay (assistive aes sestten sae seoes 225 POAC OM OCCHPUL 25 caa-th oo. ee rwaeen epsmed cates culanesevesamaniasose leds wiec arden due sien da 270 PCM OL PALGeCIpIEAN PEOCESSES...Siasy.nvyede 4iedes tubs 50 aeensted (eh rarasdedaveorsoeds 65 Though the type of Aphelops ceratorhinus represents a large rhino- ceros, portions of a skull, vertebrze, and limb-bones of another indi- vidual (Carnegie Museum Catalogue of Vertebrate Fossils, No. 842, from the Madison Valley) indicate a very much larger animal, possibly belong- ing toa different species. The greater portion of the humerus is preserved and an outline is given in Fig. 6. The measurements are as follows: | Length of humerus from upper articular oo Suriace tO Gishal Gd o.cccdecceoncaeseee es 487 Whole length of humerus, estimated... 500 Width of distal end, transverse........... 178 Diameter of distal end antero-posterior. 124 Thickness of shaft above distal trochlea 95 Teleoceras ? sp.? Carnegie Museum Catalogue of Ver- tebraté Fossils, No. 840. This specimen is the top of a skull with the nasals and occipital crest complete. It came from the bluffs on the east side of the Lower Madison Valley a mile or two farther north than the type of Aphelops ceratorhinus. Like most of the remains of rhin- oceroses from this region, the speci- men was found in a bed of sand near the bottom of the Upper Miocene (Loup Fork) beds. ‘The portion of the skull which is preserved suggests a somewhat different rhinoceros from yy, 6 Humerus of Aphelupst aus any other that has been named, but 1, (No. 842.) there is not sufficient material on which to found a species. The following are some of the more noticeable characteristics : 264 ANNALS OF THE CARNEGIE MUSEUM. Size medium ; top of skull flattened ; occiput somewhat elevated ; sagittal crest wanting ; nasals rather short, turned upward in front, and possessing rugosities for a terminal horn. This specimen is much smaller than the types of Aphelops ceratorhinus and 4. montanus. The nasals are short, laterally compressed, and turned upward toward the points, which are roughened, evidently for the accom- modation of one terminal horn. ‘They are very convex transversely on top and correspondingly concave beneath. They are not codssified. From the tips backward the borders trend upward for a short distance, then downward and outward and then curve backward. ‘They are —————$—$S= ————_———, EE a a Fic. 7. Lateral and superior views of fragment of skull of Ze/eoceras ? sp.? +. (No. 840. ) thickened posteriorly yet the inner edges are sharp and there is a beveled border facing inward and downward. The frontal region is flattened, but somewhat concave antero-posteriorly as the supra- occipital border is elevated. The cavities above and antero-internal to the orbits are large. There is no sagittal crest. The supra-tem- poral ridges are low and broad. A large concavity with one low median ridge and two lateral ridges occupies the posterior upper por- tion of the occiput. The lateral ridges end at the supero-external portion of the occipital crest. re ee 5 %, } Pa ee ee ee ee ee eee See DouGLAss : RHINOCEROSES FROM NorRTH DAKOTA AND MONTANA 265 MEASUREMENTS. mm, Length of skull from tips of nasals to middle of occipital crest............ 457 PeEnMO OOCKDM aN CHEE. \ .Lesnavspasnsetisanaande sac Wee daa pda pwnd sedges oss ak’ 165 Width of skull at supero-anterior borders of orbits. ...............06 seeseeee 218 Length of free nasals........ sinli's Auch EDV s Negi gtiaa ad pak cla bey oaks A saaie tan tiene 122 Pra a On ined, DASAalS, WOStELION .....,-cuctvensnacdadswnatwsantantoatsededandat~csedes 83, Seen WO EMCOSIES: 100 MOLT, . «ads uesnanenvenivetaeacill ibedsWesnda tides neonenere 35 enouy Ok cngosities for DOL. ....29: <0 s-de sad ahits dows seieges osiaed soiiedele cvecesy ass 35 ACERATHERIUM. In August, and again in November, 1905, the writer obtained, at White Butte and in the Little Bad Lands in North Dakota, many remains of rhinoceroses. Among them are two exceptionally complete skulls, one of which has the mandible attached. Though there are some differences in details in the two skulls yet I place them both, provisionally, in the species Aceratherium tridactylum. Aceratherium tridactylum Osborn. (Carnegie Museum Catalogue of Vertebrate Fossils, No. 1585.) This specimen consists of a skull and mandible which are nearly complete. Small portions of the left angle of the mandible and of the occipital crest had weathered away. It was found in a hard, heavy, green sandstone concretion, or part of a sandstone stratum, just above the nodular Oreodon layer. A stratum of sandstone a little distance away contained many bones and teeth of rhinoceroses and some teeth of horses. Above was a gray, rather soft sandy stratum about fifteen feet in thickness containing the jaws of rodents and remains of crocodiles. The specimen differs somewhat from the specimen figured and de- scribed by Osborn in his ‘‘ Memoir on The Extinct Rhinoceroses.’’ * The nasals are shortened and truncate, not narrowing to a point, or suddenly contracting forward at a point a little distance posterior to the apex, as in Osborn’s figure. There are two incisive alveoli in the premaxillaries, the posterior is nearly as large as the anterior alveolus, and the two are very close to each other. ‘The paroccipital processes are prismatic, having three sides. The mandibular symphysis is quite long and the canine fairly large. ‘The most striking peculiarity of this specimen is its shortened truncate nasals. 1 Memoirs of the American Museum of Natural History, Vol. I, Part III, April, 1898, p. 158, Pl. XVII. ) 266 ANNALS OF THE CARNEGIE MUSEUM. MEASUREMENTS. mm Length of skull from premaxillaries to occipital condyles.................. 512 Length. of skull from -nasals:to eeciput:...).... Jenna cadeesaoeeaseeeeeaane eee 477 Length of free nasals(,. canis V2.0 eit ee Cece rae eee ee 80 Height. of skull at.orbits not including tethy, ..2.2c.ccsace~ paves teens eee 140 Height of skull at obcipul in. 0.b ses rosa dey taeeenbaee ee eee ee ee ae 175 Liength of upper series of cheek teeth... ..i.c * a n "ies : . ; >. a y ‘ wird adel “ait ; « S » ee at ¢ 6 “ ~ ? : t . Mii Ss a Ps bes ; « : 7 = ’ i aoe , ‘ { => Ria Tad y bo ‘ 1 a | a.) hs : ‘ a * eo Le fa) re ye fe . y “Ir Ga 4 ; ANNALS CARNEGIE MUSEUM, Vol. IV. Plate LXVII. Miocene Horses from Montana. (*ozIs [VaAN}VU J[VY-duUO sainsy 19Yy}O [[v ‘ azis [einjeu LZ puvg‘sSiz) “vuRpUOT|W WOT SdSIOT] dUdDOITN = Mh mnihy < \Wie 1 eek Po ASS ATAX T° 88 le } “AL “ISA (WN3SAW SISANYVO STVNNY Dovuctass: Fosstt Horses. 275 Remet of symiplaysis: Of; MANIC, ...c2-cerayensaccawetednrwiersoragizcres ess 49 Length of femur exclusive of proximal symphySis...............ss.0eeeeees 233 rere FCT RAISE 6 ous cialonaiadiuscowmn aman ens pte eee ite Sslensitknn esr naive tn Re 218 ieeat antero-posterior diameter, of femiUP.........c.iensecsssexevecescos acs ves 30 Beenie Ob Medial MetAPOCIAl gc; ssebasnenadserardenshowscbhauasescd vars aidie 194 Transverse width of shaft of median metapodial.................0cseeeeees 17.5 Peer ECOL NMA PNAAMX .... .. Length: of ‘centrum ys. sv duaceineceebesntiedaeadaransee eee arene 72 Fifth Cervical. Transverse diameter of centrum, anteriorly............... 16 Fifth Cervical. Vertical diameter of centrum, anteriorly................... 15 Fifth Cervical. Vertical diameter of centrum, including keel, posteriorly. 23 Sixth Cervical. Greatest: length’ ©... ils.¢ stash past poeta tes eth ee 67 Sixth Cervical. Transverse diameter of centrum, anteriorly............... 17 Sixth Cervical: Vertical diameter of centrum, anieniorly..2>.., 205 ere 17 peventh Cervical. Soreatest Leme tlh n:sc.