Ny NAVA iy iy FES Hee pia shes Wh awo a Siw Asam | sar LIBRARY oF | GFOLOG Dinos ustiial MW nivetsity, CHAMPAIGN, ILLINOIS. BOOKS ARE NOT TO BE TAKEN FROM THE LIBRARY ROOM. LIBRARY USE ONLY Return this book on or before the _ Latest Date stamped below. A charge is made on all overdue books. U. of I. Library 11148-S i PROFESSIONAL PAPERS OF THE ENGINEER DEPARTMENT, U.S. ARMY. No. 18. | gd 2 2 OM a RA OF THE GEOLOGICAL EXPLORATION OF THE FORTIETH PARALLEL MADE BY ORDER OF THE SECRETARY OF WAR ACCORDING TO ACTS OF CONGRESS OF MARCH 2, 1867, AND MARCH 3, 1869, UNDER THE DIRECTION OF BRIG. AND BVT. MAJOR GENERAL A. A. HUMPHREYS, CHIEF OF ENGINEERS, BY CLARENCE KING U. 8. GEOLOGIST. Ay = ica es : Tete SE hte ae Ar ahs 7 we ? " oa : ; ~ — .% a | Fi ‘%, : = FF ' OO ee ae aa ee | aa = > . + 4 .. = a a> 2 YS) “ , Se >) ee at —_ le = my = ais ‘2 1A) ’ ANY AL Digitized by the Internet Archive in 2011 with funding from University of Illinois Urbana-Champaign http://www.archive.org/details/reportofgeologicO/unit UNITED STATES GEOLOGICAL EXPLORATION OF THE FORTIETH PARALLEL, CLARENCE KING, GEOLOGIST-IN-CHARGE. ODONTORNITHES: MON OGRAPEH ON THE EXTINCT TOOTHED BIRDS OF NORTH AMERICA; WITH THIRTY-FOUR PLATES AND FORTY WOODCUTS. BY OTHNIBHL CHARLES MARSH, PROFESSOR OF PALZ ONTOLOGY IN YALE COLLEGE. SUBMITTED TO THE CHIEF OF ENGINEERS AND PUBLISHED BY ORDER OF THE SECRETARY OF WAR UNDER AUTHORITY OF CONGRESS. WASHINGTON: GOVERNMENT PRINTING OFFIOE. 1880. 4 mm enone Apert VaG a in ae i MRL OH des pA Ve ais mrs) (2 fl awe - ee ee aaa <= > > = Tar je ay, sae A oem vr F Rg Wh j ng a } Sa = SP . Ait oy? ; ae TABLE OF CONTENTS. UOTRROMM VOUUME etre © an Sacre ene eee ae oth eee See eee a SARTO OIE CON TENTS 2 Sasa ae 2 eye 2k Ae bs eae oo Ia TORD TEMG ATE A TIONS ea 8 srs. See a a ea eae wo Sa ce eS oases JES MONG TH es Se se, Pr ee ee (ENTE ODU CUIONs eet e eras nee eee eet serene e as cbasosh ees wuss PART L—ODONTOLC, Cuarter I. THe Skuny OF JIESPEROENIG;- 2-2 22.5222222522.2.5.2-25202- e II. Ta, Temrn:or HEsPERORNIS, ===. —-2-5222.20522- 2222224 25 =~ ee Il, Tue Presacrat VERTEBRE OF HESPERORNIS,---------------- EV: vo Tuer SHOULDER-GIRDLE AND Riss or HESPERORNIS, ------_-__--- = WV. Tam Prnvic Ancu or Hrsprrornis,..-..-.-------.+.---.--- ce VI. Tue Cavupat VEerRTEsRaz oF HEsPERORNIS,.----------------- as VIL Tue Lzes anp Freer or HesPrrornis,---.--------.---------- &s VII. Tue Restoration or HEespERORNIS,-.-...-------------------- PART IL—ODONTOTORM. Cuarrer IX. Tux Sxvurz anp Treru or Icuruyornis,..-._.-.------------ e xX. Tue PRESACRAL VERTEBR# OF ICHTHYORNIS,--.------------- + XI. Tuer SsHovurper-Girpie or IcuTrHyornis AND APATORNIS, -- ---- A XII. Tue Wines or Icutuyornis anp APATORNIS,_--.._---------- os XU. Tur Petyic Arcu or Icutuyornis anD APATORNIS,----_----- = XIV. Tuer Caupat VERTEBRE OF IcHTHYORNIS,----_..-------.---- oc XV. Tue Lees anp Freer or Icurnyornis anp APATORNIS, ---- --_- se XVI. Tue Restoration or IcHTHYORNIS,...-...--.----.---------- + BRAV Alle: CON GEUBION sam erin ee = se a Sa ea aS ae ok ace en Sass ApPENDIX.—SyNopsis oF AMERICAN CreTacrous Birps,---------------------- 111 119 127 141 149 161 166 171 177 185 191 Pty ae Gua oe ~ a ee me ' » s . a we saas 4 line maT PSE obigee web a» wae clralh deb abies 2 olmnin i iaari Z sonia Tee itaantenedg eet ogi nd ig) Ghz 6g a mee os = ee ee j day es ee Pee ee pind ce Sneet hake ey Sere : 2 ere : : ee eee eee ee le i. ss F EE To i i Cd ed ‘ oF ee ae ¢ Per th Ra ee eee oe Riad : =o f te < ji 14 = ia oui “1. * “ g TPN. ae Piyp r ’ -, ota b ' i Mi ‘| a. e ee bo vaics od nhl ty pep ietiee gy per +O 8 ae 26 2. fe

8» ar a ee “poe . Tr) uj Ny a i ' ¥. «te vat ' ‘ of PN 4 <8 4 J s » = ’ a, 4 * if j ‘ { F aa ’ ** 4 * tay : Rea jun - ‘ csuce ' " ght ihle “ds eit . . Pai i ie ‘ aw, ¢ ie et woe ff ~< nS s é a s “J ere eo . * , ? * - “7 2 “ b ii 41 a> ‘ + ‘ ‘ - ail 3 ‘ . ’ ; 7o-Ti NE a ' 2 « mibn ate . ® i ¢ t f, } Jara” D ; Pes i Tae st ie | ‘uti Ere ard wt NE Tg aa 24 ns ? foie opleongen dl neon oy tat) ~ igs ¢ ans as eS. oh aoe ret « - a7 > oe Ci Tue present volume is the first of a series of Monographs designed to make known to science the Extinct Vertebrate Life of North America. In the investigation of this subject, the writer has spent the past ten years, much of it in the field, collecting, with no little hardship and danger, the material for study, and the rest in working out the characters and affinities of the ancient forms of life thus discovered. During this decade, the field work, extending from the Missouri River to the Pacific Coast, has so predominated, as the subject unfolded, that a plan of gradual publication became a necessity. The more important discoveries were briefly announced soon after they were made, but only where the specimens on which they were based admitted of accurate determination. The principal characters of the new groups were next worked out systematically, and published, with figures of the more important parts. When the investigation of a group is completed, the results, with full descriptions and illustrations, will be brought together in a monograph. This system has been carried out with the Odontornithes, and will be continued with the other groups. The investigation of several of these is now nearly completed, and the results will soon be ready for publication. xiil xiv ODONTORNITHES. The material is abundant for a series of monographs on the marvel- lous extinct vertebrates of this country, and the results already attained are full of promise for the future. A somewhat careful estimate makes the number of new species of extinct vertebrates, collected since 1868, and now in the Yale College Museum, about 1,000. Nearly 300 of these have already been described by the writer, and some have been noticed or described by other authors, but at least one-half remain to be investigated. Among the new groups brought to light by these researches, and already made known by descriptions of their principal characters, are the following, which will be fully described in subsequent volumes of the present series. The first Pterodactyles, or flying reptiles, discovered in this country, were found by the writer in the same geological horizon with the Odontorni- thes, described in the present memoir. These were of enormous size, some having a spread of wings of nearly twenty-five feet; but they were espe- cially remarkable on account of having no teeth, and hence resembling recent birds. They form a new order, Pteranodontia, from the type genus Pteranodon. Of this group, remains of more than six hundred individuals are now in the Yale College Museum—ample material to illustrate every important point in their osteology. With these fossils, were found also great numbers of Mosasauroid reptiles, a group which, although rare in Europe, attained an enormous development in this country, both in numbers and variety of forms. Remains of more than fourteen hundred individuals, belonging to this order, were secured during the explorations of the last ten years, and are now in the Museum of Yale College. The most interesting discoveries made in the Jurassic formation were the gigantic reptiles belonging to the new sub-order Sauropoda, including by far the largest land animals yet discovered. Another remarkable group of large reptiles found in the same formation were the Stegosauria. Other Dinosaurs from the same horizon, the ‘“ Atlantosaurus beds,” show that this was the dominant form of vertebrate life in that age, and many hundred specimens of these reptiles are now in the Yale Museum PREFACE. XV In a lower horizon of the same formation, the “ Sauranodon beds,” were found the remains of a peculiar new group of reptiles, the Sawranodontia, allied to Ichthyosaurus, but without teeth. In the Eocene deposits of the Rocky Mountains, the writer discovered a new order of huge mammals, the Dinocerata. Remains of several hundred individuals were secured, and a monograph on the group will follow the present memoir. In the same formation were found the remains of another new order of mammals, the Tillodontia, in many respects the most remarkable of any yet discovered. In the same Eocene deposits were secured the first remains of fossil Primates known from North America, as well as the first Cheiroptera, and Marsupialia. Abundant material also was found in the same region to illustrate the genealogy of the Horse, and a memoir on this subject is in course of preparation. To General Sherman and General Sheridan, of the United States Army, my grateful acknowledgments are especially due, since without their continued assistance the investigations—of which the present volume is the first fruits—could not have been made. To the many other officers of the Army who aided me in the field during my various explorations, often in regions dangerous from hostile Indians, my sincere thanks are likewise due. For important aid in the preparation of the present volume, I desire to express my full acknowledgments to Mr. Oscar Harger, Assistant in Paleontology, in the Museum of Yale College. My thanks are also due to Mr. George Bird Grinnell, Assistant in Osteology, for much valuable assistance, especially while the volume was in the press. Oe. M. Yate Cottece, New Haven, Conn., June 16th, 1880. bath ee Chae LS ISI ie PGE PME. aie Preis} 5 a 7 ae q i — ‘ od A pe 7, 3 waa . o> +17 thik o GATT tty Ts uta ts Reriigterat Bea eon GT ollie 20 Wyatt: _ eee ag nang eat eas 8 wl Eat 1 aaa. jad signet er, a baila eh Lt ie re 7 Fy oe Bt at ta wargatt outa os cae Sega bag caeirh ta tala eee ea poe stehe Abd 4 i 2, eae SR ee Pn Thigh ite Gh vee Pees ay: oh ap SCY ep eae aa Lu eng ais p63 glen ae al ae ae oe YORE, St Qwa vite US? Cebedat See eee ce a sh OS Sr he eee 207 ae wins hadby : vue Lincs, we "pica cht Byun diet dy ate fous igi ao fire it ees i aye " ° ¢ - r ‘ = Lvs id -, a PST CM Ch SI, oe BGs Oar ak wuts Shree HOM eS “4 s | 4 *, ie Se Wnddo BE WE Joni ier Alia ie aoeene ore Beg se = ~~. ¥ . - > a 4 “~ = ; ba F’ * iat ; rigs) veld yl thal P ie, Dea + Taunt eS) : ‘ m . P : os ee ak ae ee Se a Y 7 ie 14 t > +3 sondt Pas LAG Ye STeme Pe geet .. « - ae > tock) ania liesy See ee at cx sy aL ’ ertul a8 Mp a a Rr whe > = i bos We “— ANB. hia bets date Ole i dain SON prank j Pah ay Neo Nac eet Mes Bt oo ~ , A See OP ‘Wale rs +h ye oe ee , } it) ae Led P te 4 “fe ae { as z ‘ - by za i ee 1 aeoat gi Ae Way Dae De ori te Y < = Pet MaDe Lt AON ae ee € Hse ‘d : . ~~ 5 . c < iJ = «. ° *) ‘ ~~ & wr « INTRODUCTION. Tuer remains of Birds are among the rarest of fossils, and very few have been discovered except in the more recent formations. According to present evidence, the oldest known Birds were imbedded in the Jurassic deposits of Europe, which have yielded three individuals belonging to the genus Archeopteryx, so well preserved that the more important characters can be determined. The only other remains of birds found in the Mesozoic of the Old World are a few specimens from the Cretaceous of England, which are too fragmentary to throw much light on the extinct forms they represent. The earliest traces of Birds hitherto found in the strata of this country are from the Cretaceous, although we may confidently predict their discovery in the Jurassic beds, if not at a still lower horizon. There is at present no evidence whatever that any of the three-toed impressions in the Triassic, described as the footprints of Birds, were made by Birds; and the proof now seems conclusive that nearly all of them are the tracks of Dinosaurian reptiles, bones of which occur in the same deposits. In the Cretaceous beds of the Atlantic Coast, and especially in the green-sand region of New Jersey, various remains of Birds have been found, and described by the writer! These fossils, although often in excellent preservation, occur mainly as isolated bones, and hence their near affinities have not as yet been determined with certainty. ‘See Synopsis at the end of this volume. 1 ; 1 2 ODONTORNITHES. Along the eastern slope of the Rocky Mountains, and especially on the adjoining plains in Kansas and Colorado, there is a series of Cretaceous strata remarkably rich in vertebrate fossils. The deposits are all marine, and, away from the mountains, they lie nearly horizontal. They have suffered much from erosion, and are still wasting away, especially along the river valleys. These beds consist mainly of a fine yellow chalk and calcareous shale, both admirably adapted to preserve delicate specimens, and here have been found the extinct Birds which form the subject of the present memoir. The geological horizon of the known Odontornithes is in the Middle Cretaceous, and corresponds to the strata named by the writer the “ Pteranodon beds.” The latter are included in sub-division number three, in Meek and Hayden’s section. The accompanying fossils are Mosasauroid reptiles, which are very abundant; Plesiosaurs, allied to Pliosaurus ; Pterodactyles, of the genus Pteranodon ; and many Fishes. With these occur Rudistes, and occasionally Ammonites, Belemnites, and various other Cretaceous invertebrates. The first Bird fossil discovered in this region was the lower end of the tibia of Hesperornis, found by the writer in December, 1870, near the Smoky Hill River in Western Kansas. Specimens belonging to another genus of the Odontornithes were discovered on the same expedition. The extreme cold, and danger from hostile Indians, rendered a careful explora- tion at that time impossible. In June of the following year, the writer again visited the same region, with a larger party, and a stronger escort of United States troops, and was rewarded by the discovery of the skeleton which forms the type of Hesperornis regalis, Marsh. Various other remains of Odontornithes were secured, and have since been described by the writer. Although the fossils obtained during two months of exploration were important, the results of this trip did not equal our expectations, owing in part to the extreme heat (110° to 120° Fahrenheit, in the shade) which, causing sunstroke and fever, weakened and discouraged guides and explorers alike. INTRODUCTION. 3 A considerable part of these Cretaceous deposits still remained unex- plored, and in the autumn of 1872, a third expedition through this territory was undertaken by the writer, with a small party. Additional specimens of much interest were secured, including the type of the genus Apatornis, and one nearly complete skeleton of Hesperornis,—an ample reward for the hardship and danger we incurred. The specimens thus secured by these various expeditions have since been supplemented by important additions, collected in the same general region by different parties equipped and sent out by the writer, who no longer could give his personal supervision to work in that field. The fossil Birds procured in this region since 1870, by these different expeditions, include remains of more than one hundred different individuals of the Odontornithes. These are all in the Museum of Yale College, and form the material on which the present volume is based. A study of this extensive series of Bird remains brings to light the existence in this class of two widely separated types, which lived together during the Cretaceous period, in the same region, and yet differed more from each other than do any two recent birds. Both of these types possessed teeth, a character hitherto unknown in the class of Birds, and hence they have been placed by the writer in a separate sub-class, the Odontornithes. One of these groups includes very large swimming birds, \ without wings, and with the teeth in grooves ( Odontolce ), and is represented by the genus Hesperornis. The other contains small birds, endowed with great powers of flight, and having teeth in sockets (Odontotorme ), and biconcave vertebree ; a type best illustrated by the genus Ichthyornis. Other characters, scarcely less important, appear in each group, and we have thus a vivid picture of two primitive forms of bird structure, as unexpected as they are suggestive. A comparison of these two forms with each other, and with some recent birds, promises to clear away many difficulties in the genealogy of this class, now a closed type; and hence they are well worthy of the detailed description and full illustration here devoted to them. 4 ODONTORNITHES. The fossil birds now known from the Cretaceous deposits of this country are included in nine genera, and twenty species. These have all been described by the writer, and are represented, at present, by the remains of about one hundred and fifty different individuals. This is evidence of arich and varied avian fauna in America during Mesozoic time, and likewise indicates what we may expect from future discoveries. OLIN NO Nei ALES. eee ee lea ea LOVER Gr “EOE Crs. (Plates I-XX.) CHAPTER “1. THE SKULL OF HESPERORNIS. (Plates I-II and XX.) Ir is most fortunate for science that Hesperornis regalis—with a single exception, the oldest bird known—should now be represented by remains as complete as any fossil skeleton yet discovered, even in the later formations. Nearly all the bones of the specimens obtained were almost as perfect as in life, when first found in the matrix; although the more delicate parts were sometimes unavoidably broken in removal, and occasionally small fragments were lost. Many of the bones were near their natural position when discovered, and in such cases a special effort was made to preserve this position, or retain a record of it, by drawings. However difficult such a method of collecting really is, in the region explored, its importance will be fully appreciated by anatomists. Various remains belonging to about fifty different individuals of Hesperornis are now in the Museum of Yale College, and the most important of them are described in the present memoir. 6 ODONTORNITHES., Tue Sxuty. (Plates I-II.) The skull of Hesperornis regalis, Marsh, is long and narrow, the rostral portion forming about two-thirds of its entire length (Plate I, figure 1). Viewed from above, the outline of the whole skull is wedge-shaped; and from the side, it presents a similar form, but somewhat more acute. It has a general resemblance to the skull of Colymbus torquatus, Briinnich (figure 2, page 9); but the brain-case is smaller, and the facial portion more produced. The position and size of the orbits, and of the anterior nasal apertures, are similar; but the likeness soon ceases, for the type of cranial structure is essentially different in the two genera. In its more important characters, the skull of Hesperornis resembles that of the Ratite, or Struthious birds, and we shall find other striking evidences of affinity with this group in various portions of the skeleton. The base of the skull shows nearly all the cranial characters which Tuxley, in his invaluable memoir on the Classification of Birds, lays down to distinguish the Ratite,! namely : (1.) The posterior ends of the palatines, and the anterior ends of the pterygoids are very imperfectly, or not at all, articulated with the basisphenoid rostrum. (2.) Strong “basipterygoid” processes, arising from the body of the basisphenoid, and not from the rostrum articulate with facets which are situated nearer the posterior than the anterior ends of the inner edges of the pterygoid bones. (3.) The upper, or proximal, articular head of the quadrate bone is not divided into two distinct facets. The occipital condyle in Hesperornis is cordate in outline, and its articular face is only moderately convex (Plate II, figure 4.) It is much less rounded than the condyle in the Ratite, and has a longer base. The articulation looks backward, and there is a shallow median groove on the upper half. The foramina in the base of the skull have the same general position as in recent Struthious birds. * Proceedings Zoological Society of London, 1867, p, 418. SKULL OF HESPERORNIS. 7 The palatines resemble those of the Ostrich. They are long, slender bones, extending from their union with the pterygoids, parallel with the axis of the skull, and joining the premaxillaries. In one specimen, they lie nearly in position, the left palatine being immediately beneath the left maxillary. In front of their junction with the pterygoid, and just anterior to their widest expansion, there is a deep pocket, very similar to that in the palatine of the Ostrich. The anterior half of the bone is slender, and tapers gradually to the extremity. The vomers in Hesperornis are separate, as in lizards and a few existing birds. They are smaller than the palatines, and resemble the vomers of Rhea, more than those of the other Ratite. They are broadest at the base, which is obliquely truncated, and they taper gradually to the pointed extremity in front (Plate II, figure 8.) The thick posterior end may possibly have united with the pterygoid, as well as with the palatine. Both vomers are preserved in the skull figured in Plate I, but are displaced. The bones of the brain-case are ankylosed, but in other parts of the skull the sutures are distinct, and many of them are open, as in the Ostrich and other Ratite. The orbits are large, and placed near together. There is a well-ossified interorbital septum, which is perforated by a large fenestra. (Plate I, figure 1, of) The superior temporal arch is wanting, and the orbit was closed below by the strong quadrato-jugal bar. There was a large temporal fossa, bounded in front by a distinct postorbital process. There are well marked glandular depressions extending along the roof of the orbits, as in Colymbus and some other recent aquatic birds. The quadrato-jugal bone is slender, and its articular head fits into a deep pit in the lower end of the quadrate. There is a large triangular lachrymo-nasal fossa, between the orbit and the anterior nasal opening. The latter is long and narrow, and closed behind by the nasal bone. The suture between the frontals and the beak was quite open. The lachrymal bone is distinct, and articulates with the frontal, nasal, -and maxillary. The nasal is deeply excavated in front for the anterior narial aperture, and below joins both the maxillary and premaxillary. 8 ODONTORNITHES. The premaxillaries are elongate, and separate throughout their posterior two-thirds, (Plate I, figure 5.) Their extremities touched the frontals. Their sides are deeply excavated for the anterior nares, and in front they are ankylosed, and form a long pointed beak, the end of which is somewhat decurved. This extremity, back to the nasal openings, has its surface pitted with irregular vascular foramina, indicating, apparently, that it was once covered with a horny bill, as in modern birds. The various cranial characters above described may nearly all be found in recent birds, if we search through different groups; but in Hesperornis the stout maxillary bones were armed with well developed teeth, a feature unknown in this class, before the discovery of the remains described in the present volume. These teeth were set in a-deep continuous alveolar groove, with only faint indications of separate sockets (Plate II, figure 1.) They resemble most nearly, in form and structure, the teeth of reptiles, and are fully described in the following chapter. There were no palatal teeth, and none in the premaxillaries. Tue Brat. The brain of Hesperornis was quite small, and more reptilian in type than in any adult bird hitherto examined. On page 9, figure 1, the skull in this genus is represented, with the outline of the brain-cavity in position. The skull of the Loon is added in figure 2, also with a cast of the brain in its natural position, and life size. A comparison of the two, places the relative magnitude and proportions of the brain in each in strong contrast. The reptilian type is shown on the same page in figure 3, which represents in natural size a cast of the brain-cavity of a young Alligator. In Hesperornis regalis the olfactory lobes (figure 1, 0!) were large and elongate, and their nerves passed out of the cranium by separate orifices, one on each side of the interorbital septum. The cerebral hemispheres (figure 1, c) were of very moderate size, much smaller, proportionally, than in any existing birds, and strongly resembling the corresponding part in some reptiles. The two lobes were narrow, and sub-ovate in outline, SKULL OF HESPERORNIS. 9 Fig. 1. “Fira. 2. Figure 1. — Outline of skull and brain-cavity of Hesperornis regalis, Marsh; seen from above; three-fifths natural size. Ficure 2. — Outline of skull and brain-cavity of Loon, (Colymbus torquatus, Briinnich); same view; natural size. Figure 3. — Cast of brain-cavity of young Alligator ; seen from above; natural size. ol. olfactory lobes; c. cerebral hemispheres; op. optic lobes; cb. cerebellum; /. flocculi; m. medulla. 2 10 ODONTORNITHES. and were separated above by an osseous median crest, depending from the roof of the brain case. The optic lobes (figure 1, op) were large, and very prominent, and only slightly covered above by the cerebral hemis- pheres. Their similarity in size and position to the optic lobes of reptiles is especially noteworthy. The optic nerves were large. The cerebellum (figure 1, cb) was quite large, and reptilian in its general features. The flocculi were well developed, and lodged in distinct cavities. In contrasting the brain-cavity of Hesperornis with that of Colymbus, as shown together in figures 1 and 2, a striking difference is seen in the size, the latter being about three times the bulk of the former. The two skulls are represented of the same absolute length, for the purpose of direct comparison. If, moreover, the relative size of the entire skeleton in each case be likewise considered, the brain of Hesperornis would have even less than one-third the relative capacity of that of the Loon. As the two birds were evidently similar in shape, and habits, the comparison seems to be a fair one. Another marked difference in the brain of the two genera is seen in the relative size of the cerebral hemispheres, as represented in figures 1 and 2. The cerebral lobes of Colymbus are very large, and much expanded transversely ; and it is in this portion of the brain that the real difference of size is most apparent. These facts are especially important, since they tend directly to show that the essential principles of the law of brain-growth, established by the writer in extinct mammals, applies also to birds. This law, briefly stated, is as follows: 1. All Tertiary mammals had small brains. 2. There was a gradual increase in the size of the brain during this period. 3. This increase was confined mainly to the cerebral hemispheres, or higher portion of the brain. 4. In some groups, the convolutions of the brain have gradually become more complicated. 5. In some, the cerebellum and the olfactory lobes have even diminished in size. SKULL OF HESPERORNIS. 11 6. There is some evidence that the same general law of Dbrain- growth holds good for Birds and Reptiles, from the Cretaceous to the present time.! The concluding suggestion was in part based on facts now published for the first time in the present memoir. Tue Lower Jaws. (Plate I, figures 2--4.) The lower jaws are long and slender, and were thickly set with teeth. The rami were united at the symphysis in front only by ligament, a feature unknown in modern adult birds. There is an imperfect articulation “between the splenial and angular elements, which probably admitted of some motion ; and all the other sutures are open, or distinguishable. There was apparently a mandibular foramen. There is a well marked shallow groove on the outer superior margin of each dentary bone, for the reception of the maxillary teeth, when the jaws were closed. (Plate I, figure 3, b.) The angle of the mandible extends backward but a short distance beyond the articular face for the quadrate, and the extremity is obliquely truncated. The following measurements of the skull of Hesperornis regalis are taken mainly from the very perfect specimen figured in Plate I, and recorded as number 1206, in the Geological Catalogue of Yale College Museum. The other dimensions are derived from number 1207, of the same Catalogue, to which reference is made in all the numbers of specimens given in the descriptions that follow : Measurements of Skull. (Hesperornis regalis, No. 1206.) Length of skull, from occipital condyle to end of premaxillary,.---.----------- 257.07" Greatest transverse diameter, behind temporal fosse,-----..------------------ 61.0 Least transverse diameter, between temporal fossw, -.--.--------------------- 30.0 Greatest transverse diameter, between temporal fossze and orbits,--.--.------- 42.0 Least transverse diameter, between orbits,..-..... --.-------.------------- 9.0 Transverse diameter of premaxillaries, at anterior nares,_ .------------------- 9.0 Antero-posterior diameter of anterior nares, _--.-----.---------------------- 55.0 ? American Journal of Science and Arts, vol. viii, p. 66, July, 1874; and vol. xii, p- 61, July, 1876. 