X} a UJ a tH cr o~ -IT UJ DEPARTMENT OF THE INTERIOR U 8. GEOLOGICAL AND GEOGRAPHICAL SURVEY OF THE TERRITORIES F. V. HAYDEN, GEOLOGIST-DJ-CHABQE MISCELLANEOUS PUBLICATIONS— No. 12 HISTORY OF NORTH ^ ME RIO A. 1ST PINNIPEDS A MONOGRAPH OF THE WALRUSES, SEA-LIONS, SEA-BEARS AND SEALS OF NORTH AMERICA. BY JOEL ASAPH ALLEN Assistant in the Museum of Comparative Zoology at Cambridge Special Collaborator of the Survey WASHINGTON GOVERNMENT FEINTING OFFICE 1880 PREFATORY NOTE. UNITED STATES GEOLOGICAL AND GEOGRAPHICAL SURVEY OF THE TERRITORIES, Washington, D. (7., July 1, 1880. The present series of monographs of the 'North American Pinnipedia, by Mr. J. A. Allen, may be considered as a second installment of the systematic History of North American Mam- mals, of which the Fur-Bearing- Animals by Dr. Elliott Cones, U. S. A., forming No. 8 of the Miscellaneous Publications of the Survey, was published as a specimen fasciculus. The first monograph of this series, treating of the Walruses, was prepared nearly three years since for publication in the Bulletin of the Survey, but before it was quite ready for the press, Dr. Coues, owing to his pressing engagements in other directions, invited Mr. Allen to extend his treatise to embrace the entire suborder of the Pinnipeds, to which he had already given special atten- tion, with a view to its incorporation as a part of the proposed general History of North American Mammals. Since, how- ever, considerable time must elapse before the whole work can be completed, it has been thought best not to delay the publi- cation of the part already prepared relating to the Pennipeds. As nearly all of the species belonging to this group found in the northern hemisphere are members of the North American fauna, the present treatise is virtually a monograph of all the species occurring north of the equator, and includes incideutally a revision of those of other seas. The literature of the whole group is not only reviewed at length, but the economic pha.se of the subject is treated in detail, embracing, in fact, a general history of the Sealing industries of the world. The technical treatment of the subject is based rnainjy on the rich material of the National Museum, supplemented at many important points by that of the Museum of Comparative Zoology of Cam- bridge, which, through the kindness of the directors of these institutions, was generously placed at the author's disposal. That contained in the other principal museums of the country in IV PREFATORY NOTE. was also examined, so that so far as the species of the northern hemisphere is concerned the amount of material consulted doubtless far exceeds that ever before studied by any single investigator of the group. For the biographical part, to which much space has been allotted, matter has been freely gathered from all available sources. In addition to the results here first published, the work may be considered as a compendium of our present knowledge of the subject. In regard to the need of a work like the present, it may be stated that with the exception of Dr. Theodore Gill's important "Prodrome" of a proposed monograph of North American Pinnipeds, published in 1866, there has been no general treat- ment of the species since the excellent compilations of Drs. Harlan and Godman appeared, now more than half a century ago. Respecting foreign works, nothing has been recently published covering the ground here taken beyond a very gen- eral synopsis of the technical phases of the subject. The best accounts of the species occurring along the shores of Europe are in other languages than English, while no general history of the economic relations of the subject exists. In relation to the important Fur Seal Fisheries of Alaska, the author has been able to present in extenso the results of Captain Charles Bryant's long experience at the Fur Seal Islands, where for nearly ten years he was the government agent in charge of the islands. Although not received until the article on this species had been transmitted to the printer, it proves to be, to only a small degree, a repetition of the account given by Mr. Elliott, also reproduced at length. The history Captain Bryant gives of the changes in the numbers and relations of the different classes of these animals at the rookeries, under the present system of management of the Fur Seal business, forms a valu- able basis for generalization in regard to the future regulation of this industry, and is also an important contribution to the life-history of the species. The cuts, some thirty in number, illustrating the cranial char- acters of the Walruses, were drawn for the present work by Mr. J. H. Blake, of Cambridge, and engraved by Messrs. Eussell and Richardson, of Boston. The Survey is indebted to Professor Baird, Secretary of the Smithsonian Institution, for a series of six- teen original figures, engraved by Mr. H. H. Nichols, of Washing- ton, from photographs on wood, illustrating the sknlls of CallorU- nus ursinus, Peale's "Halwhcerwcmtarcticus," Cystopliora cristata, PREFATORY NOTE. V and Macrorhinus angustirostris ; also to the Zoological Society of London for electros of Gray's "Halicyon richardsi," and of a series of historic figures of the walrus published in the Society's "Proceedings," by the late Dr. Gray, and to the proprietors of " Science Gossip," for electros of the full-length figures of seals. These were received through Dr. Coues, who also furnished the full-length views of Eumetopias stelleri and CaUorhinus ursinus. Mr. Allen desires me to express, in this connection, acknowl- edgments of his indebtedness to Prof. Spencer F. Baird, Secre- tary of the Smithsonian Institution, and to Prof. Alexander Agassiz, Director of the Museum of Comparative Zoology, for the liberality with which they have placed at his service the rich material relating to this group of animals contained in the museums respectively under their charge ; to Prof. Henry A. Ward, of Rochester, N. Y., for the use of much valuable material relating to the Walruses that he would not otherwise have seen ; and to Captain Charles Bryant, late special agent of the United States Treasury Department, for his report, kindly prepared at the author's request, for the present work. Also to Dr. Elliott Cones, Secretary of the Survey, for the use of many of the cuts, for valuable suggestions during the prepa- tion and printing of the monograph, and revision of the proof- sheets. F. V. HAYDEN, United States Geologist. LETTER OF TRANSMITTAL. CAMBRIDGE, MASS., May 25, 1880. SIR : I have the honor to transmit herewith for approval and for publication the " History of North American Pinnipeds," being a monograph of the Walruses, Sea-Lions, Sea-Bears, and Seals of North America. Very respectfully, your obedient servant, J. A. ALLEN. F. V. HAYDEN, Washington, D. C. vn TABLE OF CONTENTS. Page. Title ............................................................................ I Prefatory note ................................................................... HI Letter of transmittal ............................................................. V Table of contents ............................................................ VH List of illustrations ................... ................... Characters of the PINNIPEDI^ Family ODOB^NID^— Walruses Synonymy ................... ................... General observations and characters of the group Genera Synopsis of the genera Genus OcoBjENUs Synonymy and history ............ •. Species ... ODOB^NUS KOSMARUS — Atlantic "Walrus Synonymy and bibliographical references External characters Sexual differences Individual variations and variations dependent upon age Measurements of skulls Dentition Fossil remains ____ ................................. ........... . Geographical distribution, present and past Ooast of North America Coast of Europe Nomenclature Etymology Literature General history Figures Habits and the chase Products Food Functions of the tusks ......................................... 137-138 Enemies ......................................................... 138-139 Domestication ................................................... 140-147 ODOB.ENUS OBESUS — Pacific "Walrus ................................. 147-186 Synonymy and bibliographical references ....................... 147 External characters and skeleton ................................ 147-155 Measurements of skeleton .................................. 14iM5U Measurements of skulls ..................................... 155 Differential characters ........... . ............................... 156-170 Nomenclature ................................................... 170-171 General history .................................................. 171-172 Figures .......................................................... 172-174 Geographical distribution ........................................ 174-178 Habits, food, commercial products, and the chase ............... 178-186 Family OTARIID^— Eared Seals .......................................... 187-411 Synonymy and characters of the group ................................. Technical history ........................................................ 188-207 Higher groups ....................................................... 188-190 Genera .............................................................. 190-193 Species .............................................................. 193-207 IX 1 fr-186 5 5-12 12-14 14 14 14-17 17-23 23-147 23-26 26-38 38-43 43-45 46 47-57 57-65 65-79 65-71 71-79 80 80-82 82-107 82-92 92-107 107-133 133-134 134-137 X TABLE OF CONTENTS. Page. Characters of the PINNIBEDIA — Continued. Family^OTAmrD^— Eared Seals. Synopsis of the genera and species 208-213 Mythical and undeterminable species 214-216 Geographical distribution 216 Fossil Otaries 217-221 Milk dentition 221-224 Irregularities of dentition 224 Position of the last upper permanent molar 225 General observations 225-227 Habits 227 Products 228 Destruction of Fur Seals for their peltries : 229-231 Genus EUMETOPIAS 231 EUMETOPIAS STELLERI— Steller's Sea-Lion 232-274 Synonymy and bibliographical references 232 External characters 232-236 External measurements 236 Skull 237-238 Measurements of skulls 238 Teeth 239 Skeleton 240-244 Measurements of skeleton 242-244 Sexual, adolescent, and individual variation 244 Geographical variation 244 Comparison with allied species 244-247 Measurements of skulls of OTARIA JUBATA 247 Geographical distribution 248 General history and nomenclature 248-254 Habits ' 254-274 Genus ZALOPHUS 275 ZALOPHUS CALIFOHXIANTS — Califomian Sea- Lion 276-312 Synonymy and bibliographical references 27C External cliura.-trrs 276-278 Young 278 Pelage 278 Size 278-283 External measurements 279-280 Measurements of skeleton of female 281-283 Skull 283-285 Measurements of skulls 285 Dentition ... 286 Sexual differences 287 Variation with age 287-289 Comparison with allied species 289 Geographical distribution 289-291 General history and nomenclature 291-296 Habits 296-312 Genus C ALLORHINUS • 312-410 CALLORHINUS URSINUS — Northern Fur Seal 313 Synonymy and bibliographical references 313-314 External characters 314 Color 314 Pelage 315 Size 316-319 External measurements 319 Ears 320 Fore limbs 320 Hind limbs.. 320 TABLE OF CONTENTS. XI 1 * ' t ' rt • Characters of the PINMPEDIA — Continued. Family OTAKIID2E— Eared Seals. Genus CALLOKHLXUS. CALLOBHINUS URSINUS— Northern Fur Seal. Skull 320-323 Measurements of skulls 323 Teeth 324 Skeleton 324-326 Measurements of skeleton 325 Sexual differences 325-327 Differences resulting from age 327 Individual variation 328 Comparison with allied species 329-331 Measurements of skulls of ARCTOCEPHALUS AUSTRALIS 331 Geographical distribution and migration 332-335 General history and nomenclature 335-339 Figures 339-341 Habits 341-371 The chase • 371-U7S Mode of capture 372-378 History and prospects of the Fur Seal business at the Prybilov Islands 378-381 Enemies of the Fur Seals 381 History of the Fur Seal Fishery at the Prybilov Islands, Alaska, from 1869 to 1877, by CHAKLES BRYANT 382-411 Preliminary ami general observations 382-388 Recent changes in the habits and relative numbers of the different classes of Seals 388-398 Cause of the changes in the habits of the Seals, &c 398-102 Albinos and sexually abnormal individuals 403 Description of the young ; variation in color with age, &c. . . . 403 Molting 404 Sexual organs, &c 405 Power of suspending respiration 406 Natural enemies 406 Effect of climatic influences 407 Number of Seals required for the subsistence of the natives . . 409 Winter resorts and habits of the Seals 410 Family PHOCID^— Earless Seals 412-756 Characters of the group 412 Technical history 412-460 Higher groups 412—114 Genera 414-421 Species 421-460 Classification 460-467 Synopsis of sub-families and genera 461-463 Synonymatiolist of the species 463-467 Geographical distribution 467—469 Fossil remains 469-481 North America 469-476 Europe 476-481 Milk dentition 481-484 General habits and instincts 484-486 Food 486 Enemies 487 Migrations 487-491 Locomotion on land 491^96 Seal hunting 496-546 XII TABLE OF CONTENTS. Page. Characters of the PINNIPEDIA — Continued. Family PHOCID^— Earless Seals. Sealing districts 496-522 "West Greenland 497 Newfoundland 497-199 Jan Mayen or "Greenland" Seas 499-511 Nova Zembla and Kara Sea 511 White Sea 511-513 Caspian Sea 513-517 North Pacific 517 South Pacific and Antarctic Seas 517-522 Methods of capture, &c 522 Shore hunting 522-530 . Esquimaux methods 522 By means of nets 523-528 The seal-hox 528 The seal-hook 529 . The " Skrackta" 529 Ice hunting 530 In the Gulf of Bothnia 530-534 Off the coast of Newfoundland 534-540 In the Jan Mayen Seas 540-542 Dangers and uncertainties of ice hunting 543-545 Species hunted 545 Abundance of Seals at particular localities 546 Products .... 546-549 Preparation of the products 549-551 "Wasteful destruction of Seals 551-553 Decrease from injudicious hunting 553 Seals and Seal hunting in the olden time in the Gulf of St. Lawrence. 553-557 Sab-family PHOCINJE 557 Genus PHOCA . . 557-559 PHOCA VITULINA— Harbor Seal 559-507 Synonymy and bibliographical references 559-562 External characters . 562-565 Distinctive characters 565-571 Individual and sexual variation 571-573 Measurements of the skulls 574 General history and nomenclature 575-584 Geographical distribution 584-588 Habits 588-597 PHOCA (PU6A) FCETIDA— Ringed Seal 597-629 Synonymy and bibliographical references 597-600 External characters . . 600-603 Individual variation and variations dependent upon sex and age. 603-605 Measurements of the skull 606 Differential characters 607-614 Phoca (Pusa) caspica 609-610 Phoca (Pusa) sibirica 612-613 Geographical distribution 614-616 General history and nomenclature 616-619 Habits 619-629 PHOCA (PAGOPHILUS) GHCENLANDICA — Harp Seal 630-654 Synonymy and bibliographical references 630-632 External characters 632-637 Sexual and individual variation and variations dependent upon age 637 Measurements of the skull 638 General history and nomenclature 639 TABLE OF CONTENTS. XIII • Page. Characters of the PINNIPEUIA — Continued. Family PHOCID^. Sub-family PHOCIN.ZE. Genus PHOCA. PHOCA (PAGOPHILUS) GRCENLANDICA — Harp SeaL Geographical distribution 640 Migrations and breeding stations 641-647 Habits 647-051 Enemies 651 Food 652 Hunting and products 652-654 Genus ERIGNATHUS 654 ERIGNATHUS BAKBATUS— Bearded Seal 655-675 Synonymy and bibliographical references 655-657 External characters , 657 Skull and skeleton 658-661 Measurements of the skeleton 660 Measurements of the skull 661 General history and nomenclature 662-666 Geographical distribution 666-670 Habits, products, and hunting 670-675 Genus HISTRIOP.HOCA 675-676 HISTRIOPHOCA FASCIATA— Bibbon Seal 676-682 Synonymy and bibliographical'references 676 External characters 670-678 Size 678 General history 678-681 Geographical distribution 681-682 Habits 682 Genus HAUICHCERUS 682-689 General history and discussion of the ' ' Genus PUSA " of Scopoli . . 683-689 HALICHCEEUS GRYPUS— Gray Seal 689-706 Synonymy and bibliographical references 689-690 External characters 690-693 Measurements of skulls 694 Geographical distribution 695-696 General history and nomenclature 696-698 Habits 699-706 Genus MONACHUS 707-708 MONACHUS TROPICALIS — "West Indian Seal 708-723 Characters 708 Dampier's account 708-710 Hill's and Gosse's accounts, 1843,1851 710-715 Gray's accounts, 1849, 1874 175-718 Gill on the "West Indian Seals, 1866 718 Analysis and discussion of the foregoing 718-720 Affinities of the Jamaican or Pedro Seal 720-721 Geographical distribution 721-723 Sub-family CYSTOPHOKHIN2E 723 Genus CYSTOPHORA 723 CYSTOPHORA CRISTATA— Hooded Seal 724-742 Synonymy and bibliographical references 724-726 External characters 726-729 Skeleton and skull 730-733 Measurements of skulls 732 Measurements of the skeleton 733 Geographical distribution and migrations 733-737 General history arid nomenclature 738-740 Habits 740-741 Hunting and products . . : 741-742 XIV TABLE OF CONTENTS. ^ Page. Characters of the PINNIPEDIA — Continued. Family PHOCID^. Sub-family— CYSTOPHORHIN.E. Genus MACRORHINUS MACRORHINUS ANGUSTIROSTRIS — Californian Sea Elephant 743 Bibliographical references 743 External characters 743-746 Skull 746-749 Measurements of skulls 748 Measurements of the skeleton of Macrorhinus leoninus 749 Comparison with the Southern Sea Elephant 749-751 Geographical distribution 751-752 General history 752-753 Habits 753-755 Chase and products 755-756 APPENDIX. A. Material examined 757-764 Family ODOB^ENID^; 757-758 Odobaenus rosmaraa 758 Odobsenns obesus Family OTARIIU-S: 758-760 Eumetopias stelleri 758 Zalophus californianus Callorhinus nrsinus 760 Family PHOCID.E 761-764 Phoca vitulina 761 Phoca fcetida 762 Phoca groenlandica Erignathus barbatus 763 Histriophoca fasciata Halichoerus grypus . . . Cystophora cristata Macrorhinus angustirostris ' B. Additions and Corrections 765-774 Family ODOB.ENID.E 765-769 Odobcenus rosmarus — Atlantic Walrus 765-768 Additional references 765 Size a iid external appearance Geographical distribution 766-767 Nova Zembla 766 Franz-Josef Land 766 Abundance in Wolstenholme Sound 766 Spitzbergen, &c 766 Iceland 766 Supposed presence of "Walruses in the Antarctic Seas 766 The "Walrus a formidable antagonist Curiosity and fearlessness of the "Walrus 767 Locomotion ; use of tlie tusks in climbing 767 Figures of the "Walrus 768 Odobcenus obesus — Pacific "Walrua 768 Distribution Family OTARIID.E 769-774 Otaries at the Galapagos Islands Fossil Otaries 770 Capture of Sea Lions for menageries 770 Zalophus californianus— California Sea Lion 771 Period of gestation Callorhinus ursinus 772 Breeding off the coast of "Washington Territory Family PHOCID^: Extinct species • LIST OF ILLUSTRATIONS. Fig. 1, p. 41. Odobcenus rosmarus, skull of female in profile and lower jaw from above. Fig. 2, p. 42. Odobacenus rosmarus, skull of female from above. Fig. 3, p. 43. Odobcenus rosmarus, skull of female from below. Fig. 4, p. 93. Olaus Magnus's " Rosinarus seu Moreus Norvegicus." Fig. 5, p. 93. Olaus Magnus's "Porous Monstrosus Oceani Germanic!." Fig. 6, p. 94. Gesner's "Rosmarus." Fig. 7, p. 94. Gesner's "Vacca marina" (Addenda to Icones Animal). Fig. 8, p. 94. Gesner's "Rosmarus" (Icones Animal., 1560). Fig. 9, p. 95. De Veer's "Sea Horse," 1609. Fig. 10, p. 96. Hessel Gerard's ""Walruss," 1613. Fig. 11, p. 100. Martin's " "Wall-ross, " 1765. Fig. 12, p. 101. Buffon's " Le Morse," 1765. Fig. 13, p. 153. Odobcenus obesus, three views of head. Fig. 14, p. 156. Odobacenus obesus, skull in profile. Fig. 15, p. 157. Odobcenus rosmarus, skull in profile. Fig. 10, p. 158. Odobcenus rosmarus, skull from front. Fig. 17, p. 158. Odobcenus obesus, skull from front. Fig. 18, p. 159. Odobcenus rosmarus, occipital view of skull. Fig. 19, p. 159. Odobcenus obesus, occipital view of skull. Fig. 20, p. 160. Odabcenus rosmarus, skull from above. Fig. 21, p. 161. Odobcenus obesus, skull from above. Fig. 22, p. 162. Odobcenus obesus, young skull from above. Fig. 23, p. 162. Odobcenus rosmarus, young skull from above. Fig. 24, p. 163. Odobcenus obesu-s, young skull from front. Fig. 25, p. 163. Odobcenus rosmarus, young skull from front. J£ ig. 26, p. 164. Odobcenus rosmarus, skull from below. Fig. 27, p. 165. Odobcenus ubesus, skull from below. Fig. 28, p. 166. Odobcenus rosmarus, lower jaw from above. Fig. 29, p. 166. Odobcenus obesus, lower jaw from above. Fig. 30, p. 167. Odobcenus rosmarus, lower jaw from side. Fig. 31, p. 167. Odobcenus obesus, lower jaw from side. Fig. 32, p. 168. Odobcenus rosmarus, lower jaw of young from above. Fig. 33, p. 168. Odobcenus obesus, lower jaw of young from above. Fig. 34, p. 169. Odobcenus rosmarus, lower jaw of young from side. Fig. 35, p. 169. Odobcenus obesus, lower jaw of young from side. Fig. 36, p. 173. Odobcenus obesus, Cook's figure of the animal. Fig. 37, p. 259. Eumetopias stelleri, figures of animal Fig. 38, p. 317. Callorhinus ursinus, figures of animal. * Fig. 39, p. 321. Callorhinus ursinus, skull of female in profile. Fig. 40, p. 321. Callorhinus ursinus, skull of female from above. Fig. 41, p 322. Callorhinus ursinus, skull of female lower jaw. Fig. 42. p. 322. Callorhinus ursinus, skull of female from below. Fig. 43, p. 563. Phoca vitulina, animal. Fig. 44, p. 580. " Halichcerus antircticus," Peale, skull in profile. Fig. 45, p. 580. " Halichcerus antarcticus," Peale, skull from above. Fig. 46, p. 581. " Halichcerus antarcticus, " Peale, skull from below. Fig. 47, p. 582. "Halichcerus antarcticus," Peale, lower jaw. Fig. 48, p. 583. " Halicyon richardsi," Gray, skull in profile. Fig. 49, p. 601. Phoca fcetida, animal. XV XVI LIST OF ILLUSTRATIONS. Fig. 50, p. 633. Phoca groelandica, animal. Fig. 51, p. 691. Halichcerns grypus, animal. Fig. 52, p. 727. Oystophora cristata, animal. Fig. 53, p. 728. Cystophora cristata, skull in profile. Fig. 54, p. 729. Oystophora cristata, skull from above. Fig. 55, p. 730. Oystophora cristata, skull from below. Fig. 56, p. 731. Oystophora cristata, lower jaw. Fig. 57, p. 744. Macrorhinus angustirostris, skull in profile. Fig. 58, p. 745. Macrorhinus angustirostris, skull from above. Fig. 59, p. 746. Macrorhinus angustirostris, skull from below. Fig. 60, p. 747. Macrorhinus angustirostris, lower jaw. HISTORY OF THE NORTH AMERICAN PINNIPEDS. The Pinnipeds, or Pitmipedia, embracing the Seals and Wal- ruses, are commonly recognized by recent systematic writers as constituting a suborder of the order Fercv, or Carnivorous Mammals. They are, in short, true Carnivora, modified for an aquatic existence, and have consequently been sometimes termed liAm/pMbiou8 Carnivora." Their whole form is modified for life in the water, which element is their true home. Here they display extreme activity, but on land their movements are confined and labored. They consequently rarely leave the water, and generally only for short periods, and are never found to move voluntarily more than a few yards from the shore. Like the other marine Mammalia, the Cetacea and Sirenia (Whales, Dolphins, Porpoises, Manatees, etc.), their bodies are more or less fish-bike in general form, and their limbs are transformed into swimming organs. As their name implies, they are fin- footed. Generally speaking, the body may be compared to two cones joined basally. Unlike the other marine mammals, the Pinnipeds are all well clothed with hair, while several of them have, underneath the exterior coarser hair, a thick, soft, silky under-fur. In contrasting them with the ordinary or terrestrial mammals, we note that the body is only exceptionally raised, and the limbs are confined within the common integument to beyond the knees and elbows, and are hence to only a slight degree serviceable for terrestrial locomotion. The first digit of the manus is generally lengthened and enlarged, as are both the outer digits of the pes. As compared with other Ferce, they present, in osteological characters, many obvious points of difference, especially in relation to the structure of the skull, limbs, and pelvis, and in dentition. The skull is distinctively characterized by great compression or constriction of the inter- orbital portion, the large size of the orbital fossa?, in the lachry- mal bone being imperforate (without a lachrymal canal) and contained within the orbit, and in the presence (generally) of Misc. Pub. No. 12 1 1 2 CHARACTERS OF PINNIPEDIA. considerable vacuities between the palatine and frontal bones and the tympanic and exoccipital bones. The deciduous den- tition is rudimentary, never to any great extent functional, and frequently does not persist beyond the fostal life of the animal. In the permanent dentition, the canines are greatly developed, sometimes enormously so; the lower incisors are never more than four in number, and sometimes only two ; the upper incisors usually number six, but sometimes only four, or even two ; the grinding teeth (premolars and molars) are generally simple in structure, and usually differ from each other merely in respect to size, or the number of roots by which they are inserted. The pelvis differs from that of the terrestrial Ferce in the shortness of the iliac portion and the eversion of its anterior border ; the ischiac bones barely meet for a short distance in the male, and are usually widely separated in the female, the pelvic arch thus remaining in the latter permanently open ventrally. The existing Pinnipeds constitute three very distinct minor groups or families, differing quite widely from each other in important characters : these are the "Walruses, or Odobcenidce, the Eared Seals, or Otariidce, and the Earless Seals, or Phocidce. The first two are far more nearly allied than are either of these with the third, so that the Odobccnidcc and Otariidw may be together contrasted with the Plwcidw. The last named is the lowest or most generalized group, while the others appear to stand on nearly the same plane, and about equally remote from the Pho- cidcv. The Walruses are really little more than thick, clumsy, obese forms of the Otariau type, with the canines enormously developed, and the whole skull correlatively modified. The limb-structure, the mode of life, and the whole economy are essentially the same in the two groups, and, aside from the cran- ial modifications presented by the Odobccn idw: which are obvi- ously related to the development of the canines as huge tusks, the Walruses are merely elephantine Otariids, the absence or presence of an external ear being in reality a feature of minor importance. The characters of the suborder and its three families may be more formally stated as follows : * * The characters here given are in part those collated by Dr. Theodore N. Gill in 1873 ("Arrangement of the Families of the Mammals." Smithsonian Miscellaneous Collections, No. 230, pp. 56, 68, 69), by whom the distinctive features of these groups were first formulated. They have, however, been carefully verified and further elaborated by the present writer, while the families are here quite differently associated. CHARACTERS OF PINNIPEDIA. D Limbs pinniform, or modified into swimming organs, and enclosed to or beyond the elbows and knees within the common integument. Digits of the mauus decreasing in length and size from the first to the fifth ; of those of the pes, the first and fifth largest and longest, the three middle ones shorter and subequal. Pelvis with the iliac portion very short, and the anterior border much everted; ischia barely meeting by a short symphysis (never anchylosed) and in the female usually widely separated. Skull generally greatly compressed interorbitally ; facial portion usually short and rather broad, and the brain-case abruptly ex- panded. Lachrymal bone imperforate and joined to the maxil- lary, enclosed wholly within the orbit. Palatines usually sepa- rated by a vacuity, often of considerable size, from the frontals. Tympanic bones separated also by a vacuity from the exoccipit- als. Dentition simple, generally unspecialized, the molars all similar in structure. Deciduous dentition rudimentary, never truly functional, and generally not persistent beyond the foetal stage of the animal. Permanent incisors usually £ or -J> some- times -f- (Cysiopliora and Macrorliinns), or even f (Odoba'ints'); canines 1 ; molars * £, £, or jj- PINNIPEDIA. A. Hind legs capable of being turned forward and used in terrestrial loco- motion. Neck lengthened (especially in family II). Skull with the mastoid processes large and salient (especially in the males), and with distinct alisphcnoid canals. Anterior feet nearly as large as the posterior, their digits rapidly decreasing in length from the first to the fifth, without distinct claws, and with a broad cartilaginous border extending beyond the digits. Hind feet suceptible of great expansion, the three middle digits only with claws, and all the digits terminating in long, narrow, car- tilaginous flaps, united basally. Femur with the trochanter minor well developed GRESSIGRADA. I. Without external ears. Form thick and heavy. Anterior por- tion of the skull greatly swollen, giving support to the enor- mously developed canines, which form long, protruding tusks. Incisors of deciduous (fo3tal) dentition § ; of permanent denti- tion §. No postorbital processes, and the surface of the mastoid processes continuous with the auditory bulls; Odobcenidce. II. With small external ears. Form slender and elongated. Ante- rior portion of the skull not unusually swollen, and the canines not highly specialized. Incisors of deciduous dentition £ , only the outer on either side cutting the gum ; of permanent denti- tion |, the two central pairs of the upper with a transverse groove. Postorbital processes strongly developed. Surface of the mastoid processes not continuous with the auditory bullai Otariidce. B. Hind legs not capable of being turned forward, and not serviceable for terrestrial locomotion. Neck short. Skull with the mastoid * In view of the uncertainty respecting the proper notation of the grind- ing teeth, they will in the present work be designated simply as molars, with no attempt at distinguishing "premolars" from "molars." 4 CHARACTERS OF PINNIPEDIA. processes swollen, but not salient, and without distinct alisphe- noid canals. Anterior limbs smaller than the posterior, the first digit little, if any, longer than the next succeeding ones, all armed with strong claws, which are terminal. Hind feet ca- pable of moderate expansion, short ; digits (usually) all armed with strong claws, and without terminal cartilaginous flaps. Femur with no trace of the trochanter minor.. .REPTIGRADA.* III. Without external ears. Postorbital processes wanting, or very small. Incisors variable (f, f , or f ). Deciduous dentition not persistent beyond foetal life .. PhotidcB. The Pinnipeds present a high degree of cerebral develop- ment, and are easily domesticated under favorable conditions. They manifest strong social and parental affection, and defend their young with great persistency and courage. They are car- nivorous (almost without exception), subsisting upon fishes, mollusks, and crustaceans, of which they consume enormous quantities. The Walruses and Eared Seals are polygamous, and the males greatly exceed the females in size. The ordinary or Earless Seals are commonly supposed to be monogamous, and there is generally little difference in the size of the sexes. The Walruses and Eared Seals usually resort in large numbers to certain favorite breeding grounds, and during the season of re- production leave the water, and pass a considerable period upon land. The Earless Seals, on the other hand, with the exception of the Sea Elephants, do not so uniformly resort to particular breeding grounds on land, and leave the water only for very short intervals. They usually bring forth their young on the ice, most of the species being confined to the colder latitudes. Only one of the various species of the Pinnipedia appears to be strictly tropical, and very few of them range into tropical waters. As a group, the Pinnipeds are distinctively character- istic of the Arctic, Antarctic, and Temperate portions of the globe, several of the genera being strictly Arctic or Subarctic in their distribution. The Walruses are at present confined mainly within the Arctic Circle, and have no representatives south of the colder portions of the Northern Hemisphere. The OtariidcB and PJiocidce, on the other hand, are abundantly represented on both sides of the equator, as will be noticed more in detail later. * For the suggestion of the terms Grcssigrada and Eeptigrada I am indebted to my friend Dr. Elliott Coues. FAMILY Walruses. " Trichetidce, GRAY, London Med. Repos., 1821, 303" (family). Apud Gray. Trichechidce, GRAT, Ann. of Philos., 1825, 340; Ann. and Mag. Nat. Hist., 3d ser., xviii, 1866, 229; Hid., 4th ser., iv, 1869, 268; Suppl. Cat. Seals and Whales, 1871, 5 (family). Trichecina, GRAY, London's Mag. Nat. Hist., i, 1837, 538; "Zool. Erebus and Terror, 3 " (subfamily). In part only, or exclusive of Halich&rus. TrichecUna, GRAY, Cat. Mam. Brit. Mus., pt. ii, 1850, 29; Cat. Seals and Whales, 1866, 33 (subfamily). In part only = Trichecina Gray, 1837. " TrichecMda; sen Campodontia, BROOKES, Cat. Anat. and Zool. Mus. 1828, 37." Trich-eclwidea, GIEBEL, Fauna der Vorwelt, i, 1847, 221: Saugeth., 1855, 127 (family). Ti-ichecina, TURNER, Proc. Zool. Soc. Lond., 1848, 85, 88 (subfamily). Rosmarida;, GILL, Proc. Essex Institute, v, 1866, 7, 11 ; Families of Mam., 1872, 27, 69, 70 ( = " Tricliecliidos Brookes, Gervais").— ALLEN, Bull. Mus. Comp. Zool., ii, 1870, 21. Eosmaroidea, GILL, Fain. Mam., 1872,70 ("superfainily" = Bosmaridce Gill). Broca, LATREILLE, Fani. Reg. Auim., 1825, 51 (family). Les Morses, F. CUVIER, Dents cles Mam., 1825, 233; Diet, Sci. Nat., lix, 1829, 465 (family). GENERAL OBSERVATIONS. Among the distinctive features of the Odobccnidce are the enormous development of the upper canines, and the consequent great enlargement of the anterior portion of the skull for their reception and support, the early loss of all the incisors except the outer pair of the upper jaw, the caducous character of the posterior molars, and the molariforui lower canines. The Wal- ruses share with the Eared Seals the ability to turn the hind feet forward, and consequently have considerable power of loco- motion on laud. This is further aided by a greater freedom of movement of the fore feet than is possessed by the Earless Seals. The Walruses differ from the Eared Seals by their much thicker bodies, shorter necks, and longer caudal vertebrae, the dorsal and lumbar vertebrae remaining of proportionately the same length. In consequence of their obesity, the ribs and the proximal segments of the limbs are longer in the Walruses than in the Eared Seals, while the distal segments of the limbs are relatively shorter. The scapula is long and narrow, instead of short and broad, as in the Otariidw, and its crest is placed 5 6 FAMILY ODOB^NHXE. more anteriorly. Accordingly, in respect to general form, we have slenderness of both body and limbs in the one contrasted with great thickness of body, and distally a disproportionate reduction of the extremities in the other. The most striking- differences, however, exist in the cranial characters, resulting from the great development of the upper canines in the Wal- ruses, and the consequent modifications of the facial portion of the skull. In the Otariidce, the general contour of the skull is strongly Ursine ; in the Odobcenidce, it is unique, owing to its great expansion anteriorly. In respect to other cranial features, the Walruses differ from the Eared Seals in having no post-orb- ital processes, and in the mastoid processes being not separated from the auditory bullae. The teeth are all single-rooted, and have in the permanent dentition no distinct crowns. On comparing the Odobccnidw with the Phocida', the differ- ences in general structure are found to be far greater than ob- tain between the Walruses and Eared Seals, especially in regard to the himl extremities ; these in the Phocida' being directed backward, and useless as organs of terrestrial locomotion. Hence, in so far as the Odobanida' and Otariidcc agree in liinb- aud skull-structure, they both similarly depart from the Phocine type. As already indicated in the synopsis of the suborder Pin- nipcdia, the Phocida' differ far more from either the Odobcenidw and Ota>-iida> than do these latter from each other. This differ- ence is especially emphasized in the skull ; for while the Odo- bwnida' and Otariida' agree in all important cranial characters, aside from the special features correlated with the immense enlargement of the upper canines in the former, they widely differ from the Phocida'. This is especially seen in the absence in the latter of an alisphenoid canal, in the greatly swollen audi- tory bulla?, the position of the carotid foramen, and the non- salient character of the mastoid processes. The few points in which the Walruses differ in myology from other Pinnipeds, Dr. Murie states to be "the presence of a co- raco-brachialis, a flexor brevis nianus, a pronator quadratus, an opponens pollicis, and a palmaris brevis," in the possession of which it differs both from Otaria and Phoca, but that in other respects they " muscularly present general agreement." " Com- pared with the Seals \Plioca £] there are two extra peronei and a flexor brevis hallucis."1 " Though deficient in concha, the auri- cular muscles are remarkably large."* :r Proc. Zool. Soc. Lond., 1870, p. 545. GENERAL OBSERVATIONS. 7 "Considering the very different attitudes assumed by the Tri- cJiecMdcc and Otarudce as compared with the Pliocidce? he further adds, " it is remarkable how very little deviation follows in the muscular development. The two former, as might be antici- pated, present a general agreement, especially iu the mode of implantation of the muscles of the hind leg, and in this respect recede from the Seal, yet but slightly." * In respect to the position and character of the viscera, a gen- eral agreement has beeen noted with those of the other Pinni- peds, and they present nothing that calls for special notice in the present connection. As Dr. Murie has stated, there is little appreciable difference exhibited throughout the Pinnipeds in the construction of the alimentary canal. " It is simply that of a Carnivore, with, however, a moderate-sized coecum. The great glandular superficies and correlated large lymphatics point to means of speedy and frequent digestion ; and in the Walrus these apparatus are extraordinarily developed."! In accordance with the characters already given (p. 3), if any subdivision of the Pinnipeds into groups of higher rank than families is to be made, it seems evident that the Odobcenidce and Otariida} are to be collectively contrasted with the Phocidcc; in other words, that to unite the Otariidce and PliocidcK as a group of co-ordinate rank with the Odobcenidw is to lose sight of the wide differences that separate the two first-named fami- lies, as well as of the many important features shared in com- mon by the Odolcenidcc- and Otariidce, by which both are trench- antly separated from the Pliocidcc. Although the Walruses are now very generally recognized as constituting a natural family of the Pinnipeds, ranking co-ordi- nately with the Eared Seals on the one hand and with the Earless Seals on the other, the affinities of few groups have been more diversely interpreted. As early as the thirteenth century, the author of the " Speculum Regale",— one of the earliest works re- lating to natural history, in which the Walrus is mentioned,— stated distinctly that the Walrus was an animal closely related to the Seals ; and we find that nearly all natural-history writers prior to the middle of the eighteenth century who referred to the Walruses, gave them the same association. It was the technical systematists of the last half of the eighteenth century who broke up this natural juxtaposition, and variously grouped * Trans. Zoo!. Soc. Loud., vol. vii, 1872, p. 459. t Trans. Zoo'l. Soc. Lornl., vol. vii, 1872, p. 4H1. FAMILY ODOB^ENHXE. them with forms with which they had no relationship. In the infancy of science, nothing was perhaps more natural than that animals should be classified in accordance with their mode of life, their habitatr or their external form, and we are hence not surprised to find that Eondelet, Gesner, Aldrovandus, Jonstou, and other pre-Linnsean writers, arranged the Pinnipeds, as well as the Sirenians and Cetaceans, with the fishes, or that, other early writers should term all four-footed creatures " Quadru- peds," and divide them into "Land Quadrupeds" and "Quadru- peds of the Sea." While all marine animals were by some early writers classified as "fishes,"* the Pinnipeds were much sooner 'disassociated from the true fishes than were the Cetaceans and Sirenians, the mammalian affinities of which were not at first recognized by even the great Linne himself, who, as late as the tenth edition of his " Systema Xatune " (1758), still left them in the class u PiscesS"1 In view of the several excellent descriptions and very credit- able figures of the Atlantic Walrus that appeared as early as the sixteenth and seventeenth centuries (a detailed account of which will be given later), it is surprising that the early sys- tematic writers should display such complete ignorance of some of the most obvious external characters of this animal, as was notably the case with Linne, Klein, Brisson, Erxleben, and Gmeliu, who strangely associated the Walrus and the Manatee as members of the same genus, and grouped them with such diverse creatures as Sloths and Elephants. Linne, it is true, in the earlier editions of the " System a ISTatura?," placed the Wal- rus with the Seals in the genus Phoca, in his order Ferce, — a near hit at their true affinities. Later, however, following probably Klein and Brisson, he fell into the grave error of removing them to nearly the most unnatural association possible. In this con- nection, it may prove not uninteresting to sketch, in brief out- line, the strange history of the classification of this singular group of fin-footed Carnivores. As already stated, Linne's first allocation of the group was the natural one. Brisson,t in 1750, led in the long role of error by forming his third "order" of mammals of the Elephant, the *Most modern languages still retain relies of this ancient custom, as evinced, for example, in such English words as shell-fish, cray-Jixh, whale- JlsJuri/. Ncid-fmliery, etc., while hvalfisli (Swedish), wahisch (Danish), i (German), etc., are common, vernacular names applied to Cetaceans. tRegnu Animal, 1756, p. 4~. GENERAL OBSERVATIONS. AValrus, and the Manatee, the two last named constituting his ••genus Odobcitus." This was a marked retrocession tVom even i IK- system of Klein,* of a few years'1 earlier date, who brought together as one family the Seals, Otters, Beaver, AYalrus, and Manatee. Linne, in 17GO,t not only removed the AYalrus from the genus Phoca, in which he had previously placed it, to 2V/- chechus, but also transferred it from liis order Fern* to Brnta, which thus contained not only the AValrns, but such a diverse assemblage as Elephants, Sloths, and Anteaters. Linne's genus Tricht.rhnn, as at this time constituted, was equivalent to Bris- son's genus •• Odobcnux." Erxleben,| who recognized no higher groups than genera, placed the Walruses and Seals together im- mediately after the Carnivores. Schreber,§ at about the same date (1777), adopted a similar classification, the Walrus stand- ing next after the Elephant and preceding the Seals. Schre- ber's genus Trichirliitx contained also the Dugoug and the Manatee. Gmelin,|| in 17kC'. Bruta" (Ornifhortyynchus and Triclieclius). This is essentially also the arrangement proposed by Klein in 1751. The first step toward dismembering the unnatural conglo- meration known previously under the names TricJiechus and Odobenuft was made by Eetzius** in 1794:, wrho divided the genus TrichechH* of former authors into three genera, namely, Ulanatus, for the Manatee; Hydromalis, for Steller's Sea-Cow (= Rhytina Illiger, 1811) 5 and Triclieclius^ the last embrac- 1. Disp. Brev. Hist. Nat., 1751, pp. 40, 02. tSyst. Nat., ed. 12, 17G6, p. 40. t Syst. Reg. Anim., 1777, p. 50::. §Saugeth., ii, [1776?]- P- 2! Id. || Syst. Nnt.:i, 50. HHaiidb. d. Naturgesch., 17£S, p. 14','. and later editions. * Koiigl. Vetensk. Aead. iiya Handling., xv, 1704, pp. 28I1-300. 10 FAMILY ODOB^ENIDJE. ing both the Walrus and the Dugong. While this was in the main a most important and progressive innovation, Eet- zius seems to have labored, like several still earlier writers. under the impression that the Walrus, like the Dugong, had )<<_> hind feet. Ozeretskovsky,* about a year later, and probably ignorant of Eetzius's paper, also placed, as curiously happened, the Walrus and the Dugong together in the genus Tnclieclim^ because he supposed the Dugoug liad liindfeet, like the Walrus ! These curious antithetical mistakes indicate how little was known by systematic writers about the structure of these ani- mals as late as the close of the last century. The elder Cuvier, t in 1798, while retaining the Walrus and the Sirenians in the genus Triclieclim, separated them from some of their former unnatural entanglements by again associating Tri- clieclius and Plioca in his group "Mainmiferes Amphibies," which he placed between the " Solipedes " and " Maminiferes Cetaces.7' He divided this group into "I. Les Phoques (Phoca)" and "II. Les Morses (Trichecus, L.)"; the latter including "1. Tn'checus rosmarus"; "2. Trichecus dugong"; "3. Trichecus manattis." As already shown, Eetzius nearly disentangled the Walrus from the Sireuians, leaving of the latter only the Dugoug in the genus Tricheclius. G. Fischer, \ in 1803, completed the sep- aration by removing the Dugoug and the Manatee, to which he gave the generic names respectively of Platt/stomm (=Halieore, Illiger, 1811) and Oxystomus ( = ]\Ianatus, Eetzius, 1704), leaving only the Walrus in Triclieclim. The genus Triclieckus, howe^-er, as first instituted by Artedi (1738) and Linne (1758), as will be shown later, did not relate in any way to the Walrus, being applied exclusively to the Manatee. It was not till 1700 that the term was first made to cover both the then known Sireuians and the Walrus, although the embroilment of the two groups began with Brisson, ten years earlier. The Pinnipeds and Sirenians, collectively considered, were first separated as distinct groups by Illiger § in 1811, who raised them to the rank of orders, they forming respectively his orders Pinnipedia and Natant-ia. The former consisted of two genera, Phoca, embracing all the Seals, and Trichechus, containing only the Walruses. They were regarded as forming a single family, "Nova Act. Acacl. Petrop., xiii, 1796, pp. 371-375. tTabl. Element,, p. 172. t Das National-Museum der Naturgeschichte, ii, 1803, pp. 344-358. § Prodromus Svstematis Mammaliiim et Aviuni, 1811, pp. 138, 139 ; Abhaudl der Akad. Wissensch. zu Berlin, 1804-1811, (1815), pp. 39-159, passim. GENERAL OBSERVATIONS. 11 equivalent in extent with the order Pinnvpedia. The propriety of the changes introduced by niiger was not speedily recognized by contemporary writers ; Cuvier, and many subsequent syste- rnatists for half a century, placing the Pinnipeds among the Carnivora and the Sirenians among the Cetacea, with the rank respectively of families, the family PJiocidce embracing all the Pinnipeds. Dr. J. E. Gray, in 1821,* and again in 1825,f widely separated the Walruses from the Seals as a family, TrichecMdcc, Avhich he most strangely placed (together with the Sirenians) in the order Cete. Later, however, in 1837, £ he reunited the Wal- ruses and the Seals into the single family Phocidce, which he divided into five subfamilies, Trichechina being the third and central group, and embracing the genera Halichcerus and Triclie- chus. This highly artificial classification he retained till 1866, when, following other systematists, he again raised the Wal- ruses to the rank of a distinct family. Latreille, § in 1825, not only treated the Pinnipeds as an order (Amphibia), but separated the Walruses from the others as a distinct family (Broca), the Seals forming his family Cyno- morpha. In 1829, F. Cuvierll divided the Pinnipeds into the Seals proper ( " les Phoques proprenient dits"), and theWalruses ("les Morses") . Brookes, ff in 1828, again recognized the Walruses as forming a family (" TrichecMdw sen Campodontia") distinct from the < »ther Pinnipeds. Wagler,** in 1830, made the Walruses merely a genus of his order Ursi. Xilsson, ft in 1837, divided the Pin- nipeds into two sections, the second of which embraced not only Triehechits, but also Halichcerus, Cystopliora, and Otaria. Tur- ner, || in 1848, from a study of the skulls, separated the Pinni- peds into three natural groups, considered by him to hold the rank of subfamilies, namely : Arctoccplialina, embracing Otaria and Arctoceplialus ; Tricliccina, consisting of the genus "Triche- cus " $ and Phocina, embracing all the other Seals. Gill, §§ in I860, * "London Med. Repos., 1821, p. 302," aptid Gray. t Annals of PMlosophy, 2cl ser., vol. x, 1825, p. 340. t London's Mag. Nat. Hist., vol. i, p. 583. vS Fam. Reg. Anim., p. 51. || Diet. Sci. Nat,, t. lix, p. 367. H "Cat. of Ms Anatoin. and Zool. Mus., p. 36," apud auct. **Naturl. Syst. Amph., p. 27. tt Vctcusk. Akad. Haudl., 1837, 235; Wiegmann's Arch. f. Naturg., 1841,. p. 306 (trausl.). tt Proc. Zool. Soc. Lond., 1848, pp. 85, 88. ^ Pror. Essex Institute, vol. v, p. 7. 