12 ODONTORNITHES. Transverse diameter of anterior nares, .-..----.-2 =e = eee ee nee ae eee Extent of premaxillary in front of anterior nares,.. -.---------------------- Transverse diameter of occipital condyle, .-.- ----222---2--ee=. == =~ --~--- === Vertical diameter of occipital condyle)- 2-25 sneer ea Dransverse diameter. of foramen Magnum) 2 sose= es -- === eee a ee Measurements of Skull. (Hesperornis regalis, No. 1207.) Transverse diameter. of oceipitalicondyle essa -- = eaaeee= ne oan eee eee Vertical diameterof. occipital (condiylesea ese es te Transverse diameter of foramen magnum,.--.---.-=.-----------. =---=------ Distance between external margins of quadrate articulation, -....------------ Greatest width of skull, behind quadrate articulations, ......-.-. ----------- Transverse diameter of quadrate articulations,..._...---------------------- Antero-posterior diameter of quadrate articulations, ----.--..----------- Ja25 Length /of quadratic, o-oo oso eee en eee ee ee er Ree eee ee ‘Transverse diameter otuproximaliend, sas sne = = eee eee eee Antero-posterior diameter of proximal end,_--.------.------=----.--------- Transverse diameter of shaft, at) middle,----622*_=23: 2 -- =o. =o eee Verticali diameter of pit for quadrato-jugaljs=2 22) 22 enan ee ee Antero-posterior diameter of pit for quadrato-jugal, ..--..--------.---------- Longest diameter of inferior condyle of quadrate, .-----..-.----------------- Measurements of Lower Jaw. (Hesperornis regalis, No. 1206.) bengthiof lowerjaw, 2-2. 2c. -22 ss see estes sees yas ee ee ee eee Greatest vertical diameter/of lower jaw,-=.-- -------------s-======——==eeene Extent ofealveolarierooves-<- 2. 322562 yoo eso eon ha ee ee Depth of alveolar rro0ve, on = 22 seems cae a ne ene ee Ee Greatest diameter of posterior facet for articulation with quadrate, .---.------- Least diameter of posterior facet for articulation with quadrate,........----.-- 9,9™™ 88.0 10.0 6.0 9.0 2.000 8.0 12.5 49.0 55.0 8.0 6.5 32.0 9.0 7.0 4.0 6.0 4.0 12.9 257.0" 25.0 ——~ OHAPTER. 1 THE TEETH OF HESPERORNIS. (Plates I, II, and XX.) Tue absence of teeth has long been regarded as one of the best distinctive characters of Birds; since teeth are present in some members of all the other classes of vertebrates. All existing Birds have the jaws covered with a horny sheath, which is usually smooth. In some groups, this beak-covering is more or less sérrated, and in a very few forms the elevations correspond to slight projections of the bone beneath, but no indications of true teeth have yet been found. This may all be said like- wise of existing Turtles. Geoffroy Saint-Hilaire and others have, indeed, detected in the young of some recent birds vascular papille which resembled those of embryonic teeth; but they were apparently portions of the undeveloped horny beak. The teeth of Hesperornis are true teeth, with their distinctive charac- ters as well marked as those of any reptile. In the upper jaw, they are confined to the maxillary bone alone, the premaxillary being entirely edentulous. In the lower jaw, the teeth extend from very near the anterior extremity of the ramus along the entire upper border of the dentary bone. The teeth above and below were implanted in a continuous groove, somewhat like those of Ichthyosaurus. From the sides of the groove, slight projections extend between the teeth, thus forming faint indications of sockets; but these projections are not sufficient to materially lessen the width or depth of the groove. (Plate I, figure 11). 13 14 ODONTORNITHES. The teeth were evidently held in position by cartilage, which permitted some fore and aft movement; but lateral motion was much restricted by the depth and narrowness of the groove, and the large size of the fangs. With the decay of this cartilage after death, the teeth readily lost their erect position, and became more or less displaced. In the best preserved specimen of Hesperornis regalis (number 1206), most of the teeth had fallen out of the grooves, when found, and were lying scattered along beside the jaws, as often seen in specimens of Ichthyosaurus. The embryonie sockets, in this specimen, indicate that there were fourteen functional teeth in the maxillary bone, and thirty- three teeth in the corresponding ramus of the lower jaw, as shown in Plate I, figures 1-4. In Hesperornis crassipes (number 1474), a number of teeth were likewise found in position in the jaws, but the exact number originally in each could not be determined. The teeth of Hesperornis were gradually replaced by successional teeth, and this took place in a manner very similar to that in some reptiles. The germ of the young tooth was formed on the inner side of the fang of the tooth in use, and, as it increased in size, a pit for its reception was here gradually made by absorption. The old tooth at last became undermined, and was expelled by the new one, which occupied the same position, the number of teeth thus remaining the same. The teeth of Hesperornis have conical pointed crowns, covered with smooth enamel, and supported on stout fangs. In form of crown and base, they closely resemble the teeth of Mosasauroid reptiles, one of which is represented on page 15, figure 7, for comparison. The outer and inner surfaces of the crown are separated by sharp ridges, which are without serrations. The outer side is nearly plane, and the inner surface strongly convex, The crowns of the teeth are mainly composed of firm dentine, invested with a layer of enamel. The relative proportions of these are shown in figure 4. The pulp cavity was large, and in the specimen above repre- sented was filled with calcite. The coronal walls of this cavity are smooth, and well defined. The fang consists of osteo-dentine. TEETIT OF HESPERORNIS. 15 QQ Ae ar Vicurs 4. — Vertical section of tooth of IZzsperornis regalis, Mars; (No. 1206). Fieuae 5. — Horizontal section of same tooth; both enlarged thirty-two diameters. Ticure 6.— Tooth of Hesperornis regalis (No. 1206); enlarged eight diameters. Tigtr2 7.— Tooth of Mosasaurus princeps, Marsh ; half natural size. a. enamel of crown; b. dentine; c. pulp cavity in crown; 0’ root of tooth; c’. absorbed cavity in root, d. youag tooth; ss. intersection of horizontal and vertical sections. 16 ODONTORNITHES The layer of enamel gradually increases in thickness from the base of the tooth to the apex, as shown in figure 4, a. It is also somewhat thicker over the anterior cutting edge than in other portions of the base of the crown, as indicated in figure 5. The line of junction between the enamel and the dentine is everywhere sharply defined. The enamel is dense and hard, the calcification having proceeded so far, in the specimens examined, that the constituent fibres could not be distinguished. The external surface of the enamel is nearly smooth, but marked by delicate strie. There is no indication of cement on the coronal surfaces. The dentine, which forms the mass of the crown, shows a well marked structure in both the vertical and horizontal sections, figures 4 and 5, b. It is firm and compact, and the calcigerous tubes are well defined. Near the base of the crown, they radiate horizontally, and in transverse section appear nearly straight, as shown in figure 5. Higher up in the crown, these tubes curve upward, diverging from the axis of the tooth less and less, until beneath the apex they become nearly parallel. In the dentine, there are distinct concentric lines of growth. These are seen near together and numerous in transverse section, figure 5, while those which appear in the vertical section, figure 4, are fewer in number, and more strongly marked. The teeth of Hesperornis, taken by themselves, appear to resemble more nearly the detached teeth of Mosasauroid reptiles, than any others known, not excepting the teeth of Ichthyornis, which will be fully described in a subsequent chapter. The teeth in the latter have compressed crowns, and are implanted in distinct sockets. In all their main features, the teeth of Hesperornis are essentially reptilian, and no anatomist would hesitate to refer them to that class, had they been found alone. Combined with the other reptilian characters of Hesperornis, noted elsewhere in the present volume, they clearly indicate a genetic connection with that group. On AP TLE Re: LEY. THE PRESACRAL VERTEBRA OF HESPERORNIS. (Plates III-V and XX.) In the type specimen of Hesperornis regalis, a number of vertebrae were preserved from various parts of the series, including nearly all of the caudals. Other individuals of this species, since discovered, have furnished all the missing vertebrae, except the atlas. An examination of the occipital condyle and of the anterior parts of the axis leaves little doubt as to the form and proportions of the atlas, so that the entire axial skeleton can now be determined with almost absolute certainty. The presacral vertebrae of Hesperornis resemble in their more important characters the corresponding vertebrae of existing birds. The articular faces of the centra conform strictly to the modern ornithic type, an interesting fact, as we shall see, when we compare them with the vertebree of another group of Odontornithes, from the same geological horizon. The neck of Hesperornis regalis was long and slender. Including the atlas and axis, thero wore seventeen cervical vertebrae ; and twenty-three in all between the skull and sacrum. The last three cervicals have free ribs, and would be called cervico-dorsals by some anatomists, and dorsals by some others. As the eighteenth vertebra is the first united to the sternum by the intervention of a sternal rib, it seems best to regard it as the first dorsal, and this view will be adopted in the present volume. 3 17 18 ODONTORNITHES. There are fourteen vertebrz in the coéssified sacral series represented in Plates X and XI, and there were twelve more in the tail, as shown in Plate XII, making the entire column consist of forty-nine vertebrae. This is a very large number for this class, and is equaled in but very few recent birds. None of the vertebrae of Hesperornis were ankylosed together, except in the sacral series, and at the extremity of the tail. None of the vertebra contained pneumatic openings, although some were lightened by medullary cavities. THe Atias anp Axis. (Plates III and XX.) The atlas was evidently a short ring of bone, very similar to the first vertebra in modern aquatic birds. Its articular cup for the reception of the occipital condyle was not deep, and was cordate in outline. The atlas articulated with the axis by well developed zygapophyses.above, and by a crescent-shaped tubercle below. The centrum of the atlas is codssified with the axis, as the odontoid process. The axis (Plate III, figure 1) is well preserved, and resembles the corresponding vertebra of the Loon (Colymbus torquatus, Briin.); but it is one-third larger, and a much stouter bone. The posterior articular face is much less oblique, being nearly vertical. The odontoid process is relatively shorter and broader, and its articular face is more extended backward, along the median line below. This process is slightly concave above, in prolongation of the floor of the neural canal. The centrum of the axis is expanded at each end, for its articular surfaces. The anterior expansion is considerably greater, and somewhat more gradual, than the posterior; and, at the least expanded portion, its thickness is less than two millimeters. The anterior articular surface is reniform, the upper outline being emarginate, to make room for the base of the odontoid process. The articulation is concave in both directions, but much more deeply concave vertically than transversely. The anterior portion of the centrum of the axis is deeply excavated at the sides, nearly on a level with the floor of the neural canal, for the transmission of the vertebral artery. A very delicate bridge of bone CERVICAL VERTEBRA OF HESPERORNIS. 19 may have existed here, completing the foramen. In the succeeding cervical vertebra, this foramen is well developed, and may conveniently be termed the lateral foramen. It is formed by the diapophysis above and the parapophysis below, and closed externally by their union with the intervening pleurapophysis. In life, this aperture probably protected both the vertebral artery and vein, and the main trunk of the sympathetic nerve, as in most recent birds. A hypapophysis extending over about the posterior half of the centrum is indicated, but is not preserved. The posterior articular surface is nearly square in outline, with rounded corners. It is only slightly oblique, the lower portion projecting but little beyond the upper margin. As in ordinary birds, it is distinctly concave vertically, and convex transversely. The neural arch was surmounted by a well developed neural spine, which is broken off in the present specimen. The pre-zygapophyses are small, oval, convex facets, slightly raised above the surrounding bone, and looking outward and upward at an angle of about 45°. The post- zygapophyses are much larger. They are concave, and look nearly directly downward. The neural canal is somewhat depressed, perhaps in part by crushing. The post-zygapophyses extend within about two milli- meters of the posterior articular surface of the centrum. They are strongly supported above by a ridge arising upon the side of the neural arch, and running outward and backward over the post-zygapophyses. No vascular or pneumatic foramina are visible upon this vertebra. This description of the axis of Hesperornis regalis is derived mainly from the specimen, number 1207, in which the series of presacral vertebree is nearly complete. The following measurements are taken from the same vertebra : Measurements of Axis. (No. 1207.) Length of centrum, including odontoid process,---..--------------- Oe ear 29.0™™ Pepin Gf GGNbNNIN, 26. is 5 who. 2o-- =~ -- as = nae e--------------------- 21.0 Least transverse diameter of centrum (approximate),-~--..--.----------------- 2.0 engthyof adontoid process). 2-2 22=- 22222-—. 2-222 -. 25-262 ----2-------- 8.0 Transverse diameter of odontoid process, at base, --..------------------~----- az Vertical diameter of odontoid process, at base, .-....----------------------- 4.0 Transverse diameter of anterior articulation of centrum, .----.---------------- 13.0 20 ODONTORNITHES. Vertical diameter of anterior articulation of centrum,.........-.-.---------- 6.0™™ superior, ..---=---- 9.0 Transverse diameters of posterior articulation of centrum, / median, ----------- 7.0 inferior =se- se == 8.0 Vertical diameter of posterior articulation of. centrum,..-.------.----------- 8.0 Transverse diameter of vertebra, across pre-zygapophyses,-.---.------------- 13.0 Transverse diameter of vertebra, across post-zygapophyses (approximate), ------ 24.0 Greatest diameter of pre-zygapophyses,---..- -----+----------+------------ 5.5 Least diameter of prezyeapophyses,--...--.-- 4222-2 5ee2 == ee eee 4.0 y, ‘ : == -{ b Diameters of post-2ye0popbySe® | otzontal, --ennsae-scosscsae 18 hence th ofiflooniofi meural canal =. ses. Sere eee ee 30.0 Length of roof of neural canal (approximate), _--...._..-__--.--..-_----. =. 26.0 Transverse diameter of anterior opening of the neural canal, ..--------------- 7.5 Vertical diameter of anterior opening of the neural canal, __.--_-.--.-------- 6.0 Transverse diameter of posterior opening of the neural canal,_-----------.---- 8.8 Vertical diameter of posterior opening of the neural canal, .....------------- 5.0 Tue Txurrp VertTesraA. (Plate III, figure 2.) The third vertebra of Hesperornis regalis resembles that of the Loon, but has the articular faces of the centrum less oblique. The centrum is compressed laterally, as in the axis, becoming quite narrow at and behind the middle; then expanding suddenly, for the posterior articulation. The anterior articular face is supported by a more gradual expansion. The face itself is somewhat oblique, looking downward at an angle of less than 30°. It is concave transversely, and convex vertically, as in ordinary birds. Immediately behind this articulation, the under surface of the centrum is excavated. Beyond the excavation, the hypapophysis gradually becomes prominent, quite thin in front, but expanded and flattened into a button-shaped process, below the posterior articulation. The latter face is very similar in shape to that of the axis, but is somewhat larger, as shown by the measurements, and slightly more oblique. The neural arch bore a less powerful spine than that of the axis. The pre-zygapophyses are much longer, and larger than in the axis, and they look inward and upward, instead of outward, as in that vertebra. They are, however, turned only slightly inward. The post-zygapophyses are less elongated than the pre-zygapophyses; and, as in the axis, are strengthened by a ridge beginning about the middle of the upper part of the CERVICAL VERTEBRA! OF HESPERORNIS. 21 neural arch, and continued outward and backward. Between this ridge and the neural spine, the surface of the vertebra is excavated posteriorly, but the posterior edge of the neural arch is considerably thickened, and rises in a ridge connecting the post-zygapophyses with the base of the neural spine. Over the post-zygapophyses, this ridge is developed into a projecting tubercle. The lateral foramen for the vertebral artery is small, and over-arched by a strong bridge of bone, from which a short, stout, obtuse pleurapophysis projects backward, and somewhat downward. The floor of the foramen is continued as a groove for a short distance backward upon the side of the vertebra, bounded below by a well developed nearly horizontal ridge. A similar, but much smaller, ridge is seen above; and, over this, a slight ridge appears on some specimens, extending a little farther back on the vertebra. The measurements given below show the principal dimensions of the third cervical vertebra of Hesperornis regalis, in the type specimen, number 1200, and in two other skeletons: Measurements of Third Vertebra. (No. 1200.) enpiiwonCenttumys ssa see se nooo) Fh oow = So Sanaa ses esse cec eas 4 hrs 22.5™™ Transverse diameter of anterior articulation of centrum,--.------- -------- 9.0 Vertical diameters of anterior articulation of centrum,’ .---.--------------- 6.0— 8.0 Transverse diameters of posterior articulation of centrum,--------. -------- 9.8— 8.2 Vertical diameters of posterior articulation of centrum, -----.-.------------ 7.0— 9.6 Transverse diameter of vertebra, across pre-zygapophyses, - - .------------- 19.0 Transverse diameter of vertebra, across lateral foramina, ------.----------- 20.0 Transverse diameter of vertebra, across post-zygapophyses, .---- ----------- 22.5-21.0 Diameters of pre-zygapophyses,......-.--.-.-----=----=-------=-----.----- 5.0- 9.0 Diameters of post-zygapophyses, .------.---..------ ---------------- -- 7.0- 8.0 Length of floor of neural canal, .-.....------.-----------------+-------- 22.0 Length of roof of neural canal, --.------------------------------------- 21.0 Length of hypapophysis,. .... ----.00 0+ eceae- 9-2 none e ee ee ee eee eee 6.5 Measurements. (No. 1206.) Rene Of MANEMNIO, ioe Lis io. S2 2 = Za ~--2. ---+.sodee05. a beeecsebeeeeeoees 29.0 Length of free portion of pleurapophysis, ......-...-.------------------- 21.0 Measurements. (No, 1207.) PEPSI ge Unie Cee CONTENT o ear ae hate ee ene Oe NS ks ES hse a 1 30.0™™ Transverse diameters of anterior articulation of centrum,-_------.-------- 11.5-14.0 Vertical diameters of anterior articulation of centrum,..-.....----.----.. 5.5— 9.0 Transverse diameter of posterior articulation of centrum, -.-.-.--.-.-------- 16.0 Transverse diameter of vertebra, across pre-zygapophyses, -- ---. --.---.--- 28.0 Transverse diameter of vertebra, across post-zygapophyses,......-. -.---- 27.0 Greatest diameter of pre-zygapophyses,......----.-------.--..--.------ 9.0 Least diameter of pre-zygapophyses, - -- - -- Se eee nance we «2 tee an 5.2 Greatest diameter of post-zygapophyses,.---.--. ------------------------ 10.0 Transverse diameter of anterior opening of neural canal,... ---.--------- 10.0 Vertical diameter of anterior opening of neural canal, ...-.-----------..-- 7.0 Bent nmote ple tra pO plysis, 04 ~ 8a .)tate st ts52 2c swocca d= 2eaed abe aueds 23.0 Length of free portion of pleurapophysis,.....-....-------------------- 18.0 Diameter of free portion of pleurapophysis, at base, . Se ee Pere ee no 5.5 iMextouldiameror of lateral foramen) -.---. .---scc-0 sees eae nce wero-=ce 5.8 Transverse diameter of lateral foramen,-.-....-.---.---.-.------------- 4.0 Tue Seventu Vertesra. (Plate III, figure 6.) The seventh vertebra is slightly longer and larger than the sixth. The centrum has the lateral ridge above the groove for the vertebral artery less developed than in the sixth, and not coalescing with the ridge below; which, on the other hand, is well developed, and clearly bounds the lower surface of the centrum. This surface is broader than in the sixth vertebra, and widely excavated longitudinally, throughout, the excavation being much deeper and wider anteriorly than in the sixth vertebra, and nowhere disappearing on the surface of the centrum. Just behind the outer inferior margin of the anterior articular face on each side, is a tubercle, somewhat elongated longitudinally with the vertebra, and projecting downward, and slightly inward. These tubercles become a prominent fea- ture on some of the sueceeding vertebrae of the series, but only the faintest trace of them is visible on the sixth vertebra, and they may be considered 28 ODONTORNITHES. as commencing upon the seventh. A similar structure is seen in the cor- responding vertebra of the Loon. In Rhea, these tubercles suddenly appear upon the sixth vertebra. The elongated tubercles in which the lateral ridges terminate posteriorly are smaller and less conspicuous in the seventh than in the sixth vertebra. The articular surfaces are larger, and some- what more expanded below. The pre-zygapophyses are nearly flat, being only slightly convex. The post-zygapophyses are slightly concave, especially near the internal angle. The base of the neural spine extends over nearly the whole length of the neural arch, but was probably shorter antero-posteriorly above, than in the sixth vertebra. The lateral foramen is slightly larger, and the diapophysis upon its outer wall is greater, more produced down- ward and forward, and farther from the pre-zygapophysis than in the sixth vertebra. A short ridge rising into a crest over the post-zygapophysis is developed in one specimen (number 1208). The posterior outline of the neural arch is notched behind at the termination of the neural spine, as in the preceding vertebra, but less deeply. The pleurapophyses are longer than in the sixth vertebra. The measurements of two specimens of the seventh cervical vertebra of Hesperornis regalis are given below: Measurements of Seventh Vertebra. (No. 1206.) Teneth of centrum) 222 <2 = S822. oa om a ae ee re =35' 32.0™™ Transverse diameters of anterior articulation of centrum, ------------------ 10.0-16.0 Vertical diameters of anterior articulation of centrum,--.--------.-------- 7.0-10.0 Transverse diameter of vertebra, across pre-zygapophyses, . - -- - --- ---- ---- 25.0 Transverse diameter of vertebra, across post-zygapophyses, - -------------- 27.0 Diameters of post-zygapophyses,2- <= <= 22 2222S ee ee eee oOo TotalJencthiof pleurapophynis;-=_-220---0 ae oe eee mt 31.0 Length of free portion of pleurapophysis,-----..-....-...--------------- 23.0 Measurements. (No, 1207.) IDeA toil Ch ing tt Se eee ns ea ye ad ee $2,.0™™ Least transverse diameter of base of centrum,_-....._....---------- ae 8.5 Transverse diameters of anterior articulation of centrum,_---..-..------- 11.0-16.0 Vertical diameters of anterior articulation of centrum, ..---.----.---------- 6.5-10.0 Transverse diameters of posterior articulation of centrum, .------.-----..- 11,0-15.5 Vertical diameters of posterior articulation of centrum,....... -...------- 6.5-10.0 Transverse diameter of vertebra, across pre-zygapophyses,-._....--------- 26.5 CERVICAL VERTEBRA OF HESPERORNIS. 29 Transverse diameter of vertebra, across post-zygapophyses, ---.----------- 27.07™ Greatest diameter of pre-zygapophysis,-...--.-.----------------------- 10.5 Tess: diameter of pre-zygapophyals,_ -.---_.- -. 2 c2c- 2 oon we me eens ene =e 6.4 Greatest diameter of post-zygapophysis,...-..----.--.---. ------------- 12.0 Least’ diameter of post-zygapophysis, - --.... /--:---------7-<2.-2------- 8.0 Transverse diameter of neural canal, at anterior opening, ----------------- 9.0 Vertical diameter of neural canal, at anterior opening, -------------------- 6.5 MUN PUNLOE PleUranOpRyAIse= .osaa6 Joa tate tan ea ate ere ee oe 31.0 Diameter of. plourapophysiss ate bases. oa-cc4 =k sae eat oo see eee oe 6.5 Wentioal diameter of. lateral foramon,.--. 22. 220.2. wos sceen see ot ese se 6.5 ‘brankverse diameter of: lateral forameny. 2 2c .22 2.55225 -22428 ee 3.4 Tue Eieuta Vertesra. (Plate III, figure 7.) The eighth vertebra is the largest cervical of the series, slightly exceeding both the seventh and the ninth in length. It closely resembles the seventh, being slightly larger in nearly all its dimensions The ventral surface of the centrum is more broadly and deeply excavated anteriorly than in the seventh, and the tubercles, or processes, on each side of this excavation are more developed, and considerably incurved. They stand somewhat wider apart at their bases than in the seventh vertebra, and the whole under surface is more widely excavated. The two tubercles near the posterior end of the lower surface are small. The lateral ridge, start- ing at the upper end of the foramen for the vertebral artery, fades out upon the side of the centrum earlier than in the preceding vertebra. The diapophysis is somewhat stronger and larger than in the seventh vertebra, and the foramen itself is larger. The neural spine is shorter than in the seventh cervical. It tapers rapidly almost to a point above, and terminates posteriorly at the margin of the neural arch. The latter shows no median notch at this point, as in the preceding vertebree, the margin being gently and evenly excavated between the zygapophyses. The pre-zygapophyses are slightly convex, and the post-zygapophyses concave. The latter are strengthened by a ridge running along the side of the neural arch, and rising and becoming more prominent over the zygapophyses. The articular surfaces of the centrum are larger, and especially broader, below, than in the seventh vertebra. The pleurapophyses are of nearly the same length as in that vertebra. 30 ODONTORNITHES. The principal dimensions of the eighth vertebra, as determined from two specimens of Hesperornis regalis, are as. follows: Measurements of Eighth Vertebra. (No. 1206.) Length of centram, ..-. 22-2 /s<=2s=q— ae eee oe ee ee eee 33,0™™ Transverse diameters of posterior articulation of centrum, ..------------- - 11.0-15.0 Vertical diameters of posterior articulation of centrum, - - -- - ee ee 8.0-12.5 Transverse diameter of vertebra, across post-zygapophyses (approximate), _- 22.5 Diameters of pre-zygapopbyses, -- -.--.- -=22 2 soe yee oo eee eee OLED Diameters of spost-zygapophyses,=—- .4--5- S= - -6 ee ee ee ee 1 Length’ of roof of neural/canal;=.3. =~ ee =. os sy ee 30.0 Total length of plenrapophysis,.2--2-2=--5---4--=5- -25—= eae ee ee 33.0 Length of free portion of pleurapophysis, ------------------------------- 25.0 Measurements. (No. 1207.) hengthweok (centrum, =: 325 oes 2 See. 2) Se sb nia ae eee ee 33.0" iLeagt diameter of centrum. (at base) = 222 62-55 ee ae eee 9.8 Transverse diameters of anterior articulation of centrum,---------- ------- 12.0-17.0 Vertical diameters of anterior articulation of centrum, -------.------------ 7.0-11.0 Transverse diameter of posterior articulation of centrum, --.---.---------- 16.0 Vertical diameter of posterior articulation of centrum, ----.----------- eanee 9.0 Transverse diameter of vertebra, across pre-zygapophyses,---.------- ---- 30.0 Transverse diameter of vertebra, across post-zygapophyses,----.---.------ 29.0 Greatest diameter of pre-zygapophysis,......-- -.-. -=-----. -----=--.=--- 10.5 Least diameter of, pre-zygapophysis,...- --- = 2- == <2 5522-5 2 2- eee 7.8 Greatest diameter of post-zygapophysis,-....---. ----------------------- 12.0 Len¢throf roof of neural’canal;= (2.222 2 2 a ce ee ee 28.2 Transverse diameter of neural canal, at anterior opening, ----------.------ 9.0 Vertical diameter of neural canal, at anterior opening, ---------.--------- 6.5 Total length of pleurapophysis; =. -2=s2= 6-0 = -=452- eee (ee ee 32.0 Length of free portion of pleurapophysis,-----.. ---------.------.------- 21.0 Diameter of pleurapophysis, at base, ---2- a 2=- ae eae = 9.0 ‘Vertical diameter: of Jateraliforamen.]--o-4- see eee ee ee ere 8.0 ‘Transverse diameter of lateral foramen, ---- 2-8) nee oe eee 4.5 Tue Ninra Vertesra. (Plate III, figure 8.) The ninth vertebra has a centrum slightly shorter than the eighth, but somewhat broader and stouter. The under surface of the centrum is slightly broader, and is more deeply and widely excavated anteriorly. This excavation, as in the preceding vertebra, is somewhat over-arched by two descending ineurved processes, which are longer and larger than in the eighth vertebra, and are also set somewhat farther forward, so as to be CERVICAL VERTEBRA OF HESPERORNIS. 31 on a line with the posterior margin of the articular surface. The tubercles at the posterior end of the lower surface are small, and scarcely noticeable. The ridge above the arterial groove upon the side of the centrum is much as in the eighth vertebra. The articular surfaces of the centrum are slightly broader below than in the preceding vertebra. The ridge strengthening the posterior zygapophysis is well developed, and rises into a tubercle above the zygapophysis. This tubercle is espe- cially developed in number 1206. The neural spine extends over slightly less of the neural arch than in the eighth vertebra, and probably was not as high. The outline of the arch is evenly concave between the post- zygapophyses, and not notched on the median line. The diapophysis is developed about as in the eighth cervical. The pleurapophyses are longer than in the preceding vertebra. The dimensions of the ninth vertebra of Hesperornis regalis are shown in the tables given below: Measurements of Ninth Vertebra. (No. 1206.) Length of centrum,......------------.--- ---------------------------- 32.5™m Transverse diameter of anterior articulation of centrum, ---. .------------- 16.8 Transverse diameter of posterior articulation of centrum, .------. -------- 12.2 Vertical diameter of posterior articulation of centrum, . ------------------ 7.0 Transverse diameter of vertebra, across pre-zygapophyses, ------------ --- 30.0 Transverse diameter of vertebra, across post-zygapophyses, - --- ----------- 33.0 Diameters of post-zygapophyses, - --- ---- ------------------------------- 10.5— 8.0 Length of floor of neural canal,----.---------------------------------- 33.0 Length of pleurapophysis, - .-.-------- .---.-------- -------------------- 31.0 Length of free portion of pleurapophysis, - - -- ---------------------------- 22.0 Measurements. (No. 1207.) Length of centrum,......------.--------------- ---- ---+-----------+-- 31,0™™ Least diameter of centrum, at base,..-. .-..----.------ ---------------- 10.0 Transverse diameters of anterior articulation of centrum, -- .------.--------- 13.0-16.0 Vertical diameters of anterior articulation of centrum, -----.-------------- 8.0-13.0 Transverse diameter of posterior articulation of centrum, -------.---------- 12,2 Vertical diameter of posterior articulation of centrum,-- ----------------- 7.0 Transverse diameters of vertebra, across pre-zygapophyses, ---- ----------- 27.0-34.0 Transverse diameter of vertebra, across post-zygapophyses, - -------------- 31.0 Greatest diameter of pre-zygapophysis,.- --------.---.----------------- 10.5 Least diameter of pre-zygapophysis, .- .-------------------------------- 8.0 32 ODONTORNITHES. Greatest diameter of post-zygapophysis,......-------------.---------.-- 12.052 Gength:of floor of neural canal, -<- 2-2 oe os nen ae ee $1.5 Length of roof .of neural canal) -2= = =o- ee 28.0 Transverse diameter of neural canal, at anterior opening, .---..------------ 8.5 Vertical diameter of neural canal, at anterior opening, ....-_.-------------- 6.5 Length of plenrapophysis,---= -=--=.5 see ae ee a . ©3810 Length of free portion of pleurapophysis, -----.-.--..--.---.------------ 20.0 Diameter of pleurapophysis;at) base, =. 5 = ea eee eae Sa 9.0 Vertical diameteror/aterall foram ens soe ee 7.0 ‘Transverse diameter/of=lateral foramen, s. se = == —- ee 5.5 Tue Trento Vertesra. (Plate IV, figure 1.) The tenth vertebra is shorter and stouter than the ninth. The under surface is perceptibly broader, and somewhat more deeply excavated in front. The descending incurved processes, also, are longer than in the ninth vertebra, but the tubercles at the posterior ends of the lateral ridges are scarcely apparent. The ridge on the side of the centrum, above the vertebrarterial groove, is very similar to the one on the last vertebra. The neural spine is short, small, and low. The post-zygapophyses occupy a slightly more anterior position in this vertebra than in the preceding ones, a change that becomes more marked in the following vertebrae, where the zygapophyses are brought nearer together, in order to facilitate the dorsal flexure of the neck, near its base. The diapophysis is larger and more prominent than in the ninth vertebra, and the pleurapophyses are somewhat longer. The proportions of the tenth vertebra of Hesperornis regalis are fully given in the accompanying measurements from two different specimens : Measurements of Tenth Vertebra. (No. 1206.) Lenpth ‘of centrom,..=<--<=- ss 2425- Se snc ace ees een ee eae a ee 32.0™™ Transverse diameters of anterior articulation of centrum,. -.-------.------- 12.0-17.5 Vertical diameters of anterior articulation of centrum,_. ..--..---.-.------- 7.0-11.5 Transverse diameters of posterior articulation of centrum, .-.--.----------- 12,0-16.0 Vertical diameters of posterior articulation of centrum, _--...-.----------- 7.5-11.0 Transverse diameter of vertebra, across pre-zygapophyses, ....------------ 34.0 Transverse diameter of vertebra, across post-zygapophyses, -.---..----.---- 32.0 Greatest. diameter of pre-zygapophiyais,------ .--- <<< ~ 2-- 26.) o~ <2 cn-- soo au == se 8.0 Tue ExveventH Vertesra. (Plate IV, figure 2.) The eleventh vertebra is shorter than the tenth, and appears much shorter when seen from above, owing to the approach of the zygapophyses toward each other, to aid in the flexure of the neck. Its inferior surface is slightly less excavated than in the preceding vertebra, and the anterior descending processes occupy a more anterior position. The diapophysis is here developed into a prominent, tubercular, oblique ridge, running backward, downward and outward, and forming the antero-lateral margin of the vertebral outline, as seen from above. The neural spine is short, and small. The ridge running from the side of the vertebra to the post- zygapophysis rises into a prominent tubercle on each side of the upper surface of the vertebra. The post-zygapophyses are more oblique, and the pleurapophyses rather shorter than in the preceding vertebra. The size and proportions of the eleventh vertebra of Hesperornis regalis are given in the measurements below : 5 34 ODONTORNITHES. Measurements of Eleventh Vertebra. (No. 1207.) Length of centrum, ..-.-.-. .-- osa0cedeeese ee eee ae 29.0=™ Least. diameter of centrum, at base)... _---ee) see eee ee eee 9.0 Transverse diameters of anterior articulation of centrum,.---.---.---.---- .- 14,0-17.0 Vertical diameter of anterior articulation of centrum,_--. .-..-.---------- 8.0 Transverse diameters of posterior articulation of centrum, -----------.----- 11.0-14.0 Vertical diameters of posterior articulation of centrum, ...--.--.-.--.------ 7.5-12.0 Transverse diameter of vertebra, across pre-zygapophyses, ---------------- 39.0 Transverse diameter of vertebra, across post-zygapophyses, .- --.--------- 32.0 Greatest diameter of pre-zygapophyses,....... .--...--.--=-=-----=----- 13.0 Least: diameter of pre-zygapophyses,o- 22 —- eos see ee eee 8.0 Greatest diameter of post-zygapophyses,..--.--..--.--=------------ ---= 13.0 Least diameter of post-zypapophyses,.... =-----. =--=-2--=---=+---=----- 9.5 eng thiof floor of neural’ canal) -2=- Ge oes ee ee ee 27.5 Lengthrof roof, ofmenralicanalss-a- sees oo ee en ee 21.0 Transverse diameter of neural canal, at anterior opening, .--. .---------- - 9.0 Vertical diameter of neural canal, at anterior opening,------. ------ ------ 5.5 Transverse diameter of neural canal, at posterior opening, ----------.---.-- 10.0 Total lenpth of: pleurapophysis,.--=-<-=<2 == 5 22" a= denen 30.0 Length of free portion of pleurapophysis,-.-----------.----------------- 24.0 Diameter of ;pleurapophysis, at base,----. =~ ee 7.2 Wransyerse diameter/oflateral toramen,<—-e— = Sees ae s 7.0 Werlical diameter of lateral foramens= 2s se ee ee 8.0 Tue TwetrrH VreRTEBRA. (Plate IV, figure 3.) The twelfth vertebra is shorter than the eleventh, and the zyga- pophyses are more approximated. The post-zygapophyses are tubercular above, and the neural spine is rudimentary. The diapophysis is more projecting, but a little less elongated, than in the eleventh vertebra. The lateral foramen for the vertebral artery is larger, and the descending processes stronger. The pleurapophyses are somewhat shorter than in the preceding vertebra. The dimensions of a nearly perfect twelfth vertebra of Hesperornis regalis are given in the table which follows : Measurements of Twelfth Vertebra, (No. 1207.) Length of :centrom,= 222.0. s2-55-6 seo <5 5 ah ee ee Oe ee 24.0"™ Least diameter of centrum) at base, 22-2 se-. ew en eee eee 7.0 Transverse diameters of anterior articulation of centrum,.--- .....-....---- 12.0-14.2 Vertical diameters of anterior articulation of centrum, ---- -- aR ee ae a= eo Transverse diameter of vertebra, across pre-zygapophyses, ..-- .--.-----.--- 46.5 CERVICAL VERTEBRA OF HESPERORNIS. 35 Transverse diameter of vertebra, across post-zygapophyses, ---- -------.--- 32.6" Greatest diameter of pre-zygapophyses,..---- ---- --------------------- 13.0 Least diameter of pre-zygapophyses, ..-.-- .--- -.-----.2--. se---02------ 10.2 Greatest diameter of post-zygapophyses, ----.--..----------------------- 12.2 Least diameter of post-zygapophiyses,......--.----.-----.---<----------- 10.0 Menminon HOON OL neural OAS 2. <<. os - occ ene oee eens aa wpe eee ene 22.0 Pecounioieroor Ormeuraloanal.<21 o8s.2 252 Gens oon een ae oa eae aeoee oe 16.0 Transverse diameter of neural canal, at anterior opening, .--.--. -.-------- 9.0 Vertical diameter of neural canal, at anterior opening, ---- ---- .--- -------- 6.6 Transverse diameter of neural canal, at posterior opening,-- .-------------- 9.0 Vertical diameter of neural canal, at posterior opening,.. .-------. -.----- 6.0 Rotalengtb ofspleurapophvels.<- 96 coe on nob noe coon nan soe 25.0 Leneth of free portion of pleuarapophysis,; . -.-...--.-.-------.-~.-. ----- 15.0 rameter Or micnrapoplnyeis: a6 WAG, 2—- =~. a= — oem oan eee eee see 7.0 rareverseqiameter on IAteral 1OLamMen..-.— =... ces ---5 225-52 cese 52-5 10.0 Mertrcaldrameterof lateral foramen,-<<- 225-2 -=* s-==52-.5.5-----=-- 7.5 Tue THrrTeenTH VertTEBRA. (Plate IV, figure 4.) The thirteenth vertebra in many respects resembles the twelfth, and isiof about the same length. It is rather more excavated at the sides, and has apparently a somewhat larger lateral foramen. The neural spine is but slightly indicated, and in this and the next vertebra reaches its mini- mum of development. The anterior and posterior zygapophyses approach still nearer to each other, to aid the flexure of the neck, which is greatest at this point. The descending processes are short and stout, and the pleurapophyses much as in the twelfth vertebra. The figures given below represent the anterior and posterior aspects of the vertebra here described. Figure 8. — Thirteenth vertebra of Hesperornis regalis, Marsh; (No. 1207), front view; natural size. Figure 9.— The same vertebra; posterior view. d. diapophysis; p. parapophysis; /f. lateral foramen ; ne, neural canal; s. neural spine; 2. pre-zygapophysis; 2’. post-zygapophysis. 36 ODONTORNITHES. The dimensions of the thirteenth cervical vertebra in one series of Hesperornis regalis ave given below: Measurements of Thirteenth Vertebra. (No. 1207.) Length of centrum,.--.-..----------------------- ----=- ---~-----=------ 25.0™™ Least diameter of centrum, at base, .--..------. --------.-------------- 6.0 Transverse diameters of anterior articulation of centrum, -----.---.--------- 14.0-15.5 Vertical diameters of anterior articulation of centrum, - --- -------- -------- 7.0-11.0 Transverse diameters of posterior articulation of centrum, ---. ------------- 10.0-14.0 Vertical diameters of posterior articulation of centrum, ---.--------------- 9.0-13.0 Transverse diameter of vertebra, across pre-zygapophyses, - - -- -------- ---- 49.0 Transverse diameter of vertebra, across post-zygapophyses, -----.------- -- 31.5 Greatest diameter of pre-zygapophyses,---- -----~----------------------- 14.0 Least diameter of pre-zygapophyses, - ------- ---------------------- ane o 9.0 Greatest diameter of post-zygapophyses,-..--- --.---------------------- 12.0 Least diameter of post-zygapophyses, -- .------------------------------- 10.0 Length of floor of neural canal,..-.--.- .--- ------------ ---------------- 25.0 Length of roof of neural canal, .....-.-----.----------- -----=----.----- 17.0 Transverse diameter of neural canal, at anterior opening, -- .-------------- 8.8 Vertical diameter of neural canal, at anterior opening,.--. ---------------- 7.0 Transverse diameter of neural canal, at posterior opening, ---- ------------ 8.5 Vertical diameter of neural canal, at posterior opening, ---- .-.. ----------- 9.0 Length of hypapophysis below centrum, ---------------.---------------- 9.0 Tue FourreentH Vertesra. (Plate IV, figure 5.) The fourteenth vertebra measures a little less than the thirteenth along the centrum. The zygapophyses are approximated, and the neural spine is rudimentary, as in that vertebra. The lateral foramen appears to have been large, and enclosed by bone, which is in part broken away in the specimen represented in Plate IV. The most conspicuous feature of this vertebra is the great development of the inferior descending processes, or ““catapophyses,” as they are sometimes called. These are much flattened, with nearly parallel sides, and rounded ends. They arise near together on the antero-lateral lower part of the centrum, and descend obliquely for- ward, and a little outward. They are somewhat strengthened along their inner faces at the base by a low, broad, rounded ridge, which at its base forms, with its fellow of the opposite side, a pulley-shaped surface. They do not appear to have been quite symmetrical in their distal portions on the opposite sides. CERVICAL VERTEBRA! OF HESPERORNIS. 37 The tubercle below the pre-zygapophysis begins in this vertebra to assume the form of a true diapophysis, or upper transverse process, but presents no articular surface for a rib. The centrum is somewhat excavated at the sides, and the excavation is bounded posteriorly by a rounded ridge. The latter begins near the posterior end of the inferior surface, and runs forward, upward and outward to the base of the diapophysis, on the lower part of which it is continued. Above this ridge, and behind the dia- pophysis, there is a shallow fossa. This vertebra would be regarded by some anatomists as the last true cervical. Its dimensions taken from three well preserved specimens of Hesperornis regalis are given below: Measurements of the Fourteenth Vertebra, (No. 1207.) Ten yuhe GIF Ga IE SS SE ee 23.0™™ Meas iametenon Centrum. At; DAS. 5602-2 aca ome eee e-em on 7.0 Transverse diameters of anterior articulation of centrum,.--.. = -.--- ----- 12.5-18.0 Vertical diameters of anterior articulation of centrum, ----.-.--.---------- 7.0-11.5 Transverse diameters of posterior articulation of centrum, ---------------.- 11.0-15.0 Vertical diameters of posterior articulation of centrum,-_--. ------------- 8.0-11.0 Transverse diameter of vertebra, across pre-zygapophyses, - - ---- ---- ------ 58.0 Transverse diameter of vertebra, across post-zygapophyses, - - ---- Se 30.0 Diameters) of pre-zygapophyses, .-....------=. ~---~----=--.-- =----=- 12.0-11.0 Diameters of post-zygapophyses, --.---------- SRE ee en meee 0 alt ene thion.taor of neural: canal 222-0 5- 4 2-222. secse eso - ee = ---- 20.5 Menmunvor OOM Ol Neural canal 2222 222 ~ sono ee en 16.0 Transverse diameter of neural canal, at anterior opening, ------- ----------- 9.5 Vertical diameter of neural canal, at anterior opening,-- .----------------- 7.0 Transverse diameter of neural canal, at posterior opening, - -----------.---- 8.0 Vertical diameter of neural canal, at posterior opening, -- ------------.---- 7.0 Length of inferior descending processes,..--.-------------. ------------- 17.0 Measurements. (No. 1476.) Dee a i aa sae es Ae eee eee 24,0™™ oan eiameter Gm Ceninairase! 222 soos as oe Sooo 5 sci = Se sees 8.0 Vertical diameter of anterior articulation of centrum, --------.----------- 8.8 Transverse diameters of posterior articulation of centrum, . --- ---.--------- 10.5-15.0 Vertical diameters of posterior articulation of centrum, -.------ ---- ------- 9.5-13.0 fae aon voor Ob Mental canal == soe 2 so in aaa ams ae ~ 55 ~--- 21.0 PenrenGSEeroOe Cl NeUtal CANS anno cece aaa wm Sem Sanwa eee ~s wae 16.0 Transverse diameter of neural canal, at anterior opening, .---------------- 9.0 Vertical diameter of neural canal, at anterior opening, .--------------- ---- 8.0 Transverse diameter of neural canal, at posterior opening, ---- ------------ 9.0 Vertical diameter of neural canal, at posterior opening, --.--.-.---.------ 8.0 38 ODONTORNITHES. Measurements of Fourteenth Vertebra. (No. 1477.) Lengthof-centrum,..c. 202-23 23. dae se ee ee eee = ae 24,.07™ Transverse diameters of anterior articulation of centrum,.--.-----.-------- 18.0-19.0 Vertical diameters of anterior articulation of centrum,----.--------------- 7.0-11.5 Transverse diameter of posterior articulation of centrum, -----.--.---.-- ---- 11.0 Vertical diameter of posterior articulation of centrum, ----~-----------. --- 9.0 Transverse diameter of vertebra, across pre-zygapophyses, .--.------------ 39.0 Transverse diameter of vertebra, across post-zygapophyses, .--- ------------ 33.0 Diameters of pre-zygapophyses, === -22- 22 == see eee eee eee) 00-1420 Diameters of post-zygapophyses, ---- .------- - Ee eee eee meee. US, Length.of floor of meurall'canal) = 22 == see a 22.0 Length of roof-of neuralicanal?:22222) seq eee eee eee 15.0 Transverse diameter of neural canal, at anterior opening, .- .------- -------- 9.0 Vertical diameter of neural canal, at anterior opening, ---.--------------.-- 8.0 Transverse diameter of neural canal, at posterior opening, ---------------- 8.5 Vertical diameter of neural canal, at posterior opening, ------------------- 9.0 Distance between extremities of anterior and posterior zygapophyses, .----- 31.0 Tue Firreenta Vertesra. (Plate IV, figure 6.) The fifteenth vertebra is shorter than the fourteenth, and of peculiar shape. The centrum is deeply excavated at the sides, and its articular faces are transversely elongated. Below, it sends downward and forward a strong prismatic, or elongate pyramidal, hypapophysis, as long as the centrum. The shaft of this process is quadrate in horizontal section, with the anterior face excavated, and bearing a prominent pointed tubercle on its right margin, near the end. This peculiar hypapophysis is bifurcated at the extremity, and the prongs are divergent. They are somewhat unsymmetrical, the left one being slightly longer than the right, and directed a little more outward. There is no lateral foramen, but a distinct subtriangular facet for the head of the first rib is borne on a short parapophysis, projecting to the anterior margin of the articular face of the centrum, and somewhat below it. The small rib supported by this articular face is represented in Plate VIII, figure 1. The diapophyses are broken off, just below the articular faces for the tubercle of the rib. The zygapophyses are rather less approximate than in the preceding vertebra, and the neural spine, though still rudimentary, is larger. CERVICAL VERTEBRA OF HESPERORNIS. 39 In addition to the figures of this vertebra given in Plate IV, the accompanying cut shows the posterior aspect of the same specimen. Ficure 10.—Fifteenth vertebra of Hesperornis regalis, Marsh; (No. 1207), posterior view; natural size. d. diapophysis; p. parapophysis; %. hypapophysis; nc. neural canal; s. neural spine; 2. pre-zygapophysis ; 2’. post-zygapophysis. The measurements given below are from three vertebre, in different individuals of Hesperornis regalis. Measurements of Fifteenth Vertebra, (No. 1207.) Length of centrum,......-. -.---..-- --- -------- =------- 2+ -------- 27,05" Transverse diameters of anterior articulation of centrum,-.--. .------------ 14.0-22.0 Vertical diameters of anterior articulation of centrum, -- ---------- ---- ---- 6.5- 7.5 Transverse diameters of posterior articulation of centrum, ---------------- 14,0-19.0 Vertical diameter of posterior articulation of centrum, ---~.- .------------- 13.0 Transverse diameter of vertebra, across pre-zygapophyses, .- -- ------------ 36.0 Transverse diameter of vertebra, across post-zygapophyses, - - -- ------------ 33.0 Diameters of pre-zygapophyses, - - --------------------- ----------------- 12.5-12.0 Diameters of post-zygapophyses, .... .----------------------- ------- --- 18.0-110 Length of floor of neural canal, .--.------------------------------------ 22.0 Length of roof of neural canal, --.-.--------------- -------------------- 19.0 Transverse diameter of neural canal, at anterior opening, - -----.---------- 9.0 Vertical diameter of neural canal, at anterior opening, ---- -------- -------- 7.5 Transverse diameter of neural canal, at posterior opening, ---------------- 8.0 Vertical diameter of neural canal, at posterior opening, ------------------- 9.0 Length of hypapophysis,-....-.-.------- .--- ------- ------------------- 20.0 Length of bifid portion of hypapophysis,.--.--- ..---.-----------+------ 5.0 Diameters of articulation for head of rib,..------ .--..------------------ 5.2— 3.5 40 ODONTORNITHES. Measurements of Fifteenth Vertebra. (No. 1476.) Mengrth Of Gem trun, «=== 5 = fam a mp a Vk Vaca Transverse diameter of anterior articulation of centrum,.. .--- ------------ 22.0 Vertical diameters of anterior articulation of centrum,_---.--.-----.------ 6.0- 9.0 Transverse diameters of posterior articulation of centrum, --------------- 14.0-19.5 Vertical diameters of posterior articulation of centrum, .---....---------- 8.0-14.0 Transverse diameter of vertebra, across pre-zygapophyses, -- .. ------------- 36.0 Transverse diameter of vertebra, across parapophyses, ---- ---------------- 30.5 Diameters of pre-zygapophy Ses, soe oa ee LO.0 LO Length of floor of neuralicgnal) === eae ee 23.0 Length’ of ‘roof of menralcanaly= 9) =e 2 a ee ee 20.5 Transverse diameter of neural canal, at anterior opening, oo epee aA 9.0 Vertical diameter of neural canal, at anterior opening, ---------- ---------- 7.5 Transverse diameter of neural asa at posterior opening, - eg oR e eA a= 8.0 Vertical diameter of neural canal, at posterior opening, ------------------- 8.0 Diameters of articulation\for headof rib) 22-222 -- - sees e eee 6.0— 5.0 Measurements. (No. 1477.) ihength\oficentrum, .-) 2255 =) eee so ee 22a Transy erse diameters of anterior articulation of (centrom,. eee see 14,5-22.0 Vertical diameters of anterior articulation of centrum,--------.---------- 6.0-— 7.5 Transverse diameters of posterior articulation of centrum, ---- ------------ 15.0-22.5 Vertical diameters of posterior articulation of centrum,--.---.-----.------ 8.0-14.5 Transverse diameter of vertebra, across pre-zygapophyses,---- ----------- 34.0 Transverse diameter of vertebra, across post-zygapophyses, -- .. --~-------- 32.5 Diameters of pre-zygapophyses,..--..-.... ..-.. -.---.--=---------=— = 90-1320 Diameters of post-zyeapophyskes, 5222-5 oe ee = ee 10.5-14.0 Tiength of floor of neural jcanal;.-.---. £-\2 355-22. 485 oe 21.0 Tieng thom roof ofumermalltc smalls ee ae 20.0 Transverse diameter of neural canal, at anterior opening, ---- ---- ---- am 10.0 Vertical diameter of neural canal, at anterior opening, - -- - .-------------- 8.0 Length of hypapophysis;; 22 25: = -— 28 ae ee re 15.0 Diameters of articulation for head of nb, Ae a a PEN RE 7.0- 7.5 Distance between extremities of anterior and posterior zygapophyses, - . . - -- 37.5 Tue SrxteentH VertTesra. (Plate [V, figure 7.) The sixteenth vertebra differs considerably from the fifteenth, and begins to assume the distinctive features of the dorsal vertebra. It is shorter than the fifteenth, and is the shortest vertebra between the atlas and the sacrum. The centrum is excavated at the sides, and its articular faces are laterally expanded, but of little vertical extent. A strong hypa- pophysis descends from the inferior surface, rather more vertically than in the preceding vertebra. ‘The articular face for the head of the second rib CERVICAL VERTEBRA OF HESPERORNIS. 41 is sub-oval in outline, witn the long axis nearly vertical. The para- pophysis supporting it is less prominent than in the fifteenth vertebra, and does not extend in front of, nor below, the anterior articulation of the centrum. The diapophyses are well developed, and somewhat ascending (Plate IV, figure 7, b). They are flattened above, and strengthened by a rounded ridge below. This ridge is united with the posterior portion of the diapophysis, but anteriorly they are separated by a rounded groove, which is shallow distally, but continues proximally into a deep excavation under the anterior zygapophysis. The neural spine is well developed, and tubercular at the top, as in the succeeding vertebra, but of less antero-posterior extent. It is irregularly excavated at the base posteriorly by a fossa for the attachment of a liga- ment. The post-zygapophyses are less projecting, and somewhat nearer together than in the preceding vertebra. The posterior aspect of the sixteenth vertebra is well represented in the following figure : Ficure 11.—Sixteenth vertebra of Hesperornis regalis, Marsh; (No. 1477), posterior view; natural size. d. diapophysis; p. parapophysis; hk. hypapophysis; nc. neural canal; s. neural spine; 2. pre-zygapophysis ; 2’. post-zygapophysis. Full measurements of three specimens of the sixteenth vertebra of Hesperornis regalis will be found in the following tables: 6 42 ODONTORNITHES. Measurements of Sixteenth Vertebra. (No. 1207.) Length of centrum,.-.-------------------- --------------- ------------ 18.07™ Transverse diameters of anterior articulation of centrum, _-----.----------- 17.0-24.0 Vertical diameters of antericr articulation of centrum,---.-.----..-------- 5.0-10.5 Transverse diameters of posterior articulation of centrum,.----.-----.----- 13.0-19.0 Vertical diameters of posterior articulation of centrum, ------.~------------ 7.5-14.0 Transverse diameter of vertebra, across pre-zygapophyses,------------ --- 30.0 Transverse diimeter of vertebra, across diapophyses,.-..--------- --.---- 61.0 Transverse diameter of vertebra, across post-zygapophyses, - --- --------- -- 27.0 Diameters of pre-zygapophyses, --.- -------.-----------. ----------- ---- 18.8-11.0 Diameters of post-zygapophyses, .--- -------- --------------------------- 18.0-10.0 Length of floor of neural canal,..-----..---..-------------- ---- ------= 20.0 Length of roof of neurall canal, -__- -_------ --=- == == = 20.5 Transverse diameter of neural canal, at anterior opening, ---. ---. --------- 10.0 Vertical diameter of neural canal, at anterior opening, ----- ---- ----------- 8.0 Transverse diameter of neural canal, at posterior opening, . --------.------- 8.0 Vertical diameter of neural canal, at posterior opening,-. --- -------------- 8.0 Diameters on articulation ton bead of Toss = a= ses. ae nee ee 8.8- 6.0 Diameters of articulation for tubercle of rib,-----------------.----------- 7.0- 4.0 Height of neural spine above floor of neural canal, .---------------------- 29.0 Antero-posterior diameter of neural spine, ---- .--.------------- -------- 12.0 Measurements. (No. 1476.) Meng tof icon trans sae ate a 18.0™™ Transverse diameter of anterior articulation of centrum, --------------.---- 24.0 Vertical diameters of anterior articulation of centrum,..------- --------- 5.0-10.0 Transverse diameters of posterior articulation of centrum,- ---------------- 16.0-20.5 Transverse diameter of vertebra, across pre-zygapophyses, - -----.---------- 35.0 Transverse diameter of vertebra, across parapophyses,----------- -------- 33.5 Transverse diameter of vertebra, across diapophyses, -- -- ---- ---- --------- 65.0 Transverse diameter of vertebra, across post-zygapophyses, - - ------------- 30.0 Diameters of pre-zygapophbyses,----------------------------- --------- 12.5-12.5 Diameters of post-zygapophyses,--.. ----------------.--- = === 10.0-11.5 Dengthiof floorof nenrallicanaly = =- == == eee eee 23.0 Length of roof of neural canal, ---------- ee on Se ee ee oe een eeee 21.0 Diameters of articulation for head of rib,.-----.----.--.