12 GENERA OF fHE FAMILY. was the next author who recognized the Walruses as forming a distinct family, which he termed Rosmarida\ In this step, he was immediately followed by Gray,* and by the present writer t in 1870. Lilljeborg,! in 1874, also accorded them family rank, as has been the custom of late with various other writers. Gill, § in 1872, raised them to the rank of a " superfamily " (Bosma- roidea), treating them as a group co-ordinate in rank with his " Phocoidea," consisting of the Phocidcc and Otariidce. Their final resting-place in the natural system has now prob- ably been at last reached, the majority of modern systematists agreeing in according to them the position and rank of a family of the Pinnipedia. To Illiger seems due the credit of first dis- tinctly recognizing the real affinities of both the Pinnipeds and Sirenians to other mammals, and with him originated the names by which these groups are now commonly recognized, the chief modification of ILliger's arrangement being the reduction of the Pinnipedia from a distinct order to the rank of a suborder of the Ferce. GENERA. The family Odoba'nidw (Triclicchidcc Gray and Brookes = .Ros- Mdt'ida' Gill) includes, so far as at present known, only the existing genus Odobccmts (= Triclieclms of many authors, not of Artedi nor of Linne) and the two extinct genera TrichecJiodon and Alacthcrium, recently described from fossil remains found in Belgium. Alactlierium.,\\ while evidently referable to the Odo- b(?nid(e, differs quite strikingly from the existing Walruses. The parts known are the left ramus of the lower jaw, the greater portion of the cranium (the facial portion and teeth only want- ing), several cervical vertebrae, a portion of the pelvis, and vari- ous bones of the extremities. The rami of the lower jaw are not anchylosed as in the Walrus, and the dentition is quite dif- ferent from that of Odobwnus, that of the lower jaw being I. 2, C. 1, M. 4. The symphysis occupies nearly half of the length of the jaw. Van Beneden describes the skull as resembling in some characters the skull of the Otaries, and in others those of the Morses. The molar teeth he says could not be easily distin- guished from those of the Morse if they were found isolated. * Aim. and Mag. Nat, Hist,, 3d ser., vol. xviii, 1850, p. i>29. t Bull. Mus. Comp. Zool., vol. ii, p. 21. t Fauna ofer Sveriges ocli Norgos Eyggr., p. 674. ^ Arrangement of Families of Mammals, 1872, p. 69. || Van Beneden, Aim. Mus. d'Hist. Nat. de Belgique, i, 1877, p. 50. GENERAL OBSERVATIONS. 1 o oaiiines were found at Anvers, but Yau Beuedeu is strongly of the opinion that the teeth described by Kay Lankester,* from the Red Crag of England, in 1865, and named Trichechodon hux- leyi, are those of his Alactlierium cretsii. The other bones re- ferred to Alacthcrium bear a general resemblance to the corre- sponding bones of the existing Walrus, but indicate an animal of much larger size. The femur and some of the other bones bear also a resemblance to corresponding parts of the Otaries. A cast of the cerebral cavity shows that the brain was not much unlike- that of the existing Walruses and Otaries, but with the cerebel- lum smaller. Alactlierium thus proves to have been a Pinniped of great size, closely related in general features to the Walruses of to-day, but presenting features also characterizing the Eared Seals as well as others common to no other Pinniped. The genus Trichechodon of Van Beneden (probably not = Trichechodon, Laukester, 1805) is much less well known, the only portion of the skull referred to it being part of a right ramus. The other bones believed to represent it are nine vertebrae, part of a pel- vis, a hnmerus, a femur, several metatarsal, metacarpal, and phalaugeal bones, etc., and part of a tusk. Says Van Beuedeu : " Tine branche de maxillaire est tout ce que nous possedons de la tete. Les dents rnanquent, inais le bord est assez coinplet pour qu'on puisse bieii juger de leurs caracteres par les alveoles. Xous pouvons, du reste, fort bieu aussi apprecier la forme de cet os, distinguer sa syinphyse et sa brievete. " L'os est brise a son extremity anterieure, la syinphyse est fort courte et 1'os n'a pas plus d'epaisseur sur la ligne mediaue que sur le cote. Les alveoles sont comparativement fort grandes : les trois dernieres sont a peu pre semblables, I'antexieure est la plus petite. C'est 1'inverse dans le Morse. La canine devait etre fort grande. II n'y a qu'une seule alveole pour une dent inci- sive. "Le corps du maxillaire est reniarquable pour sa courbure. Toute la partie poste"rieure qui constitue la branche du maxillaire manque. On voit sur la face externe trois trous mentonniers. " En comparant ce maxillaire a celui du Morse vivant, on voit que la symphyse est toute differente, qu'il existe une grande alveole pour la dent canine et des traces d'une petite alveole pour une incisive qui restait probablement cachee sous les gen- cives. Dans le Morse vivant, il n'y a pas de place pour une canine [graude] au maxillaii'e inferieur." * See beyond, ]>. &2. 14 THE GENUS ODOB^ENUS. » The otner bones are described as more or less resembling those of the Walrus, and do not much exceed them in size. Some of them are also said to closely resemble corresponding- parts of Alactlierium. Yan Benedens descriptions and figures of the lower jaw fragment indicate features widely different from those of the corresponding part in the Walrus, especially in the shortness of the symphysis and in the curvature of the part represented, but above all in. the number, relative size, and form of the alveoli, and particularly in the large size of that of the canine, which must have been almost as highly specialized as in the Sea Lions. That the tusks referred to it by Van Beneden (those described by Lankester especially, as well as the fragment he himself fig- ures) belong here, there seems to be at least room for reason- able doubt.* The differences presented by the jaw fragment of Trichechodon as compared with the corresponding part of Alactlierium are even still more marked. The more obvious characters distinctive of the three genera of the Odobcenidce, as at present known, may be briefly indicated as follows : of tlie <+cnera.\ 1. ODOB^ENUS. — Eanii of lower jaw firmly anchylosed, even in early life ; sympliysis short. Incisors (in adult) 0; canines 1 — 1: molars 3 — 3, the last much smaller than the others. 2. ALACTHERIUM. — Kami of lower jaw not anchylosed ; symphysis very long. Incisors (in adult) 2 — 2; canines 1 — 1; molars 4 — 4, the last smaller than the preceding ones. 3. " TKICHECHODOX " (Van Beneden). — Kami of lower jaw (apparently) unanchylosed. Incisors 1 — 1?, very small; canines 1 — 1, highly spe- cialized; molars 4 — 4, the first small, the last three much larger and suhequal. GENUS ODOBJENUS, Linne. Odolenus, LUSTSTE, Syst. Nat., i. 1735 (ed. Fe"e), 59 (applied exclusively to the Walrus in a generic sense). — BEISSON, Regne Aniin., 1756, 48 (used strictly in a generic sense, but embracing "1. La Vache marine — OfZoZ>e»i Denies plan! in utraque maxilla. Dors urn iiupenue. Fistula . . . . The cita- tions under Tricliechus embrace no allusion to the "Walrus, but relate wholly to Sirenians, or to the Manatee, as the latter was then known-! Artedi's description of the Manatee is quite full and explicit, but includes also characters and references belong- ing to the Dugong. § Trichechus forms Artedfs "genus LI." and is placed in his ''order Y, Playiurl" (embracing the Ceta- ceans and Sirenians, the other genera of this order being Pliy- seter, Ddplilnus, Balccna, Monodon, and Catodon), and is hence * Respecting the proper generic name of the Walruses, Wiegmauu, in 1838, thus forcibly expressed his views : ' • Die Gattung Odobenus [von Brissou,1756] hiitte beibehalten werdeu mitsseu, da der ganz abgeschmackte Name Triclie- cli us gar nicht dem Walrosse, sondern iirspriinglich dem Mauati angehort, und von Artedi fur diesen gebildet war, um die bei einem Fische oder vielmehr Wallfische auffalleude Behaarung zu bezeichueu." — Archiv fur Naturge- schichtc, v. Jahrg., Band i, 1838, p. 116. tlchthyologia, 1738, pars i, p. 74; pars iii, p. 79; pars iv, p. 109. lu Arte- di's work the name is twice written Trichechus and twice Tliriclieclius. On p. 74 of pars i. where it first occurs, its derivation is given, namely: "Triche- chus a -&pf^ crinis $• l%do<; iriscis quia solus inter pisces fere hirsutus sit." t The references in a general way appear to include all the Sireuians then known. ^E. g., "Dentiuni duo utrinque eminent, longitndine spithauiiu crassitu pollicis." 16 THE GENUS ODOB^ENUS. equivalent to the Cete of Liime (Syst. ISTat., ed. x, 1758). Linney in 1758, first introduced Artedi's genus Triclieclnia, at \yliich time he placed in it only the Manatee, Dugong, and ^teller's Sea Cow, leaving the Walruses still in Plioca. His diagnosis of the genus* embraced none of the distinctive characters of the Walrus. In 1766 (12th ed., Syst, Nat.), lie transferred the Walrus from Plioca to TricJiecltm, making it the first species of the genus. The diagnosis, though slightly changed ver- bally, has still little, if any, reference to the characters of the Walruses, unless it be the phrase " Laniarii superiores solitarii,"! which is equally applicable to the Dugong, and is not at all the equivalent of "Phoca dentibus canines exsertis, " previously ascribed to the Walrus in former editions, when the Walruses were placed under Plioea. Hence, to whatever the generic- name Tricheclius may be referable, it certainly is not pertinent to the Walrus. This being settled, the question arises, 'What generic name is of unquestionable applicability to the Walruses ? Here the real difficult}* in the ease begins, for authors who admit the inapplicability of TricJiechus to this group are not agreed as to what shall be substituted for it. Scandinavian writers, as Mahngren (1804) and Lilljeborg (1874), and Peters (1864) among German authorities, have for some years employed Odobcenus, a name apparently originating with Liune (as Odobe- nus) in 1735, and adopted in a generic sense by Brisson in 1750. A modified form of it (Odontobwnux) was also employed by Snn- devall in 1859. Gill, in 1866, and other recent American writers. have brought into some prominence the name Rosmarus, first used in a generic sense by Klein in 1751, by Scopoli in 1777, by Pallas $ in 1831, and by Lamout § in 1801 ; while the great mass of English and Continental writers still cling to Tricliecltus. The genera Odobenoiherium and Trichechodon, based on fossil remains of the Walrus, have also been recently introduced into the literature of the subject, the former by Gratioletin 1858, and the latter by Lankester in 1865 ; but these (especially the first) *"Dentes primores nulla, laniarii superiores solitarii, molares ex osse rugoso utrinque inferius duo. Laliia geniiata. Pedes posteriores coadivnati in pinnam." — Syst. Nat., ed. x, i, 1758, p. 34. tThe second diagnosis of Triclieclius is, in full, as follows: "Dentcs prinio- res nulli ntrinque. Laniarii superiores solitarii. Molares ex osse rugoso utrinque ; inferius duo. Labia genimata. Pedes posteriores corupedes co- adunati in pinuam." — Syst. Nut., ed. xii, 1766, i, p. 48. tZool. Eosso-Asiat., vol. i, 269. $ Seasons with the Sea-horses, pp. 141, 167. THE SPECIES OF THE GENUS. 17 appear to be referable to the existing Walruses, and of course become merely synonyms of earlier names. Consequently the choice evidently lies between Odobccnus and Rosmarus. Odobcc- nus has sixteen years' priority over Rosmarus, if we go back to the earliest introduction of these names into systematic nomen- clature. * It is true that Eosmarus was the earliest Latin name applied to the Walrus, its use dating back to the middle of the sixteenth century, when it was employed interchangeably with Mors and Morsus by Olaus Magnus, Gesner, Herberstain, and others, but only in a vernacular sense. Although used by Klein systematically in 1751, Gill adopted it from Scopoli, 1777, proba- bly because Klein was not a " binomialist." Linne" used Odobcenus generically in 1735, as did also Brisson in 1756. The whole question turns on what shall be considered as the proper start- ing-point for generic nomenclature, about which opinion is still divided. If the early generic names of Artedi, Klein, Brisson, and Linne (prior to 1758) are admissible, as many high author- ities believe, then Odobcenus is unquestionably the only tenable generic name for the group in question, of which Rosmarus is a synonym, t SPECIES. The existing Walruses have been commonly considered as belonging to a single circumpolar species. A few authors have recognized two, or deemed the existence of two probable, while one appears to have admitted three. Altogether, however, not less than six or seven specific names have been given to the ex- isting species, besides several based on fossil remains of the Atlantic Walrus. In the present paper, the attempt will be made to establish the existence of two j but before entering further upon the discussion, it may not be out of place to glance briefly at the views previous authors have held respect- ing the point in question. Pennant appears to have been the first to call attention to the probable existence of more than a single species of Walrus, who, in 1792, in speaking of the Walruses of the Alaskan coast, says : " I entertain doubts whether these animals [of " Uualascha, Sandwich Sound, and Turnagain Eiver"] are of the same species * Odobcemts, Linu6, "Digit! ant., post. 5, palmipes. Eoss Morsus. Dentes intermedii superiores longissirni." — Syst. Nat., 1735 (ed. Fe"e), 59. — Rosmarus, Klein, Quad. Disp. Brev. Hist. Nat., 1751, 40, 92. tin accordance with custom in similar cases, the name of the family be- comes Odobccnidw, — neither Eosmaridce nor Trichechidce being tenable. Misc. Pub. ISTo. 12 2 18 THE GENUS ODOB.ENUS. with those of the Gulph of St. Laurence. The tusks of those of the Frozen Sea are much longer, more slender, and have a twist and inward curvature."* Shaw, a few years later, thought that the Walrus described and figured in the account of Cap- tain Cook's last voyage, though perhaps not specifically distinct from those of the Arctic shores of Europe, should be regarded as belonging to a different variety. t He appears, however, to have based his opinion wholly on figures of the animals, and par- ticularly on those given by Cook and Jonston (the latter a copy of Gerrard's, at second-hand from De Laet). Illiger, in 1811, formally recognized two species in his " Ueberblick der Saug- thiere nach ihrer Yertheilung liber die Welttheile,"| namely, Trlclieclms rosmarus, occurring on the northern shores of (West- ern ?) Asia, Europe, and North America, and T. obesus, occur- ring on the northwestern shores of North America and the ad- joining northeastern shores of Asia. While I do not find that he has anywhere given the distinctive characters of those two species, he, in the above-cited paper, also named the animal described and figured by Cook, T. divergens. F. Cuvier, in 1825, in describing the dentition of the "Morses," says : " Ces dents ont etc de'crites d'apres plusieurs tetes qui semblent avoir appartenu a deux especes, a en juger du moius par les proportions de quel- ques unes de leurs parties, et non settlement par Fe'tendue de leurs d6fenses, caractere qui avait deja fait soup9onner a Shaw * Arctic Zoology, vol. i, 1792, pp. 170, 171. tHe says: "An excellent representation is also given in pi. 52 of the last voyage of our illustrious navigator, Captain Cook. It is easy to see, however, a remarkable difference between the tusks of this last, and those of the former kind figured in Jonston, and it clearly appears, that though this difference is not such as to justify our considering them as two distinct species, yet it obliges us to remark them as varieties ; and it should seem, that, in the regions then visited by Captain Cook, viz. the icy coasts of the American continent, in lat. 70, the Walrus is found with tusks much longer, thinner, and far more sharp-pointed, in proportion, than the common Wal- rus ; and they have a slight inclination to a subspiral twist : there is also a difference in the position of the tusks in the two animals ; those of the variety figured in Captain Cook's voyage curving inwards in such a manner as nearly to meet at the points, while those of the former divaricate. These differences appear very striking on collating different heads of these ani- mals. Something may, however, be allowed to the different stages of growth as well as to the difference of sex. In order that these differences may be the more clearly understood, we have figured both varieties on the annexed plates "—General Zoology, vol. i, 1800, pp. 236, 237, pis. 68, 68*. tAbhaudl. der Akad. der Wissensch. zu Berlin, 1804-1811, p. 64. Bead before the Academy Feb. 28, 1811, but apparently not published till 1815. THE SPECIES OF THE GENUS. 19 IVxisteiice de deux especes de morses.'7* Fremery, iii 1831, having before him a series of eleven skulls, distinguished three species, namely, Tricliechus rosmarus, T. long-idem, and T. cookl. The first ( T. rosmarus) was principally characterized by having diverging tusks, about as long as the length of the whole head, faintly grooved on the outside, and with two distinct grooves on the inside ; by the possession of five back teeth, the last two very small ; by the lower edge of the nasal opening being but little produced ; by the occipital crest being strongly developed ; and by the great specific gravity of the bones of the skull. The second (T. longidens) was principally characterized by the tusks equalling or exceeding in length two-thirds of the length of the skull, with a single deep groove on the inner side ; by having only four back teeth, the last one small; a smaller develop- ment of the occipital crest (except in old animals /) ; and a lighter specific gravity of the bones. The third (T. cooJci), considered as a doubtful species, was based wholly on Shaw's plate 68 (from Cook), already noticed, and hence is the same as Hliger's T. diver gens. Wiegmann, von Baer, Stannius, and most subsequent writers, have properly regarded Frernery's characters of his T. rosmarus and T. longidens as based merely on ordinary indi- vidual or sexual differences. Wiegmann, and also Ternniinck, according to Fremery, believed the female to be distinguishable from the male by its longer and thinner tusks, with the crests and ridges of the skull less developed, while other differences, as the relative prominence of the bony lower edge of the nasal opening, were differences characteristic merely of different indi- viduals.t Stannius, however, in 1842, after passing in review * Dents des Mainmiferes, p. 235. t Wiegmann, in commenting upon Fremery's supposed specific differences, observes as follows respecting probable sexual and individual differences in the tusks and skulls of Walruses: "Hr. Fremery fiihrt an, dass Hr. Temminck einen (nach Deutlichkeit der Nahte) nocli jungen Schiidel des Reichsmuseums mit ausgezeichnet langen diinnen Stosszahnen fur den eines Weibchens gehalten habe. Ich erinnere micli aucli von Gronlands- fahrern gehort zu haben; dass sich das Weibchen durch langere, dtiunere, dass Miinnchen durch kiirzere, aber viel dickere Stosszahne auszeichne." The alleged difference in the specific gravity of the bones of the skull he be- lieves also to be a sexual feature, as possibly also the difference in the num- ber of molar teeth. Respecting the prominence of the lower border of the nasal opening he says: "Die mehr oder minder starke Hervorragung des unteren Randes der Naseniiffnung kann ich dagegen uur fur erne individuelle Verschiedenheit halten, da ich sie bei einem Schiidel mit kurzeu Stoss- zahnen, der die iibrigen vom Verf. hervorgehobenen Merkmale besitzt, sehr stark, und mngekehrt bei einem alten Schiidel mit langen Stosszahnen kaum fiir Naturgesch., 1838, pp. 128, 129. 20 THE GENUS ODOB^NUS. the characters assigned by Freinery as distinctive of several species, and after mentioning at length other features of varia- tion observed by him in a considerable series of skulls, describ- ing several of his specimens in detail, and arriving at the con- clusion that up to that time all the supposed species of Walrus constituted really but a single species, added another, under the appropriate name Trichechus dubius. This with subsequent authors has shared the fate of Fremery's species,* being consid- ered as based merely on individual variation. As will be more fully noticed later, two nominal species have been founded on the fossil remains of the "Walrus, namely, Tri- chechus virginianuSj DeKay, 1842, and Odobenoth&rium larte- tianum of Gratiolet, the former based on remains from Accoinac County, Virginia, and the latter on remains from near Paris, France. Lankester, in 1865, added still another, based on tusks from the Eed Crag of England, under the name Triche- chodon liuxleyi. Dr. Leidy, in 1860, in a paper on fossil remains of the Wal- rus from the eastern coast of the United States, again noticed the differences in the size, length, and curvature of the tusks in specimens from the northwest coast of North America and the common Walrus of the North Atlantic. He says : "In the course of the preceding investigations [referring to previous portions of his paper], I was led to examine a specimen, in the cabinet of the Academy of Natural Sciences [of Philadelphia], consisting of the stuffed skin of a portion of the head envelop- ing the jaws of a species of Walrus apparently differing from the true Trichccus rosmarus, of which, as characteristic, I have viewed the figures of the skull and skeletons as given by Dau- benton, Cuvier, and De Blainville. The specimen was pre- sented by Sandwith Drinker, Esq., of Canton, China, and was probably derived from the Asiatic shore of the Arctic Ocean. From the worn condition of the upper incisor and molars, it appears to have belonged to an old individual ; and in the case of the lower jaw, the teeth appear to have been entirely worn out. The tusks are very much larger and are narrower than in " Giebel, in 1855, referred to Fremery's and Stanuius's species as still need- ing confirmation : " Die von Fremery uacli der Beschaffenheit der Ziihne unterscMedenen Arteu, Tr. longidensundTr. Cookl; sind liiugst als unhaltbar erkannt worden und auch die YOU Staimius auf Schadeldifierenzen Tbegriin- dete Art, Tr. dubiits, cutbehrt uoch der weitern Bestatiguiig." — SciuyetMerc, p. 128, footnote. THE SPECIES OF THE GENUS. 21 the T. rosmarus, and they curve downward, outward, and in- ward, instead of continually diverging as in this species. At their emergence from the alveoli the tusks are two and three- quarter inches apart, near the middle five and a quarter inches, and at their tips only one inch. Their length is twenty-two inches and then1 diameter at the alveolar border antero-posteri- orly two and a quarter inches, and transversely one and a half inches. Towards their lower part they are twisted from within, forwards and outwardly." After quoting Pennant's remark (already given, see p. 17) about similar differences noted by him, he adds that "the superior incisor and molar teeth are also very much smaller than in the fossils of T. rosmarus," and he gives measurements showing this difference. He then says : " The hairs of the upper lip of the T. rosmarus are stated by Shaw, to be about three inches long, and almost equal to a straw in diameter.