-------.---.---- 5.8- 8.6 Measurements. (No. 1477.) Tength, of centrum, 222-2. hoe an oe a ee eee 19,0™™ Transverse diameters of anterior articulation of centrum,.. -~--------.----- 13.0-22.5 Vertical diameters of anterior articulation of centrum, ------------.---.-- 6.0-10.5 Transverse diameters of posterior articulation of centrum,. --------------- 15.0-23.0 Vertical diameters of posterior articulation of centrum,-----.--.--------- 9.0-13.5 Transverse diameter of vertebra, across diapophyses,--------- -. -------- 46.0 Transverse diameter of vertebra, across parapophyses,------.------------ 30.0 Diameters of pre-zygapophyses, . -...--. ---- .--- ---- «+--+ e202 ----+------- 10,5-14.0 CERVICAL VERTEBRA OF HESPERORNIS. 43 Diameters of post-zygapophyses,.... .....--- --.- ----<<------ ------=+2-- 9.0—11.0%™ Length of floor of neural canal,...-..-.------.----.-------- ------------ 22.0 Length On Inypanopn ysis, <5 5 = sao - Soar lemeie tn teen pee ls 18.0 Diameters of articulation for head of TDs. o2u:s sec aks Se eee 027 9.0 Tue SEVENTEENTH VeRTEBRA. (Plate V, figure 1.) The seventeenth vertebra, the last true cervical, is slightly longer than the sixteenth, and has assumed the more general characters of the dorsal vertebre. The excavations at the sides of the centrum are rather less pronounced than in the preceding vertebra. The hypapophysis is strong, directed vertically downward, and is expanded at its distal end, as shown in the figure below. Fic. 12. Fiaurr 12.—Seventeenth vertebra of Zesperornis regalis, Marsh; (No, 1477), posterior view ; natural size. The posterior articulation of the centrum has a greater vertical extent than in the sixteenth vertebra. The articular surface for the head of the rib is larger, and a little more elevated, but not more prominent than in that vertebra. The diapophyses are directed somewhat upward, and supported beneath by a strong rounded ridge, above which as in the preceding vertebra, is a groove, ending in a deep pit below the pre-zyga- pophysis. The neural spine has a somewhat greater antero-posterior extent, and is higher than in the last vertebra. It is thickened and truncated at the top, as in the succeeding vertebre. The posterior zygapophyses are less prominent, and nearer together, than in the sixteenth vertebra. 44 ODONTORNITHES. Full measurements of the seventeenth vertebra in Hesperornis regalis will be found below: Measurements of Seventeenth Vertebra, ’ Dength of centrum, <== = ee ee Transverse diameters of anterior articulation of centrum, - - (No. 1207.) Vertical diameters of anterior articulation of centrum, ------------------- Transverse diameters of posterior articulation of centrum, - Vertical diameters of posterior articulation of centrum, --_- Transverse diameter of vertebra, across pre-zygapophyses, - Transverse diameter of vertebra, across diapophyses, - - -- - - Transverse diameter of vertebra, across post-zygapophyses, Diameters of pre-zygapophyses, - --- ---- ----------------- Diameters of post-zygapophyses, - ----------------------- Length of floor of neural canal, ------------------------- Length of roof of neural canal, --------.-_-_--------__-- Transverse diameter of neural canal, at anterior opening, - - Vertical diameter of neural canal, at anterior opening, - --- - Transverse diameter of neural canal, at posterior opening, - Vertical diameter of neural canal, at posterior opening, - --- Diameters of articulation for head of rib, ---------------- Diameter of articulation for tubercle of rib,-------------- Height of neural spine above floor of neural canal, _---_--- Antero-posterior diameters of neural spine, - --- ----------- Measurements. (No. 1476.) Temprib. ok sera trary 2 ee Least diameter of centrum, _-..+..---2--2 252222552 2--5=- Transverse diameters of anterior articulation of centrum, - - Vertical diameters of anterior articulation of centrum, ---- Transverse diameters of posterior articulation of centrum, - Vertical diameters of posterior articulation of centrum, - - - - Transverse diameter of vertebra, across pre-zygapophyses, - Transverse diameter of vertebra, across parapophyses, - - - - - Transverse diameter of vertebra, across post-zygapophyses, - Diameter of pre-zygapophysis, - - ------------------------ Diameters of post-zygapophyses, - - - - - ------------------- Dength’ of floor of neural’ canal,---=-==2-------- Length of roof of neural canal, .--.-.-------------------- Diameter of articulation for head of rib,-------.--------- Measurements. (No. 1477. Wiser ta OF OIG a a ee Transverse diameters of anterior articulation of centrum, - - Vertical diameters of anterior articulation of centrum, - ~~ - Transverse diameters of posterior articulation of centrum, - - - 90.07 16.0-27.0 6.0—10.0 13.5-19.0 9.0-13.0 8.5— 6.0 7.0 29.0 13.5-18.5 22.0"™ 7.5 15.0-21.0 5.0-10.5 13.0-19.0 10.0—14.0 25.0 20.0™™ 13.5-25.0 5.0-10.0 15,0-21.0 DORSAL VERTEBRA OF HESPERORNIS. 45 Vertical diameters of posterior articulation of centrum, - ------------------ 9.0-14.5™™ Transverse diameter of vertebra, across pre-zygapophyses, - - -- ------------ 29.0 Transverse diameter of vertebra, across diapophyses, - - -- ----------------- * 63.0 Transverse diameter of vertebra, across parapophyses, - - - -- --------------- 32.0 Transverse diameter of vertebra, across post-zygapophyses, - -------------- 27.0 Diameters of pre-zygapophyses, --.------------------------------------- 9.5-11.0 Diameters of post-zygapophyses, --------------------------------------- 7.5— 9.0 Length of floor of neural canal, ---.-.--------------------------------- 23.0 Length of roof of neural canal, -----.-.--------------------------------- 21.0 Transverse diameter of neural canal, at posterior opening, - ---- ------------ 8.0 Vertical diameter of neural canal, at posterior opening, ---- --------------- lity AL: Length of hypapophynis)--2.22-- -- =... -2-- <2 = see 52-2 -- 2-25-52 ---2-5 19.0 Diameters of articulation for head of rib,-.---.------------------------- 7.5- 9.0 Tue ErcuTeentu, or First Dorsat, Vertesra. (Plate V, figure 2.) The eighteenth vertebra is a little longer than the seventeenth, and its centrum is stouter, and less excavated at the sides. The hypapophysis descends vertically, is longer than the centrum, and bears an oblique un- symmetrical expansion at the end. The articular face for the head of the rib is a little more elevated than in the preceding vertebra, but no larger. The rib supported by this vertebra is represented in Plate VIII, figure 4, and the adjoining sternal rib, in figure 16. The diapophyses are much as in that vertebra, and are preceded by a deep pit, extending beneath the posterior part of the pre-zygapophysis. A much smaller excavation is situated just behind the base of the diapophysis. The neural spine was somewhat larger than in the preceding vertebra, but is broken off in the specimen figured. (Plate V, figures 2, 2¢.) The post-zygapophyses are a little more approximate than in the seventeenth vertebra. The size of the present vertebra and of its various parts in Hesperornis regalis is given in detail below: Measurements of Eighteenth Vertebra. No. 1207.) TLS QUT ee i a I es ees 22.0™™ Transverse diameters of anterior articulation of centrum, ----------------- 17.0-26.0 Vertical diameters of anterior articulation of centrum, ------------------- 7.5-11.0 Transverse diameter of posterior articulation of centrum, ----------------- 13.0 Vertical diameter of posterior articulation of centrum, -------------------- 10.0 Transverse diameter of vertebra, across pre-zygapophyses, - - -- ------------ 25.0 Transverse diameter of vertebra, across diapophyses, - --- ----------------- 61.5 46 ODONTORNITHES. Transverse diameter of vertebra, across post-zygapophyses, - - ------------- 20.0™™ Diameters of pre-zygapophyses, - - -- -----------------------------_------ 9.5— 7.0 Diameters of post-zygapophyses, ------------------------ == =n 9.5— 8.0 Length of floor of neural canal,__---------------------- === 23.0 Length of roof of neural canal,_-------_____ =~ = 23.0 Transverse diameter of neural canal, at anterior opening, ---- ------------- 10.0 Vertical diameter of neural canal, at anterior opening, ------------------- 7.0 Transverse diameter of neural canal, at posterior opening, - --- ------------ 7.0 Vertical diameter of neural canal, at posterior opening, ---- -------------- 7.0 Diameters of articulation forshead of rib,.-=----------+-=-=----=-=-=---- 8.2— 6.5 Diameter of articulation for tubercle of rib,_--------------------------- 6.5 Measurements of Eighteenth Vertebra. (No. 1476.) Ire ye Oem 8 Ra ee a pete Soe nee SsesSsensase 25.0™™ ‘Least (diameter of centrum . -25 2.5 eee ee = eee ee 23.0 Transverse diameter of neural canal, at anterior opening, ----------------- 8.0 Vertical diameter of neural canal, at anterior opening, -------------------- 5.5 Transverse diameter of neural canal, at posterior opening, --------.------- 8.0 Vertical diameter of neural canal, at posterior opening, ------------------- 5.5 Diameters of articulation for head/of ribsssese see eee ee 7.0— 8.0 Antero-posterior diameter of neural spine, at base, ------------------------ 19.0 Tue Twenty-First VerTEBRA. (Plate V, figure 5.) The twenty-first vertebra is slightly longer than the twentieth, and the centrum is stouter, and less excavated at the sides. The hypapophysis is short and weak. The parapophyses are very short, so that the articula- tion for the head of the rib is scarcely raised above the general surface of the vertebra. It is situated nearly on a level with the neural canal, but DORSAL VERTEBR2 OF HESPERORNIS. 51 extends somewhat below. The diapophyses in this vertebra begin to shorten, though but slightly, and are directed horizontally outward, and somewhat anteriorly. Neither in front of, nor behind their base, is there any proper pit in the vertebra. The neural spine is somewhat higher, but of less antero-posterior extent, than in the preceding vertebra, and the surfaces before and behind near the base are rather more roughened for the attachment of ligaments. To indicate the range of variation in the vertebrae of Hesperornis regalis, measurements of the twenty-first vertebra in five individuals are given in the tables below. Number 1200 is the type of the species. Measurements of Twenty-first Vertebra. (No. 1200.) PRN FeR EO Le GER ENUM oe ee ee aes Se ee a once 24,0™@ meat transverse diameter of cetitramy._.--------2--.225----------=4-=~ 9.0 Transverse diameters of anterior articulation of centrum, ------------------ 16.5—23.0 Vertical diameters of anterior articulation of centrum, -------------------- 9.0-13.0 Transverse diameters of posterior articulation of centrum, ---------------- 14.8-22.0 Vertical diameters of posterior articulation of centrum, ------------------- 10.5-16.5 Transverse diameter of vertebra, across pre-zygapophyses, ---------------- 20.0 Transverse diameter of vertebra, across parapophyses, . --- ---------------- 27.0 EIMeLErs/ OM Nre-7iy PAPOPNVRGS, 2 - = 92a on = on eg news == n= 9.5-8.5 Manpulmetsioorotmeural: Ganal) <9 2265 2 8 52 ooo enw qe nn ew ae - = saan 25.5 Manminrofroors OseMGUral Galal tsi. se 8s 2 gen an one ge aa sae o =n 23.0 Transverse diameter of neural canal, at anterior opening, ----.------------ 7.0 Vertical diameter of neural canal, at anterior opening, -------------------- 5.0 Transverse diameter of neural canal, at posterior opening, - --- --.--------- 8.0 Vertical diameter of neural canal, at posterior opening, -----------.------- 6.0 Diameters of articulation for head: of ribj..-2-...-.---.--....-=-..-<-.-- 7.0- 8.0 Measurements. (No. 1206.) Transverse diameters of anterior articulation of centrum, ----------------- 16.0-24.0™™ Vertical diameters of anterior articulation of centrum, -------------------- 12.0-13.0 Transverse diameter of vertebra, across pre-zygapophyses, - - -- ---- -------- 20.0 Transverse diameter of vertebra, across parapophyses, - ------------------- 25.0 Miameters Of anterior zygapophyxes;. --+ _=-- ---- ----2..-----.-.------- 9.5-11.0 Transverse diameter of neural canal, at anterior opening, ----------------- 7.5 Vertical diameter of neural canal, at anterior opening, -------------------- 6.0 Diameters of articulation for head of rib, -.-------------------------:---- 6.0- 9.0 Vos SOG digger yc) a ee Oe ee ee 14.0 Height of neural spine above floor of neural canal, --------.--------------- 33.0 Antero-posterior diameters of neural spine, - --- -------------------------- 17.0-21.0 52 ODONTORNITHES. Measurements of Twenty-first Vertebra. (No. 1207.) Menethrof centrum, - =. - .--= +3. 22522 sos ee 25.0™™ Transverse diameters of anterior articulation of centrum, -_--------------- 17.0-27.0 Vertical diameters of anterior articulation of centrum, -_------------------ 11.0-14.0 Transverse diameters of posterior articulation of centrum, ---- ------------ 15.0-23.0 Vertical diameters of posterior articulation of centrum, ------------------- 12.0-17.0 Transverse diameter of vertebra, across pre-zygapophyses, - --- ------------ 20.5 Transverse diameter of vertebra, across diapophyses, --------------------- 5.7 Transverse diameter of vertebra, across Beste G@apophyses,a-— ee ee ee 19.0 Diameters ‘of :pre-zy gapophyses; =~ 2. "2 268 - Ba Se ee nee ee 8.2— 9.8 Diameters.) 52 ce 42 ee oli koe eve ese lecl 49.0 iivainsyverse ciameter of stermal énd,<* 3. .* <6. esnc os Swen k coun seus 2.5 Measurements. (No. 1207.) Length, from scapular to sternal articulation, ..........------------------ 55.0™™ Greatest diameter of scapular and humeral articulation, ----.--------------- 12.0 Least diameter of scapular and humeral articulation, ---.---..------------- 3.4 MeminverencnineLer of slernal end, --8--.s-2--c-2ccl eee ese foc eel 3.0 60 ODONTORNITHES. Tue Srernum. (Plates VI-VIII.) The sternum in Hesperornis somewhat resembles in general form the corresponding bone in the genus Uria, but in other respects is more like that in the Ratite. It is thin and weak, and entirely without a keel. It is expanded in front, especially between the costal processes, and has two deep grooves for the reception of the coracoids. These grooves are placed obliquely, converging anteriorly, and are widely separated from each other. The sternum has a rounded mesial projection in front, which is somewhat thickened, but there is no true manubrium. A comparison of the keelless sternum of Hesperornis, represented in Plate VIII, figure 6, with that of one of the Ratite, for example, the Emeu, as shown below, will make clear the resemblance in this part of the structure between these two types. Fig. 15. Figure 15.—Sternum of Emeu (Dromeus Nove Hollandie, Latham); front view; one-half natural size. a, costal process; b. surface for articulation of sternal ribs; c. groove for left coracoid. The sides of the sternum in Hesperornis are concave in outline, and in Hesperornis regalis, there are four articular projections on each side for the attachment of sternal ribs. These processes are all on the anterior half of the sternum. Behind these, the lateral margins are nearly parallel. The posterior end of the sternum is quite thin, and had two shallow emarginations. In Hesperornis crassipes (Plate VII), the sternum repre- sented is nearly perfect, and in this species there are five articular faces on each side, for the sternal ribs. The posterior margin in the same species is less excavated than in Hesperornis regalis. SHOULDER GIRDLE OF HESPERORNIS. 61 The measurements which follow are taken from the sternum of Hesperornis regalis (number 1206), and that of Hesperornis crassipes (number 1474). Measurements of Sternum (LHesperornis regalis. No. 1206.) sirius mlenmibu(anproximaea)s—420 62. S262! 2= oe nee cane ene 200.0"™ enon anu CH MtiOns fOr TDS ao —2 a 66 ODONTORNITHES. The first sternal rib of Hesperornis regalis, uniting above by the inter- vention of the fourth vertebral rib with the eighteenth, or first true dorsal vertebra, is short and nearly straight (Plate IX, figure 16). It is obliquely truncated in front, and has an articular face for union with the sternum. The posterior end is expanded, somewhat rounded, and has on its upper posterior angle an articular facet for the vertebral rib which joins it. The second sternal rib (Plate IX, figure 17) is longer and stouter than the first, and somewhat more curved downward. Its articular faces are similar, but the upper, or posterior, half of the shaft is much more compressed. The third sternal rib is longer than the second, and somewhat less curved (Plate IX, figure 18). Its distal end is more slender, and the articular face oblique. The proximal, or posterior, portion is much flat- tened transversely, and the lower margin of this end extends beyond the articulation. The fourth sternal rib, which united with the seventh vertebral rib, is long and slender, and considerably curved downward (Plate IX, figure 19). Its anterior articular face is oblique, and the adjoining portion of the shaft slender. The posterior extremity is expanded, and projects downward, and backward, below the articular face. The next sternal rib, the fifth, did not unite directly with the sternum in Hesperornis regalis, but with the preceding rib near its base, and hence its anterior extremity has no true articular face. It is long and slender, curved downward throughout its middle portion, and its posterior end is thin and expanded (Plate IX, figure 20). This rib united with the eighth vertebral rib, and, in Hesperornis crassipes, it united also independently with the sternum, by a well developed articular tubercle. The sixth sternal rib, in Jesperornis regalis, articulated with the ninth vertebral rib, and, in front, was attached lightly to the sternal rib last deseribed. STERNAL RIBS OF HESPERORNIS. 67 In the small number of sternal ribs articulating with the sternum, Hesperornis resembles the Struthious birds, rather than the Natatores. The persistent articulation of the uncinate processes with the vertebral ribs is usual in birds incapable of flight, and has been observed in a few others. The following measurements show the principal dimensions of the vertebral and sternal ribs, and uncinate processes in Hesperornis regalis. The length of the vertebral ribs is measured in a straight line, joining their extremities. Measurements of Vertebral Ribs. (No. 1206.) Tsk 2d. 3d. 4th. 5th. 6th. ith. 8th. 9th. Length, in a straight line,_--------- Somberg en l4tbe- oe) 2, 190; 209, 22 a Distance from center of head to center of tubercle,..........-- 23. -19= 19:— 20... <- 20; Oi. Si. 23: Elevation of tubercle,_______--__-- 105, 10222 8; fue 4, te es Distance from tubercle to end of rib, 32. 83. -. 141. -. 159. 177. 194. —- Distance of uncinate process from : loweriendofeTmbse. 25.2 = . 22. ae Sis (8b25 402) “G2 em ecz5s Measurements of Uncinate Processes. (No. 1206.) Ist. 2d. 3d. 4th. 5th. 6th. Wanthn aaa ane Sto eet 25. 54. 52. 55. 50. 30.™™ rameter of articular end,..........---.-.. ie 12. 10. ahi E ih 8. LSE us 2G Ue ee 5. 1. 9. 8. 6. 5. irontesc, preadthss =5.5-2- 2-2 soc 2 5-2 ees 12. 15. 14, 15. 13. 8. Measurements of Sternal Ribs, (No. 12086.) 1st. 2d. 3d. 4th. 5th. WMenGihese = a2 see ae sa Gas 2c oe = Lea we 110. a ae a Diameter of proximal articulation, -------------- 9 1: 8 4, Least antero-posterior diameter of shaft, ---- ---- 38. 5. 4, 3. 3. Diameter of distal articulation, --...------------ 6. 8. 8. 8. vf PREMPLer OF GIRTON. cone apn oot soca accs sats 12. <4 15. 18. ie r (ive “inah ce hs acer abt se “tia , ap ; “T 5 aa). . ey NS 1..48 “<4 »* aay (ave ae ay . as a ia', teeia ap ae woe terete ® - ene aT ala) kJ tite fagn: as ‘4 ae 7 et ' nite ior wee Pie - ' — ~ é > ' ‘ r. = 7% ’ ‘ - Me ‘ +¢e ~~ % * ‘ ‘ — - » °% CTA PTE V.. THE PELVIC ARCH OF HESPERORNIS. (Plates X—XI and XX.) Tue pelvic arch of Hesperornis exhibits many features of interest, and characters more distinctly reptilian than that of any recent bird. In its general form, the pelvis of Hesperornis regalis resembles that of Podiceps. It is very long and narrow, as in that genus, and in other diving birds. The acetabulum differs from that in all known birds, in being closed internally by bone, except a foramen that perforates the inner wall, as in the Crocodiles. The ilium, ischium, and pubis, moreover, have their posterior extremities free and distinct. This reptilian character is seen, likewise, in the Emeu, as well as in Tinamus, which in other respects also shows aflinities with the Ostriches. Tue Inium. (Plates X and XI.) The ilium is extremely long and narrow. Its superior outline is gently arcuate, and this border joins the dorsal margin of the other ilium on the median line, thus forming a roof-shaped covering over nearly all the sacrum. The pre-acetabular portion of the ilium is only about one- fourth of its entire extent. From the acetabulum forward, the ilia diverge, showing at first, between their upper margins, the neural spines of the vertebree below, and then, in front, the zygapophyses of the first of the sacral series. The anterior extremity of the ilium is thin, and rounded in outline, the ventral margin extending nearly to the bottom of the first sacral vertebra. 69 70 ODONTORNITHES. The post-acetabular part of the ilium is very elongate, and its lower border somewhat curved downward. The free extremity, behind the last codssified vertebra, is thin, and turned slightly upward, and outward. On the lower portion of the outer surface, there is a strong ridge, which arises behind the upper part of the anti-trochanter, and sweeps gently down- ward until it reaches the lower margin of the ilium; whence it rises gradually, and continues backward to nearly opposite the last sacral vertebra. This ridge is shown in Plate X, figure 1, and its prominence in this specimen (number 1206), may have been somewhat increased by pressure, before removal from the matrix. The acetabulum has a well-defmed border, which is nearly circular. Above this, there is a large anti-trochanter, or articular surface to which was applied the neck of the femur. The internal face of the acetabulum was not closed merely by fibrous tissue, as in modern birds, but by bone, which is penetrated by a foramen of moderate size, (Plate XI, figure 1, af). The acetabular region of the ilium is strongly coéssified with the sacrum, while nearer its extremities the union is less perfect. Tue Iscurum. (Plates X and XI.) The ischium forms part of the acetabulum, and anti-trochanter, and then, contracting rapidly, is continued backward as a long slender bone, which is entirely free at its distal end. Viewed from the side, the upper margin curves downward, nearly parallel with the convex ventral margin of the ilium, while the lower border is very gently arcuate, with the convexity below. The distal extremity is obtusely pointed, and is somewhat turned upward and outward, terminating nearly opposite the posterior end of the ilium. The anterior half of the free portion is rod-like, and the distal half expands gradually, nearly to the extremity. The outer convex surface of the rod-like portion is continued over the distal half as an obtuse rounded ridge, which extends to the extremity. There are no processes on this bone extending upward toward the ilium, or downward to the pubis. PELVIC ARCIL OF HESPERORNIS. 71 The following figures represent the pelvic bones of two recent birds, the Emeu and Grebe, which, in this part of the skeleton, also, repeat some of the characters seen in Hesperornis. Figure 16.—Pelvis of Emeu (Dromeus Nove Hol-andie, Latham); side view; one-fifth natural size. Figure 17.—Pelvis of Grebe (Podiceps occidentalis, Lawrence); side view; natural size. Fia. 18. Figure 18.—The same pelvis (Podiceps occidentalis, Lawrence); scen from above. The sign‘fication of the letters is the same in all the above figures, viz: a. acetabulum; / ilio-sciatic foramen; i. ilium; és. ischium; p. pubis; p’. post-pubis; s. sacrum. 72 ODONTORNITHES. Tue Posts. (Plates X and XI.) The bone usually called “pubis” in modern birds, which the writer has shown to be probably the post-pubis, forms, in Hesperornis, the lower posterior portion of the acetabulum, and then extends backward as a long, slender, rod-like bone, nearly parallel with the ischium. It is some- what longer than the ischium, and its distal extremity is truncated at right angles with the shaft. This bone is of nearly equal width throughout its entire length, and sends off no processes or projections toward the ischium, or from its own lower margin. Viewed from the inner side, the bone above described is seen to pass beneath the acetabular foramen in a strong ridge, and terminate in front of the acetabulum, in an obtusely rounded tuberosity. This process may be seen in some recent birds, more particularly in those that especially use their posterior limbs. It has been called the ilio-pectineal process of the ilium, and is considered an outgrowth of that bone. The writer has shown, however, by a comparison of the pelvic elements in Dinosaurian reptiles and in birds, that this prominence should be regarded rather as probably representing the pubie bone of reptiles. The relation of the principal pelvic elements to each other in the pelvis of recent birds is illustrated in the figures given on page 71. In the pelves represented on page 73, the remnant of the reptilian pubis is still plainly to be seen, especially in Geococcyx. It is not improbable that the retention of this process may be due in part to the habits of certain species, as it seems to be best developed in running birds, and those that especially use the posterior limbs. A similar process is seen in some mammals, where it may serve a like purpose. The outer surface of the pelvis in Hesperornis, especially of the iliun, is marked by delicate vascular impressions, but the bones themselves are : ] dense, and nearly solid. ‘American Journal of Science and Arts. Vol. xvi, p. 415, Novy., 1878; and vol. xvii, p. 92, Jan., 1879. PELVIC ARCH OF HESPERORNIS. 73 The three pelvic bones of Hesperornis are firmly codssified, as in existing birds, but the ilium is ankylosed to the sacrum in the acetabular region alone. The true pubic element, or “ ilio-pectineal” process, is not larger than in many recent birds, and is much inferior in size to the corresponding protuberance in the pelvis of Geococcyx, and of Tinamus, shown below. In these two genera, moreover, another point of resemblance to Hesperornis appears in the acetabulum, the inner margin of which is materially narrowed by ossification, although a large perforation remains. In Hesperornis, the acetabular foramen is much smaller than in any known bird, and during life was doubtless closed by the round ligament. This reduction of the acetabular opening by ossification strengthens the pelvis at this point, and thus is of service where powerful action of the posterior limbs is required. Figure 20.—Pelvis of Zinamus robustus, Sclater and Salvin; seen from the left; natural size. a. acetabulum; é ilium; ts. ischium; p. pubis; p’. postpubis. 10 74 ODONTORNITHES. Tue Sacrum. (Plates X and XI.) The sacrum in Hesperornis regalis is very long and narrow, and the vertebree which compose it are usually well codssified. The number of vertebre in the true sacrum cannot be accurately determined, but in the ankylosed series, illustrated in plates X and XI, there are fourteen. This series may be conveniently described as the sacrum, although some of the anterior vertebre are evidently sacro-dorsals, and some of the posterior should be regarded as urosacrals or caudals. The first vertebra of the series is the twenty-fourth in number, count- ing from the skull. It is the largest of the sacral series, and is entirely enclosed between the ilia. Its transverse processes bear no articulated rib, but are expanded at the distal end, and abut directly against the ilium. The line of union of this vertebra with the succeeding one is strongly marked, and in one specimen of this species (number 1477), the vertebra is free, and hence it is represented in the figures below, and its measure- ments are given separately. The junction between the twenty-fifth and twenty-sixth vertebre can be traced on the lower surface, but behind this the lines of union are obliterated, except two or three near the posterior end of the series. The figures below represent the twenty-fourth vertebra taken from the series in which it is distinct. Figure 21.—Twenty-fourth vertebra of Hesperornis regalis, Marsh; (No. 1477) front view; natural size. Fiaure 22.—The same vertebra; posterior view. The neural spine in this specimen is not preserved. d. diapophysis; ne. neural canal; 2. pre-zygapophysis. PELVIC ARCH OF ITESPERORNIS. 