* In the specimen under consideration, the hairs of the moustache are stiff-pointed spines, not more than one line long at the upper part of the lip, and they gradually increase in size until at the lower and outer part of the lip they are about one inch in length." He further adds, in the same connection : " Since presenting the above communication to the Society, the Academy has received from Mr. Drinker, of Canton, an entire specimen of the Walrus from Northern Asia. In this individual, which measures in a straight line eight feet from the nose to the tail, the tusks are ten inches long, and diverge from their alveoli to the tips, where they are five and a half inches apart, but they are slender, as in the stuffed head above mentioned, and appear as if they would ultimately have obtained the same length and direction. Per- haps the peculiarities noticed may prove to be of a sexual char- acter."t As will be shown later, we have here the more prominent ex- ternal differences characterizing the two species of Walrus for the first time explicitly stated from direct observation of speci- mens. If Dr. Leidy had had at that time good skulls of the two species for comparison, the other important cranial differ- ences (noted beyond) could not have escaped him, and he per- haps would have been led to formally recognize the Pacific Wal- rus as a species distinct from the Atlantic Walrus. I have met with nothing further touching this subject prior to Mr. H. W. Elliott's report on the Seal Islands of Alaska, * "Shaw's Zoology, vol. i, pt. i, p. 234." t Trans. Aiuer. Phil. Soc. Phila,., vol. x.i, pp. 85, 86. 22 THE GENUS ODOB.ENUS. published in 1873, in which, tinder the heading " The Walrus of Bering Sea, (Eosmams arcticus) " he says: — "I write 'the "Walrus of Bering $ra', because this animal is quite distinct' from the Walrus of the North Atlantic and Greenland, differing from it specifically in a very striking manner, by its greater size and semi-hairless skin."5" This is all he says, however, respect- ing their differences, no reference being made to the really dis- tinctive features. Thus the matter rested till, in 187G, Gill for- mally recognized two species in his " List of the Principal Use- ful or Injurious Mammals,"! in a catalogue of a " Collection to Illustrate the Animal Resources of the United States" in the ex- hibit of the National Museum at the International Exhibition of 1876, held in Philadelphia. This is merely a nominal list, in which appears, under " Bosmaridcv," the following, which I here fully and literally transcribe : ROSMARUS OBESUS, (Illiger,) Gill. The [Atlantic] Walrus. Atlantic Coast. ROSMARUS COOKII, (Freniery,) Gill. The [Pacific] Walrus. Paci:ir Here is simply a nominal recognition of two species without expressed reasons therefor. In an article on the Rosmaridce, published in 1877, Dr. Gill again says : " Two species appear to exist — one (B. obesns] inhabiting the northern Atlantic, and the other (B. Coolri-i) the northern Pacific." \ Van Beneden, on the other hand, in 1877, distinctly affirms his disbelief in the existence of two species. In referring to the subject he says: "Nous ne croyons pas que les Morses du de"- troit de Behring different specifiqueinent de ceux de la mer de Baffin ou de la Nouvelle-Zenible, et c'est it tout, a notre avis, que Freniery a essaye de les repartir en especes distinctes d'apres les modifications de leurs dents." He further adds the testimony of von Baer as follows: " Von Baer s'est occupe en 1835 de cette question a 1' Academic de St. P6tersbourg et 1'il- lustre naturaliste m'ecilvait, pen de temps avant sa mort, au sujet de la difference legere des Morses, a 1'Est et a 1'Ouest de * Report on the Prylrilov Group or Seal Islands of Alaska, 1873 (not paged). Also, Report on the Condition of Affairs in Alaska, 1875, p. 160. IThis "List" is anonymous, and is hence, perhaps, not properly quotable in this connection, although its authorship is known to the present 'writer. t Johnson's New Universal Cyclopaedia, vol. iii, 1877, p. 1725. SYNONYMY. 23 la nier Glaciale, qu'il regardait les differences coinme des mo- difications locales*. Ce n'est pas 1'avis de Henry W. Elliot, qui considere le Morse du nord du Pacifique comme un animal distinct." t In another connection he refers to the subject as follows : " H y a des auteurs qni pensent que le Morse du Xord Pacifique est assez different de celui du Greenland, pour en faire une espece distincte. Nous ne partageons pas cet avis. Les modifications sont assez pen importantes et nous croyons pouvoir le niettre sur le compte de variations locales." J ODOB^NUS EOSMAEUS, Holmgren. Atlantic Walrus. "Kosmarus, seu Morxns Xorvcgicus, OLAUS MAGNUS, Hist, de Gent. Sept. 1555, 757 (figure) " ; also later editions. Bosmartis, GESXER, Hist. Anim. Aquat., 1558, 249; also later editions. Sosmarus, Wallross, JONSTON, Hist. Nat. de Piscibus et Cetis, 1649, 727, pi. xliv (two lower figures ; upper ono from Gesner, the lower from De Laet) ; also later editions.— " KLEIN, Eeg. Anim., 1754, 67."— "Sco- POLI, Hist. Nat., 1777, — ." — ZIMMERMAXX, Spec. Zool. Geograpli. Quad, etc., 1777, 330. Equiis marinus et Hippopotamus falso dictuts, Morse or Sea Horse, RAY, Syu., 1G95, 191. WaUross, MARTENS, Spitzb., 1675, 78, pi. P, fig. b. — EGEDE, Besclir. und Natur- Gesch. Gruiilaud, 1742, 54 ; 1763, 106 ; Descrip. et Hist. Nat. du Grocnl., 1765, 61 (with a figure). — CRAXTZ, Hist, von Gronl., 1765, 165; Englished., 1768, 125.— GOETHE, "Morphol., 1, 1817, 211"; Act. Acad. Ca3S. Leop. Carol., xv, i, 1831, 8, pi. iv (dentition, etc.). — VoxBAER, Mem. Acad. St. Petersb. Math, etc., vie s6r., ii, 1835, 199 (blood- vessels of limbs). — JAEGER, Mutter's Arch, fur Anat., 1844, 70 (den- tition— Labrador specimens). Walross, MARTEXS, Zoologische Garten, xi, 1870, 283 (etymology). Wallrus sen Mors, EUYSCH, Theatr. Animal., 1718, 159, pi. xliv (figure same as Jonston's). Walrus, WORM, Mus. Worm., 1655, 289 (fig. from De Laet).— WYMAX, Proc. Bost. Soc. Nat. Hist., iii, 1850, 242 (relation to Pachyderms). — LEA, Proc. Acad. Nat. Sci. Phila., 1854, 265 (use of the skin).— WHEAT- LAND, Proc. Essex Institute, 1, 1854, 62 (remarks on a skull). — SOXXTAG, Nar. Grinnell Expl. Exp. 1857, 113 (woodcut— group of Walruses).— MURRAY, Geogr. Distr. Mara., 1866, 128, map, xxviii* (distribution; in part).— HAYES, Open Polar Sea, 1867, 404 (hunt- ing).—PACKARD, Bull. Essex Institute, i, 1869, 137 (former exist- ence in Gulf of St. Lawrence). — ATWOOD, Proc. Bost. Soc. Nat. * " Les Morses des cotes de Sibe"rie ou de 1'est de 1'Asie out les dents cani- nes plus fortes que les Morses de Spitzberg et de Greenland, me disait-il dans une lettre " t Ann. du Mus. Roy. d'Hist. Nat. Belgique, pt. 1, 1877, 45. t Ibid., p. 17. 24 ODOB^ENUS ROSMAEUS ATLANTIC WALRUS. Hist., xiii, 1870, 220 (remarks cm a skull from the Gulf of St. Law- rence).— TURNER, Jouru. Anat. and Phys., v, 1870, 115 (relations of pericardium). — RINK, Danish Greenland, 1877, 126 (distribution), 248, 252, 272 (chase). Arctic Walrus, PENNANT, Syuop. Quad., 1781, 335; Arctic Zool., 2d ed., i, 1792, 168 (in part). Fossil Walrus, BARTON, London Phil. Mag., xxxii, 1805, 98 (no locality).— MITCHELL, SMITH, & COOPER, Aun. New York Lye. Nat. Hist., ii, 1828, 271 (fossil, Accomac Co., Va. — doubtfully referred to the exist- ing species) ; Edinb. New Phil. Journ., v, 1828, 325 (abstract of the last).— HARLAN, Edinb. New Phil. Journ., xvii, 1834, 360; Trans. Geol. Soc. Penu., i, 1835, 75; Med. and Phys. Researches, 1835, 277 (same specimen). — LYELL & OWEN, Proc. Lond. Geol. Soc., iv, 1843, 32 (fossil, Martha's Vineyard, Mass.); Amer. Journ. Sci. and Arts, xlvi, 1843, 319 (same). La Vaclie marine, BRISSON, Reg. Auiru., 1756, 48. "Morsch, GMELIN, Reise durch Russlaud, iii, 1751, 165. Morse ou Vaclie marine, BUFFON, Hist. Nat., xiii, 1765, 358, pi. liv (animal). — DAUBENTON, Buffon's Hist. Nat,, xiii, 1765, 415, pi. Iv (skull).— HOL- LANDRE, Abr6g6 d'Hist. Nat. des Quad. Vivip., i, 1790, pi. xii, fig. 3.— F. CUVIER, Diet, des Sci. Nat., xxxiii, 1816, 27 ; Dents des Mam., 1825, 233, pi. xcv. Morse, HUET, Coll. de Mam. du Mus. d'Hist. Nat., 1808, 59, pi. liii (fig. from Buffon). Sea Coiv, SHULDHAM, Phil. Trans., Ixv, 1775, 249. Phoca, BONNANIO, Rerum Nat. Hist., i, (no date), 159, pi. xxxix, fig. 27 (a poor representation of De Laet's figure, with the young one omitted). Phoca rosmarus, LINNE, Syst. Nat., i, 1758, 38. Triclwchus rosmarus, LINNE, Syst. Nat., i, 1766, 49.— MULLER, Prod. Zool. Dan., i. 1776, 1.— SCHREBER, Siiugeth., ii, 1775, 262, pi. Ixxix (from Buffoii).— ZIMMERMANN, Geogr. Geschichte, i, 1778, 299; ii, 1780, 424. — FABRICIUS, Fauna Grceul., 1780, 4. — ERXLEBEN, Syst. Reg. Anim., 1787, 593.— GMELIN, Syst, Nat., i, 1788, 59.— SHAW, Nat, Miscel., 1791, pi. cclxxvi; Gen. Zool., i, 1800, 234 (in part), fig. 68, (from Jouston). — BLUMENBACH, Handb. der Naturgesch. , 1788, 142; 1821, 136; 1825, 112; Abbild. uatur. Gegenst., 1796-1810, No. 15, text and plate (from Jonston). — DONNDORFF, Zool. Beytriige, 1792, 124. — RETZIUS, Kong. Vet. Akad. Nya Handl., xv, 1794, 391 ; Fauna Sue- cicas, 1800, 48. — OZERETSKOVSKY, Nov. Act. Acad. Sci. Imp. Petrop., xiii, (1796), 1802, 371.— BARTON, Phil. Mag., xxxii, 1805, 98 (fos- sil; locality not stated). — G. CUVIER, Tableau e"le~nient., 1798, 172; Lecons d'Anat. Comp., 1800-1805, — ; 2e 'h he secured and prepared numerous specimens of both sexes, which were lost with the ill-- fated vessel. tin the National Museum there is also a female skull of the Pacific "Walrus that presents corresponding differences as compared with male skulls of the same species. VAEIATIONS DEPENDENT UPON AGE, ETC. 43 conjectures of Wieginann and Tenrminck. What other differ- ences obtain, especially in external characters, can as yet be only conjectured. It is to be hoped, however, that we shall not have long to wait for detailed accounts of the external characters of the adults of both sexes. FIG. 3. — Odobcemis rosmarus, 9- INDIVIDUAL VARIATIONS, AND VARIATIONS DEPENDENT UPON AGE. — That a wide range of individual variation obtains in this species is sufficiently evident from an examination of even a limited series of skulls. These differences have been noted in considerable detail by Frernery, Wiegmann, Stannius, and Jaeger, as will be presently noticed more in detail in pre- senting the general history of the species. Still greater differ- ences, of course, result from differences of age. These collect- ively, as will be noted later, have formed the basis of several nominal species. All the Pinnipeds appear to be subject to a wide range of variations of this character, and none more so than the Walruses. These affect to a considerable extent the general proportions of the skull, and especially the form and relative development of different bones. These latter differences are best seen in comparatively young skulls, since most of the sutures close at a rather early age. Among these variations are especially noteworthy those of the nasal bones, the inter- maxillaries, and the frontals, and to a less degree those of the base of the skull. The crests and ridges for muscular attach- 44 ODOB^NUS KOSMARUS ATLANTIC WALEUS. ineut of course greatly increase with age, and vary considerably in respect to direction, position, and relative development in different individuals. The bony crests at the junction of the intermaxillaries below the anterior nasal opening are especially variable with age, becoming gradually obliterated in adult life by the general thickening of the bones of the skull. They are certainly less prominent in old age than in youth, and the same is true of the incisive border of the interinaxillaries. The in- termaxillaries, as a rule, only meet the nasals in their upward extension, but in occasional specimens there is a narrow exten- sion of them posteriorly between the nasals and maxillaries, reaching for one-half to two-thirds the length of the nasals. This variation is seen in the skulls figured by Goethe* and by Blainville,t and has been noted in two skulls by Stannius.l In other cases the intermaxillary rises to the surface between the nasals and maxillaries only in the form of narrow isolated areas, as is seen in a skull figured by Goethe,§ and in two skulls I have myself examined. Hence Blainville, when he says, " . . . . et le pr6niaxillaire, epais, renionte jusque entre le nasal et le inaxillaire, de maniere a circonscrire avec le premier 1'orifice nasal . . . . ,"|| describes the exceptional instead of the normal condition. The nasals vary greatly in breadth and in length in different specimens, and even in the same specimen one is sometimes much wider than the other. The concavity and width of the bony palate is also subject to much variation, in this respect hardly two specimens being found to agree. In some, the con- cavity is nearly one-fourth greater than in others. *Act. Acad. Cses. Leop. Carol., xi, pt. i, pi. iv, fig. 2. t Ost6ograpliie, Des Phoques, pi. i. J Says Stannius : "Bisweilen aber, wie bei den Kieler Schadeln a und c, tritt noch eine diinne Leiste dieses Fortsatzes zwischen die das Oberkiefer- bein und das Nasenbein verbindende Liingsnalit und trennt eine Strecke weit diese Knochen. So sieht man es auch auf der in dem Blainville'schen Werke befindlichen Abbildung. Indem diese Leiste an einigen Stellen stiirker, an andern Stellen weniger stark oder gar nicM nach aussen liervortritt und zu Tage kommt, hat es bisweilen den An schein, als fiinden sich isolirte Knochen- stiickchen in der eben genaunten Nalit. "Wirklicli erwiihnt de Fremery eines zAvischen Nasenbein uud Oberkieferbein vorkommenden Ossiculum Wormi- anuin bei seinem aus Labrador stammenden Walross-Schadel." — Mullen's Arcliiv fur Anat., 1842, p. 401. §Act. Acad. Cses. Leop. Carol., Bd. xv, pt. i, 1831, pi. iv, fig. 1. II OstdograpMe, Des Phoques, p. 20. VARIATIONS DEPENDENT UPON AGE, ETC. 45 The froiitals vary greatly in form, at their posterior border, especially in respect to their interparietal extension. This por- tion has sometimes a breadth equal to that of the nasal bones, and terminates quite squarely ; at other times, it has less than half this breadth, and is rapidly narrowed posteriorly. The tusks vary considerably in length, size, and form, and more especially in direction, in specimens of the same age and sex. They become much larger in old age than in middle life, but are then more or less abraded and broken at the points. The grooving on the sides varies more or less with each indi- vidual, and even in the two tusks of the same animal. The tusks generally widely diverge, but are sometimes nearly par- allel, but appear to be very rarely convergent, while in the female they are frequently more or less convergent, and some- times touch at the points, or even overlap. In regard to external characters, considerable changes result from age, especially in respect to the size and amount of abra- sion of the tusks, and through the loss of the hair incident to old age, and the shortening of the mystacial bristles. The following table of measurements (given in millimeters) shows to some extent the variations that occur in the general size and form of the skull. 46 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. & ^S S) . _. tb do *— I i— i fl f3 p— i C3 2 s 3 _j ? f3 3 13 IS 0 2 p 13 o •BS900ld piotKuoo yv ^q§pq '.M/ef jg^ioi 00 O (M 00 CO CO •q^at .«f ^MOT CO CO O -* CO CO i-H O> O CO '^^wdjo™ oo m cq oo co t- CO CO CO CM (M CO •S9U98 ttuojuBjoni jgddn jo q^Sugi m co o oo t— •<* CO CO IO O5 00 t- •sdpj !)B fjJiBdB 90ire!,sip 'sgrnuBQ CO 00 CO CO t— -^JH CO O5 IM CM i-( •9SBq !}B S9§p9 ^BU 00 •* O 00 00 CO CD O -^ CM O l> CO O -J9^X9 U99Ai^.9q 90UB^STp 'S9HtUBQ •9SBq ye 9on9.rajnmo.ip 'sgntncQ CD 10 o r- t^ o o CM C5 t> CS t^ t^ **- ™*'««*V-K t- CNq co o o co •^lJOii9^sod qi)ptjii. 'gguoq ^BSBJJ OO -^ CM CO 3< CD •q^Sng^ 'S9noq ^BSBJJ IN C5 O t- in t- •93SSOJ O t- CO O 00 0 CD O CD CS b- OO •S9SS900jd PTO^SBUI ^B qipB9ig; b- O •** (M CO (M CO CO CO CO CO O CO •«# W (M CN CO CM Cl N "vyemoSS.'z ys q^pB9jg; 3-* oo o o in m oo t- in CQ o co IM iH iH CM CM rH rH •q^9T CO ' CO b- . CM O t~ . rH oo . cq ^ CO • CO CO . CO CO •X9g O Of CM- *o "O Of "o : : i • • £ a :;;;•• 1 r-H IS 1C( 13 ^3 13 1 8 1 2 -*-—-D O O O rH O O rt< I" ^ O r-l i-H * * * •*— H— •+— •*-*. § IS 3C K B a « O R O 03 ,^ ^Q M CM CQ CO o ^ 5^ DENTITION. 47 DENTITION. — The dentition of Walruses, for various reasons, lias been a perplexing subject, and lias engaged the attention of many eminent writers. In the adult stage it presents many abnormalities, and, besides, is subject to much individual varia- tion, both in the temporary and adult series. For a long time its deviations from the normal type were not well understood, and even now leading authorities do not appear to be quite in harmony in respect to the proper notation. As previously stated, the incisors of both jaws, except the outer pair in the upper, disappear soon after birth, and before middle life is reached the last tooth of the molar series on each side in both jaws also usually disappears. A brief history of the principal investigations, and opinions held at different times respecting the dentition of the Atlantic Walrus (for the investigations respecting the dentition of this group appear to have been based almost wholly upon this species), is herewith appended as forrn- ing a highly interesting chapter in the technical history of the species. In this historical sketch will be found noted many facts relating to the general subject, given by the authors whose papers are here briefly summarized. The dentition of the Atlantic Walrus has been discussed in greater or less detail by Eapp, von Baer, Wiegmann, Fremery, Stannius, Jaeger, Owen, Mahngren, Peters, and various other writers. Wiegmann, in 1838, pretty fully presented the early history of the subject, noting the almost total lack of informa- tion respecting the matter shown by Linne, who evidently paid little attention to the references to the subject made by previ- ous writers. The credit of first giving any definite statement respecting the number of the teeth and their character is due to Anderson, who, in 1734, gave the number of molars as four above and three below. Brisson, in 1756, gave the number as four both above and below ; while Crantz, in 17G5, again gives four above and three below, and quite fairly describes the nor- mal dentition of the adult.* In the same year, Daubenton gave also again four below on each side as well as four above. * I append hi full Crautz's description : "It had no sharp ineisores in its mouth, and none at all before, but only four teeth on each side ; on the right side of the uuder-jaAv three pretty broad concave grinders The two long tusks or horns growing out of its face above the nose, and bending down over its month, so as almost to barricade it np, seem to be more an impediment than a help to it The right tusk is about an inch longer than the left, and its whole length 48 ODOB^NUS EOSMAEUS — ATLANTIC WALRUS. No author prior to Schreber (1775) appears to have met with deciduous incisors, who found two such upper incisors on each side in a young- skull in the Museum of Erlaugen. These he correctly conjectured Avere temporary, disappearing' at a later stage of life.* To Goethe, however, is given the credit of recognizing the true character of the first tooth of the upper molariform series. Says Camper (as quoted by Wieginanu) : "Es 1st der Hr. Gothe, sachsenweimarscher Gekeimer Bath, der mir zuerst die ossa mtermaxillana des Wallrosses und der Schneideziihne desselben hat kennen lerneu, indent er mir eine vortreffliche Abhandlung mit schonen Zeichnungen dieser Knocheii ver- schiedener Thiere zugeschickt hatte." Camper, in criticising Linne's errors regarding the Walrus,! gives four incisors (2~\^ and four molars above and five below (£=|) (or sometimes ouly four below). The observations of Schreber, Goethe, and Cam- per appear to have been generally overlooked by subsequent writers, so that it was left for G. Cuvier to discover anew the presence of deciduous incisors in the young Walrus. Between the canines he recognized two incisors similar to the molars, which he says the majority of observers had overlooked, because they are not fixed in the intermaxillary, and between these again two pointed small ones in young individuals. He gave the number of molars as four on each side, above and below, and stated that there are neither incisors nor canines in the is 27 inches, 7 of which are grafted into the scull ; its circumference is 8 inches. They stand about three inches asunder in the head, and at their extremities 9 inches apart, bent a little downwards."— H iuiory of Greenland, etc., English translation, London, 1767, p. 126. * Schreber's account is as follows: " . . . . Die ERSTE Gattung, das iii- sonderheit sogennante WALLKOSS, hat zwar, ob gleich kem Schrifsteller etwas davon sagt, zween Vorderziihne in der olern Rinrilade; sie sind aber sehr klein, ragen wenig aus ihren Holen hervor, und werden allem Ansehen uach auserhalb dem Zahnfieische nicht zn bemerken seyn, zuinal da sie nicht am Eande der obern Kinnlade, sondern mehr hineinwiiits stehen. Ich finde sie an einem zur Naturaliensauinilung hiesiger Uuiversitiit gehorigen Wall- rossschiidel; und da derselbe, besage seiner Grosse, von einem juugen Thiere ist ; so glaube ich beynahe gar, dass sie bey zuuelmiendem Alter des Thieres ausfallen und nicht wieder wachsen. Sio kouinien also hier in keine weitere Betrachtung, als dass sie dem Systematiker eiueu Wink geben, dis Thier nicht zu weit von dem Robbeugeschlechte zu entfernen."— SauqetMere Th. ii, p. 260. 1 1 quote the French edition of Camper's Avorks (CEuvres, torn, ii, p. 480, Paris, 1803), tho only one accessible to me. DENTITION. 49 lower jaw.* F. Cuvier gave later also the same dental formula. He deemed tliat the peculiar or anomalous dentition of the Wal- ruses indicated that they were an isolated group, having affini- ties, on the one hand, with the Carnivora, and, on the other, with the Ruminants ! t According to Wiegmann, Budolpliif (in 1802) recognized the1 first of the series of lower grinding teeth as a canine. § Thus, as Wiegmann long since observed, the subject remained till Eapp was so fortunate, in 1828, as to have opportunity to examine a foetal specimen. In this example, he found six inci- sors in the upper jaw and five in the lower (f^). He also expressed it as his belief that the first lower molar should be regarded as a canine, because (1) it was somewhat further removed from the rest than the others were from each other; because (2) of its greater length and thickness in the adult aur- mal ; because (3) it stands close to the temporary or milk inci- sors, and shuts against the outermost of the upper incisors ; and because (4) it lacks the transverse depression seen on the inner side of the crown of the back-teeth. The dental formula recognized by him for the Walrus may be considered as T 3 — 3. n ! — !. M 4~4 A- 3lT3> °' r=Tl? ' -' 3 — 3' Freinery, in 1831, also made reference to the dentition of the Walruses ; but his paper bears mainly upon the question of whether there are one or more species of these animals, and will be further noticed in another connection. He notes partic- ularly the presence, in some of his skulls, of two small molars above, behind the large ones. Wiegmann, 1 1 in 1838, contributed facts additional to those already recorded, but his memoir is largely devoted to a discus- sion of the observations of preceding writers. He assents to Eudolphi's and Eapp's interpretation of the homological rela- tion of the first large tooth of the lower jaw; refers to finding * Begne Animal, torn, i, 1817, p. 168. t Dents des Main., p. 234. t Anatomisch-pliysiologiscne Abhandlungen, p. 145. § Wiegmann says : "... . Ueberdies 1st Rudolph! der erste, der die uuteren Eckziibne erkennt. Er bemerkt niimlich, dass der erste Backenzalin des Unterkiefers sich von den iibrigen durch seine Grosse auszeichne, und wenn auch der Form nach eiuem Backenzaline ahnlicli, doch seiner Grosse nach beinahe fur einen Eckzahn zu halten ware, was spiiter durch Kapp, dem indessen diese Notiz nnbekannt blieb, ausser Zweifel gesetzt 1st." — Arch, fiir Naturg., 1838, pp. 119, 120. || Archiv fiir Naturgesch., 1838, pp. 113-130. Misc. Pub. No. 12 4 50 ODOB^ENUS EOSMAEUS ATLANTIC WALEUS. traces of alveoli of six incisors both above and below, and to the early deciduous character of the last (fifth) upper molar, and the frequent disappearance of the fourth. He concludes that the normal number of the back-teeth is f^|, and that in early life the dentition of the Walrus is not widely different from that of other Pinnipeds. The same year (1838), Macgillivray* considered the normal dentition of the Walrus to be I. §5|j C. ^ ; Pm. + M. §^| =— = 32. His conclusion was based on the examination of a quite young specimen, of which he speaks as follows: "The normal dentition of the Walrus is shown by the skull of a young individual in the Museum of the Edinburgh College of Surgeons. In the upper jaw there are on each side three incisors, the first or inner extremely small, the second a little larger, the third or outer disproportionately large, being equal to the larger grind- ers. The socket of this tooth is placed in the intermaxillary bone, but towards its mouth it is partly formed by the maxillary. The small incisors have deep conical sockets. The canine tooth is displaced, being thrust outwards beyond the line of the other teeth, and causing the peculiar bulging of the head. The lateral incisor is on the level of its anterior margin, and the first grinder is opposite to its middle. There are five grinders, having conical obtuse sockets, and consequently single roots ; the first smaller than the last incisor, the second and third largest, the fourth much smaller, the fifth very small, all shortly conical, and blunt, with enamel on the tip only. The canine tooth is also at first enamelled at its extremity. In the lower jaw there are two very small conical incisors on each side; the canine tooth is wanting; five grinders, with single con- ical compressed roots, and short compressed conical crowns, enamelled at the point ; the first, second, and third nearly equal, the latter being a little larger, the foiirth much smaller, and the fifth very small. The tusks, or enormously developed canine teeth of the upper jaw, are compressed, conical, a little curved backward, directed downwards and a little forwards, and somewhat diverging, but in some individuals, when very long, they again converge towards the points. In adults, the incisors are obliterated, excepting the lateral pair of the upper jaw;, the fifth grinder in both jaws has also disappeared, and sometimes the fourth in one or both jaws." * British Quadrupeds, 1838, pp. 220, 221. DENTITION. 