75 The neural spines of the sacral vertebra, and of the others cobssified with them, are compressed transversely. With the exception of those of the two last, or urosacral, vertebrae, these spines do not extend above the upper margins of the ilia. These vertebrae are the longest of the series, surpassing in extent even the anterior ones. Some of the principal dimensions of the pelvic arch of Hesperornis regalis are given in the measurements which follow : Measurements of Pelvic Arch, (No. 1206.) Length of ilium,_.....---.-----------------------------------------"-- 380.0" Depth of pelvis, across acetabulum, -- -- --------------------------------- 74.0 Greatest depth of ilium, behind acetabulum, - --- ------------------------- 52.0 Distance from anterior end of ilium to center of acetabulum, - ------------- 104.0 Transverse diameter of acetabulum, ------------------------------------ 24.0 Total extent of codssified sacral vertebra, - ------------------------------ 320.0 Length of the first sacral, or twenty-fourth, vertebra, -- ------------------- 21.0 Transverse diameter of anterior articulation, ---------------------------- 25.0 Vertical diameter of anterior articulation, __.-.-------------------------- 14.0 Length of ischium from posterior border of acetabulum, ---- -~------------ 260.0 Length of pubis and post-pubis, - -- - -- -- -------------------------------- 330.0 Extent of pubis in front of anterior border of acetabulum, - --- ------------ 15.0 Extent of post-pubis behind posterior border of acetabulum, - --- ----------- 291.0 Vertical diameter of post-pubis, at base, -------------------------------- 12.0 Greatest diameter of foramen in acetabulum, - --- ------------------------ 12.0 Least diameter of foramen in acetabulum, - ---- -------------------------- 7.0 Height of anti-trochanter above acetabulum, ---------------------------- 25.0 Antero-posterior extent of anti-trochanter, ------------------------------ 22.0 Greatest distance between upper margins of anti-trochanters, -- ------------ ~ 54.0 Measurements of Twenty-fourth Vertebra. (No. 1477.) Length of centrum, .--.----------------------------------------------- 22.0™" Least transverse diameter of centrum, --------------------------------- 12.0 Transverse diameter of anterior articulation of centrum,_-- -------------- 31.0 Vertical diameters of anterior articulation of centrum, ---- ---------------- 11.0-17.0 Transverse diameters of posterior articulation of centrum, ---------------- 20.5-24.0 Vertical diameter of posterior articulation of centrum, -----.----~---------- 12.0 Transverse diameter of vertebra, across pre-zygapophyses, - - - - ------------ 20.0 Transverse diameter of vertebra, across diapophyses, - - - - - - -- - ------------ 52.0 Diameters of pre-zygapophyses, - - -- ---- -------------------------------- 7.5-10.5 Length of floor of neural canal, - --------------------------------------- 20.5 . - x . > 1} - : — * a ‘ a” ea bi 4 ray y ee wits J.) A. , - : - ’ = wy _ all yd = ~y’ Zz cn od yun hoe ee s _ ¢ : aes 7 will y ad = Le 36 . A - 7 4 ? * o . eo ; - 7 . CHAPTRE VI. THE CAUDAL VERTEBRA OF HESPERORNIS. (Plates XII and XX.) Tue tail of Hesperornis regalis presents some peculiarities of structure not before seen in birds. It was composed, apparently, of twelve vertebr, and all of these are preserved, with the exception of that portion of the last which formed the extreme end of thé tail. The number of caudal vertebre in Hesperornis regalis exceeds those in any known recent bird, with the possible exception of the Great Auk (Alca impennis, Linn.), now nearly or quite extinct. The number falls far short, however, of that in the tail of Archeopteryx, which also differs essentially in its general structure from the caudal extremity of Hesperornis. The anterior free caudals of Hesperornis are short, with high neural spines, and moderate transverse processes. They are opisthoccelian, as in the corresponding vertebrae of Pavo and Geococcyx. The middle and posterior caudals have very long, and horizontally expanded, transverse processes, which restricted lateral motion, and clearly indicate that the tail was mainly moved vertically, evidently as an aid in diving. The last three or four caudal vertebra are firmly codssified, forming a flat, horizontal, terminal mass, analogous to, but quite unlike, the plough-share bone of modern birds. The caudal vertebree of Hesperornis are composed of compact osseous tissue, without pneumatic openings. None of them possess zygapophyses, 77 78 ODONTORNITHES. THe First CaupaL VERTEBRA. (Plate XII, figure 1.) The first free caudal vertebra, which is the thirty-eighth in number, counting from the skull, is much shorter than the preceding one in the codssified sacral series, but considerably longer than those which follow. In its natural position, it was entirely included between the posterior extremities of the ilia. The anterior articular face of the centrum is vertically elliptical in outline, and somewhat convex. Immediately behind this, the centrum is constricted, and then expands regularly to the posterior articular face, which is much larger than the one in front, and moderately concave. It is sub-circular in outline, but its transverse diameter exceeds its vertical, and the upper and lower margins are slightly notched on the median line. The transverse processes extend downward, and slightly outward, and their external surfaces, which are nearly in the same plane as the sides of the neural arch, were attached to the divergent iliac bones by cartilage. The neural spine was elevated, and much expanded in a fore and aft direction. The neural canal, at its anterior orifice, is much com- pressed transversely; its posterior outlet is triangular, and very large, indicating a marked expansion of the spinal cord in this region of the vertebral column. This feature was doubtless directly connected with the powerful movements of the tail. The principal dimensions of the first free caudal, or thirty-eighth vertebra, in Hesperornis regalis, are given below: Measurements of First Caudal Vertebra. (No. 1206.) ength: of Gentrim,< -s2< sack = oe ae ere ee ee es 13:05 ‘Least transverse: diameter or centrum... sees eae ee ee ee 6.0 Transverse diameter of anterior articulation, ......-.----.--------------- 6.5 Vertical diameter of anterior articulation, _-_-- --.-----------.--.-.---+--- 8.0 Transverse diameter of posterior articulation, ---...---------------------- 11.0 Vertical diameter of posterior articulation, -----.------------------------- 8.0 Diameter of vertebra, across transverse processes, - ----------------------- 20.0 Extent of transverse processes below centrum, - --- - ---------------------- 7.0 17.0 Length of floor of neural canal... 92.2. 22. 2 ee ee eee eee Vertical diameter of anterior opening of neural canal, -------------------- Vertical diameter of posterior opening of neural canal, - ------------------ 7.0 CAUDAL VERTEBRA OF HESPERORNIS. 79 Tue Seconp Caupat Vertesra. (Plate XII, figure 2.) The second free caudal vertebra is much shorter than the preceding, and its centrum is much more massive. The anterior articular face is somewhat convex, and has a depressed sub-circular outline, with the lower half of the margin excavated. The posterior articular face is nearly flat, but somewhat concave near the centre. The transverse processes are directed outward and slightly downward, terminating a little below the ventral surface of the centrum. They are longer than those of the preceding vertebra, but have less antero-posterior extent. The neural spine is stout and elevated, and has its superior extremity expanded into a round knob. The neural canal is large, and vertically oval in outline, and its lower margin cuts into the upper border of the anterior articular face. The posterior margin of the neural arch, when seen from the side, is parallel with that of the posterior articular face of the centrum, and slightly in advance of it. Measurements of Second Caudal Vertebra, (No. 1206.) Penijein Gf Ganerutns=--— 2-2 - 252 222222<+---2----555<2----------------- 15:02" Transverse diameter of anterior articulation, ---.--..---------------------- 14.0 Vertical diameter of anterior articulation, ------------------------------- 10.0 Transverse diameter of posterior articulation, -- --.----------------------- 13.0 Vertical diameter of posterior articulation, ------------------------------ 11.0 Diameter of vertebra, across transverse processes, - - -- - --- ---------------- 29.0 Length of transverse processes from centrum, --------------------------- 8.0 Length of floor of neural canal, - -------------------------------------- 12.0 Vertical diameter of anterior opening of neural canal, - --- ---------------- 5.0 Vertical diameter of posterior opening of neural canal, -------------------- 7.0 Height of neural spine above centrum, ---------------------------------- 21.0 Tue Turrp Caupat Vertesra. (Plate XII, figure 3.) The third caudal vertebra of Hesperornis regalis, or the fortieth in the column, is the shortest of the series, with the exception of those at the end of the tail. It is also the first caudal which is entirely behind the ends of the ilia, and its shortness is probably due to the fact that here the flexure of the tail essentially began. 80 ODONTORNITHES. The anterior articular face of the centrum is conyex, broadly ovate in outline, and slightly notched above, below the neural canal. The posterior articular face is similar in form, and slightly concave. The transverse processes are stouter than in the last vertebra, and their distal ends are thickened, and abruptly truncated. The axes of these processes, if continued to the axis of the neural spine, would meet in the neural canal, and form equal angles with it and with each other. The neural spine is elevated and massive, and its posterior margin is nearly vertical. Its apex is bent forward and expanded, and is cordiform in superior outline. The neural arch covers the whole of the centrum, and the canal is large, and vertically ovate in outline. Measurements of Third Caudal Vertebra. (No. 1200.) Menpthrotncentrum) 2: .< 525. 2eas) eck 8 o_o ee ee 120R Least diameter, below transverse processes, ------------------------------ 9.0 Transverse diameter of anterior articulation, ..-....-...-.--------------- 13.0 Vertical diameters of anterior articulation, --------------_----2_---_--=-- 10.0-11.0 Transverse diameter of posterior articulation, -_....---.------------------ 12.5 Vertical diameters of posterior articulation, .----.-.-----_.--_--_---__-_- 10.0-11.0 Diameter of vertebra, across transverse processes, - - ---------------------- 32.0 Length of floorcf neuralicanalt= <2) Svar 2 ee 10.0 engthof roofofmenralicanal 2a. — 2 e= eee oe oe ee 8.5 Vertical diameter of anterior opening of neural canal, -------------------- 5.0 Vertical diameter of posterior opening of neural canal, ------------------- 5.0 Transverse diameter of posterior opening of neural canal, ----------------- 4.0 Height of neural spine, above floor of neural canal,_-------.-------------- 22.0 Transverse diameter of shaft of neural spine, .--------------------------=- 3.0 Transverse diameter of head of neural spine, ---------------------------- 6.0 Antero-posterior diameter of diapophysis, at base,-..-.....-------------- 6.5 Tue Fourtn Caupau Verresra. (Plate XII, figure 4.) The fourth caudal vertebra is similar in many respects to the one last described, but the centrum is materially longer. The anterior articular face is considerably smaller than in the third vertebra. It is somewhat concave, and so deeply notched above for the neural canal as to be sub- cordate in outline. The posterior face is similar to that of the last vertebra, but the transverse processes are stouter, and extend more directly backward. The neural arch, when seen from the front, is very similar to that in the preceding vertebra, but its spine is shorter, and its cordiform apex is broader, and more turned forward. CAUDAL VERTEBRA OF HESPERORNIS. 81 Measurements of Fourth Caudal Vertebra. (No. 1200.) PIKeRtEsh GORLSM Nee vase OS Ao ls, busts Gee wane sae eee ates 12.0™™ Least diameter, below transverse processes, ------------------------------ 7.5 Transverse diameter of anterior articulation,.-.........-..-------------- 11.5 Vertical diameters of anterior articulation, .......----.------------------ 9.2-11.0 Transverse diameter of posterior articulation, --.-.-...--.---------------- 10.2 Vertical diameters of posterior articulation, .........-------------------- 9.5-10.0 Diameter of vertebra, across transverse processes, - - ---------------------- 32.0 HSAUIrEL OX UO0T) Off UBUTA CRNA a- 2 Sosa sens on on Sook ee em ae one 10.5 banpiby of root OF Neural cana) 2-52-3822 aoe oh re eat oe 8.2 Transverse diameter of anterior opening of neural canal, - ----------------- 4.0 Vertical diameter of anterior opening of neural canal, -------------------- 5.0 Transverse diameter of posterior opening of neural canal, __--..-.---------- 4.0 Vertical diameter of posterior opening of neural canal, ------------------- 4.5 Height of neural spine above floor of neural canal, ----------------------- 19.0 Least antero-posterior diameter of transverse processes, - ------------------ 6.5 Transverse diameter of shaft of neural spine, ---------------------------- 3.0 Transverse diameter of head of neural spine, ---------------------------- 7.0 Tue Fiera Caupat Vertesra. (Plate XII, figure 5.) The fifth caudal vertebra of Hesperornis regalis is somewhat longer than the fourth, but its centrum is narrower, and its articular faces perceptibly smaller. The general form of the anterior articulation is sub-quadrate in outline, with the corners rounded. It is notched above for the neural canal, and has its lower border also indented on the median line. Both articular facets are somewhat concave. The transverse processes are much longer than in the last vertebra, and their distal ends are expanded horizontally. The axes of the descend- ing portion of these processes bear the same relation to the axis of the neural spine, as in the third vertebra, the three meeting in the neural canal, and including equal angles with each other. The lower surface of the transverse process extends somewhat below the ventral face of the centrum, and then turns abruptly upward and outward, nearly to the extremity, which is slightly deflected. Seen from above, the ends of these transverse processes appear rounded, and their posterior margins project behind the articulation of the centrum. The ventral surface of the body of this vertebra is concave in a fore and aft direction, and convex transversely. j1 82 ODONTORNITHES. A small free hypapophysis was articulated with the posterior lower margin of this vertebra, and also with the adjoining margin of the succeed- ing centrum. This bone resembles somewhat the corresponding part in the Loon, and is clearly homologous with the chevron bones in reptiles. The neural arch of the fifth caudal extends over nearly the whole centrum. The posterior margin of the spine curves backward above the neural canal, and the summit of the spine is directed forward. The neural canal is materially smaller than in the fourth vertebra, but is similar in outline. Measurements of Fifth Caudal Vertebra. (No. 1200.) Ihengthyof-contram:2o-=- 6 sen == een eee ee ee ee ie 13.0" Transverse diameter of anterior articulation, -—.------------=--=-==-==----= 11.0 Vertical diameters of anterior articulation, ----....------.--=--------=--- 9.0-10.2 Transverse diameter of posterior articulation, ---------------------------- 11.0 Vertical diameters of posterior articulation, ---...--.-------------------- 9.0— 9.5 Sreneth:of floor of meural canaljo22=-2 525-2 = sas eee a= ae eee 12.0 rene thio£ roof ofmenral cm all yee eee ee eee 9.5 Transverse diameter of neural canal, at anterior opening, ------------------ 3.2 Vertical diameter of neural canal, at anterior opening, -------------------- 4.0 Transverse diameter of neural canal, at posterior opening, - ---------------- 4.0 Vertical diameter of neural canal, at posterior opening, ------------------- 4.0 Diameter of vertebra, across transverse processes, --.--------------------- 39.0 Antero-posterior diameter of transverse process, at base, ------------------ 7.0 Tue SrxtH CaupaL Vertesra. (Plate XII, figure 6.) This vertebra differs considerably from the fifth. Its centrum is longer, and much more slender. The transverse processes are much longer, and so expanded at their distal ends as to greatly restrict lateral motion; which was still further prevented by the same means in the succeeding vertebree. The under surfaces of the transverse processes have a marked prominence at the point where the outward bend occurs. These prominences extend below the ventral surface of the centrum. ... cok cone co eee 13.0 Length of ridge between tibia and fibula,....................--.----___- 25.0 Measurements, (No. 1207.) Left. Right. OSE ao) ESS eee Oe ee ee 98.5™™ 101.0™™ Antero-posterior diameter of head of femur,..-._........--._--.__-__- 19.0 18.5 Transverse diameter of proximal end,............--.-.--.---_-...--- 52.0 52.0 Least transverse diameter, near middle,_....___.___.._--____.....__.- 22.0 22.0 Transverse diameter of distal end,_....._.._............_.......___. 52.0 53.0 Antero-posterior diameter, near middle,_-.._............-------_____- 18.0 18.0 Greatest diameter of distal articulation,...._.......____...._....____- 53.0 53.0 Greatest diameter of tibial articulation,...._._...__............_____- 40.0 40.0 Transverse diameter of fibular articulation,..-__.-.....__..__..._____- 15.0 16.0 Greatest diameter of inner condyle,................----------------- 21.0 21.0 Least diameter of inner condyle,..................--.-.------------- 13.0 13.0 12 90 ODONTORNITHES. Measurements of Femur. (No. 1476.) Leneth of right, femur, -...--..+ =) ae eee ee 105.0™™ Diameter of head of femur: = 22: 22222 ee ee ne 19.2 ‘Transverse diameter of.proximal)end) = 22-2 ee ee ee 54.0 Transverse diameter. of shaft neanmiddle- ==) = 2-2 ee ee ee 21.0 ‘Transverse diameter of distal end>s2= "== se ee ee 53.0 Vertical: diameter, neanthe middle.2 3e = ee 19.0 Greatest diameter of distal iarticnlation=-5- > oe ee eee 53.0 Greatest: diameter of: tibial articulation,» 9 ee ee eee 38.0 Transverse| diameter.offibular articulation, 22) -- 222 22- e sone eee ee eee 15.5 Greatest: diameter of ;nnen condyle;=>-2 = a ee ee 20.0 Least:diameter of inner condyle; 2.2. . -5322-=- > 22225 sse ee a eee ee 13.0 Length of ridge between tibia and fibula,_-.......---------------------- 25.0 Measurements. (No. 1477.) Length of xight femur)... == 4 - Soo Son eee eee 97.0™= Antero-posterior diameter of head of femur, _-------_-------------------- 19.0 Greatest proximal diameter of shaft,......--..--..-----------u2.-.-- ees 53.0 Greatest’ distal:diameter ‘of shaft, $os = = 5 ee eee ee eee 51.0 beast transverse diameter/of Shalt, .—-- eee = ae ee ee 23.0 Antero-posterior diameter of shaft, near middle, ---__-.-.---------------- 20.0 Greatest diameter of distal artienlation.=—-— sss] ee 52.0 Greatest diameterof tibial: articulation. === == =n ee 38.0 Transverse diameter of fibular ‘articulation; 92-22 -2-) =e se en 15.0 Greatest. diameter of imnercondyle;=sses5-5 342-2225 see aa eee 20.0 Least diameter ‘ofnner condyle; = 2-22222 2-223 oe nn oe ee a 12.5 Measurements. (Hesperornis crassipes, No. 1474.) Maximum diameter of distal end of right femur, _-.--......-------------- 51.0™™ ‘Transverse diameter of shatt,-.-- 24- soe eee ee 21.0 Vertical diameter of ‘shaftso=: 25 oe ee a ee ee ee 21.0 Gréatest diameten) of shait,vatmiddle,- 2-222" 2-2) 55 hae een 25.2 \ Tue Trp1a. (Plates XIV and XX.) The tibia of Hesperornis regalis is very long and powerful, and much the largest bone in the skeleton. In its general features, it most resembles the tibia in the genus Podiceps. It is somewhat expanded at its proximal end, where it articulates with the femur, and the articular faces are placed obliquely, and inclined outwards. The inner articular face is much the larger, and is slightly concave. The outer face is somewhat convex, and is only partly separated from the other by a constriction of the articular surface. The outer articulation projects considerably beyond the external TIBIA OF HESPERORNIS. of margin of the shaft, and is supported below by a strong rounded ridge, which, lower down, passes into the true fibular ridge. There is a strong epicnemial crest on the anterior surface of the proximal end, which, above the articulation, rises into a powerful tuberosity, or enemial process. To the outer posterior surface of this spine, the patella was attached (Plate XX). The shaft of the tibia is nearly straight, with its transverse, slightly greater than its fore and aft, diameter. Seen from in front, the outer margin appears slightly concave, and the inner margin convex. The fibular ridge is strongly developed, and can be traced along the proximal three-fourths of the bone. Just below the middle, it disappears for a short distance, being cut off by an oblique groove for the artery, which here passed between the tibia and fibula. In the type specimen, number 1200, the surface of the shaft is marked by delicate vascular impressions, which are imperfectly indicated in Plate XIV, figure 1. In the specimen itself, these impressions are as perfect as in life, and well illustrate the marvelous degree of preservation characteristic of the vertebrate fossils from this geological horizon. The surface of the tibia shows no pneumatic foramina, but there is a rather large medullary cavity in the shaft. The size of this cavity is shown in the accompanying cuts, which represent transverse sections of the tibia near the middle, and near the distal end. Fig. 26. Figure 26.—Transverse section through middle part of tibia of Hesperornis regalis; (No. 1207), top view; natural size. Figure 27.—Transverse section through same bone; (No. 1491), near distal end; same view; natural size. m. medullary cavity ; /. fibular ridge. The distal end of the tibia resembles that of the same bone in Podiceps, but the inner condyle is here much the larger, and the two are only moder- ately expanded. The inner condyle extends somewhat below the outer one, but its fore and aft diameter is much less. The transverse diameter of 92 ODONTORNITHES. the two condyles, when seen from the front, is sub-equal, and the groove between them does not correspond with the axis of the bone, but is some- what inclined inward. There is no osseous bridge in front, over the canal through which passes the tendon of the tibialis anticus muscle. The absence of this osseous bridge is noteworthy also in Podiceps, as it is apparently present in all other known aquatic birds. The groove between the posterior projections of the two condyles is broad, and flat, but narrows rapidly above. In the tables below will be found detailed measurements of the tibia of Hesperornis regalis, in the type, and in one other specimen: Measurements of Tibia. (No. 1200.) Left. Right. Ihenp thor tibins 52 sateen eee ena eel 320.0™™ 320.0™™ Mlevation oflcnemial' process;= 25s ) Ae ee OF a ee (4) FEET OF HESPERORNIS. 103 Measurements. (THesperornis gracilis, No. 1478.) SOAR IRODMIER GC WATS. a ao. foe cas oa Spo nk sn ane eee 41,0™™ Vertical diameter of proximal articulation, --_--......-«...---------.-------- 17.0 Transverse diameter of proximal articulation, ........-.----=----------------- 15.0 ACE CCR GIES CCNA 8, i) -———— ae aer e ere Ae e e re 20.0 enniibaneeerse wiamotaot: snatt, .- <2. ~ 2a. censeeusndhesnar otue uae pele 12.5 MENnECOnUIGs OTaMGver Or Alatine: 25 <. i ssa Las 5 oak eae heen eee a ee 10.0 ‘ransyverse.diameter of distal’ articulation; =.= -.<..<.-.2.<.-2-<<-.=---- 2252522 14.0 Vertical diameter of distal articulation,.................-------- pio See ee 15.0 Diameter of." per? mi. distal artienlation).-.-.=-2---2.4525----222s5s-2c~eceoen 6.5 Measurements of Second Phalanx. (Hesperornis regalis, No. 1200.) Three GV? ROYCE AOL S ihe 2 OR Se RR a a 39.5™™ Groncenincinmetenn aly proximal 6nd, 3-- = -= 822 Sees. =o 5588 Jb oe ose 16.5 DenAOAMeLan Au? PLOkUN Al ONO ona soon eo Soe ena ke pee een onan nem 13.0 Transverse diameter of proximal articular face, ---..-..----------------.----- 13.0 Vertical diameter of proximal articular face,_..._.......--.---------<-------- 13.0 INGHATSOIAIMCLGly OLENA ert 2 oe ee whe oe eR Bk So DN) 12.0 erreatent OMmInfler an aIntAn@le, fas. 2 scons 22~ as oben esos ees co See eee 15.0 Maxmunn diameter OF distal articular face,..-=..==---2<---s-.-22--ssee cue oe 14.0 Transverse diameter of “peg” in distal articular face, ..-.-....--------------- 4.6 Measurements. (No. 1206.) Neri iG AGCONG) PNAIANK omit oo whe Salon 3 an oa Sateen soon sous ese 40.0™™ Greatest, diameter/of proximal end, --.- 2 2222. os oe. sob. Sonn s- = 225528 16.5 Meanq oanieten Of proximal 6nd... —.\-26-5-. 5. cones seca sec eces dene eee 13.5 Transverse diameter of proximal articular face,.----------------------------- 14.0 Wertical diameter of proximal articular face,..2==.-..---.--5---+--+---s-=2-= 15.0 Hecan mG ram etd ee Sant ers 2p eto et See OE ee ie ee Se ee 8.0 ieanwstOre aud art Ginever OL siakise== 58-2 5 oon eee ease ce 12.0 Greatest diameter of distal end,.......-..--..-....--..<-= BP aay So Sat aie 15.0 Greatest cimmeLer Ob distal articular ta0e,. 2-22-62 2S Ste nase 14.0 Drametar ol. “neo all discal artaoulur fae; - 2 =- 2 - 252-8 os a eee ence 4.5 Measurements of Second Phalanx. (Hesperornis crassipes, No. 1474.) Ue iene ea DLE) oh See ne ke pe ee a fe ee 38.0™™ SeOOBLOAGGNBINGLer Au PIOEIMAl CNG see —— goo en eee ee ane wanes 17.0 LOE ER Gre Eg Nge cn CG DS ee 13.0 Transverse diameter of proximal articular face, ........_.-.------------------ 12.5 Vertical diameter of proximal articular face, -_.......-....-.-.--------------- 14.5 eRe NIG UG OIAIIGLEr Gr SNetG: 2 2 oe) oo ce ee ee een 12.0 ieaneraiaiiered OF ANALG, Ob GIRGAL ONG... 3.2260 no ook ES cc enigwecee nase on 14.0 104 ODONTORNITHES. Measurements of Third Phalanx. (Hesperornis regalis, No. 1200.) Length of third phalanx, .--... == 055. 268 oe ees sae eae 40.0" Greatest diameter ‘of proxinial'end 2-2 see 18.0 Least diameter of proximaliend,.<--_¢- 222222 e= ee eee ee 9.0 Greatest diameter of proximal articular/surface,_-=-=---==-~—- == =e = 15.0 Least ‘diameter of proximal articular surtaces-= === =p eee 8.0 Least fore and aft diameter of shatt, = s9e= sooo. ee 11.0 Greatest diameter of ‘distalend;.- -" 5.2. 22 eee ee 12.5 Least diameter, of ‘distal’end;=_./ 2... ee ee 5.5 Greatest diameter of distabaxticnlarsurtaces92—- =) =e ee eee 10.0 Diameteriof “Spex sini distaltarticulanjsactaces se see 4.8 Measurements. (No. 1206.) Leng th:of sthird phalanx: 2.26 32 se ee ee eee 41.0™™ Greatestidiameter of. proximal end) =2=2 22 "= 52 oo ee ee ee 17.0 Least diametertofsproxum alien dpe ee oe 10.0 Greatest diameter of proximal articular surface, _---.--£-.---+-.---=---------- 15.0 ‘Least diameter of proximal articular surface, == =2- 22222-2225 se see sea 10.0 Least fore.and attidiameterion shart. a. eee a ee 11.0 ‘Least diameter of shaft, 28-222 hee ee eee ee ee 6.0 [east diameter of distal’end) 22255. 222¢ #222862) oe ee ee eee 6.5 Diameter‘of “peg” in distal“articular surface,---=-22- == === eee eee 3.0 Measurements of Fourth Phalanx. (Hesperornis regalis, No. 1200.) Greatest diameter of prommaliend) 22's sere ane ee ee 14.0"™ Least diameter'of proximal jend:.- >= 2252-25225. oe ee a eee 7.5 Greatest diameter ‘of proximal articulation, ---2—--==-2- = 2= sso 2 aoe 12.0 Least diameter of proximal articulation;=—-=——-- o-— == =e gee eee 6.5 THe Tuirp Dieir. (Plates XVIII and XX.) The third digit was very much smaller than the fourth, and was sup- ported by the small middle metatarsal element. It consisted of but four phalanges, and, when at rest, was directed forward and slightly inward. The first phalanx (Plate XVIII, figure 5) is shorter than the first in the outer toe, and much more slender, the whole bone being greatly compressed transversely. The proximal extremity is triangular in outline, and its articular face is divided by a sharp vertical ridge. The shaft is somewhat sigmoid longitudinally, and the articulation of the distal end is divided into two sub-equal portions by a nearly vertical groove. FEET OF HESPERORNIS. 105 This articulation shows an approach to the crescent and peg structure, but in its main features resembles the usual phalangeal joint in birds. The second phalanx in this digit (Plate XVIII, figure 6), is much shorter than the first, somewhat less compressed, but in other respects, very similar. The shaft is nearly straight, and but very slightly excavated on its outer inferior surface. The third phalanx (Plate XVIII, figure 7) is materially longer than the second, and much more slender. It shows the same tendency toward horizontal flattening, seen in the penultimate phalanx of the outer toe, the advantage of which has been explained above. The proximal articulation is subcircular in outline, and moderately concave. The distal extremity is much compressed. The fourth, or terminal, phalanx of this digit has not been recovered, but the articulation which supported it, and the shape of the ultimate phalanx of the second toe, indicate its form and general proportions, which are represented in Plate XX. The principal dimensions of the phalanges in the third digit of Hesperornis regalis and Hesperornis crassipes are as follows: Measurements of First Phalanx of Third Digit. (Hesperornis regalis, No. 1200.) UBEN YC A igiaie GINS etn see LEN re eee CoN cel | Se en py Se a a 41.0™™ Greatest vertical diameters of proximal end,--_----..