51 Stamiius. in 1842, further contributed to the subject by adding observations respecting variation in the number of the teeth resulting from age, describing in detail the incisive dentition of a series of four skulls of different ages. In two young skulls, the outer temporary incisor of the upper jaw on either side remained; the alveoli of the second pair were still distinct, while the alveoli of the middle pair were nearly oblit- erated. In another, the alveoli of the inner pair of incisors were wholly obliterated ; those of the second pair were barely recog- nizable, while those of the outer pair were distinct, the teeth having fallen later. He was also able to recognize- the alveoli of six incisors in the lower jaws of the skulls just mentioned, and states that he thought Eapp's view of the homology of the first lower back-tooth (considered as a canine) was probably correct. He further takes exceptions to the value of the char- acters assumed by Fremery as the basis of several species of Walruses.* He adds, in respect to the tusks, that in old age they become wholly solid to the base. In regard to the upper molars, he notes the presence of five in several instances, and finds that, as a rule, the fourth disappears before the fifth, or, at least, that its alveolus becomes sooner obliterated. He also confirms the statement previously made by Wiegmann, that the alveoli become filled by depositions of bony matter in concentric layers at the bottom and on the sides. "These alleged specific characters he notices in detail, and considers them, as dependent upon age. He says: "Das Missliche dieser Charaktere erhellt schou aus deui Umstande, dass dieselben nur fur vollig ausgewachsene Thiere, nicht aber fiir junge anwendbar sind, denn das starkste "Wachsthuru der Eckzahne fallt erst offenbar in eine spatere Lebensperiode in welcher nainentlich die beiden innersten Schneidezahne und die beiden letzten Back- zahue jeder Seite der oberen Kinnlade schon geschwunden sind. Hierzu kommt noch der Umstand, dass auch die Eckzahne bei sehr alten Thieren an der Spitze bedeutend abgenutzt sind, demnach in spiiteren Lebeussta- dien an Lange wahrscheinlich wieder abnehmen. Endlich scheint es ja selbst, als ob die Lange dieser Ziihne je nach den Geschlechteru verschieden •ware. " Eben so wenig Gewicht niochte ich auf die Fnrchungen dieser Ziihne legeu. Ihrer Zahl, wie ihrer Starke nach sind sie bei verschiedeuen iibri- gens nicht von eiuander abvreichendeu Individuen verschieden, wie ich mich dnrch Yergleichung einer grossen Anzahl von Walrosszahnen iiber- zeugt habe ; ja diese Furchen sind bisweilen an beiden, bisweilen nur an Einein dieser Z aline spurlos versclrwrinden." Respecting Triclieekus cooki, he adds: " Auch an einem Schadelfragmente des Kieler Museums finde ich etwas convergirende Eckzahne, mochte aber zweifelu, ob dieser Umstaud eine Artunterscheidung rechtfertigt." — Mailer's Archiv far Anaiomie, etc., 1842, pp. 398, 399. 52 ODOBJEXUS ROSMAEUS ATLANTIC WALRUS. Jaeger,* in 1844, described the dentition of three rather young- Walrus skulls from Labrador. In the youngest skull (8£ inches long, G£ inches broad, Paris measure), which had the canines about two inches long, he notes that the last upper molar had already fallen from the right side, but still remained 011 the left, behind which, as well as behind the alveolus of the fourth molar on the right side, was a little shallow pit, in which, dur- ing foetal life, a tooth had perhaps stood. In front of these beforementioned teeth were three molars on each side, and in front of these a conical incisor, and the alveoli of the others were traceable, although already filled with a spongy substance. In the lower jaw, there were five teeth on each side, with traces of three already fallen foetal incisors on one side and of two on the other. The second skull (9f inches by 7|) was somewhat older, the canines being about five inches long. There were present in this skull three upper molars on each side, and a filled-up alveolus behind them. Of these teeth (as also in the other skull), the middle one was the largest and most worn.t In front of these, and somewhat distant from them (5"'), was an incisor on each side, and in addition to these another pair of small conical incisors. In the lower jaw of this skull were, on each side, four 'teeth homologized as C. 1 — 1, M. 3 — 3; Jaeger thus recognizing, as had Eudolphi, Kapp, Wieginann, Freniery, and Stannius, the first of the lower-jaw series as a canine. Behind these were traces of the alveoli of the fourth pair of molars. In the third skull (length 12i inches, breadth 10£), still older, with tusks a foot long, were three upper " back-teeth" on each side, close together, the middle one being the largest, and in front of these a cutting-tooth. The lower jaw had also four teeth on each side, homologized as before. In this skull, there remained no trace of the middle incisors. Another still older skull had the same dental formula as the last. Owen,| in 1853, gave the following formula for the deciduous dentition of the Walrus: 1. 1=| ; C. i^J; M. |^-o=18. This was based on the examination of a young animal, which had died in the Zoological Gardens of London. He, at the same time, proposed the following as the formula of the normal or func- tional dentition of the Walrus: I. l^-J; O.J=-Jj Pin. |^|=18. "* Mailer's Arch., 1844, pp. 70-75. t In the young skulls described by Stannius, the middle molar is mentioned as being uniformly the largest and most worn. t Proc. Zool. Soc. Lond., 1853, pp. 105, 106. DENTITION. 53 Professor Owen, in referring to instances of deviation from this formula, dependent on differences of age and sex, stated "that occasionally a small tooth was found anterior to the normal series of four, and more commonly in the upper than the lower jaw ; and that, more rarely, a small tooth was superadded behind the nor- mal four, in the upper jaw, and still more rarely in the lower jaw; the formula of the dentition of such varieties, in excess, being,— 1. 1=|, C.J=-J, Pm.|=f, M. ^=26." Owen here makes no reference to the literature of the subject, and evidently gave a very erroneous interpretation of the dental formula. In his later references to the subject he gives an entirely different in- terpretation, and one more nearly agreeing with that now com- monly accepted. In his latest reference to the subject,* he writes : " In the Walrus (Triclieclius rosmarus) the normal incisive formula is transitorily represented in the very young animal, which has three teeth in each preniaxillary and two on each side of the fore part of the lower jaw; they soon disappear except the outer pair above, which remain close to the maxil- lary suture, on the inner side of the sockets of the enormous canines, and commence the series of small and simple molars which they resemble in size and form. In the adult there are usually three such molars on each side, behind the permanent incisor, and four similar teeth on each side of the lower jaw ; the anterior one passing into the interspace between the upper incisor and the first molar The canines are of enormous size Their homotype below retains the size and shape of the succeeding molars." The formula of the normal dentition apparently here recognized is : I. |=|; C. j=^ j M. |=|=^=26. Giebel,t in 1855, gave six incisors both above and below as the number existing in the young before and for a short tune after birth. Of these, the lower are said to soon fall out, their alveoli then becoming filled with a bony .deposit. Of the upper inci- sors, the inner pair first disappear, and soon after them also the middle pair, leaving only the outer pair, which begin the rnola- riform series, and to which they are often referred, this outer pair persisting till late in life. The upper canines, he says, are never cast. \ In the lower jaw, the first permanent tooth is regarded as a canine, because it is thicker and rounder than the posterior teeth, and lacks the cross-furrow that marks the oth- * Anatomy of Vertebrates, vol. iii, p. 338. tOdontog., p. 82, pi. 36, fig. 7; Sangeth., p. 1'29. t They are, however, as shown by Malrngren (see beyond), preceded in the embryo by temporary teeth. 54 ODOB^ENUS EOSMARUS - ATLANTIC WALRUS. ers. The young animal lias five upper molariform teeth (" Back- zahne"), the last two of which. are smallest and early disappear, and also later the third, leaving only two behind the canine, and an anterior molariform. incisor. In the lower jaw there are only four "Backzahne" on each side, of which the last and smallest very soon falls away. The dental formula given is as follows: "(§5^) + i + (Hi)" = temporary dentition: 1. 1=|, C. fEij M- iEi> adult dentition: I. £=-J, C. i=-J, M. §=-=. While Giebel accepts the first permanent tooth of the molariform series of the upper jaw as an incisor, and the first in the lower jaw as a canine, he recognizes only two persistent molars on each side, above and below. Malmgreu,* in 1864, figured tile dentition from a foetal speci- men, and published an elaborate paper on the dentition of the Walrus, in which he reviewed at some length the history of the subject, noticing quite fully the writings of the early authors, from Crantz to the Cuviers, and the papers of Eapp, Owen, Wieginann, Mlsson, and other later writers. The formula he presents as that of the permanent dentition is : I. ^~, C. ~^, M- s5|=f =185 and for the deciduous dentition: I. |=|, C. }-=£, The specimen figured shows both the permanent and decidu- ous dentition. The deciduous teeth are most of them separately figured, of natural size, as minute, slender, spindle-rooted teeth, with short, thickened crowns. The permanent teeth are already in place, although even the upper canines had probably not pierced the gum. The middle pair of incisors of both jaws had already disappeared, leaving only their distinctly recognizable alveoli. His specimen appears to have had but a single cadu- cous molar behind the permanent series, from which he assumes the number of upper molars to be 4 — 4 instead of 5 — 5. The following year, Peters $ referred to Malmgren's paper, publishing a plate illustrating the dentition as existing in a some- what older skull (received from Labrador) than that figured by Malmgren. Peters here takes exception to Malmgren's assumed number of back-teeth, which, in accordance with the views of Eapp and Wieginann, Peters behoved should be |^|, instead of | ~. * Ofversigt af Konyl. Vet.-Akad. Forhaudl., 1863, pp. 505-522, pi. vii. t The paper bein.u; published in Swedish, I am iiuable to follow him ill his* discussion of the subject. t Moiiut.sb. K. P. Akad., 1865, pp. 685-687, pi. facing p. 685. DENTITION. 55 According to Peters, Malmgren, from not finding more than four upper back-teeth in any of the many skulls of various ages he had examined, concluded that when a fifth is present it is abnormal. The young skull figured and described by Peters, however, has in the upper jaw the fourth and fifth back- teeth still in place on the right side, and the fourth on the left side, with an alveolus of a fifth. This Peters considered as affording new proof of the correctness of Wiegmann's formula. As already noticed, five molars have been recognised by Fremery, Eapp, Giebel, and Owen, and, though perhaps not always pres- ent, are frequently to be met with. The dental formula of the Walrus, as determined by Eapp and Wiegmanu, has been adopted by Van der Hoeven * and Blasius, t as well as by Peters, and essentially by Giebel. Gie- bel, however, gives only four deciduous lower incisors, instead of six. Owen, in his later works, agrees in this point with Gie- bel, but takes apparently no cognizance of the deciduous fourth and fifth molars, to which he refers, however, in his earlier papers. Gray, f in 1866, although quoting the formula given by Eapp, adopts the following : " Cutting teeth -| in young, £ in adult 5 grinders ~^ in adult, truncated, all single-rooted ; ca- nines, upper very large, exserted." He, however, quotes Eapp's formula, and also that given by Owen in his " Catalogue of the Osteological Series of the Museum of the Eoyal College of Sur- geons" (1853, p. 630). Professor Flower, § in 1869, gave a diagram of the dentition of the Walrus based on many observations made by himself and on " those of others, especially Professor Malmgren," in which both the temporary and permanent dentition is indicated as fol- lows : Milk dentition : I. |=|, C. i=J, M. J^ j permanent denti- tion: I. ±=±. C. i=^, M. |=|. He adds that « it is probable that 0 — 0' 1 — X' o — <> an anterior rudimentary incisor is developed in the upper if not in the lower jaw," making the temporary incisors hypotheticaUy f^f . " I believe," he says, " that the rudimentary milk teeth never cut the gum, and are absorbed rather than shed. This process commences before birth, The rudimentary teeth, however, in front of and behind the large teeth are not * Lehrbuch der Zoologie, 1856, p. 733, English ed. t Sikigcthiere Deutschlands, 1857, pp. 261, 262. t Cat. Seals aiid Whales, p. 35. $ Jouru. Auat. and Phys., iii, p. 272. 56 ODOB^ENUS EOSMARUS ATLANTIC WALRUS. unfrequently persistent to extreme old age, although commonly lost in macerated skulls. These rudimentary teeth are usually described as l milk-teeth '; even the posterior ones are some- times so called, but it appears to me an open question whether they do not rather represent permanent teeth in a rudimentary or aborted condition." Huxley, in his "Anatomy of Vertebrated Animals" (pp. 300, 361), published in 1872, adopts the following as the dental for- mula of the Walruses : « I. i=-J, C. £=-J, p. m. m. |=| + ^J." He says : " The dentition of the Walrus is extremely peculiar. In the adult, there is one simple conical tooth in the outer part of the premaxilla, followed by a huge tusk -like canine, and three, short, simple-fanged teeth. Sometimes, two other teeth, which soon fall out, lie behind these, on each side of the upper jaw. In the mandible there are no incisors, but a single short canine is followed by three similar, simple teeth, and by one other, which is caducous." * Both here and in the formula no reference is made to the deciduous incisors, although the caducous molars are recognized. In the foregoing resume, we have seen how vague was the information bearing on this subject possessed by all writers prior to about the beginning of the present century ; IIOAV the earlier notices of the existence of incisors in the young were overlooked and rediscovered by later writers, as well as how slowly the first permanent tooth of the molariform series of the upper jaw came to be generally recognized as a true incisor and not a molar ; how, later, the number of incisors in the young was found to be six in the upper jaw and six in the lower jaw, with, as a rule, two small caducous molars on each side in the upper jaw, and one on each side in the lower behind the permanent grinding teeth; that the first permanent molariform tooth of the lower jaw was a canine and not a molar; and that by dif- ferent writers the number of incisors recognized in the lower jaw has been sometimes four and sometimes six, and the cadu- cous upper molars regarded sometimes as one and sometimes as two. Finally, that the true formula of the full dentition of the Walrus is I. |=|; C. ±=-\; Pm. M. f5|=g=34. It hence appears that the dentition of the Walruses is peculiar and some- what abnormal in four features, namely, (1) the early disap- pearance of all the incisors except the outer pair of the upper * Aiiat. Vertebr. Anim., pp. 3GO, 361. FOSSIL EEMAINS. 57 jaw, (2) the enormous development of the upper canines, (3) the slight specialization of the lower canines, and (4) the caducous character of the two posterior pairs of molars of the upper jaw and the posterior pair in the lower jaw. The early dentition of the AValrus differs mainly from that of most other Pinnipeds in having six lower incisors instead of four, the incisive formula of other Pinnipeds, as generally recognized, being usually |^|, fre- quently f^J, and sometimes (as in MacrorMnus and Cystophora) |=|? — never, at least in the permanent dentition, |=|, but I am far from sure this number may not sometimes appear in the deciduous dentition. In the Sea Otter (Enliydris], there are said to be six lower incisors in the young, while only four are present in adult life. The middle pair of lower incisors so early disappear that even in very young specimens they are some- tunes wanting. Eapp found in a foetal specimen three on one side and only two on the other, and quite a number of promi- nent writers on the subject have recognized two pairs of lower incisors as the normal number. In many specimens, the alveoli of three pairs have been found, and, in addition to the instances already given, I may add that there is a young skull in the Museum of Comparative Zoology that shows decided traces of three pairs, the outer incisor on one side being still in place. In view of all that is at present known respecting the sub- ject, I adopt the following formula} as being well-established, — premising, however, that they are substantially in accord with the view of the case presented by Professor Flower in 1869 : — Temporary dentition: I. |=|; C.^; M. ^ = g = 32. Permanent dentition : I. ^ ; 0. }={ j M. f^f = g = 26 ; the last two upper molars and the last lower one on each side being rudimentary and often absent. FOSSIL EEMAINS. — Remains of the Atlantic Wajrus, in a fossil state, have been found at various points along the Atlantic coast from Maine to South Carolina, and in Europe as far south as England and France. The first noticed from American locali- ties was thus mentioned by Barton in 1805, but the locality is not given. He says: "The bones of one of these large animals have been found. These appear to have belonged to a species of trichechus; perhaps to the triclicclms rosmariis or morse."* Messrs. Mitchill, Smith, and Cooper described, in * London Phil. Mag., vol. xxxii, 1805, p. 98. 58 ODOB^NUS ROSMARUS ATLANTIC WALRUS. 1828,* a specimen consisting of the anterior portion of a skull, found on the sea-beach in Accomac County, Virginia. The same specimen was also described later by Harlan.t These writers all considered it as bearing the closest resemblance to the corre- sponding portion of the skull of the existing Walrus, to which they doubtfully referred it ; but later it was regarded by DeKay as representing a distinct species, to which he gave the name Triclieclius mrginianus.\ In 1844, Lyell described a tusk ob- tained from the Tertiary Clays of Gardiner, Maine, which Owen regarded as probably belonging to an extinct species.§ Lyell 1 1 also refers to a skull he obtained at Martha's Vineyard, Massachusetts. He describes this skull as "differing from skulls of the existing species (TricJiechus rotund run, Linn.), with which it was compared by Professor Owen, in having only six molars and two tusks, whereas those of the recent have four molars on each side, besides occasionally a rudimentary one. The front tusk is rounder than that of the recent walrus." fl In 1857, Dr. Leidy** described and figured a skull found on the sea-beach at Long Branch, Monmouth County, New Jersey, where it was obtained by Prof. J. F. Frazer in 1853, and refers to another specimen (consisting of the facial portion of a skull) discovered at the same locality by Prof. Geo. H. Cook. The first-named specimen, says Dr. Leidy, " has lost a portion of the cranium proper, and the exserted portion of one tusk, but other- wise, except being a little water-worn, is in a good state of preservation. It is unchanged in texture, and nearly so in colour; and it belonged to an old individual, as all the sutures are completely obliterated. The form of the facial portion of this specimen corresponds with that of the specimen from Vir- ginia, [described by DeKay and preceding writers,] above men- tioned ; and the entire skull closely resembles that of the recent Walrus, Trichechus rosmarus, as represented in the figures of Daubenton, Cuvier, and De Blainville ; and its measurements are also sufficiently near those given by the first-named author to recognize it as the same species. * Ann. Lye. Nat, Hist. New York, vol. ii, 1828, p. '271. tEilinb. New Phil. Joimi., vol. xvii, 1834, p. 360. iNat. Hist. New York, Zoology, pt. i, 1842, p. 56, pi. xix, figs. 1, a, I. 5 See Packard, Mem. Bost. Soc. Nat. Hist,, vol. i, 18G7, p. '240. ||Amer. Jouru. Sci. and Arts, vol. xlvi, 1844, p. 319. IT As is well known, the existing Walrus has occasionally only Ihe number of teeth found in the Martha's Vineyard specimen. ** Trans. Am. Phil. Soc., vol. xi, 1857, p. 83, pis. iv, v, tig. 1. FOSSIL KEMAINS. 59 " The tusks in tlie fossil curved downwardly in a diverging manner, and were about four inches distant from each other at their emergence from the alveoli, and ten inches at their tips. The remaining tusk in the specimen is thirteen inches long from its alveolar border, and in this latter position it is three inches in diameter antero-posteriorly and one and three quar- ters inches transversely . . . ." The other specimen, from New Jersey, mentioned above, he says is also "unchanged from its original texture, but is brown from the infiltration of oxide of iron. It also belonged to an old individual, as all the sutures are obliterated, and the third molars together with the greater part of their alveoli are gone. In its anatomical de- tails the specimen agrees with the corresponding portion of Professor Frazer's specimen, except it is an inch and a half broader in the position of the canine alveoli, and the antero- posterior diameter of the tusk is rather less."* Of both these specimens, Dr. Leidy gives figures, and they agree entirely with corresponding parts of the existing North Atlantic species. Dr. Leidy, however, notes differences between these specimens and those of the Walrus of the North Pacific. Dr. Leidy adds : " An important question now arises in rela- tion to the age or geological period to which the three Walrus skulls, thus discovered oil the coast of New Jersey and Virginia, belong. As they appear to be of the same species as the recent Triclieclius ros-marus, which once lived in great numbers in the Gulf of St. Lawrence, they are most probably the remains of individuals that were once floated upon fields of ice southerly, and left on the present United States coast. Or, perhaps they may be the remains of the same species which probably during the glacial period extended its habitation very far south of the latitude in which it has been found in the historic period.'1! In view of the now well-known former extension of the habitat of the Moose, Caribou, Eeindeer, Musk Ox, and other northern mammals, southward to Kentucky, the latter hypothesis seems the more probable one, and that the species in glacial times inhabited the eastern coast of the United States southward to Virginia, if not even beyond this point. More recently, Dr. Leidy has announced the occurrence of Walrus remains in the phosphate beds of Ashley Eiver, South Carolina, and has described and figured a tusk from that locality. * Trans Anier. Phil. Soc., vol. xi, pp. 83, 84. tlbid., p. 84. 60 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. "This specimen," lie says, "is as black as ebony, dense, heavy, and brittle, and is nearly complete, except at the thin border of the pulp cavity. The curvature is slight, and it indicates the tooth to be of the left side." He gives its dimensions or length externally, following the curvature, thirteen inches; near the root it has an antero-posterior diameter of three and five-eighths inches, and a transverse diameter of one and three-fourths inches, and at the middle the transverse diameter is two and one-eighth inches, while the antero-posterior diameter is about the same as at the base. "In robust character," he adds, "the tusk quite equals those of the largest mature recent skulls which have come under my observation, but is much shorter and more ab- ruptly tapering. The specimen looks like what we might sup- pose the tusks of the living animal would be were they broken off near the middle and then worn away little more than one- fourth the length in a curved line deflected from the course of the anterior longitudinal convexity to the tip. The comparative brevity of the tusk and its worn condition at the end may per- haps have depended upon just such an accident and subsequent wear. In a mature skull from the shore of Sable Island, and preserved in the Museum of the Academy, the tusks, which are of the usual size, are worn in the same manner as the Ashley specimen for more than half their length." After describing in detail the fluting of the tusks, and the variation noticeable in this respect in different skulls of the liv- ing Walrus, he concludes that, while the fluting differs some- what in the fossil tusks from that usually seen in the tusks of the existing animal, these differences cannot be considered as having specific value. In referring to DeKay's " Triclieclms vir- gin ianus," he says: "No remains of an undoubtedly extinct species known to me have been discovered anywhere." He finally adds, respecting the Ashley fossil, that " it is an inter- esting fact to have learned that this [the living] or a closely re- lated species formerly existed so far south as the Ashley Eiver, South Carolina."* The discovery of the greater part of the skeleton of a Walrus, including the skull, with the tusks over five inches long, and all the teeth except two, in the Quaternary Clays at Portland, Me., was made during July of the present year (1878). It was found in excavating for the foundation of the new "Boston & Maine" transfer station, at about seven feet from the surface. * Journ. Acacl. Nat. Sci. PMla., 2d ser., vol. viii, 1877, pp. 214-216, pi. xxx, tig. 6. FOSSIL EEMAINS. 61 " It was partially imbedded in a layer of blue clay a foot in thick- ness, overlaid by a layer of lighter clay two feet two inches thick, containing- casts and shells of My a arenaria, Macoma subulosa, Mytylus cdulis, Cardium (Serripes) grcenlandicum, Astarte trun- cata, Saxicava distorta. Nucula antiqua, Leda tenuisulcata, L. truncata, Natica clausa and pusilla, and Balanus. The skeleton is in the Museum of the Portland Society of Natural History."* In Europe, Walrus remains were reported by Cuvier t as found at Angers, France, but Gervaisf found later that the only por- tion of those remains accessible to him belonged not to the Walrus, but to the Halitlierium. In 1858, however, a part of a cranium was described by Gra- tiolet, from the diluvial deposits of Montrouge, near Paris. He, however, considered it as distinct from the existing species, even generically, and gave it the name Odobenotlierium lartetianum.§ In 1874, a nearly entire skull was described by Defrance, from similar deposits near the village of Sainte-Menehould, Marne, which he not only considered as identical with the living species, but also referred the fragment previously described by Gratiolet to the same species. Eespecting these specimens he says : " En comparant entre elles les tetes du Triclieclms rosmarus de nos iners, de V Odobenotlierium Lartetianum et du Triclieclms de Sainte-Menehould, on leur trouve une resseinblance aussi com- plete que possible, sauf en ce qui concerne la forme et le vo- lume de I'apophyse niastoi'de, point qui presents des differences assez sensible. On sait que dans le T. rosmarus cette apophyse est tres-grande, presque verticale, et saillante la partie infe- rieure du crane; celle de 1' Odobenotlierium, e'galement tres- volumineuse, se prolonge presque horizontalernent en arriere, sans de"passer le crane inf6rieurement ; celle du Triclieclms de Sainte-Menehould pr6sente un volume plus consid6rable encore que dans les deux autres, sans se prolonger en arriere coinme dans 1' Odobenotlierium , rnais inferieurement comme dans le Triclieclms actuel. Ces nuances 16geres indiquent evideminent une e"troite parent^ entre ces trois individus ; aussi est-il diffi- cile de conrprendre que Gratiolet ait voulu 6tablir un nouveau genre sur des particularites pen accentue"e que celles que lui pre"sentait la portion de crane dont il etait possesseur, et qui ne " American Naturalist, vol. xii, p. G33, Sept. , 1878 ; see also Portland (Maine) Argus, of July — , 1878. tOssem. Foss. tZool. et Pal6out. Frangaises, 1859, p. 88. {Bull. Soc. Ge"ol. de France, 2« s6r., xv, 1858, p. 624. 62 ODOB.ENUS EOSMARUS — ATLANTIC WALRUS. sont d'ailleurs que des particularity relatives pour la plupart & Page et au sexe, ainsi que 1'a etabli M. Gervais." * Van Beuedent refers to Gratiolet's specimen at some length, giving its full history and exposing its true character. H e says : " On a trouve a Montrouge, pres de Paris, il y a quelques an- nees, uu crane dont on s'est beancoup occupe et que Gratiolet a decrit sous le nom d' 'Odobenotlidre. Lartet Pavait remis a Gratio- let. Nous avons examine cette tete avec tout le soin ne"cessaire et nous partageons completeinent 1'avis que M. Paul Gervais a expriine" a son sujet dans la Zoologie et la Pdleontologie frangaises (p. 88), c'est-a-dire, que ce crane fracture" et qui a subi Faction du feu, n'est autre chose qu'un crane de Morse vivant qui e"te" rapport6 du Nord. "Nous avons e~tudie cette piece avec M. Paul Gervais, ay ant devant nous tons les e"16nients de comparaison que possede le Museum, et c'est apres avoir serieusemeut h6site~ si POdobe"- uothere est un Morse ou non, que nous nous soinme range de Pavis de notre savant confrere. " Get Odobenothere repose sur un fragment de crane dont la cavite" ce"rebrale a ete utilisee pour uu usage quelconque et qui aura ete apporte dans cet etat par quelque pecheur du Nord. C'est le cote" droit et non le cote gauche qui est conserve. " Celui qui se trouve devant ce fragment de crane et qui a devant lui un choix de sections des diverses regions de la tete, comprend aisement comment a pu se tromper. " L'importance que Le Hon a attach^e a la presence de cette tete dans le Diluvium rouge, tombe ainsi completemeut ; a propos de la periode glaciaire, Le Hoii avait accord^ une grande valeur a cette pr^tendue de~couverte de gratiolet."| Lankester, in 1865, described fossil tusks, from the Eed Crag of England, of an animal evidently closely allied to the Walrus. He enumerates no less than twelve or fifteen specimens of these remains, mostly fragments, collected from various localities, all from the so-called " Eed Crag " formation of England, or its equivalent. The principal localities are Button, Felixstow, and Bawdsey, in England, but he refers also to their occurrence at different points in Belgium. The majority of the specimens of the tusk obtained, writes Mr. Lankester, " are its pointed -Bull. Soc, G(Sol. de France, 3e ser., ii, 1874, pp. 169, 170. tDescrip. des Ossements Fossiles des Environs d'Anvers, Ann. Mus. d'Hist. Nat. de Belgigue, i, 1877, pp. 40, 41. t "LE HON, L'liominc fossile, 1867, p. 304. — ID., Mouvement des mers . . . . , p. 48, 1870." FOSSIL REMAINS. G3 terminations ; but other specimens, of the base and intermedi- ate portions, have come to light. Throughout its length," Mr. Laukester continues, " which in some examples must have been fully three feet, the tusk is slightly curved ; but in those which appear to be fully grown the curve is considerably greater towards the terminal point, the direction of the curve probably giving the tusk, if its Pinuigrade affinities be established, a retroflected position, as in the Ditiotherium. The Crag tusk is very much compressed laterally, so that its transverse sec- tion has an elliptical outline, whilst that of the D mother ium- tusk is nearly circular. The amount of lateral compression is, however, extremely variable, as it is also in the living Wal- ruses; the amount also of the lateral as well as the antero- posterior flection of the tusk appears to vary, as in the recent TrichectiSj the variability of which in the size and form of its tusks is well known. A single large furrow on the outer sur- face, two on the inner, and one on the inner curved margin, ex- tend along the whole length of the tusk in many specimens, exactlv similar to those on some tusks of Walrus ; but in both »^ / the recent and fossil specimens they are subject to much varia- tion, in their major or minor development. Xo appearance of any wearing of the point of the tusks by use during life is observa- ble ; and indeed the greater backward curvature of that part seems to result from its freedom from usage, since in the Walrus the point of the tusk is rapidly worn away, which of course checks any tendency to curvature which might become appar- ent if the tusk were not used against such hard substances as rocks and blocks of ice. " From an examination of the general contour and form of the tusks, without regard to their substance or structure, one would unquestionably be led to regard them as belonging to an animal similar to the existing Walrus, inasmuch as it is in this animal alone that this form of tusk, with its longitudinal furrows, great length, and gentle curvature, is found." After describing in detail the structure of these fossil tusks, as shown in sections and as revealed by the microscope, Mr. Lankester further observes: "In its microscopical structure, the dentine of the fossil tusks presents a complete resemblance to that of the Walrus.* .... The dentinal tubes are very "Their microscopical structure, as well as external form, are illustrated by numerous ligtires, forming plates x ami xi, accompanying Mr. Laiikes- ter's paper. 64 ODOBJSNUS EOSMARUS ATLANTIC WALRUS. nearly of the same size, and equally closely packed, and are connected with stellate lacunae in some numbers near the peri- phery of the tooth. This structure, which is not peculiar to the Walrus, is, nevertheless, a test of affinity, inasmuch as the form of the lacunae varies in different animals. They are not met with in the tusks of the Proboscidea or the Hippopotamus, but occur in the curious incisors of the Dugong. The ' dentiual cells 'of the Crag tusks also resemble those of the Walrus In structure the cement exactly resembles that of the Walrus, displaying vascular canals, bone-lacunae, and canaliculi, of the same form and disposition • but the proportion which it bears to the thickness of the other tooth-tissues appears to be larger in the Walrus than in the fossil. "From the foregoing remarks it will be apparent that we have in these fossil tusks characters which ally them most closely to the large canines of the genus Trichecus." After enumerating the points of form and structure which distinguish these tusks from those of other animals, and those which assimilate them to those of the Walrus, he thus generalizes the results of his investiga- tions : " Lastly, they resemble the large canine tusks of the living Tr-ichecus in their curvature, varying lateral compression, large surface-furrows, short and wide pulp-cavity, globular ' osseo-dentine', and every detail of minute structure. They differ from them in their greater curvature at the point of the tusk, their greater lateral compression, and minor development of cement. " I accordingly propose to establish the genus Trichecodon to receive the animal thus indicated. The justification of a gen- eric separation must be sought in the fact of the great antiquity of the Bed Crag, and the consequent probability of the associ- ation of other and more distinctive attributes with those of the tusks." As regards its geological position and associations, Mr. Laii • kester adds: "It appears that the Tricliecodon Huxley i, like the Cetacean remains of the Crag and large Sharks' teeth, is a derived fossil in the Bed Crag, belonging properly to the Middle Crag, which is not now observable in this country [Eng- land], but is well developed at and near Antwerp."* It thus appears that Mr. Lankester was as much, or more, in- fluenced in his generic differentiation of these fossils from their * Quarterly Journal of the Geological Society of London, xxi 18P5 pp 22G-231. GEOGRAPHICAL DISTRIBUTION. 65 nearest living allies, by the geological evidence of their antiq- uity as by the actually observed and admittedly slight differ- ences of form and structure. Mr. Lankester does not inform us respecting the locality whence came his specimens of the tusks of the living Walrus with which lie compared the fossil tusks. In this connection it may be added (see further on this point the account of Odobcenus obesus given beyond) that the tusks of the Pa- cific species (Odobccnus obesns) are not only longer and slenderer than those of the Atlantic species (0. rosmarus), but are sharper- pointed and more incurved, and do not present the worn and broken appearance so often (indeed, usually) seen in the tusks of old individuals of the latter. Whether or not they present differences of structure has not, so far as known to me, been microscopically determined. The tusks of the Pacific species, furthermore, sometimes attain the size indicated for the tusks of " Tricliccodon huxleyi." For the present I must consider Lan- kester's Trichecodon liuxleyi as certainly not generically separ- able from the existing Walruses, although it may have differed from the existing Atlantic species in larger size and possibly in other characters, as so often happens among the immediate pro- genitors of existing species in other groups of mammals. Van Benedeu has recently reviewed at considerable length the history of the supposed and actual fossil remains of the Walrus,* showing that most of those reported as found in differ- ent parts of France and Germany were really those of different species of extinct Sirenians or other animals than the Walrus. Van Beneden, however, describes and figures a dorsal vertebra he considers as that of the Walrus, found near Deurne, and a sca- phoid bone from Anvers. GEOGRAPHICAL DISTRIBUTION, PRESENT AND PAST. — 1. Coast of North America. — As already shown (antea, pp. 57-61), the Walrus, like the Musk Ox, the Caribou, and the Moose, ranged during the great Ice Period much beyond the southern limit of its boundary at the time the eastern coast of North America was first visited by Europeans. While its remains have been found as far south as New Jersey, Virginia, and even South Carolina, there is no evidence of its existence on the New Eng- land coast within historic time, or during the last three hun- dred and fifty years. During the last half of the sixteenth cen- tury they are known to have frequented the southern coast of * Ann. Mus. d'Hist. Nat. de Belgique, i, 1877, pp. 39-42. Misc. Pub. No. 12 5 66 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. Nova Scotia, as well as the shores and islands to the northward ; but this appears to have been at that time their southern limit of distribution. In May, 1534, they were inet with by James Cartier, about the island of "Ramea" (probably Sable Island), who thus refers to them: "About the said Island [Rarnea] are very greate beastes as great as oxen, which have two -great teeth in their mouths like unto Elephants teeth, & live also in the Sea. We saw them sleeping upon the bauke of the water : wee think- ing to take it, went with our boates, but so soone as he heard us, he cast hmiselfe into the sea." * They were afterward hunted here for their tusks and oil. Thus Richard Fischer, in speaking of the same island, says: "On which Isle [of Ramea] are so great abundance of the huge and inightie Sea Oxen with great teeth in inoneths of April, May and June, that there have bene fifteene hundreth killed there by one small barke, in the year 1591." t The same writer tells us that George Drake, two years later, "found a shippe of Saint Malo three parts freighted with these fishes." Another writer says that he had seen a "dry flat full at once" of their teeth, "which are a foote and sometimes more in length." They also, at about the same time, frequented the so-called "Bird Islands" off Cape Breton. Says Charles Leigh : " Upon the lesse of these Islands of Birds we saw greate store of Morsses or Sea Oxen, which were a sleepe upou the rockes : but when we approached nere unto them with our boate they cast themselves into the sea and pursued us with such furie as that we were glad to flee from them." It is later said that the number of these " Sea Oxen" was '/about thirty or forty ."f From the accounts of other writers we learn that these "Sea Oxen" were accustomed to resort to these various islands during April, May, and June, for the purpose of bringing forth their young. Thus, "Thomas James of Bristoll," in speaking of the "Isle of Raniea," says it was situated "in 47 degrees, some fiftie leagues from the Grand Bay, neere Newfoundland: and is about twentie leagues about, and some part of the Island is flat Sands and shoulds : and the fish conuneth on bankc (to do their kiude) in April, May & June, by numbers of thousands, which fish is very big : and hath two great teeth : and the skinue of them is like Buffes leather : and they will not away from their yong ones. The yong ones are ? Hakluyt, Voyages, vol. iii, p. 254. t Ibid. , p. 238. t Ibid. , pp. 242, 249. GEOGRAPHICAL DISTRIBUTION. 67 as good meat as Veale. And with the bellies of five of the saide fishes they make a hogshead of Traine, which Traiue is very sweet, which if it will make sope, the king of Spaine may burue some of his Olive trees." * Charlevoix also alludes to the Walrus fishery at Sable Island, which the English at one time established there, but says it was soon abandoned, being found unprofitable.! Molineux Shuldhain has left us quite a full account (and one that has been often quoted) of the habits of these animals, and of the wholesale destruction by which they were speedily extirpated from the Atlantic coast south of Labrador. This account, writ- ten in 1775, says: "The sea-cow is a native of the Magdalen Islands, St. John's, and Anticosti in the Gulph of St. Lawrence. They resort very early in the spring to the former of these places, which seems to be by nature particularly adapted to the wants of these animals, abounding with clams of a very large size, and the most convenient landing-places, called Echouries. Here they crawl up in great numbers, and sometimes remain for fourteen days together without food, when the weather is fair; but on the first appearance of rain, they immediately retreat to the water with great precipitation. They are, when out of the water, very unwieldy, and move with great difficulty. They weigh from 1500 to 2000 pounds, producing, according to their size, from one to two barrels of oil, which is boiled out of a fat substance that lies between the skin and the flesh. Im- mediately 011 their arrival they calf, and engender again about two months after; so that they carry their young about nine months. They never have more than two at a time, and seldom more than one. " The echouries are formed principally by nature, being a grad- ual slope of soft rock, with which the Magdalen Islands abound; about 80 to 100 yards wide at the water side, and spreading so as to contain, near the summit, a very considerable number. Here they are suffered to come and amuse themselves for a con- siderable time, till they acquire a boldness, being at their first landing so exceedingly timid as to make it impossible for any person to approach them. In a few weeks they assemble in great numbers ; formerly, when undisturbed by the Americans, to the amount of seven or eight thousand ; and the form of the echotirie not allowing them to remain contiguous to the water, the foremost ones are insensibly pushed above the slope. When * Hakluyt, Voyages, vol. iii, p. 237. t Charlevoix, vol. v, p. 21G. 68 ODOBJ3NUS ROSMARUS ATLANTIC WALRUS. they are arrived to a convenient distance the fishermen, having provided the necessary apparatus, take the advantage of a sea wind, or a breeze blowing rather obliquely on the shore, to prevent the smeUing of these animals (who have that sense in great perfection, contributing to their safety), and with the as- sistance of very good dogs, endeavour in the night time to sepa- rate those that are the farthest advanced from those next the water, driving them different ways. This they call making a cut, and is generally looked upon to be a most dangerous process, it being impossible to drive them in any particular direction, and difficult to avoid them ; but as they are advanced above the slope of the echourie, the darkness of the night de- prives them of every direction to the water, so that they stray about and are killed at leisure, those that are nearest the shore being the first victims. In this manner there has been killed fifteen or sixteen hundred at one cut. They then skin them, and take off a coat of fat that always surrounds them, which they dissolve by heat into oil. The skin is cut into slices of two or three inches wide, and exported to America for carriage traces, and to England for glue. The teeth is an inferior sort of ivory, and is manufactured for the same purposes, but soon turns yellow."* According to Dr. A. S. Packard, jr., its bones are still found at the localities mentioned by Shuldham. "According to tra- dition," he further says, uit also inhabited some of the harbors of Cape Breton ; and I have been informed by a fisherman in Maine, whose word I do not doubt, that on an islet near Cape Sable, Nova Scotia [probably the "Isle of Kamea" of the early voyagers already quoted], its bones are found abundantly on the sandy shore, fifteen to twenty feet above the sea. In the St. Lawrence Gulf they were exterminated during the middle of the last century. The last one seen or heard of in the Gulf, so far as I can ascertain, was killed at St. August- ine, Labrador, twenty-five years since. One was seen at Square Island fifteen years since, and two shortly before that, and another was killed at the same place about eight years since. I saw the head of a young Walrus, which was found floating, dead, having been killed, apparently by a harpoon, in the drift ice north of Belle Isle."t Dr. J. Bernard Gilpin, writing a few years later (in 1869), in referring to the former occurrence of the Walrus on the shores * PMl. Trans., vol. Ixv, p. 249. t Proc. Bost. Soc, Nat. Hist,, vol. x, 1866, p. 271. GEOGRAPHICAL DISTRIBUTION. 69 and islands of the Gulf of St. Lawrence, says: "At Miscou, Bay Chaleur, Perley found only their bones, but in such numbers as to form artificial sea beaches. These were doubtless victims of 'The Royal Company of Miscou', founded during the earlier part of the seventeenth century, by the King of France, and whose ephemeral city of New Rochelle, numbering at one time some thousands, has passed away leaving no sign. The mur- dered Sea-horses have left a more enduring monument than the murderers." He further adds: "Though we have no accounts later than the seventeenth and eighteenth centuries of their inhabiting Sable Island, yet it is very probable that they continued to resort there until they entirely left these latitudes. Its difficulty of access, its being uninhabited, and its sandy bars fringed with a ceaseless surf, point it out as their last hold."