---------------------- 18.-19.0 Least vertical diameter of proximal end, ..............-....-.--.------------ 14.0 FLOABHVerne GismeLer Gu Proximal;endy.- 28 22 2. Sse oe 5 ke ee kee te 10.0 Greatest vertical diameter of proximal articulation, ----__-..----.-.----------- 14.0 Least vertical diameter of proximal articulation, --_-----......-..------------ 12.0 Transverse diameter of proximal articulation, ..........-.-------------------- 8.5 aun @eEkhione diam Geenorm ANREt 222 22. eee oes 8 ee eee caen 8.0 eG CURVersa ia MeLer OL SNAtio aoc. 5.2 2o<5 oes oes ee ee oe ace 6.2 Greatest diameter of shaft, at distal end,.....=........-.--..---.--....---..- 10.0 Memnuugmeran On quate, a0) GINUAMOUGs <5 —- 222 eagn ee lee on ote 8.0 Greatest vertical diameters of distal articulation,._____.____________-___-- _. 8.-10.0 Least vertical diameter of distal articulation,__.....................--------- 7.5 Transverse vertical diameter of distal articulation, 14 106 ODONTORNITHES. Measurements of First Phalanx of Third Digit. (Hesperornis crassipes, No. 1474.) Tength of first phalanx, ----2-----=- 226 = === =e 39.0™™ Transverse diameter of proximal end, ----_-------------------_--___------... 9.0 Transverse diameter of proximal articulation, -------------------------------- 9.0 Least vertical diameters of distal articulation, ---------------------------- 8.5-10.0 Least ‘vertical: diamleterio£ shat. os = == ere ee 8.0 Teast transverse diameter Of Siait ess ac ae ee 7.0 Greatest diameter of shaft, at distal end,-=2 2222-2 = 2" "222 =e ee ee 10.0 Least diameter of shatt, at distal terse =e se eee ee 6.5 Least vertical diameter of distal articulation, --------------------------------- 7.5 Transverse diameter of distal articulation, --------------------. -------------- 7.5 Measurements of Second Phalanx. (Hesperornis regalis, No. 1200.) Length of second phalanx, - -.----------------------------------==---------- 30.0 Greatest vertical diameter of proximal end, ---- ------------------------------ 14.0 Transverse diameter of proximal end, --------------------------------------- 8.0 Vertical diameter of proximal articulation, ---- ------------------------------ 11.0 Transverse diameter of proximal articulation, ----_--------------------------- 7.0 ‘east tore and aft diameter Of Shabt,2 9) === 2 ae 7.0 Mies tematnim tm Glare tere 0 Sas at tra 4.0 Greatest diameter ofndistal end me = see 8.0 Tieast ameter co fy clus teal en lose ee 6.5 Greatest diameter of distal articulation, ----2--.---.--------------=-- =~ = 7.8 Transverse diameter of distal articulation, ----------------------------------- 6.0 Measurements of Third Phalanx. (Hesperornis regalis, No. 1200.) Greatest diameter of proximal end, ----------------------------------------- 10.0" Least diameter of proximal end, - - -- -------- -------------------------------- Deb Greatest diameter of proximal articulation, ---- ------------------------------ 9.0 Least diameter of proximal articulation, ---- --------------------------------- 5.0 Tur Seconp Dierr. (Plates XIX and XX.) The second digit was much smaller than the third, and composed of three phalanges. It was supported by the short free extremity of the inner element of the tarso-metatarsal bone, and was directed inward and forward. The proximal end of the first phalanx (Plate XIX, figure 1) was raised above the general plane of the foot. Its proximal end is trihedral in outline, and the distal end also, although the shaft is twisted so that the boundary lines of the two are not parallel. The two extremities are FEET OF HESPERORNIS. 107 truncated obliquely, making the outer margin of the bone considerably shorter than the inner. The shaft of the bone is depressed, and the outer and inner margins converge distally. The second phalanx of this digit (Plate XIX, figure 2) is somewhat similar to the first, but shorter, and much more slender. The extremities are less obliquely truncated, and the shaft and distal end are much more flattened. The articular faces of this phalanx, as well as of the others in this digit, admit of considerable rotation, but have the crescent and peg articulation only partially developed. The terminal phalanx of the second digit (Plate XIX, figure 3) is short, much depressed, and pointed at its distal end. Seen from above, it is wedge-shaped in outline, with the proximal articulation at right angles to the axis of the bone. In its natural position, this phalanx had its superior surface turned more or less outward, and during recovery after the back- ward stroke it was turned edgewise, so that this face looked entirely outward. The measurements of the phalanges of the second digit in Hesperornis regalis, and Hesperornis crassipes, are given in the tables below: Measurements of First Phalanx of Second Digit. (Hesperornis regalis, No. 1206.) GroRtentyGiameter Of Proximal CNG, 2 222 <2 5a. o 26S s nS nose sees een Sa awe sae 14.5™™ Warnicall diameter of proximal ends 22 522262. 22 2e 8 oo Soon on cease cece 12.0 Wrangverse diameter-of proximal end,=-.-- 52-22 <2. 2 25 see bao ne nce ee 13.0 Vertical diameter of proximal articulation, -----...---.--.------------------- 11.0 Transverse diameter of proximal articulation, ----.-..------------------------ 11.0 Measurements. (No. 1476.) Aerie Csi Psy PNALANT SE coe renee ees See fe SO SB oe a ae 42,0™™ Greaves GuaMGter OL ' PrOeIMAUGUOs ena. 92s sc 252 oe on oe eee 15.0 MiarcnL CiaMeLer OL PrOxMMal GN a3 ae see oa kee oon oto sack 13.0 Transverse diameter of proximal end, -------- ee ee ort i eee Re 13.8 Vertical diameter of proximal articulation, ----.............-.......---.----- 11.0 Transverse diameter of proximal articulation, --------............-.-.-------- 10.2 Peni waruial Gismetor Of AUBTt,7.~. =e sane eee eee seek else el 5 eke 6.5 Permnnyerra ameter Of ANGtin=s. oo eee ee ee ok 8.5 Greatest vertical diameter of distal end,.-.-._---.__-- Cnt ee Ei eae 9.0 Least vertical diameter of distal end,..........__...___-_- Lh 5 Sct Pay’ RR Ae OED 8.0 Transverse diameter of distal end,\ ~~~ _ ~~. ---- a. ae — “OO 108 ODONTORNITHES. Measurements of First Phalanx of Second Digit. (Hesperornis crassipes, No. 1474.) Length of first phalanx, ------------ 9-928 8 en nn nae 40.57" Greatest diameter of proximaliend) === sense ones ee eee 14.5 Vertical diameter of proximal! ends sees ee ee eee 12.5 Transverse diameter of proximal end) eee eee nee eee 13.0 Vertical diameter of proximal articulation, ------........-.------------------ 12.0 Transverse diameter of proximal articulation, ------....---------------------- 10.0 Greatest.vertical’ diameter-of distal endl S252 >= 2 ee ee eee 8.5 Least vertical. diameter ofdistal’ end) 22222 5- = "ee ee eee ee ee 4.5 Transverse diameter of distalwend,=-22s2-9 oe on ee a eee ees Pere Measurements of Second Phalanx. (Hesperornis regulis, No. 1476.) Tenpthiof Become pba eK ee 41,0™™ Greatest (transverse) diameter of proximal end, ------------------------------ 10.0 SViertical ad rare bern Ote po S00 kL NN ee ee 9.0 Transverse diameter of proximal articulation, -------------------------------- 9.0 Vertical diameter of proximal articulation, ----------------------------------- 7.2 east transverse diameter of shaft, -- 222-252) 2" Sees 6.2 Wiertioal diameter otishalteee =. -s-]2. acer = See ee a ee eee 3.6 Transverse Giameter of distaliend, .--...-- --=- 2-4. 2 seen en aa ee eee ea 2 Vertical ‘diameter ‘of distal) endsses22 2 =e = ee ee ee 3.0 Transverse diameter of distal articulation, _.--..-.__--.._._..__--.-=----=-=<- 6.5 Vertical diametersiof distal articulations =25--.-- = 2oes oon a eee eae 2.5-3.0 Measurements of Third Phalanx. (No. 1476.) Length of third phalanx, -------------------------------------------------- 15.0™™ Transverse diameter of proximal end,-—----=—-2 = oon 7.0 Vertical diameter of proximal end, ----------------_------------------------ 3.3 Tur Firsr Dierr. (Plates XVIII, XIX and XX.) The metatarsal rudiment of the hallux, or first digit (Plate XIX, fizure 4) has already been described. It supported ‘below a first phalanx (Plate XIX, figure 5), much flattened, and of moderate length. ‘The digit was completed by a short, pointed, terminal phalanx (Plate XIX, figure 6), very similar in its general form and proportions to the ungual phalanx of the second digit. The hallux was directed forward and inward, like the first digit of Colymbus, and not backward, as in most modern birds. The diminutive size of this digit in Hesperornis, and the rounded loose joints between the phalanges show that it was of little service in locomotion. FEET OF HESPERORNIS. 109 The following are the principal dimensions of the phalanges of the hallux in Hesperornis regalis : Measurements of First Phalanz of First Digit. (Hesperornis regalis, No. 1476.) Least transverse diameter of first phalanx, .......-.......-------------------- 6.0™™ Nanrnicalr ommeLenegn abatteie: 165 25C0. - csn Cel eccug aces Macs eeccsuee nee emes 3.0 Transverse diameter of distal articulation, -...............---.--------------- 6.0 Vertical diameter of distal articulation,................-.---.--.--..-------- 2.0 Measurements of Second Phalanx, (No. 1476.) Ut ge 2 On ae a ee a ee ame ees eae 14.0" Pun V Clea Olam cares Blo Tita GN yearn es on eee ee wee eee ae 7.0 Werioalrdiamerenoneproxunal end so. S25. 2 oh sono wae as came asso ssse=-se 3.0 Grentente LeanaVerseL OIaINGlOkve ssa ce so oho ee ae ce gee See aea enna ana ae 7.6 a ie aide "A — j= O5 yoy $a _ * + | ‘ ? g x . * z * > ® i dl s Pe e . pat . ? : , ? . * A] CHAPTER VIII. THE RESTORATION OF HESPERORNIS. (Plate XX.) Tuer foregoing description of the individual bones of Hesperornis, together with the full illustrations given in the accompanying plates, will make clear to anatomists all the important points in the osseous structure of this ancient bird, in many respects the most interesting member of the class yet discovered. With the exception of one or two terminal toe-bones, and the extreme point of the tail, every part of the skeleton is preserved in one or more specimens, and this ample material has been used in the restoration represented in Plate XX. Nota few of the bones were nearly in their natural position, when discovered, and the remainder have been assigned to their appropriate places in the skeleton, after much careful comparison of the fossils with the nearest allied living forms. It is therefore confidently believed that the field of conjecture has been reduced to a minimum in the skeleton as restored. The restoration of Hesperornis regalis in Plate XX represents the skeleton one-half natural size, and in a position which the bird doubtless sometimes assumed when on land, although it is probably more erect than was habitual. On the water, the body was of course more nearly horizontal, the neck more bent, and the legs usually much farther behind. 111 112 ODONTORNITHES. In the preceding Plates, I-XIX, the bones of Hesperornis regalis have been represented in natural size, and the magnitude of the whole bird may be judged from the fact that the skeleton, if extended, would measure about six feet (1.8") from the point of the bill to the end of the toes. Hesperornis crassipes was somewhat larger, and Hesperornis gracilis, so far as known, was apparently smaller, and of more delicate proportions. When on Jand in the position represented in Plate XX, Hesperornis regalis would be rather more than three feet in height. Hesperornis was a typical aquatic bird, and in habit was doubtless very similar to the Loon, although, flight being impossible, its life was probably passed entirely upon the water, except when visiting the shore for the purpose of breeding. The nearest land at that time was the sue- cession of low islands which marked the position of the present Rocky Mountains. In the-shallow tropical sea, extending from this land five hundred miles or more to the eastward, and to unknown limits north and south, there was the greatest abundance and variety of fishes, and these doubtless constituted the main food of the present species. Hesperornis, as we have seen, was an admirable diver, while the long neck with its capabilities of rapid flexure, and the long slender jaws armed with sharp recurved teeth formed together a perfect instrument for the capture and retention of the most agile fish. As the lower jaws were united in front only by cartilage, as in Serpents, and had on each side a joint which admitted of some motion, the power of swallowing was doubtless equal to almost any emergency. Having thus shown what the skeleton of Hesperornis is, and what its mode of life must have been, it remains to consider the more important question of how the peculiar combination of general and specialized characters manifested in its structure originated. The two most striking features of Hesperornis are the teeth, and the limbs, and an inquiry in regard to them first suggests itself. The teeth of Hesperornis may be regarded as a character inherited from a reptilian ancestry. Their strong resemblance to the teeth of RESTORATION OF HESPERORNIS, 113 reptiles, in form, structure, and succession, is evidence of this, and their method of implantation in a common alveolar groove (Holcodont), con- forms strictly to what we have in one well known group of reptiles, exem- plified by Ichthyosaurus. This method of insertion in the jaw is a primitive dental character, quite different from what we should naturally expect as an accompaniment of the modern style of vertebra, and is a much lower grade than the implantation of the teeth in distinct sockets (Thecodont), a feature characteristic, as we shall see, of another group of Odontornithes, of which Ichthyornis is the type. These teeth indicate unmistakably that Hesperornis was carnivorous in habit, and doubtless was descended from a long line of rapacious ancestors. In considering the limbs of Hesperornis, two explanations of their peculiar modifications naturally suggest themselves. The rudimentary wings, viewed in the light of modern science, clearly indicate that Hesperornis was in this respect a degraded type. The Struthious characters which we have noticed in various parts of the skeleton might be regarded, not as evidence of close relationship, but rather as general reptilian characters, common to the two groups through inheritance from a remote reptilian ancestry. According to this view, the wings may have been gradually lost by disuse, after the aquatic life was assumed. In proportion as the wings diminished, the legs and feet increased in size, for their work increased. This change would be strictly in accordance with the law of compensation, and the well known economy of nature. We may suppose, moreover, the ancestors of Hesperornis to have been at one time on an equality with the Loon, and later with the Penguin, in respect to means of flight and swimming. As the wings slowly diminished in size, first came the loss of flight, while the wings retained, doubtless for a long time, their power of propulsion through the water. As this too became gradually restricted, the legs and feet gained proportionally. The power derived from them, aided indirectly by the tail, in time so predom- inated, that the wings became entirely aborted, a remnant of the humerus alone remaining. 114 ODONTORNITHES. During the life-history as thus indicated, Hesperornis would exemplify in the waters of the Cretaceous period the evolution that has recently taken place in ocean navigation, in the gradual change of the side-wheel steamer into the modern propeller. Another explanation seems on the whole more reasonable, and more in accordance with the known facts. The Struthious characters, seen in Hesperornis, should probably be regarded as evidence of real affinity, and in this case Hesperornis would be essentially a carnivorous, swimming Ostrich. The diminutive wings and very large posterior extremities would then have been acquired on land, by the same means that have given similar characters to the Ratite, and subsequently have been adapted to an aquatie life. Against this view, the carnivorous character of Hesperornis would be no valid objection. The long neck and peculiar jaws and teeth would be equally effective in seizing prey on the land, and many of the herbivorous cotemporaries would doubtless have been easy victims. This would be precisely analogous to what we have among the corresponding eroups in the Dinosaurs. There is to-day no evidence that any of the Struthious birds, or their ancestors, ever possessed the power of flight, although this is generally assumed. The case is even stronger with Hesperornis, as this genus stands much nearer the ancestral type, both in structure and in time. The absence from the sternum of any trace of a keel is alone strong proof against flight; the peculiar Dinosauroid union of the scapula and coracoid, unlike that of any volant bird or reptile, confirms this; and other testimony bearing in the same direction is not wanting. All Carinate birds, moreover, so far as known, indicate by their embryology that they have passed through the Struthious, or lower stage ; and some of them, Tinamus, for instance, still retain one or more of its distinctive characters. There are, indeed, various flightless birds, recently extinet, which do not belong to the Ostrich group, but are truly Carinate in all their essential features. The Dodo (Didus), Solitaire (Pezophaps ), Cnemiornis, and Notornis are well known examples; but these all show in their shoulder-girdle unmistakable traces of the lost power of flight. The RESTORATION OF HESPERORNIS. 115 characters necessary to volant movements, once attained, would appear never to be completely lost, and this alone seems to furnish a crucial test. When such suggestive indications are wanting in the skeleton, we may fairly challenge any assumption of previous flight. Although Hesperornis' may thus, like its Reptilian ancestry, have always been incapable of flight, the anterior limbs may have long continued limited aids to locomotion. Whether used actively in the air, like the wings of the Ostrich, or of young swimming birds, or passively, like the sail-set pinions of a Swan, or later as imperfect paddles, the wings of Hesperornis were certainly not well fitted for diving, and hence they gradually became useless, and virtually disappeared. We may imagine among the reasons for the gradual loss of wings, the fact that they were too weak to be of much service under water, while from their position they added greatly to the resistance, especially during rapid diving. To diminish this resistance, they would naturally be applied closely to the side, and from such disuse, would gradually suffer atrophy. In this great swimming bird, as thus modified, we have presented to us an interesting problem in animal mechanics. The wings may be regarded as wanting, since the remnant of the humerus was attached closely to the side, as in the Apteryx, if not entirely concealed beneath the skin, like a scapula. The locomotion was therefore entirely performed by means of the posterior limbs, a specialization here seen for the first time in aquatic birds, recent or fossil. Those who have observed a Penguin or a Loon swimming beneath the water know what a vigorous use such birds then make of their wings, however useless these members may appear to be on land. Not only do the wings, in such a case, assist in the forward movement through the water, but they are of much service in steering. A Penguin, when in swift sub-aqueous flight, can turn around, by the aid of its wings, while moving twice its length. Hesperornis had no such aid, but the legs and feet were far superior, for swimming and diving, to those of the Penguins, not merely in power, but in the more perfect adaptive mechanism. This was doubtless the main reason why the posterior limbs of Hesperornis became so predominant. 116 ODONTORNITHES. The tail of Hesperornis was clearly of great service in its aquatic life. In the number of vertebrae and length, it exceeds nearly all known birds, and it is unique in its widely expanded transverse processes, and in its depressed, horizontal, plough-share bone. This broad horizontal tail reminds one of that of the beaver, and was undoubtedly of great assistance in steering, and in diving. Whether it was, like the beaver’s tail, destitute of feathers, or like the tail of Plotus was furnished with long stiff rectrices, so as to act as a rudder, cannot at present be determined ~ with certainty, although the latter view seems more probable. That Hesperornis was provided with feathers of some kind, we can hardly doubt. The surrounding circumstances were evidently very favorable to Hesperornis for along period. ‘There was apparently during this time an absence of enemies in the air above, and an abundance of food in the water. Hesperornis was more than a match for the gigantic toothless Pterodactyles, which hovered over the waters here in such great numbers, and the other inhabitants of the air all appear to have been small. The ocean in which Hesperornis swam teemed with fishes of many kinds, and thus a great variety of food was at hand, and obtained with little effort. In this aquatic paradise Hesperornis flourished, disturbed only by the serpentine Mosasaur, which, even without tradition, we may imagine, caused its banishment, if not its destruction. In the preceding description, the writer has compared Hesperornis with the Ratits, or Ostrich group, and also with the diving birds as exemplified by Colymbus and Podiceps, and the more noteworthy points of resemblance or difference have been stated. It will hardly be profitable to extend the comparison to other groups of modern birds, as the similarity in points of structure is mainly of a general nature. The genus Archeopteryx from the Jurassic of Europe, the only bird at present known more ancient than Hesperornis, presents some marked points of difference, for example, its elongated tail; as well as some peculiar resem- blances, which will be discussed in the concluding part of this volume. RESTORATION OF HESPERORNIS. 117 The genera Ichthyornis and Apatornis, which are from the same geological formation as Hesperornis, and are very fully described in the following chapters, differ widely from that genus in having biconcave vertebre, as well as in other most important characters. Hesperornis really stands alone among all known birds, and the special characters which thus separate it will be stated in the table of classification at the end of the present memoir. a See a eee ; be it & a ee i SON TORNITEES. PP ALR be. Ob. ON TT OT Or MA. (Plates XXI-XXXIV.) Crear irik. EX. THE SKULL AND TEETH OF ICHTHYORNIS. (Plates XXI, XXVI, and XXXIV.) Tue birds included in the second order of Odontornithes, so far as now known, were all of small size, and possessed powerful wings, and very small legs and feet. They differed widely therefore in form and habit from those described in Part I, and, as we shall see, exhibited various significant characters, which distinguish them more strongly from the Odontolce than any existing birds are separated from each other, Some of these characters, as for instance their biconcave vertebrae, separate them widely from all birds recent and extinct, and point back unmistakably to a very lowly ancestry, even below the reptiles. The remains of this group preserved are more or less pneumatic, and this fact, in connection with their small size, is perhaps the main reason why so few have been discovered. As might naturally be expected, the hollow bones of flying birds, being filled with air, enable the carcass to float upon the water much longer than it otherwise would, and it is thus liable to be destroyed by fishes or other animals. Hence, the chances of the entombment of a complete skeleton are greatly diminished. Such delicate bones, moreover, even after their entombment, require a favorable combination of circumstances for their preservation in good condition. 119 120 ODONTORNITHES. Fortunately, the bottom of the Cretaceous ocean, in which the remains of these birds were embedded, left nothing to be desired in this respect, since in its fine calcareous sediments the most delicate vascular impressions of the bones were often preserved. A most careful search in these deposits, as they are now exposed on the Plains east of the Rocky Mountains, conducted by the writer in person, with the aid of other members of the various expeditions already men- tioned, has resulted in bringing to light various remains representing no less than seventy-seven different individuals of this group of the Odontornithes. These remains are all in the Yale College Museum, and form the material on which the following descriptions are based. An investigation of these fossils shows that they are included in two well marked genera, Ichthyornis and Apatornis, the former represented by several species, and the latter by only one. These were all small birds, scarcely larger than a Pigeon. In their powerful wings and small legs and feet, they remind one of the Terns, Ley > and, according to present evidence, they were aquatic birds of similar life and habits. Tue Sxutut. (Plate XXII.) In Ichthyornis dispar, the type of the genus Ichthyornis, and of the order Odontotorme, the skull was very large in proportion to the rest of the skeleton. This disproportion is shown in the restoration on Plate XXVI. The cranial portion of the skull is quite short, but the facial part is much produced. The occipital condyle is very small, and directed backward. Above the condyle, the occipital portion of the skull was nearly vertical. The lateral margin of this surface is bounded by a sharp ridge which separates it from the temporal fossa, and this ridge united with its tellow above was continued forward on the median line, as a sagittal crest, between the temporal fosse. The latter were large and deep, and were separated from the orbit by a moderate post-orbital process. The orbits were very large, and near together. The quadrate is well preserved in one species, and its articular head has only a single facet, as in ITesperornis, and the Ratite. SKULL OF ICHTHYORNIS. 121 The quadrate of Ichthyornis victor is represented in the cuts below, figure 28. It is of large size, and in its general form does not differ essentially from the quadrate in some modern birds. The undivided articular head (figure 28, h) is its most important feature. Fig. 28. Figure 28.—Quadrate bone of Ichthyornis victor, Marsh; twice natural size. a. posterior view; b. front view; ¢. inner view; d. outer view; h. head; p. articulation for pterygoid; j. depression for jugal. The skull of Zchthyornis has deep glandular depressions along the roof of the orbits, similar to those on the cranium of Hesperornis, and some recent water birds. The premaxillary bones are separate throughout their posterior half, but were doubtless firmly united in front. The ante- rior part of these bones has not been recovered, but the lower jaws would seem to indicate that in some respects it resembled the beak of Hesperornis. In the type specimen (number 1450), the occipital condyle is sub- trihedral in outline, when seen from behind. In another specimen (num- ber 1459) this condyle is cordate in form, resembling the corresponding part in Hesperornis. In most of its features, the base of the skull is more like that of Hesperornis than of any modern birds. Tue Bra. The brain of Ichthyornis was remarkably small, and in its main features, strongly reptilian. In form and proportions, it resembled the brain of Hesperornis more nearly than that of any other known bird. The figure given below represents an outline of the skull of Ichthyornis victor, with a cast of the brain-cavity in position. A comparison of this figure with that of the skull and brain of Hesperornis, represented on page 9, will show the main points of resemblance. The most noticeable reptilian features in the two brain-casts are the elongated form, and prominent optic lobes. 