* Dr. Gilpin also records the capture of a Walrus in the Straits of Belle Isle, Labrador, in March, 1869, which was dragged on the ice for five miles, and then taken by ship to St. John's, Newfoundland, and thence to Halifax, Nova Scotia, where it was described and figured by Dr. Gilpin. t Mr. Reeksf states that a "specimen was driven ashore in St. George's Bay," New- foundland, about 18C8, and alludes to the frequent occurrence of their bones along the Newfoundland coast. It is still an inhabitant of the shores of Hudson's Bay, Davis Strait, and Greenland, where, however, its numbers are annu- ally decreasing. In Greenland, according to Mr. Robert Brown (writing of its distribution in 1867), "it is found all the year round, but not south of Rifkol, in lat. 65°. In an inlet called Irsortok it collects in considerable numbers, to the terror of the natives, who have to pass that way. ... It has been found as far north as the Eskimo live, or explorers have gone. On the western shores of Davis's Strait it is not uncommon about Pond's, Scott's, and Home Bays, and is killed in considerable numbers by the natives. It is not now found in such numbers as it once was ; and no reasonable man who sees the slaughter to which it is subject in Spitzbergen and elsewhere can doubt that its days are numbered^ It has already become extinct in several places where it was once common. Its utter extinction is a foregone conclusion."§ *Proc. and Trans. Nova Scotia Inst. Nat. Sci., vol. ii, pt. 3, pp. 126-127. tlbid.. pp. 123-127, with a plate. |. ZoGiogist, 1871, 2550. §Proc. Zool. Soc. Lond., 1868, p. 433. / A ,X \ 70 ODOB^NUS EOSMARUS ATLANTIC WALRUS. Kane and Hayes, during the years 1853 to 1855, found the Walrus very abundant about Port Foulke, on the western coast of Greenland, in latitude 79°, but they seem to have, since that date, greatly decreased in numbers along the whole of the Greenland coast. Captain Feilden, in his paper on the "Mam- malia of North Greenland and Grmnell Land," observed in 1875 by the British Arctic Expedition, after alluding to their former abundance about Port Foulke, as observed by Kane and Hayes, says: " Curiously enough, we did not see one of these animals in the vicinity of Port Foulke nor in Smith Sound, until we reached Franklin Pierce Bay. There, in the vicinity of Nor- man Lockyer Island, we saw several Walruses, and killed two or three. . . . Near Cape Fraser I saw a single Walrus; but as far as my observation goes, it does not proceed further north than the meeting of the Baffin Bay and Polar tides near the above mentioned Cape."* Mr. Ludwig Kumlien, naturalist of the Howgate Polar Ex- pedition of 1877, states :t "The Walrus is quite common about Cape Mercy and the southern waters of Cumberland Sound,, but at the present day rarely strays up the Sound. Their re- mains, however, are by no means rare, even in the greater Kingwah, and many of the old Eskimo hut foundations contain the remains of this animal. The Eskimo say they got mad and left. Certain it is, they are found around Annanactook only as stragglers at the present day. Considerable numbers were observed on pieces of floating ice near Cape Mercy, in July. About Nugumeute they are largely hunted by the Eskimo living there." Eespecting their occurrence more to the southward, on the Greenland coast, Dr. Sink states: "The Walrus is only rarely met with along the coast, with the exception of the tract between 66° and 08° N. lat., where it occurs pretty numerously at times. The daring task of entering into contest with this animal from the kayak on the open sea forms a regular sport to the natives of Kangamiut in GG° N. lat. The number yearly killed has not been separately calculated, . . . but they can hardly exceed 200." t The westernmost point at which it has been observed is said to be the western shore of Hudson's Bay. Mr. J. C. Eoss states it to be an inhabitant of the west coast of Baffin's Bay and *Tho Zoologist, 3d ser., vol. i, p. 360, September, 1877. t Iii MSS. notes lie has kindly placed at my disposal. t Danish Greenland, its People and its Products, pp. 126-127, 1877. GEOGRAPHICAL DISTRIBUTION. 71 Repulse Bay, and to be occasionally met with in the northern part of Prince Regent's Inlet, but says it is unknown to the natives of Boothia.* Dr. Richardson says : "The Walruses were very numerous at Igloolik and on the other parts of the coast to the eastward of the Fury and Hecla's Strait. They are riot found, however, at the month of the Copper Mine River, although the black whale had been sometimes drifted thither."! He also refers to its being unknown to the Eskimos of the Coppermine and Mackenzie Rivers, f No species of Walrus appears to have ever been seen on the Arctic coast of America between the 97th and 158th meridians, or for a distance of about sixty degrees of longitude. 2. Coast of Europe. — On the western shores of Europe the Wal- rus has been taken at no remote date as far south as Scotland,§ and Mr. Robert Brown, in 18C8, stated that he suspected it to be a "not unfrequent visitor " to the less frequented portions of the Scottish shores, he considering it probable that " not a few of the i Sea-horses ' and ' Sea-cows ' which every now and again terrify the fishermen on the shores of the wild western Scottish lochs, and get embalmed among their folk-lore, may be the Walrus." 1 1 Fleming states that one was killed in the Sound of Stockness, on the east coast of Harris, in December, 1817, fl while another, according to Macgillivray and others, was killed in Orkney in June, 1825. ** Mr. R. Brown adds that one was seen in Orkney in 1857, and another in IsTor' Isles about the same time, tt It appears to have never occurred in Iceland, ex- cept as a rare straggler. Many years ago they are said to have lived on the shores of Fiuinark, and at a much later date to have abounded on some of the islands off this coast. Mr. Lainont says: "We learn from the voyage of Ohthere, which was per- formed about a thousand years ago, that the Walrus then abounded 011 the coast of Finniarken itself; they have, however, abandoned that coast for some centuries, although individual stragglers have been occasionally captured there up to within * Ross's 3d. Voy., App., 1835, p. xxi. t Suppl. Parry's 2d Voy., p. 338. t Zoology of Beecliey's Voyage, Mara., 1839, p. 6. $ Hector Boece's History of Scotland, as quoted by British zoologists. || Proc. Zool. Soc. Loud., 1868, p. 433. II British Animals, p. 19. **Edinb. New Phil. Jouru., vol. ii, p. 389; British Quad., Jard. Nat. Libr., Mam., vol. vii, p. 223. See also Bell, Hamilton, etc., 1. c. ttProc. Zool. Soc. Lend., 1868, p. 433. 72 ODOBvENUS ROSMARUS ATLANTIC WALRUS. . the last thirty [now about forty-six] years. [*] After their desertion of the Fiumarken coast, Bear Island [or Cherie Island, lying about two hundred and eighty miles north of the North Cape] became the principal scene of their destruction ; and next the Thousand Islands [southeast of Spitsbergen], Hope Island [a little further north, but still in the southeast corner], and Byk Yse Island, which in their turn are now very inferior hunting-ground to the banks and skerries lying to the north of Spitzbergen. " Fortunately for the persecuted Walruses, however, these lat- ter districts are only accessible in open seasons, or perhaps once in three or four summers, so that they get a little breathing- time there to breed and replenish their numbers, or undoubt- edly the next twenty or thirty years would witness the total extinction of Eosmarus trichecus on the coasts of the islands of Northern Europe. " The Walrus is also found all round the coasts of Nova Zem- bla, but not in such numbers as at Spitzbergen; and he under- goes, if possible, more persecution in those islands from some colonies of Eussians or Samo'iedes, who, I am told, regularly winter in Nova Zenibla for the purpose of hunting and fish- ing.»t " The war of extermination," says Mr. Lamout, in his later work, "which has been carried on for many years in Spitzber- gen and Novaya Zeinlya has driven all the Arctic fauna [mam- mals] from their old haunts, and, in seeking retreats more inac- cessible to man, it is probable that they have had in some degree to alter their habits. For example, up to about twenty years ago it was customary for all W^alrus-hunters to entertain a reasonable hope that by waiting till late in the season all for- mer ill-luck might be compensated in a few fortunate hours by killing some hundreds on shore ; in fact, favorite haunts were well known to the fishers, and were visited successively before finally leaving the hunting-grounds. Now, although the Arctic seas are explored by steamers and visited annually by as bold and enterprising hunters as formerly, such a windfall as a herd of Walruses ashore is seldom heard of. " Each year better found vessels and more elaborate weapons * Mr. Lament has since reported the capture of a large bull " in Magero Sound near the North Cape about 1868."— Yachting in tlie Arctic Seas, p. 58, footnote. t »Seasons with the Sea-horses, pp. 167, 168. GEOGRAPHICAL DISTRIBUTION. 73 are sent out to harry the Walrus ; as a consequence every sea- sou there is greater difficulty in obtaining a cargo — for two rea- sons, those animals which have ventured into what was safe feed- ing-ground last year meet their enemy, and half are killed, while the other half escaping will be found next year a step farther away. This intelligent retreating of the Walrus before a supe- rior enemy will, I believe, preserve the species after its scarcity in accessible waters renders it no longer an object of sport and commerce. That the Walrus, ... is being driven from every district where the hand of man is felt, is certain."* Mr. Alfred Newton, writing in 1864, respecting their former presence on the coasts of Fiumark, and their distribution at that date, observes : "I see no reason to doubt the assertion, or per- haps it would be safer to say the inference, that in former days Walruses habitually frequented the coasts of Fininark. In the sixteenth and seventeenth centuries they were certainly abund- ant about Bear Island ; they are spoken of there as l lying like hogges upon heaps/ [t] . . . ; yet for the last thirty years prob- ably not one has been seen there. Now they are hemmed in by the packed ice of the Polar Sea on the one side and their merciless enemies on the other. The result cannot admit of any doubt. .... Its numbers are apparently decreasing with woful ra- pidity. The time is certainly not very far distant when the Trielieehus rosmarus will be as extinct in the Spitzbergeu seas as Eliytina gigas is in those of Behriug's Straits." | In Eichard Chancellor's account of his " discoverie of Mos- covia," in 1553-1554, we read : u To the North part of that Coun- trey are the places where they have their Furres, as Sables, Marterns, greese Bevers, Foxes white, blacke, and red, Minkes, Ermines, Minivers, and Harts. There are also a fishes teeth, which fish is called a Morsse. The takers thereof dwell in a place called Postesora, which bring them upon Harts to Lam- pas to sell, and from Lampas carrie them to a place called Col- niogro, where the high Market is holdeu on Saint Nicolas day."§ On Hondius's map of Russia accompanying this account Lam- pas is placed on the White Sea, near the mouth of the Dwina Eiver. * Yachting in the Arctic Seas, 1876, pp. 59, 60. « t [" It seemed very strange to us," says Jonas Poole, in his account of his visit to Cherie Island in 1604, " to see such a multitude of monsters of the Sea, lye like hogges upon heapes." — Purchas Ins Pilgrimes, vol. iii, p. 557.] * Proc. Zool. Soc. Lond., 1864, p. 500. § Purchas his Pilgrhnes, vol. iii, pp. 213, 214. 74 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. The Walruses appear to have been first met with oil Cherie Island in 1003, and to have become nearly exterminated there within a very few years. The history of their destruction there and at Spitsbergen during the early part of the seventeenth century is given in the following excerpts : "In the yeare 1003. Stephen Bennet was imployed by the Companie,* in a Ship called the Grace, to those parts Northwards of the Cape ["of Norway"], and was at Cherie Hand and killed some Sea-horses, and brought home Lead Oare from thence . . . " Heere it is to bee understood, that the Companie having by often resort and imployment to those parts, observed the great number of Sea-horses at Cherie Hand, and likewise the multi- tude of Whales, that shewed themselves upon the coast of Greenland [now Spitzbergen]; They lirst applyed themselves to the killing of Morces, which they continued from yeere to yeere with a Ship or two yeerely, in which Ships the Companie ap- pointed Tltomas Wehlcn Commander, and in the yeere 1609. the Companie imployed one Thomas Edge their Apprentice, for their Northern Voyage, and joyned him in Commission with the fore- sayd Welden. Now the often using of Cherie Hand, did make the Sea-horse grow scarce and decay, which made the Conipauie looke out for further Discoveries."! During the expedition of 1604, Jonas Poole, who has left an account of the "Divers Voyages to Cherie Hand in the yeeres 1604, 1605, 1606, 1608, 1609," says that as they approached Cherie Island, "We had not furled our Sayles, but we saw inany Morses swimming by our ship, and heard withall so huge a noyse of roaring, as if there had beene an hundred Lions. Immediately wee manned our Boate. . . . wee landed, and saw abundance of Morses on thes hoare, close by the Sea-side," etc. They attacked them with muskets, " not knowing whither they could runne swiftly or seize upon us or no." Owing to in- experience, they succeeded in killing only fifteen out of " above a thousand," but secured a hogshead of teeth, which they picked up on the shore. Two days later they found, on another part of the island, " ueere a thousand Morses," of which they killed " thirtie or thereabouts, and when wee had taken off their heads, went aboard." The next day they went on shore again and * Incorporated sonic time prior to the year 1556, under the name "The chants of England," and called also the " Muscovia Merchants '' and rlv "Mnscovia Companie." t Pnrchas his Pilgrimes, vol. iii, p. 464. GEOGKAPHICAL DISTRIBUTION. 75 " fell a killing of the beasts. . . . We killed that jday six- ty Morses, all the heads whereof were very principall." They departed soon after for England. The next year (1605) they returned to Cherie Island. On the 8th of July, says the account, " we entred into a Cove, having all our men on shoare with shot and javelins, and slue abundance of Morses. The yeere before we slue all with shot, not think- ing that a javelin could pierce their skinnes: which we found now contrarie, if they be well handled, for otherwise, a man may thrust with all his force and not enter : or if he doe enter, he shall spoyle his Lance upon their bones ; for they will strike with their fore-feet and bend a Lance round and breake it, if it bee not all the better plated. They will also strike with their Teeth at him that is next them : but because their Teeth grow downward, their strokes are of small force and danger." They took in " eleven tunnes of Oyle, and the teeth of all the beasts aforesaid." The following year (1G06) they again set out for Cherie Island, arriving there July 3. They found the ice still about the island, and the Walruses not yet on shore ; " For their nature is such, that they will not come on land as long as any Ice is about the land." On the 14th they perceived on shore " of the beasts sufficient to make our voyage, wee prepared to goe killing. Master Welden and Master Bennet appointed mee to take eleven men with mee, and to goe beyond the beasts where they lay; that they and wee might meet at the middest of them, and so enclose them, that none of them should get into the Sea, . . . . and before six houres were ended, we had slayne about seven or eight hundred Beasts. . . . For ten dayes space we plyed our businesse very hard, and brought it almost to an end." They took in " two and twentie tuns of the Oyle of the Morses, and three hogsheads of their Teeth." In 1608 they again reached Cherie Island toward the end of June, and onthe22d "came into a Cove where the Morses were, and slew about 900. or 1000. of them in less than seven houres: and then we plied our business untill the second of July : at what time we had taken into our ship 22. tunnes and three hogs- heads of Oyle." On their return they took with them two live young Walruses, one of which lived till they reached London.* The voyage in 1609 was less successful. They slew at one- time eighty, at another one hundred and fifty, and at still an- * Purchas his Pilgrimes, vol. iii, pp. 557-560. 76 ODOB^ENUS KOSMARUS ATLANTIC WALEUS. other time forty-five; but they lost most of them in consequence of bad weather. " In the yee're 1610. the Companie set out two Ships, viz. the Lionesse for CJierie Hand, Thomas Edge Com- mander ; and the Amitie, for a Northerns Discoverie, the Mas- ter of which ship was Jonas Poole : who in the nioneth of May fell with a Land, and called it Greenland, this is the Land that was discovered by Sir Hugli Willoughby long before [Spctsberg of the Hollanders], which Ship Anitie continued upon the coast of Greenland, discovering the Harbours and killing of Morces [the first killed by the English on Spitsbergen], untill the moneth of August, and so returned for England, having gotten about some twelve Tunnes of goods, and an Unicornes home. a In the yeere 1611. the Companie set foorth two Ships, the Marie Margaret Admirall, burthen one hundred and sixtie tunues, Thomas Edge Commander ; and the Elisabeth, burthen sixtie tunnes, Jonas Poole Master, well manned and furnished with all necessarie Provisions, they departed from Blackwall the twentieth of Aprill, and arrived at the Foreland in Green- land in the Latitude of 79. degrees, the twentieth of May fol- lowing, the Adinirall had in her six Biskayners expert men for the killing of the Whale : this was the first yeere the Compa- nie set out for the killing of Whales in Greenland, and about the twelfth of June the Biskayners killed a small Whale, which yeelded twelve Tunnes of Oyle, being the first Oyle that ever was made in Greenland. The Companies two Shalops looking about the Harbour for Whales, about the five and twentieth of June rowing into Sir Thomas Smith his Bay, on the East side .of the Sound saw on the shoare great store of Sea-horses: after they had found the Morses they presently rowed unto the ship, being in crosse Eoad seven leagues off, and acquainted the Cap- tayne what they had found. The Captayue understanding of it, gave order to the Master, Stephen Bennet, that he should take into his Ship fiftie tunnes of emptie Caske, and set sayle with the Ship to goe into Foule Sound. The Captayue went pres- ently away in one Shallop with sixe men unto the Seainorse, and tooke with him Lances, and comming to them they set on them and killed five hundred Morses, and kept one thousand Morses riving on shoare, because it was not profitable to kill them all at one time. The next day the Ship being gone unto the place & well mored where the Morse were killed, all the men. belonging to the Ship went on shoare, to worke and make Oyle of the Morses ; and when they had wrought two or three dayes, GEOGRAPHICAL DISTRIBUTION. 77 it fortuned that a small quaiititie of Ice coine out of Foule Sound, and put the Ship from her Moring. . . . The Ship being cast away without hope of recoverie, the Commander Thomas Edge gave order, that all the Morse living on shoare shook! be let goe into the Sea, and so gave over making of Oyle. . . ." Fitting up their boats as well as they could they soon after abandoned the coast of Spitsbergen ("Green- land"), and set sail for Cherie Island, where they found the "Elizabeth" and returned to Spitsbergen "to take in such Goods as the sayd Edge had left in Foule Sound, woorth fifteene hundred pounds."* As early as the year 1611, the previous persecutions of the Walruses at Cherie Island had made them very wary. Thomas Finch, in his account of a visit to this island by William Gour- don in August of that year, says: "At our comniing to the Hand, wee had three or foure dayes together very fine weather: in which time came in reasonable store of Morses, . . . yet by no ineanes would they go on those beaches and places, that formerly they have been killed on. But fortie or fiftie of them together, went into- little holes within the Eocke, which were so little, steepe and slipperie, that as soone as wee did approach towards them, they would tumble all into the sea. The like whereof by the Masters and William Go urdons report, was never done."t During the years 1612, 1613, 1614, and 1615, numerous vessels were sent out from England to Spitzbergen for the products of the Walruses and Whales, but generally met with indifferent success, being much troubled with Spanish, Dutch, and Dan- ish " interlopers." " In the yeere 1616, the Companie set out for Greenland eight Sayle of great ships, and two Pinnasses under the command of Thomas Edge, who following his course, arrived in Greenland about the fourth of June, having formerly appointed all his ships for their severall Harbours, for their making of their Voy- age upon the Whale, and having in every Harbour a sufficient number of expert men, and all provisions fitting for such a Voy- age. This yeere it pleased God to blesse them by their labours, that they full laded all their ships with Oyle, and left an over- plus in the Countrey, which their ships could not take in. They imployed this yeere a small Pinnasse unto the East- ward part of Greenland, Namely, the Hand called now Edges Hand, *Purclias his Pilgrimes, vol. iii, pp. 464, 465 tlbid., p. 536. 78 OD01LENUS EOSMARUS ATLANTIC WALRUS. and other Hands lying to the North-wards as farre as seventie eight degrees, this Piimasse was some twentie tunnes, and had twelve men in her, who killed one thousand Sea-horses on Edges Hand, and brought all their Teeth home for London." In 1017, they " employed a ship of sixtie tunnes, with twenty men in her, who discovered to the Eastward of Greenland, as farre North- wards as seventy-nine degrees, and an Hand which he named Witches Hand, and divers other Hands as by the Map appeareth, and killed store of Sea-horses there . . ."* The Dutch, Danes, and Spaniards began, in 1612, also to visit Spitzbergen in pursuit of Whales and Sea-horses, but are reported by the English to have made indifferent voyages. The company soon also had rivals in the "Hull-men," who, as well as the Dutch, did them much "ill service."! About the years 1611 and 1612, the Whale-fishery was found to be more profitable than Walrus-hunting, and subsequently became the main pursuit, not only by the English, but by the Dutch and Danes. Yet the Walruses were by no means left wholly unmolested, having been constantly hunted, with more or less persistency, down to the present day, and, as already shown, were long since exterminated from Cherie Island and other smaller islands more to the northward, and greatly re- duced in numbers on the shores of Spitzbergen. Walruses have been recently reported as occurring on the outer or northwestern coast of Eova Zembla, but as not exist- ing on the inner or southeastern coast. Von Baer, on the au- thority of S. G. Gnielin and others, gave the eastern limit of the distribution of the Atlantic Walruses as the mouth of the Jene- sei Eiver, though very rarely single individuals wandered as far eastward as the Piasina Eiver. He even regarded the Gulf of Obi as almost beyond their true home.t Von Middeudorff, however, considers von Baer's eastern limit as incorrect, and cites old Eussian manuscript log-books ("handschriftliche Schiffsbiicher") in proof of their occurrence in nuVbers in Au- gust, 1736, as far east as the eastern Taimyr Peninsula, and of their being met with in August, 1739, as far east as Chatauga Bay. Still further eastward, in the vicinity of the mouth of the Lena Eiver, he gives similar authority for their occurrence in August, 1735, and says that Dr. Figurin attests their presence *Purchas Ms Pilgrimes, vol. iii, p. 467. tlbid., pp. 472, 473. JM6m. de 1'Acad. des Sci. de St. P^tersb., vie s6r., Sci. math., phys. etnat., tome iv, 2^- pars, pp. 174, 184. GEOGRAPHICAL DISTRIBUTION. 79 •on the shores of the delta-islands of the Lena. Eespecting the more easterly coast of the Siberian Ice Sea, he says it is cer- tainly known that the Walmses of Behriug's Sea extend west- ward in great numbers to Koljutschiu Island. Only the males, liowever, reach this limit, the females not extending beyond the vicinity of the mouth of the Kolyma Elver.* It hence appears that about 1735 to 1739 Walruses were met with as far eastward as the mouth of the Lena River; but Wraugell, nearly a century later, explored quite thoroughly this whole region without meeting with them, and I have found •only one reference to their existence on the Siberian coast be- tween the Kolyma and Jenesei Elvers later than those cited by von Middeudorff. According to a recent letter t from Professor Nordenskjold, of the Swedish Northeast Passage Expedition, "two Walruses" were seen in August, 1878, a little to the eastward of the Jenesei Eiver, and that open water was found as far as the mouth of the Lena. From this it would seein that there is nothing to prevent, at least in favorable years, the Walruses from passing eastward to the mouth of the Lena. There still remains, how- ever, a breadth of some thirty degrees of longitude (between 130° and 100°) where as yet no Walruses have been seen. They appear to have been only very rarely met with to the eastward •of the Jenesei (longitude 82° E.), and to be uncommon east of the Gulf of Obi. At present the Atlantic Walrus ranges along the northeast- ern coast of North America from Labrador northward to Ee- pulse Bay and Prince Eegent's Inlet, and along the shores of Greenland ; in the Old World only about the islands and in the icy seas to the northward of Eastern Europe and the neigh- boring portions of Western Asia, where it rarely, if ever, now visits the shores of the continent. On the eastern coast of North America, Walruses have been met with as far north as explorers have penetrated, and as far as the Esquimaux live. They winter as far north as they can find open water, retiring southward in autumn before the ad- vance of the unbroken ice-sheet. Kane speaks of their remain- ing in Benssellaer Harbor (latitude 78° 37') in 1853, till the sec- ond week of September, when the temperature reached zero of Fahrenheit, $ * Von Middeudorff's Sibirische Reise, Bd. iv, 1867, pp. 935, 936. t See Nature, vol. six, p. 102, December 5, 1878. t Arctic Exploration, vol. i, p. 140. 80 ODOBvENUS ROSMARUS ATLANTIC WALRUS. NOMENCLATURE. — Several specific names have been in more or less current use for the Atlantic Walrus, or rather for the Atlantic and Pacific species collectively. Accepting Odobwnu-s as the proper generic name of the group, there is nothing to prevent the adoption of rosmarus for the specific name of the Atlantic species. It was used for this species exclusively by Linne, Erxlebeu, and other early systematic writers, the Pacific Walrus being at that time unknown to the systeniatists. If Ros- marus be used as the generic name of the group, as it has been by a few late writers, as a substitute for the wholly untenable one of Trichechus, it will be, of course, necessary to adopt some other name for the species. Dr. Gill has used obesus of Illiger ;. but as this was applied by Illiger exclusively to the Pacific Wal- rus, it cannot properly be used for the Atlantic species. It would be difficult to select a subsequent name that would not be open to objection, if one should stop short of trichechus, used (inadvertently?) in a specific sense ("Rosmarus tricheclms") by Lamont in 1861. The name longidens of Freniery, 1831, was based on what subsequent writers have considered as probably the female, but the name is highly inappropriate, inasmuch as it is the Pacific species, and not the Atlantic, that has the longer tusks. There are left mrginianus of DeKay and duMus of Stamiius : the first is objection able on account of its geograph- ical significance ; the other is only doubtfully referable to the Atlantic species. Adopting Odolccnus for the genus, leaves rosmarus available for the species, thus settling the whole diffi- culty. As already noticed (antea, p. 20), two species besides virgi- nianus have been based on fossil remains, and have been made the basis of new genera. The first of these is the Odobenotlie- rium lartetianum of Gratiolet, since referred by Defrance to the existing species ; the other is the Trichecodon Imxleyi of Lan- kester, which there is perhaps reason for regarding as the large extinct progenitor of the existing Walruses. ETYMOLOGY. — The term rosmarus was originally used by Olaus Magnus, about the middle of the sixteenth century, in a vernacular sense, interchangeably with morsus, the Latinized form of the Eussian word morsz (or morss). It was used in the same way by Gesner a few years later, as well as by numerous other pre-Linnaean authors. Eespecting the etymology of the word, von Baer gives the following : " In dern historisch-topo- ETYMOLOGY. 81 graphischen Werke : De gentium septentrionalium conditionibus cet. Romae 1555 heisst es : Norvagium littus maximos ac grandes pisces ekphantis habet, qui morsi sen rosmari vocantur, forsitan ob asperitate mordendi sic appellati, (Eine recht witzige Etyuio- logie !) quia, si quern hominem in maris littore viderint apprelien- dereque poterint, in eum celerime insiliunt, ac dente lacerant et in momento interimunt." * The same author also gives the following from Herberstain (1567): "Under andern ist auch ein thier, so grosse wie ein ochs, und von den einwonern Mors oder der Tod geheissen wird." t Hence, either from superstitious notions of the terri- ble character of this animal, or from the resemblance of the Eussian word morss to the Latin word mors, these terms be- came early confounded, and rendered by the German word To<7, or death. | In the account of the exploits of the Norman Othere, where the Walrus first finds its place in literature, it is termed Horse- wael. As noted by Martens § and other writers, equivalent words in other languages have become current for this animal, as Walross or Wallross of the Germans, Wallrus of the Dutch, * In an early (1658) English version of Olaus Magnus's work ("A Com- pendious History of the Goths, Swedes, & Vandals and other Northern Nations. Written by Olaus Magnus, Arch-Bishop of Upsal, and Metropoli- tan of Sweden", p. 231), this passage is rendered as follows: "The Norway Coast, toward the more Northern parts, hath huge great Fish as big as Ele- phants, which are called Morsi, or Rosmari, may be they are so from their sharp biting ; for if they see any man on the Sea-shore, and can catch him, they come suddenly upon him, and rend him with their Teeth, that they will kill him in a trice." From this it would appear that Morsus, as used by Olaus Magnus, might be simply the Latin word morsus, from mordere, to bite. tSee von Baer, Me"m. de Acad. des Sci. de St. Pe"tersb., vie se'r., Sc. math., phys. et nat., tome iv, 2de pars, pp. 112, 113. t Von Baer quotes a passage from the ' ' Rerum Moscoviticarum auctores varii," originally published early in the sixteenth century, in which occurs the phrase "scandut ex mari pisces morss nuncupati," which he regards as the first introduction into Latin of the Sclavic name Mop/Kb. In Western Europe it a little later became current in the form of Morsz, which was soon written Morss or Mors, from which Buffon later formed the name Morse, which has since been the common appellation of this animal among French writers. Von Baer further observes that the accidental resemblance in sound of this word to that of the Latin word for death (mors) appears to have contributed not a little to the strange conception of the terribleness of this animal which was early entertained and even still prevails in Western Europe, although the Russian accounts do not speak of it. iJZoolog. Garten, Jahrg. xi, 1870, p. 283, where the etymology of the names of the Walrus is briefly discussed. Misc. Pub. No. 12 6 82 ODOB^NUS KOSMARUS ATLANTIC WALRUS. and Walrus of the English. By the early Scandinavian writers it was termed Rosmlivalr, which later became resolved into Eos- mul, from which, perhaps, originated the Latin term Rostnarus, which has the same significance, introduced by Olaus Magnus and Gesner, and the Norwegian word Rostungr. Gesner and several subsequent writers also used the word Meerross, and we have in English the equivalent term Sea-horse, as one of the ap- pellations of the "Walrus, and also, but more rarely, Meerpferd in German, and Cheval marin in French. The current French term Morse appears, as already stated, to have been introduced by Buffon as a modification of the Kus- sian word morss, used by Michow (1517) and Herberstain (1549). Among other old vernacular names we find in English Sea Cow, in French Vaclie marine, in Latin Bos marinus, etc., while by the early French settlers in America it was commonly termed Bete a la grande dent. LITERATURE. — 1. General History. — Passing over the by some supposed allusions to the Walrus by Pliny as too vague and uncertain for positive identification, * we meet, according to von Baer, with the first positive reference to the present species in the account of the exploits of the famous Norman ex- plorer Othere, or Octher, who, about the year 871 (890 accord- ing to some authorities), made a voyage to some point beyond the North Cape, where he met with large herds of Walruses, some of the tusks of which he is said to have taken to England as a present to King Alfred, t Walruses appear to have been * See K. E. voii Baer, M6m. de 1'Acad. Imp. des Sci. de St. Pe"tersb., vime se"r., Sci. math., pliys. et nat., tome iv, 3me livr., 1836, (1837), pp. 101, 102. To this admirable monograph I am greatly indebted for information respecting the earlier publications bearing upon the history of the Walruses. To this exhaustive memoir the reader is referred for a full exposition of this part of the subject. The following short summary is based, so far as the early his- tory of the subject is concerned, mainly upon von Baer's monograph, an analysis of which will be presented at a subsequent page. (See postea, p. 88, footnote.) t Hakluyt's rendering of this account is as follows: "The principall purpose of his [Othere's] traveile this way, was to encrcase the knowledge and disco verie of these coasts and countreyes, for the more comoditie of fish- ing of horsewhales, which have in their teeth bones of great price and ex- cellencie : whereof he brought some at his returue unto the king. Their skinnes are also very good to make cables for shippes, and so used. This kind of whale is much lesse in quantitie then other kindes, having not in length above seven elles." — HAKLUYT'S Voyages, vol. i, p. 5. GENERAL HISTORY. 83 an object of chase on the coast of Finmark as early as 980, and must have been met with by the Norsemen when they visited Greenland about the end of the tenth century. Their tusks were an article of commercial value among the Mongolian and Tartar tribes as early as the twelfth to the fifteenth centuries. Aside from the various notices by Scandinavian writers, the earliest unmistakable reference to the "Walrus, other than that connected with Othere, as above mentioned, was, according to von Baer (1. c., p. 108), by Albertus Magnus, in the first half of the thirteenth century. Says this writer (as quoted by von Baer), whose account is here paraphrased : The hairy Cetaceans have very long tusks, by which they suspend themselves to the rocks in order to sleep. Then conies the fisherman and separates near the tail as much skin as he can from the underlying fat, and then attaches a cord, which has at the other end a large ring, which he makes fast to a post or tree. Then when the fish awakens (by all of these operations he was not yet awakened), they cast a huge sling-stone upon his head. Being aroused, he attempts to get away, and is held by the tail near to the place and captured, either swimming in the water or half alive on the shore. This ludicrous description von Baer believes had for its foundation misunderstood reports of the Walruses' habit of reposing upon the shore or upon ice-bergs, the use of their tusks in climbing up to these places of rest, and their deep sleep, and that the account of the mode of capture was based on an incorrect knowledge of the use of the harpoon; and that the account shows that as early as the thirteenth century the Walrus was harpooned on the coast extending from the White Sea north- wards. * * This curious legend is quoted by Gesner in his Historia Animalia Aqua-' tilia, 1558, p. 254. The following rendering appears also in the above-cited English version of Olaus Magnus: "Therefore, these Fish called Rosmari, or Morsi, have heads fashioned like to an Oxes, and a hairy Skin, and hair grow- ing as thick as straw or corn-reeds, that lye loose very largely. They will raise themselves with their Teeth as by Ladders to the very tops of Rocks, that they may feed on the Dewie Grasse, or fresh Water, and role themselves in it, and then go to the Sea again, unless in the mean while they fall very fast asleep, and rest upon the Rocks, for then Fisher-men make all the haste they can, and begin at the Tail, and part the Skin from the Fat ; and into this that is parted, they put most strong cords, and fasten them on the rug- ged Rocks, or Trees, that are near; then they throw stones at his head, out of a Sling, to raise him, and they compel him to descend, spoiled of the greatest part of his Skin which is fastened to the Ropes : he being thereby 84 ODOB^ENUS ROSMARUS— ATLANTIC WALRUS. The Walrus is also referred to by Hector Boethius in 1526, in his History of Scotland ; * by Herberstain (or Herberstein, as also written) in 1549 ; by Pare about the year 1600 ; and by Al- drovandus in 1642. Herberstain also very correctly indicates the habits of these animals, which, he says, repair to the shore in large herds to repose, and that while the herd sleeps one of their number keeps watch. He compares their feet to those of the Beaver, and refers to the value of their tusks to the Eussians, Turks, and Tartars, and observes that they called them fish-teeth, t Even before the middle of the sixteenth century, Walruses had been met with on the eastern shore of North America. In May, 1534, they were seen by Cartier, and later in the same century by Fischer, Drake, and others, on the coast of Nova Scotia and adjacent islands, and later still by other explorers on the islands in the Gulf of Saint Lawrence (see antea, p. 66), in the accounts of whose voyages \ occur interesting notices of these animals. In the year 1553, Edward VI of England sent an expedi- tion under Willoughby and Chancellor to the White Sea, which resulted in still further increasing our knowledge of the Wal- ruses, especially of their distribution eastward along the Arctic coast of Europe and Asia. Chancellor's short account § refers especially to the uses made of the skins and tusks. The earliest delineations of the Walrus appear to have been made by Olaus Magnus in his " Tabula Terrarum Septeutrio- nalium" (1555), where he has portrayed many strange and fabu- lous animal forms which there is reason to believe were based upon this animal. 1 1 Gesner a few years later (1558), in his " His- debilitated, tearful, and half dead, he is made a rich prey, especially for his Teeth, that are veiy pretious amongst the Scythians, the Moscovites, Eus- sians, and Tartars, (as Ivory amongst the Indians) by reason of its hardness, whiteness, and ponderousnesse. For which cause, by excellent industry of Artificers, they are made fit for handles for Javelins : And this is also testi- fied by Mecliovita, an Historian of Poland, in his double Sarmatia, and Paulus Joviits after him, relates it by the Relation of one Demetrius, that was sent from the great Duke of Moscovy, to Pope Clement the 7th." — Loc. cit., pp. 231,232. " "Scotorum Regui Descriptio, p. 90," as cited by various writers. t Herberstaiu, as cited by von Baer, 1. c., p. 111. I See Hakluyt's Voyages, vol. iii, ed. 1810, pp. 237, 238, 242, 249, 254, etc. § See Hakluyt's Voyages, vol. i, ed. 1599, p. 237. || Olaus Magnus's figures will be noticed later under the section devoted to the figures of the Walrus (postea, p. 92 et seq.). GENERAL HISTORY. 85 toriimi Animaliuin" (in the volume devoted to the " Animalia Aquatilia "), faithfully copied all of Olaus Magnus's figures under the heading "De Cetis," and then presents, under the name EosmantSj the figure of the Walrus from Olaus Magnus. This figure, however, he judiciously criticises, stating that the tusks should be in the upper jaw, and not in the lower, as they were represented by Olaus Magnus. This last-named author, in the later editions of his work "De Gentium Septentrionaliuin Con- ditionibus," etc. (as in that of 1563), rightly places, according to von Baer, the tusks in the upper jaw. Gesner (continues von Baer) knew only the first edition of this work, and took his figure from the above-mentioned " Tabula Terrarum Septen- trionaliuin." Also were unknown to him the accounts of the Wal- rus given by " Herberstain, Chancellor, and Othere," so that he made extracts from only Michovius and Albertus Magnus. He also knew no better than to oifer, as a figure of the Walrus, a drawing he had received from Strassburg, representing, pretty fairly, the head and tusks, while the rest was purely a fabrica- tion. Some rhymes, which he further inflicts upon his readers, show clearly how " awful" the conceptions of the Walrus then were (or, as von Baer puts it, " Wie schauerlich noch die Vor- stelluugen vom Wallrosse waren ").* In 1608, a young living Walrus was taken to England, having been captured on Bear or Cherie Island off the coast of Nor- way, t while four years later (1612) another young Walrus, with the stuffed skin of its mother, was taken to Holland. The first appears to have been very intelligently described by 2Elius Ever- hard Vorstius, whose description is quoted by De Laet. f The specimen taken to Holland was well figured by Hessel Gerard, the young one doubtless from life, the figures being published by him in 1613, § and subsequently repeatedly copied (as will be more fully noticed later). In 1625, Purchas, in his history of the voyages of the English to Cherie Island and Spitzbergen (then called "Greenland"), gives much interesting information respecting the chase of the * To show what these conceptions were, von Baer cites the passages already quoted (antea, p. 81), in reference to the singular misinterpretations given in Western Europe to the Russian name Morss. See von Baer, 1. c., p. 113. t Recueil de Voy. au Nord, 2e e"d., tome ii, p. 368. JNov. Orb. s. Dcscrip. Ind. Occ., 1633, p. 41. § See von Baer, 1. c., p. 128; Gray, Proc. Zool. Soc. Lond., 1853, p. 115. 86 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. Walrus at these islands, and in one place a quaint description and some very curious figures of the animal.* In 1675, the Walrus was again described and wretchedly figured by Martens, t who is said to have been the first " natu- ralist" who ever saw the Walrus in its native haunts. Zorg- drager, $ in 1720, supplied by far the fullest account of these animals, as observed by him in Spitsbergen, that had appeared up to that date. He gives not only a quite detailed and truth- ful account of their habits, especially under persecution, but also of their wholesale destruction at that early time in the Spitz- bergen seas, and of their extermination at some of the points at which they had formerly been accustomed to land in immense herds. He also notes the increasing difficulties of their capture owing to the great shyness of man they had acquired in conse- quence of persecution, and describes the manner in which they were captured, and also their products. Copious extracts from Zorgdrager's account of the Walrus are given by Buffon (trans- lated into French from a German edition), and he has also been extensively quoted by even much later writers. The Greenland Walrus was described by Egede § in 1741, by Anderson || in 1747, by Ellis fl in 1748, byCranz**in 1765, and by Fabricius tt in 1780, some of whom added much infor- mation respecting its habits and distribution, its usefulness to the natives and their ways of hunting it, as well as respecting its external characters. The above-cited accounts of the Walrus formed the basis of numerous subsequent compilations, and most of those last given are cited by the early systematic writers, few of whom, as pre- viously shown (see antea, pp. 8-11), had any just appreciation of even its most obvious external characters. Linn£, as already noted (antea, p. 8), profited little by what had been written by preceding authors, while Brisson, Erxleben, and Gmelin manifest a scarcely better acquaintance with this badly misrep- resented and poorly understood creature. IsTo little confu- sion has hence arisen in systematic works respecting its posi- * See antea, p. 74-78, and posted. t Spitsbergen, pp. 78-83, pi. P, fig. b. t Bloeyende Opkomst der Aloude en Hedendaagsche • Groenlandsche Vis- schery, etc., ed. 1720, pp. 1(35-172. § Det gamle Gr<|>nlauds nye Per lustration, etc., 1741, p. 45. || Nachricliten YOU Island, Gronland und der Strasse Davis, p. 258. If Voyage to Hudson's Bay, p. 134. ** Historic von Gronland, pp. 165, 167. tf Fauna Groenl., p. 4. GENERAL HISTORY. 87 tioii and affinities (see antea, pp. 7-12). The accounts by Hout- tuyn, Buffon, Pennant, P. S. L. Miiller, and Schreber are excel- lent for their time. These authors all recognized the close relationship of the Walrus to the Seals, and quite correctly indicated its external characters and habits. Some of these accounts, however, include references to both species. Daubenton, in Buffon's "Histoire Naturelle," * gave a de- scription and figure of a Walrus's skull, and made the first contribution to our knowledge of its internal anatomy, based on the dissection of a fostal specimen. Since the beginning of the present century, the Walrus has been the subject of almost numberless notices, as well as of sev- eral elaborate papers, devoted in most cases to special points in its anatomy, very few of which need be here enumerated, t The elder Cuvier, beginning with his " Le9ons d'Anatomie comparee" (1800-1805), and ending with the third edition of his "Ossernens fossiles" (1825), contributed considerably to our general knowledge of its structure and affinities, especially of its osteology ; he in 1825 J first figuring and describing its skel- eton. A paper by Sir Everard Home, § in 1824, figured and de- scribed the stomach and feet from specimens taken to England from Hudson's Bay, preserved in salt. This paper is noteworthy mainly on account of the singularly erroneous interpretation there made of the structure and functions of the feet, Home supposing that these organs were provided with sucking discs, by means of which the creature was enabled to adhere firmly to the ice in climbing. The skeleton of the Walrus was again figured and described by Pander and d' Alton || in 1826, and still later by BlainviUe fl about 1840. Von Baer, ** in 1835, published some account of the arterial system of the Walrus, based on a dissection of a young specimen. Its general anato- my, especially its limb-structure, myology, vascular and respi- ratory systems, viscera and generative organs, and external cha- * Tome xiii, 1765, pp. 415-424, pll. liv, Iv. The skull had been previously figured by Houttuyn (in 1761), as will be noticed later. t Those relating to its dentition have been already noticed in detail (see anted,, pp. 47-57) ; several others have also been specially referred to, and nearly all are cited in the references given at pp. 23-26. .t Ossem. Foss., 3e e"d., tome v, iime pt., pp. 521-523, pi. xxxiii. § Phil. Trans., 1824, pp. 233-241, pi. iv. || Skelete der Robben und Lamantine, pll. i, ii. IT Osteographie, Des Phoques, pll. i and iv. ** Mem. de 1'Acad. St. Pe"tersb., vime s6r., Sci. math., phys. et nat., tome ii.me. 1835, pp. 199-212. 88 ODOB^ENUS ROSMARUS ATLANTIC WALRUS. racters, were quite fully and satisfactorily treated by Dr. J. Murie* in 1872. Illiger, in 1811, in a paper on the geographical distribution of the mammals of the Northern Hemisphere (see anted, p. 18), first nominally recognized the Pacific Walrus as a species distinct from the Atlantic animal, while Fremery, in 1831, recognized three species, and Stannius, in 1842, admitted two,t but, as already noticed, only one species of Walrus has been commonly recognized. The matter of variation dependent upon sex, age, and individual peculiarities, has received, as already 'noticed (see antea, pp. 38-43), special attention at the hands of Wieg- mann, Stannius, Jaeger, and other writers. Unquestionably, the most important paper relating to the lit- erature, geographical distribution, and habits of the Walruses is the well-known and justly celebrated memoir by von Baer,| * Trans. Loncl. Zool. Soc., vol. vii, pp. 411-462, 8 woodcuts, and pll. li-lv. t For a notice of the literature of this part of the subject see an tea, pp. 17-23. t Anatomische und zoologische Untersuchuugen iiber das Wallross (Triche- chus rosmarus) und Vergleichung dieses Thiers mit andern See^Saugethieren. Von Dr. K. E. v. Baer. Gelesen den 6. Nov. 1835.