16 122 ODONTORNITHES. The olfactory lobes of Ichthyornis (figure 29, ol) were large and prominent. They were separated in front by the inter-orbital septum, and their crura made their exit, apparently, through separate foramina. The cerebral hemispheres were of moderate size, and proportionally less elongate than in Hesperornis. Their form and proportions, in Ichthy- ornis victor, are well shown in figure 29, c, which represents a cast taken directly from a cranium with this portion well preserved. The optic lobes (figure 29, op) are very large, and the brain at this part was nearly as wide as across the hemispheres. The cerebellum (ch) was large, and elongate. The general form and dimensions only of the posterior part of the brain of Ichthyornis are shown in figure 29, and hence are indicated by a uniform shading. The outline given was obtained from a comparison and measurements of two different specimens. Fig. 29. ‘ Fig. 30. Fiaure 29.—Outline of the skull and brain-cavity of IJchthyornis victor, Marsh; seen from above; five-sixths natural size, Figure 30.—Outline of the skull and brain-cavity of Sterna cantiaca, Gmelin; same view; natural size. ol. olfactory lobes; c. cerebral hemispheres; op. optic lobes; cb. cerebellum. SKULL OF ICHTHYORNIS. 123 Figure 30 represents the skull and cast of the brain-cavity of a Tern, and the figure of Ichthyornis, beside it, is made of the same absolute length for comparison. The Tern was chosen for this purpose because in size and structure of its skeleton it bore considerable resemblance to Ichthyornis. The mode of life, also, of the two birds was probably similar. In comparing the brain-cavity of Ichthyornis with that of Sterna, as shown in figures 29 and 30, the strong contrast in size is at once apparent, while the most marked difference is seen in the cerebral hemispheres. If the latter alone be compared, the hemispheres of Ichthyornis are less than one-fourth the size of those of Sterna. If, however, the bulk of the entire brain of each be compared, that of Ichthyornis would be less than one-third the size of that of the Tern. This would hold true after allowing a con- siderable reduction for any uncertainty in regard to the exact proportions of the posterior part of the brain-case of Ichthyornis. The result of this comparison between Ichthyornis and the Tern agrees yery closely with that obtained in contrasting the size of the brain-cavities of Hesperornis and Colymbus, as given on page 10; and hence is of special interest, as in no other instances have the brain-cavities of Mesozoic birds been investigated. Tur Lower Jaws. (Plates XXI and XXVI.) In the type specimen of Ichthyornis dispar (No. 1450), both lower jaws are preserved. The two rami are entirely separate, having been united in front only by cartilage. They are large and massive, nearly straight, and much compressed transversely, especially in the posterior portion. The upper margin of the dentary bone is nearly straight. The anterior extremity is rounded, and without any distinct symphysial sur- face. The right ramus is well shown in Plate XXI, figures 2, 3, and 4. The sutures in this jaw are nearly all obsolete, with the exception of that between the splenial and angular bones. This suture is especially open on the inner surface of the jaw (figure 2, e), and, in life, doubtless admitted of some motion. There is no mandibular foramen, and, just behind the articular face for the quadrate, the extremity is abruptly truncated. 124 ODONTORNITHES. The dentigerous portion of the lower jaw is so similar to that of some of the smaller Mosasauroid Reptiles, that, without other portions of the skeleton, the two could hardly be distinguished. Tue Treetu. (Plate XXI.) The teeth of Ichthyornis were implanted in distinct sockets, thus differing widely from those of Hesperornis. They are all sharp and pointed, more or less compressed, and strongly recurved. The crowns are coated with enamel, and the two fore and aft cutting edges are sharp and smooth, without serrations. The maxillary teeth appear to have been larger than those opposing them. Their alveolar cavities are crowded together, and yet distinctly separated from each other by a thin septum of bone. They are oval in outline, and quite shallow, owing to the small vertical extent of the maxillary (Plate XXI, figure 1, and 1a). Whether the premaxillary bones contained teeth is uncertain, but they were probably edentulous, as in Hesperornis. The dental cavities in the lower jaw are elliptical in outline, and vary in size according to position. They are in general deeper, and separated more widely from each other, than those in the maxillaries. In the right lower jaw of the type specimen of Ichthyornis dispar, there are twenty- one distinct sockets. The anterior one is very near the extremity, and contains a tooth of moderate size. The largest teeth in the lower jaw are just back of the middle of the dentary bone. From this region to the posterior end of the dentary, the teeth diminished gradually in size. The whole surface of each tooth exposed above the. jaw is covered with smooth enamel. In the right lower jaw of the type specimen, there are eight teeth still in their sockets, and from these the others have been restored in outline, as shown in Plate XXI. In Jchthyornis anceps (number 1749), the lower jaws are more slender than in the type species, and the number of teeth greater. The anterior half of the dentary contained large teeth only. The posterior eleven teeth were all small, and nearly uniform in size. SKULL OF ICHTHYORNIS. 125 A specimen of Jehthyornis victor (number 1735) had much stouter and deeper lower jaws than / dispar, but they appear to have contained the same number of teeth. In the lower teeth of IJchthyornis, the pulp-cavity passes well up into the base of the crown. The fang is compressed, and directed downward and forward. It is firmly set in a deep socket, which it nearly or quite fills. The dental succession took place vertically, as in Crocodiles and Dinosaurs; not laterally, as in Hesperornis, and the Mosasaurs, a fact of no little significance. The young teeth are much inclined when they first appear above the jaw, after the old teeth have been expelled. The following measurements of portions of the skull and jaws of Ichthyornis are from the specimens above described : Measurements of Skull. (No. 1459.) Transverse diameter of occipital condyle, ------------------------------------ Rae Median vertical diameter of occipital condyle, ---- --------------------------- 2.2 Distance between upper margins of orbits, ---------------------------------- 4.5 Measurements of Jaws. (No. 1450.) Extent of three sockets for maxillary teeth, ----.---------------------------- 5. = Length of entare lower jaw,-.--.-.---.-----------.-. .--------------------- 67. Merten OU alyEglacMUare We on ones ao - esos See aes ono = oon aan saa -- =~ == =- 41. transverse Giamuter Of ramus, <.2-c---52-40-----~ = 2-22 sen ee 5 - <= =~ == 2 Greatest vertical diameter, .........--------------------------------------- 8. Transverse diameter of jaw at articulation, - -.------------------------------ 5.5 Height of crown of tenth tooth, above jaw, -------------------------------- 2.1 Depth of ramus, below first tooth, -..--.-..-------------------------------- 4.5 Depth of ramus, béfow last tooth,_-..--.-..----.----------=--------------- 3. Measurements of Lower Jaw. (No. 1749.) Extent of ramus containing twenty-two teeth, ....-.-------------------------- 36.5™™ Depth of ramus, below second tooth, -----...-..-.------------------------=-- 2: Depth of ramus, below last tooth,.-.-..-..-...---------------------------- 4.5 Measurements of Lower Jaw. (No. 1735.) Extent of ramus containing the last eleven teeth, -..------------------------- 20 Depth of ramus, below first of above teeth, ------...------------------------ 4. Depth of ramus, below last tooth, ---.......---.---------------- ------------ 4.5 CHAPTER X. THE PRESACRAL VERTEBRA OF ICHTHYORNIS. (Plates XXII, XXVI, XXVII, and XXXIV.) The presacral vertebrae of Ichthyornis present characters more remark- able than those of any other known birds, and hence those preserved are here described in detail, and fully illustrated. The series is not complete in any one skeleton yet discovered, but the large number of vertebra secured in various specimens make clear the more important features of this part of the structure. None of the vertebrae of Jchthyornis, in front of the sacrum, are ankylosed together, and a number of them contain cavities in the sides, which are apparently pneumatic. Among existing birds, the Terns appear to bear the nearest general resemblance to Jchthyornis, and hence their vertebree are here used for comparison with those of that genus. Tue Atxas. (Plate XXVII, figure 1.) Tue atlas of Ichthyornis, as in most living birds, is a ring of bone, very short antero-posteriorly, compared with the succeeding vertebrae. In Ichthyornis victor (number 1733), it is only about one-third as long as the axis, exclusive of the odontoid process, the proportion being nearly the same as in Sterna regia, Gambel. The centrum of the atlas in Ichthyornis is coéssified with that of the axis, forming a true odontoid process, which is robust, and somewhat convex upon its upper surface. The hypapophysis presents a crescent 127 128 ODONTORNITHES. shaped articular surface in front, for articulation with the occipital condyle. The angles of this crescent are prominent on each side above the odontoid process, and during life doubtless gave attachment to a ligamentary bridge above that process, completing a nearly circular cup, adapted to the occipital condyle, as in existing birds. The hypapophysis is produced below into a somewhat wedge-shaped process, pointing slightly backward, and strengthened at the base behind by a low small tubercle on each side. The posterior articular face of the hypapophysis is imperfectly exposed in the specimen figured, but was evidently adapted to the anterior articular surface of the axis, in such a manner as to allow only a rotary movement about the odontoid process. This motion was, however, much limited by the strong zygapophyses, which are much better developed than in the Tern. They are strengthened by a ridge rising on the side of the vertebra, and running upward, backward, and somewhat outward, and ending, above the zygapophysis, in an obtusely pointed tubercle. A similar form of zygapophyses is seen in the Egyptian Stork, Ciconia alba, but in the Tern, these processes are rudimentary. The upper surface of the atlas is not well preserved in the specimen here described, but appears to have been destitute of a neural spine. Tue Axis. (Plate XXVII, figure 1-1d.) oc The axis of Ichthyornis (number 1733) is proportionally a little longer than in the Tern. The centrum is firmly coéssified in front with the centrum of the atlas, no trace of the suture remaining. Below the odon- toid process, is a large articular surface, not exposed in the present specimen, but evidently of a lunate form, somewhat concave, and permitting little motion except rotary. The lateral and inferior edges of this articular surface are distinct, and rather acute; and the articulation is borne upon an anterior expansion of the centrum, which attains its greatest diameter at the edges of the articular surface. These characters are well shown in number 1775. The ventral surface of the centrum is evenly rounded for nearly one- third of its length, then excavated on each side of a median carina, which PRESACRAL VERTEBR.Z OF ICHTHYORNIS. 129 thickens backward, and nearly acquires the character of a hypapophysis below the posterior articular surface. On each side of the centrum near the middle, is a rather acute salient tubercle, just behind which the trans- verse diameter of the centrum is at the minimum. Above and a little behind these tubercles, are large pneumatic foramina. The posterior articular face of the centrum is much smaller than the anterior, and is sub-quadrate in outline. It is distinctly concave vertically, and less so laterally; and begins to assume the peculiar character seen in the biconcave articulations of the following vertebree. The pre-zygapophyses are sessile facets on the anterior margin of the lateral walls of the neural canal. They project very slightly in front of the general contour lines, much less however than in Ciconia, while in the Tern these processes do not project at all. The post-zygapophyses are larger than the pre-zygapophyses, and in form are sub-triangular, with | rounded angles. They are concave in a lateral direction, and strengthened above by a stout ridge running out, somewhat beyond the zygapophysis on each side, into a blunt point. This projection is much weaker than in the Tern, where the post-zygapophyses are surmounted by prominent, divergent processes. The neural spine was low, and stout, more elongate antero-posteriorly, but less robust, than in the Tern. The following measurements show the principal dimensions of the atlas and axis in one specimen of Ichthyornis victor: Measurements of Atlas and Azis, (No. 1733.) (PRS he Gur AUN Teese eet eg ee Baan Pa tae ee ea eee e BY ee eri ptlitgia Genero Gia tVIn. one Saete ete. 2 eee hd ee ee Se 7.0 Length of centrum of axis, including odontoid process, --.-------------------- 9.0 PPraneverna immeceMOrl Ono Ny aids so se Hk 8 Sek to Soe se 3.5 Transverse diameter of atlas, across zygapophyses, --------.------------------ 7.0 Transverse diameter of anterior articulation of axis,._......_.---------------- 3.5 Vertical diameter of posterior articulation of axis, ---.---.-------------------- 2.7 Transverse diameter of posterior articulation of axis, -...---.--.---------------- 1.8 Transverse diameter of vertebra, across post-zygapophyses (approximate), -- -- -- 9.0 17 130 ODONTORNITHES. Tue Turrp VERTEBRA. (Plate XXVII, figure 2.) The third cervical vertebra of Ichthyornis (number 1733) differs much from the second, and in several respects resembles the corresponding vertebra in the Tern. The centrum is deeply excavated in front, and the anterior articulation is much inclined to the axis of the vertebra. Behind this excavation, the centrum abruptly becomes wedge-shaped, but the lower edge thickens posteriorly, and, below the posterior articulation, is about half as broad as the articular surface itself. The plane of the anterior articulation is nearer horizontal than vertical, being inclined at an angle of nearly or quite 60°, with the axis of the centrum. In a vertical section, this articulation is moderately convex, while laterally it is strongly concave, thus presenting an approach to the characters of the ordinary avian vertebral articulation. The posterior articular surface of the axis, and of this vertebra also, fail to show this peculiarity. This specialized feature, occurring at the first bend of the neck, gives a direct hint as to the origin of the unique articulation in the vertebre of modern birds. The full explanation of this articular structure will be given in a subsequent chapter. The posterior articulation is, like that of the axis, sub-quadrate in outline, but is more elongate vertically than in that vertebra. This face is concave vertically, and less so horizontally, except perhaps near the margins, which are rounded. On the sides of the centrum near the lower part of the anterior articular surface, rather strong parapophyses are developed, which are connected with the diapophyses above by a strong bridge of bone, or pleurapophysis. These together enclose on each side a small, vertically oval, lateral foramen, Immediately above, and a little external to the lateral foramina, are the strong and elongated pre-zygapophyses, which are directed somewhat obliquely downward, outward, and forward, and are about equally convex in both directions. They are proportionally longer than in the Tern, and look more directly upward and less inward than in that species. Below, they are strengthened by a ridge running up from the outer wall of the lateral foramen. The post-zygapophyses are shaped much as in the PRESACRAL VERTEBR& OF ICHTHYORNIS. 131 preceding vertebra, being only about half as long as the pre-zygapophyses, and looking almost directly downward. They are surmounted on each side by a blunt tubercle pointing outward and backward, but much less salient than in the Tern. The neural spine was moderately strong, but is much broken in the specimen here described, number 1733. Full measurements of the third cervical vertebra of Ichthyornis victor are given below: Measurements of Third Vertebra. (No. 1733.) LenOE OG Ces nh Ls eae aes Oe ee ee ee Bec! | o= Transverse diameter of anterior articulation of COntrw, =. 7 =... Soe 2.5 Vertical diameters of anterior articulation of centrum,_-__-----.----.---__-- 1.8-1.0 Transverse diameter of posterior articulation of centrum,__-___........-.------ 2.0 Vertical diameter of posterior articulation of centrum,__..---...----------___- 2.7 ‘Transverse diameter of neural canal,_..._.......-.--------------------2-ae-e 2.6 Vertical diameter of neural canal,_..__../..=1........-.-=----------2e-5-- 2.0 Transverse diameter of vertebra, across pre-zygapophyses,..__.......--------- 7.5 Transverse diameter of vertebra, across post-zygapophyses, ..._......_.--___-- 9.2 Transverse diameter of lateral foramen,._............----------------------- 0.8 Nertreal diameter af Interal foramian,._--........ 2... 1-.-----2L--- aces 1.1 Tue Trento Vertepra. (Plate XXVII, figure 3). A vertebra from the cervical series, in number 1733, appears to correspond best to the ninth cervical of Sterna regia, with which it may be compared. The inferior surface of the centrum is flattened laterally, but concave longitudinally. The anterior articulation is somewhat crushed below, but seems to have been concave in both directions, and of a sub- quadrate form. The posterior articulation is more rounded, but has a nearly straight upper margin. It is about equally concave in both direc- tions, and rounded on the edges. From the antero-lateral regions of the inferior surface of the centrum, are two projecting processes, or catapophyses, extending downward and somewhat forward to a distance nearly equal to the vertical diameter of the anterior articulation of the centrum. Above and external to these, on each side, are the parapophyses, united to the diapophyses by the ankylosed pleurapophyses, which 132 ODONTORNITHES, together enclose the lateral foramen of the centrum. The posterior prolongations of the pleurapophyses are broken off. The pre-zygapophyses look upward and backward at an angle of nearly 45°, and a little inward. They are elongated, being fully twice as long as broad, and are semi-oval in form. At the sides of the vertebra, they are connected by a ridge with the bases of the post-zygapophyses. Above this ridge, on each side of the vertebra, is a deep foramen, appar- ently pneumatic. The post-zygapophyses are similar in shape to the pre- zygapophyses, but look nearly directly downward, and a little outward. They are each surmounted by a low tubercle, which does not project beyond the zygapophysis, but is placed nearly above the middle of it. The upper surface of this vertebra is flattened, and destitute of a neural spine. It presents a shallow oval depression at the middle of the anterior margin, and, behind, shows a similar, but somewhat larger and deeper median pit for ligamentary attachment. The principal dimensions of the tenth vertebra of Jchthyornis victor are as follows: Measurements of Tenth Vertebra. (No. 1733.) IBS Ye NiGose (Seek ee eee ae oe en eee Sees ecee seace ost 02> Transverse diameter of anterior articulation of centrum, ----------------------- 3.0 Vertical diameter of anterior articulation of centrum, ------------------------- 2.1 Transverse diameter of posterior articulation of centrum, ---------------------- 2.8 Vertical diameter of posterior articulation of centrum, ----~-~------------------ 2.8 Transverse diameter’ of neural canaljoo2> a. ee ee oun Vertical diameter of neural canal, ---- _------ "ie AE os ae Oy pe ere ears 2.2 Transverse diameters of vertebra, across pre-zygapophyses, - ---.----------- 8.8—10.0 Transverse diameter of vertebra, across post-zygapophyses, ---- .--------------- 8.5 Transverse.diameter of lateral foramen = o-- 4a. === =ee ee ee eee 2.0 Vertical diameter of lateral foramen,--. 2. ------- --- 2c = = ee eee ee 1.8 Tae Tweitrra Vertesra. (Plate XXII, figure 1. ? a) A yertebra here regarded as the twelfth of the series (number 1450) resembles the eleventh cervical of Sterna régia. The centrum is much reduced in thickness near the middle, and is pierced on each side by a pneumatic foramen. A strong hypapophysis arises from the anterior part PRESACRAL VERTEBRA OF ICHTITYORNIS. 133 of the lower surface, and extends downward, and a little forward. Its base is about two-thirds the antero-posterior extent of the centrum, and, below, it is broadly rounded. The anterior face of the centrum is slightly inclined to the axis of the bone, so as to look somewhat downward. It is sub- quadrate in outline, with rounded angles, and is slightly broader below than above. The margin, except below, is prominent, and the face is concave in both directions. The posterior face is more nearly circular, in outline, but flattened above, and is more concave than the anterior. Strong diapophyses arise on the lower part of the anterior end of the centrum, and are directed outward, backward, and downward. They are united with slender pleurapophyses, which complete the lateral foramina by union with the parapophyses above, just below and exterior to the pre- zygapophyses. The lateral foramina are larger in proportion to the size of the centrum than in the eleventh cervical of the Tern, as is also the neural canal. The pre-zygapophyses look strongly inward, and somewhat upward and backward, and are separated from the post-zygapophyses by a distance only about equal to their length. Between the pre-zygapophyses, the neural canal is broadly open above. The post-zygapophyses look almost directly downward, and somewhat outward. They are well elevated above the floor of the neural canal by its nearly perpendicular walls. The roof of the neural canal is short on the median line, excavated by a longitudin- ally oval pit below, and bears a rudimentary neural spine, less prominent than in corresponding vertebra of the Tern. Below the neural spine, behind, is a pit for ligamentary attachment. The neural canal is large. The following are the main measurements of this vertebra in the type specimen of Jchthyornis dispar: Measurements of Twelfth Vertebra. (No. i450.) (EOS ee LE BOL ee re oo Vertical diameter of anterior articulation of centrum,__.._......._.________--- 2.5 Transverse diameter of anterior articulation of centrum,______________________- 2.8 PROG iets DADO BNYAIB SQ 2. 2-2 OL ee ee aS S. be Sees Se 2.8 Antero-posterior diameter of hypapophysis, at base,_.__________..--.-_.. 2... 3.0 134 ODONTORNITHES. Transverse diameter of vertebra, across diapophyses, - ------------------------- 10.0"™ Transverse diameter, across pre-zygapophyses, - ..--...--..---.---.--.------- 7.6 Transverse diameter of vertebra, across post-zygapophyses, -------------------- 7.0 Vertical diameter'of neural canal, 225. ps er ee eee ee 2.4 Transverse Giameter: of meurall! Caria eee eee ee ee ee 2.6 Greatest diameter of Jateral foramen eee 2.6 Least diameter ‘of, lateral’ foramen; s-2 232 gapophyses, 136 ODONTORNITHES. The SixreentH VerTEBRA. (Plate XXII, figure 3.) A vertebra of the type specimen (number 1450) resembles both the sixteenth and seventeenth of the Tern. It belongs in the dorsal series. The centrum has its least transverse diameter at the middle, and sends perpendicularly downward from the anterior half a sub-conical hypa- pophysis, two-lobed at the apex. This process is wanting in the Tern. The articular faces of the centrum are sub-cireular in outline, somewhat flattened above. Both are concave, and the posterior is more deeply so than the anterior. The centrum is very deeply excavated at the sides, above the middle, leaving only a thin wall on the median line. The articulation for the head of the rib is borne on the lateral wall of the neural canal, and is obliquely oval in outline. The diapophyses are broken away in this specimen, and the neural spine is very imperfect. The zygapophyses of the opposite sides are near together. The pre-zygapophyses project but little in front of the centrum, while the post-zygapophyses extend nearly their whole length behind it. Another vertebra (number 1733), somewhat posterior to that regarded as the fifteenth, has the centrum preserved, but is not figured. This cen- trum is rounded below, slightly carinate on the median line, and sends down three divergent tubercles near its anterior end. Two of these are lateral, and somewhat divergent. The third is smaller, and more acute, and is median in position, thus representing a hypapophysis. The anterior face of the centrum is injured in this specimen, but appears to have been, like the posterior face, sub-cireular, and somewhat concave. On each side above the middle, the body of the centrum is excavated, so as to leave only a thin longitudinal median wall to support the floor of the neural canal. The articular surface for the head of the rib is above this excavation, on the lateral wall of the neural canal. PRESACRAL VERTEBRA OF ICHTHYORNIS. 137 The sixteenth vertebra of Jchthyornis dispar measures as follows: Measurements of Sixteenth Vertebra. (No. 1450.) Rene UDTOMONUUUWl es asa 8 oo cc a sn eo ee ee ee 5.8™" panera mater ot) GANthUn. .- — at ee ee em 1.2 Vertical diameter of anterior articulation of centrum, ------------------------ 2.5 Transverse diameter of anterior articulation of centrum, ---------------------- 2.5 Vertical diameter of posterior articulation of centrum, - -----.----------------- 2.4 Transverse diameter of posterior articulation of centrum, -- ------------------- 2.4 ently Obey papo pn yeas ea sn eg 1.8 Transverse dismetamor neural canal. 26525 2 soe eo oe oe 1.8 Transverse diameter of vertebra, across pre-zygapophyses, - ---~---------------- 3.3 Transverse diameter of vertebra, across post-zygapophyses, ay pai a oa ae 2.5 Tue SevenTeentTH VERTEBRA. (Plate XXVII, figure 5.) A dorsal vertebra (number 1733) may be best compared with the seventeenth vertebra of the Tern, to which it bears some resemblance The centrum is but little contracted at the sides, and is almost evenly rounded below. It presents near its anterior end two unequal tubercles, the right one being the larger. The articular faces are slightly elongated vertically, and nearly equally concave. The sides of the centrum in this vertebra are deeply excavated above the middle for almost the entire length, so that only a thin layer of bone is left to support the floor of the neural canal. This structure is much like that in the corresponding vertebra of the Tern, and a little more pronounced than in what may have been the preceding vertebra. The articular surface for the head of the rib is situated above the anterior part of this excavation, and is supported by a ridge running upward and backward on the outer wall of the neural canal. The pre-zygapophyses are elongated, and approximate. The dia- pophyses are both broken off. The post-zygapophyses are a little nearer together than the pre-zygapophyses. The neural spine was strong, but is much fractured in the present specimen. This vertebra of Jchthyornis victor has its principal dimensions as follows: 18 138 ODONTORNITHES., Measurements of the Seventeenth Vertebra. (No. 1733.) Wem prt Of ‘cern tr a ate 62735 Tieast transverse diameter of centrum sess =e oe ee 1.8 Transverse diameter of anterior articulation of centrum,_----_--_---_---__-__- 2.5 Vertical diameter of anterior articulation of centrum,__--__--__--__--_______-- 3.0 Transverse diameter of posterior articulation of centrum, ----_---..------------ 2.5 Vertical diameter’of posterior articulation; ==- sss. = e= =e nae ne ee 3.2 Transverse diameter of vertebra, across pre-zygapophyses, --.------------------ 4.0 Transverse diameter of vertebra, across post-zygapophyses, ---.---.------------ 3.2 Tue EIGHTEENTH VERTEBRA. Another vertebra (number 1733) not figured, evidently posterior to the one above described, has the under surface of the centrum somewhat hour-glass shaped, and nearly alike at each end, being destitute of tubercles or hypapophysis, and very evenly rounded throughout below. The artic- ular faces are sub-circular, and about equally concave. Laterally, this centrum, like the one before it, is deeply excavated nearly to the middle on each side, leaving a thin wall of bone along the median line to support the floor of the neural canal. The articular face for the head of the rib is above the anterior part of this excavation, and is obliquely oval. The diapophyses, zygapophyses, and most of the neural spine, are broken away in this specimen. In one of the best preserved skeletons of Ichthyornis victor (number 1732), several vertebrae, apparently consecutive, lie near together in front of the sacrum. The first of these, or the eighteenth vertebra, has been thrown somewhat out of the line of the next two, and possibly may not have been consecutive with them. The centrum is badly crushed, but was evidently excavated at the sides above, as in the Tern, and had concave articular faces. The neural spine was about two-thirds as long antero- posteriorly as the centrum, and was proportionally higher than in the Tern. It does not appear to have been united to the adjoining spines by ossified tendons, as in that bird. The posterior zygapophyses are approxi- mate laterally, as in the sueceeding vertebrae, and look outward as well as downward, The diapophyses were broad, and directed somewhat backward, PRESACRAL VERTEBRA OF ICHTHYORNIS. 139 Ture NINETEENTH VERTEBRA. The succeeding vertebra in the above series (number 1732) is some- what crushed laterally, so that the centrum appears keeled, but it was probably rounded below. At the sides, it is excavated, leaving only a thin median wall. The diapophyses are strong, nearly horizontal, and directed somewhat backward. They are supported by a vertical ridge below, and bear at the end a small sub-circular articular face for the tubercle of the rib. Their upper surfaces are moderately broad. The neural spine is as extensive at its thickened upper end as the centrum. A vertebra not figured, which may be regarded as the nineteenth, is represented in one specimen (number 1733) by less than half of a centrum, apparently the posterior end, which has a nearly circular and slightly concave articular face. This centrum is broken near the middle, showing a transverse section very similar in form to that of a T rail inverted. A similar section might be made across the centra of the three preceding vertebrxe of this species (Ichthyornis victor.) Tur TWENTIETH VERTEBRA. Another more posterior vertebra not figured, belonging to this specimen (number 1733), has a shorter and stouter centrum than the preceding, and is less excavated at the sides. The articular surfaces are concave. The parapophysial surface for the head of the rib is small, and elevated. The neural canal is much crushed, and the zygapophyses, diapophyses, and neural spine, are broken away. The twentieth vertebra of number 1732 is broader below than the nineteenth, and rounded, without trace of hypapophysis. The sides of the centrum are excavated above the middle, and the articular faces are sub- circular, and moderately concave. The diapophyses are strong, and supported by a ridge below. They are directed somewhat backward, and terminated by a small face for the articulation of the tubercle of the rib. 140 ODONTORNITHES. Tue Twenty-First VertTesra. (Plate XXII, figure 4.) A vertebra in the type specimen (number 1450) may be compared with the twenty-first of Sterna regia. The centrum is somewhat crushed, but appears to have been moderately broad below, and destitute of any hypapophysis. It is deeply excavated at the sides, as are the other dorsals, at least those from the posterior part of the series. The articulation for the head of the rib is obliquely oval. The diapophyses as well as the neural spine are broken off; and the zygapophyses are crushed out of position, but were approximate laterally, as in the preceding vertebre. A vertebra (number 1732) adherent to the anterior end of the sacrum, and almost wholly in front of the ilia, thus corresponding with the twenty- first vertebra of the Tern, is also exposed in the matrix with that above described. The neural spine appears to have been developed much as in the Tern, but is here crushed. The diapophyses are broad, and at the end were united to those adjoining by flattened ossified tendons. The pre-zygapophyses look obliquely inward, and upward. The following measurements give the size of the twenty-first vertebra in the type specimen of Jchthyornis dispar : Measurements of Twenty-first Vertebra. (No. 1450.) henge th of Gen tain asta ae a re ee Cnet ‘Least diameter of; centrum) ..°2 2.2. cake 5. et ne ee nee 1.5 Vertical diameter of anterior articulation of centrum,__---_---_--------------- 2.5 Transverse diameter of anterior articulation of centrum, ----..----------- .----- 3.0 Vertical diameter of posterior articulation of centrum, -------.---------------- 2.4 Transverse diameter of posterior articulation of centrum, ----..---------------- 2.5 Transverse diameter of vertebra, across pre-zygapophyses, - - .------------------ 3.0 Transverse diameter of vertebra, across post-zygapophyses, -------------------- 2.5 CHAPTER XI. SHOULDER-GIRDLE OF ICHTHYORNIS AND APATORNIS. (Plates XXIII, XXIX, and XXXIV.) Tue scapular arch of Ichthyornis, and its near ally Apatornis, conforms strictly to the type seen in living carinate birds. This part of the skeleton gives no hint of the peculiar reptilian features in other portions of the structure, and has none of the Struthious characters noticed in Hesperornis. There is apparently no part of the entire shoulder girdle of Ichthyornis, or the allied genus, that might not have been used by some existing birds with strong powers of flight. Tuer Scapuna. (Plates XXIX, and XXXIV.) In the Odontotorme, the scapula presents two well marked types. In Ichthyornis (Plate XXIX, figures 9, 9a, 10-12), the bone is robust near its articular end, where it is moderately, but suddenly, expanded and thickened, much as in Ardea. Here, it presents two confluent articular surfaces. The larger of these forms a part of the glenoid fossa, and the smaller is for articulation with the coracoid. The scapular part of the glenoid fossa is flattened, slightly narrower below than above, and strongly inclined to the plane of the bone. In its largest dimension, it is slightly concave in section, while transversely it is nearly flat, but rounded at the edges. The articular surface for union with the coracoid is nearly 141 142 ODONTORNITHES. hemispherical, and adapted to the subcireular pit in that bone. The acromial process of the scapula is conical, short and pointed, and does not project beyond the articular portion of the bone. The shaft is hollow at the lower end. In the genus Apatornis, the scapula (Plate X XIX, figures 1, 1a, 2-4) is elongate, and distally flattened. In a side view, the lower half is curved, while the upper or distal half is nearly straight. In a front view, the bone appears regularly curved throughout. The proximal end is moderately and gradually expanded transversely, and bears the two customary articular surfaces. The upper and larger of these, forming the scapular portion of the glenoid fossa, is very oblique to the plane of the bone, and is distinctly convex in both directions. It increases in transverse diameter toward the base, instead of becoming narrower in that region, as in Ichthyornis. The articular surface for the coracoid is much as in that genus, being sub-hemispherical, and adapted to the pit on the face of the coracoid. This articular face is bounded internally by a distinct ridge, as shown in figure 3. The principal difference, however, between this scapula and that of Ichthyornis is shown in the development of the acromial process, which in Apatornis is highly developed, and projects far beyond the articular surfaces of the bone, as in Graculus. Anteriorty, this process is obliquely truncated, so that at the end it is blunt, and somewhat wedge-shaped. Its general direction is inclined at an angle of forty-five degrees with the direction of the shaft of the bone at its origin. The shaft of the scapula is perfectly preserved in specimen number 1734. It is here broadest just below the middle, where the upper edge becomes thin, and somewhat expanded Beyond the middle, the bone tapers to near the distal end, and the extremity is rounded. The scapula decreases pretty regularly in thickness from near the proximal to the distal extremity, where it becomes thin and blade-like. The following measurements indicate the size of the scapula in Ichthyornis and Apatornis. SHOULDER-GIRDLE OF ICHTHYORNIS. 143 Measurements of Scapula, (Ichthyornis dispar, No. 1718.) Greatest diameter, across humeral articulation,-.-.------.-------------------- 1 [9 vows Least diameter, through humeral articulation, - - ------------------------------ 2.8 Length of acromial process, from base, - --- ---------------------------------- 0.6 Measurements of Scapula, (Ichthyornis victor, No. 1452.) Greatest diameter, across humeral articulation, ----..------------------------- Pm Least diameter, through humeral articulation, ------------------------------- 3. Length of acromial process, from base, - -----------------------+------------- 0.8 The various bones in the scapular arch of IJchthyornis are shown in position in the figure below. Fig. 31. Figure 31.—Scapular arch of Jchthyornis victor, Marsh; twice natural size. s. scapula; c. coracoid; el clavicle, st. sternum; &. keel of sternum; sr. costal border; g. glenoid fossa. 144 ODONTORNITHES. Measurements of Scapula, (Apatornis celer, No. 1734.) Left. Right. ength of scapula, << = <2 soe oe eee ee 52.5™™ Greatest diameter, across humeral articulation,--------------------- bales Least diameter, through humeral articulation, _~-___-----.----------- oy Length of acromial process; irom) pase) eee a 3.5 Diameter of shaft, above humeral articulation,-___----._..._-____-- 3. 3 Tue Coracorp. (Plate XXIII and XXIX.) In the genus Ichthyornis, the coracoid is a strong bone, much widened at the proximal end, where it articulates with the sternum in an elongated groove on each side. These grooves, and consequently the coracoids in their natural position, cross the median line in front in such a manner © that the right coracoid is external to, or in front of, the left, as in the herons and their allies, and in some other birds. The proximal end is much expanded, as in Colymbus, but the bone is less robust than in that genus. The external part of the expanded portion is thin, and is short in the line of the axis of the bone. The inner angle of the expansion is produced and acute, as in Ardea. The shaft of the coracoid is flattened, nearly or quite as far as the articular surface for the humerus. This surface is nearly flat, suboval, and somewhat less than twice as long as broad. The articular surface for the scapula is a nearly hemispherical pit, or cup, in diameter about equal to the transverse diameter of the humeral articulation. It is placed just within the lower end of that articulation, instead of almost wholly below it, as in Apatornis. Just beneath the articular face for the scapula, is a foramen, per- forating the base of a strong sub-scapular process, which is crushed and broken away from most of the specimens preserved. This process is triangular in shape (Plate XXIX, fig. 16), and much like that seen in Ardea herodias, Linn. Below and beyond the articular faces, the coracoid is flattened vertically, and, at the end, it is expanded; but, as in Ardea, it presents no facet for articulation with the clavicle, which is unknown in Jchthyornis. The shaft of the bone is hollow, with thin walls. SHOULDER-GIRDLE OF ICHTHYORNIS. 145 The coracoid of Apatornis resembles in general that of Ichthyornis, but is less expanded at the lower or sternal end; and these ends overlapped much less extensively than in that genus, if indeed they overlapped at all. The articular surface for the scapula is, as in Ichthyornis, a hemispherical cup, but it is placed almost wholly below the articular surface for the humerus, or the coracoid part of the glenoid cavity. The articulation for the humerus is comparatively broader than in Ichthyornis, and is excavated in the middle, instead of being nearly flat, as in that genus. Under and beyond these articular faces, the coracoid is flattened vertically, almost to an edge, below. The shaft of the bone is bent strongly inward beyond the glenoid cavity, instead of continuing nearly straight, and there is a pronounced terminal knob, but no distinet face for articulation with the clavicle. The glenoidal, or subscapular, pro- cess is perforated by a foramen at its base, as in Ichthyornis. The shaft of the coracoid is hollow. Measurements of the coracoids in the types and some other specimens of Ichthyornis and Apatornis are given below. Measurements of Coracoid. (Ichthyornis dispar, No. 1450.) anton srertaly artioulaulon. eee =o. = esa on eo oo ce eee ii Greatest antero-posterior diameter of proximal end, -------------------------- 1.7 Antero-posterior diameter of shaft, where broken, ---------------------------- 2.2 Transverse diameter of shaft, where broken, .---.---------------------------- 4, Measurements of Coracoid. (Ichthyornis dispar, No. 1718.) Length of humeral articulation, ---..--.-.--.---------------------------------- G6. == Greatest transverse diameter of humeral articulation, --.-----.---------------- 3.4 Transverse diameter of shaft, below humeral articulation, -------------------- 3.8 Wortical diameter o£ -clavicular procgss,--..- --...-- - ss. 2s- 5 -- 5 see ee ae 5. Transverse diameter of clavicular process, ---.-....--------------------------- 2.2 Measurements of Coracoid, (Ichthyornis victor, No. 1452.) Length of humeral articulation, - - -----.------------------------------------- 6.5™™ Greatest transverse diameter of humeral articulation, - -----.------------------ 4. Least transverse diameter of shaft, below articulation, -~---~-.-.-.-.-------------- 3.5 Vertical diameter of clavicular process, --..---------------------------------- 5.2 Transverse diameter of clavicular process, - -- -------------------------------- 2.8 19 146 ODONTORNITHES. Measurements of Coracoid. (Ichthyornis victor, No. 1721.) Length of humeral articulation; 222s eee f.27= Transverse diameter of humeral articulation,...-. -£------____---------+-_._-- 4, Vertical diameter of clavicular process,--- = 2255-5 5-5 oo ase ee ee 5.5 Transverse diameter of clavicular process,..------------------------------=-- 3. Measurements of Coracoid. (Ichthyornis victor, No. 1741.) Length-of humeral ‘articulation’= 3-5. 2.2 ee 6.) Greatest transverse diameter of humeral articulation,___.__......-_---_______- 3.6 Diameter of shaft below humeral articulation,.__.___._..._.-.--_------_______- 4, Vertical diameter of clavicalar process's - 22152321228 eee, ee ae 6. Transverse'diameterof clavienlanprocess=-= 2. 2) ee ee 3. Measurements of Coracoid. (Ichthyornis victor, No. 1743.) Ihenpthr of “coraporde tenn = kre Oke en ee ee Sy er Length of jhumeral@anticulations ==" 855.9) + Sank eee en ee ee 6. Width of thumerallanticul ata ore sore se eee ne eee en 3.2 Diameter of pit for iscapulla;t soe ee oe eee ene 3. Depth of clavioular sprocess, 28: 252-55. 2522. o-oo 5. Transverse diameter of clayioular process,..---.-..--. 9 _ == 22k eee eee 3. Length ‘of plenoid. processive” “e522 ees 2 ee ee ee 4.24 Transverse diameter vor shatt.<=-— 26,8 a ee ee ee 4, Vertical diameter‘of ishaft,..2.2 2-22 80 5-42 a ee ee eee 2:5 Measurements of Coracoid. (Ichthyornis victor, No. 1745.) Length ‘of Jcoracoid} 2s e22 sane Soar Coane eee ee Se a ee ee Se Length of Jhumeral“artioulation,..<-203.02-- = oe) = Sat eee ee ee 6. Width: of, humeralartionlation,=-552.5. 222508222 ese oe ee 3.4 Dianister of pit for scapula, --- 2242 oA ee ee 3.4 Viertical diameter of ‘clavicular/sprocessje= soo oe ee oe ee 5. Transverse diameter of clavicular process, ------------- 2-2 -22---2- 5-4-2 -- 2.8 Transverse diameter of ‘shaft;...<. 0.222022 .-22-- ee ee ee 4, Antero-posterior diameter ‘of ‘Bhatt,2- 2 \2-=) a= pee ee 2.6 Measurements of Coracoid. (Ichthyornis victor, No. 1458.) Length of. humeral ‘artioulation, «-..=- ¢32 = =e a ee ee ee ee 6. mm Greatest transverse diameter of humeral articulation,._..____.__.___.________- 3.2 Least transverse diameter of shaft, below articulation,........___._.__________ 3.8 Vertical diameter of clavicular process,_-.............---------------eeece-e 5. Transverse diameter of clavicular process,.................------------------ 2.5 Diameter of scapular ‘articulation;==..—-— on. ee oe ee 3.8 JLength of ‘glenoid* proneas, 22 3 < Sess sacs ee ee ees 4, SHOULDER-GIRDLE OF ICHTIHYORNIS. 147 Measurements of Coracoid. (Ichthyornis validus, No. 1446.) Longest diameter of humeral articulation,.....-.---------------------------- $= Transverse diameter of humeral articulation, ......--------------------------- 4,2 Least transverse diameter of shaft, below articulation, ......------------------ 4.5 VerticaP diameter of clavicular process, .-.-:-=:--22--2<2- 252 sesen5c-e-s--6r= 5.5 ‘Transverse diagagter of clavicular process, _..-...---------------------------- 3. Measurements of Coracoid, (Apatornis celer, No. 1734.) Left. Right. Shenipthr ob: Goratgid;= ta ee tet a a ease aman oe a 24) 3= Length of humeral/articulation,..---....----------=-==-<<<-<-=-==6 5.5 Transverse diameter of humeral articulation, -- - -------------------- 3.8 Diameter of shaft below humeral articulation, __-------------------- 3. Vertical diameter of clavicular process, --------_..----------------- 5. 5. Transverse diameter of clavicular process, - - --------..-------------- 2.5 rAmieeer tieRUeLnAL GUC. oe 2 = 2 ee een een sae sea ae 14, Tue CuavicLte. (Plate XXIX, figure 7.) The only portion of the clavicle yet known among the Odontotorme is a fragment from the upper end of that bone, in Apatornis. This fragment is considerably flattened, probably in part by pressure. It presents no face for articulation with the coracoid, but tapers to a flattened point at the top, where it is adapted for attachment to the acromial process of the scapula. Tue Srernum. (Plates XXIII, XXIX, and XXXIV.) The sternum in the type species of Ichthyornis is deeply keeled, affording ample room for muscular attachment. The manubrium is broad, and some- what unsymmetrical. The coracoid grooves are elongated, and overlap widely across the median line, as in Ardea, and some birds of prey. As usually occurs in living birds with this peculiarity, the inner end of the left coracoid was above that of the right. These grooves become shallow near the middle, where the ridge of bone behind nearly disappears, but, toward the outer end, the ridge is again well developed, and acute. ‘The lateral margins and all the posterior part of the sternum are absent from the type specimen of Jchthyornis dispar. The specimen of Ichthyornis victor (number 1461) in which the sternum is preserved shows a strong and deep keel, robust and prominent in front. 148 ODONTORNITHES. The manubrium closely resembles that of Ichthyornis dispar. The cora- coid grooves overlap widely on the median line, the left being above the right, as in Ichthyornis dispar. The costal processes and the posterior SD b] '} p outline of the specimen are not preserved. In the genus Apatornis, the sternum was keeled, as in Ichthyornis, although the precise depth of the keel cannot be determined. The manu- brium was narrower than in Ichthyornis, and symmetrical, and ended in a blunt point. The coracoid grooves scarcely more than meet on the median line, and the left is but little depressed below the right, the asym- metry being only about as great as in Graculus dilophus, Gray. The costal process is triangular, and the costal border presents six articular faces for sternal ribs. The first of these is comparatively small and indistinct, and is well in advance of the five succeeding ones. The latter are separated by nearly equal intervals, each about two-thirds as long as the first. The posterior outline of the sternum is much broken, but appears to have been deeply emarginate. The following measurements are taken from the specimens mentioned above : Measurements of Sternum. (Ichthyornis dispar, No. 1450.) Transverse diameter, near anterior end,.-.-...-------------.____-__-....._.- $2.5™™ Depthvof: keel; asm preserva mee tee ea 10. Measurements of Sternum. (Ichthyornis victor, No. 1461.) henge tok: keel 5) sit) 50, eae eee 84, mm Maximum length of keel, 22-22-22 o0- ose se nee a aoe poe eee eee 45. IDY2yo}8 Wade) 1 Pp Se ean = ras 9 Sse Sea ee ne 13.5 Measurements of Sternum. (Apatornis celer, No. 1734.) Greatest transverse: diameter]... 52 ee ee ee 33, ™™ Least: transverse diameter, .-2.-22 5. 25 noose tooo ne eee eae nee ee ne 28, CHAPTER, AIF. THE WINGS OF ICHTHYORNIS AND APATORNIS. (Plates XXIV-XXVI and XXX-XXXI1.) } Tue wings in the two genera of Odontotorme clearly indicate very | strong power of flight, thus differing as widely as possible from the corres- ponding parts in Hesperornis. In Ichthyornis, nearly all the bones of the _wings are well preserved, and these are all remarkably like those of some earinate birds living to-day. These remains give but slight indications of the reptilian features seen in the portions of the skeleton already described. Tue Humerus. (Plates XXIV and XXX.) The humerus in Ichthyornis is strong and well developed, thus differing widely from that of Hesperornis. The most striking feature of the bone is the enormous radial crest, surpassing in comparative size that of any living bird. It is well shown on Plate XXIV, figure 3, re. This crest is, moreover, remarkable for its position, which is in a plane nearly parallel with the long axis of the head of the humerus, instead of considerably inclined to this axis, as in most birds. This fact is clearly shown by a number of well preserved examples, including the type specimen of Ichthyornis dispar (number 1450). In these two points, the humerus of Ichthyornis strongly resembles the corresponding bone in the Pterodactyles. 149 150 ODONTORNITHES. The nearest approach in size among modem birds to the radial crest of Ichthyornis is perhaps to be found among raptorial species, where its elongation along the shaft of the bone is considerable, but its height is still proportionally much less than in the present genus. In Sterna, this crest is even less than half the comparative size in Ichthyornis. The ulnar tuberosity is also well developed, and, like the radial crest, is nearly in the same plane with the head of the humerus. No pneumatic foramina are to be seen in the specimens preserved. The distal end of the humerus of Ichthyornis resembles that of the genus Ardea, rather than Sterna, the ectocondyloid process being repre- sented only by a low tubercle. The shaft of the bone is hollow, with moderately thin walls, and its length is about one-fifteenth less than that of the ulna. In the genus Sterna, the humerus is about one-fifth less in length than the ulna Dimensions of the humerus in the type specimen and in four other species of the Odontotorme are given below: Measurements of Humerus. (Ichthyornis dispar, No. 1450.) Length ‘of ‘humerus, 22 225-2 == 2 2s2 25-50 ce s= = a eno se eee ee 58.0™™ Greatest diameteriof proximall/end)22-- =) - = 99 == eee eae eee 13.0 Diameters'of proximal) articulation) === = 922222222 sneer eee eee 3.0-10.0 Diameters of ‘distalend;=- 5-2-2 o-oo ene ee een eee eee 9.6— 5.0 Least’ diametersiofishatt, 222-525. --205-- = eee eeeeere 3.0- 4.8 Length from proximal end to rise of radial crest, ------------------------ 23.5 Transverse diameter, across radial crest, .---.-----..-------------------- 12.0 Measurements of Humerus. (Ichthyornis dispar, No. 1730.) Length of humerns, < <<< 2222 cs. 5455-6 4see< 3-2 on ee ee ee 62:52" Greatest diameter'of proximal’ end, _.-< --.... =~. en en eee eee 15. Transverse diameter of proximaliend)_- = 22° 525. 1 —* 5 ons a ae eee 3.1 Diameter of: ahaft,...2<--.cc 50.5.0 23 so ete ee ee Se ee ee 5.2 Groatest diameter of, distaliend) 2... 3° 3 ee oe ee Antero-posterior diameter of ulnar condyle, ..-.............-...-...--=....- 5. Measurements of Humerus. (Ichthyornis victor, No. 1742.) Length ‘of ‘leit ‘humerus, 33 <2 2-o oe oe ee oe ee ee ee {22a Gresdtest diameter,or.proxmnal end, se: o= bo on seca ee ee ae eee nee 17. Diameters of proximal artioulation, 42----.22-5-ced- ss - eae ape ee eee eee 138-4. Diameters ot distal and 222s ence Senn eee eee oo eo ee ees 12.5-5.5 Diameter of shaft (approximate), ---.----.---.--------.--- ae ee eee A WINGS OF ICHTHYORNIS AND APATORNIS. 151 Measurements of Humerus. (Ichthyornis victor, No, 1447.) Seren ramnaierus! <-.-2 bo. ale lesen Soke oo ee ee (DT Antero-posterior extent of proximal articulation, ....-...-....-...----------- 17.6 Transverse diameter of proximal articulation,.............--.-.---.--------- 4 Transverse diameter of distal articulation, ............-.-.------------.----- 11.2 Greatest diameter of radial condyle, .--...........------------------------- 5.5 Measurements of Humerus. (Ichthyornis victor, No. 1452.) Greatest diameter of proximal articulation, .--/............----.----------- 16.3" Transverse diameter of proximal articulation, ..........---.----------------- 3.9 Gresfenty diameter ofy distal nds: te5 <2. 22 5 Soe ce eee ee ue ea eee 12.1 Greatest diameter of radial condyle, . .......-...---2--2---.-.-2-<--n--20- 6. Measurements of Humerus. (Ichthyornis victor, No. 1457.) Greatextdiametenor distal end: 2257028) tee cele Le bese 11, == Antero-posterior diameter of radial condyle,---...-.-.--..-.--.-..---------- 5.2 Monucalndiamerer OL wadiAall COn@Vle,.o2- =e: 62 a oS ease nse Whoa eee 6.5 DRSWeG hat (fe EIN 2 eS EE Se ee i cer se | 5.5 Measurements of Humerus. (Ichthyornis tener, No. 1738.) \GnbatonOiamerenon Gistal end sate ee oe SS en eee eee 15°" Antero-posterior diameter of radial condyle, --..----.----------------------- 3.3 Vertical diameter of radial condyle, --..-...----.-----------..-- fy once = ae: 4.5 Tue Rapius. (Plates XXIV and XXX.) The radius like the other bones of the wing in Ichthyornis is strong and robust. The proximal end is of moderate size, and presents a sub-elliptic, slightly concave, articular face for the condyle of the humerus, and a slight lateral facet for the ulna. Nearly opposite this facet, the bone is distinctly tuberculated, and on the side toward the ulna at a short distance from the proximal end is a distinct elongate oval facet, raised on a tubercle above the general surface of the bone. This tubercle appears to give the shaft of the bone a short proximal bend toward the ulna. It then becomes nearly straight, until beyond the middle it curves again gradually toward the ulna, and at the end is moderately expanded and thickened. The distal articular surface for the ulna is indistinct, while that for the radial carpal is, as usual, elongate and convex in both directions, 152 ODONTORNITHES. In the genus Apatornis, the face for articulation with the humerus is nearly flat, while the ulnar articular surface is more distinct, and extended upon the side of the bone, as in Graculus. The shaft of the radius is hollow in both genera of Odontotorme. The following measurements will show the principal dimensions of the radius in this group: Measurements of Radius. (Ichthyornis dispar, No. 1450.) Diameter of shalt, Stel e> 32. ae A ee a ee eee See on Greatest; diameteriofedistal end, 22.2 525258228 22 ee en eee 5.5 Least diameterof:distal\end, 2: <2 222.2 8 ts eee ee ee ee 3.0 Measurements of Radius. (Ichthyornis victor, No. 1741.) Length.of :radiuss224s. 2. ee ee ee le ee ee Ola Greatest diameteniof (proximal end) aa-ss—e = ee = ee 5.2 Least diameter of proximaliend)-.se 22-2 2=--5-5-255 sce ee oe eee 3.8 Diameter of shaft; 3-08 39-3-2 oa ee ee ee oe eee 3. Transverse diameter of «distaliend, 2224-2 = = =e ee ee 7.3 Least: diameter of sdistaliend 2 ee ne ae ee eee oe ee 4, Measurements of Radius. (Ichthyornis victor, No. 1733.) Greatest diameter of “proximal OND rs esa roa cee eee eee 260m Least’ diameter'of sproximall end) 2-22. co es = ee eee 3.5 Diameterof-shatt, 2222 8 2 ae ee ee ee 2.8 Greatest diameterof cdistaltend, 9-2 2-2 2. ee ee 6.1 Least diameter of distalend;..--5- = 0022 on oo ee 3. Measurements of Radius. (Apatornis celer, No. 1734.) Greatest diameter of ‘proximaliend)2.- 25-2 2: 2 eee 4.2™m ueast diameter of: proximaliend, 2325-28) see ee ee 3. Diameter of ‘shaft, 2.222 os 22S ok See a ee ee ee 2.2 THe Uuna. (Plates XXIV and XXX.) The ulna in Jchthyornis is a strong moderately curved bone, some- what trihedral proximally, but becoming more nearly cylindrical toward the distal end, where it is also less curved than in the proximal half. The surface for articulation with the humerus is placed more obliquely to the shaft of the bone than in Sferna, and occupies a greater proportion of the WINGS OF ICHTHYORNIS AND APATORNIS. 153 proximal end than in that genus, agreeing better in both these respects with the ulna of Ardea. The shaft of the bone is hollow, with thin walls, and presents faint but unmistakable evidence of the attachment of about a dozen secondary quill-feathers. The distal end of the ulna resembles that of Ardea, rather than that of Sterna. The dimensions of the ulna in three species of Ichthyornis are as follows : Measurements of Ulna. (Ichthyornis dispar, No. 1450.) eiipthir aie in ee ere ee ee ee ee en eo ee a (Ehaniaucrronenumitre steer eter wget Pe ee et 8 Se ee 2.6 Greatest diameter of proximal articulation, --------.------------------------- 5.0 Greatest diameter of distal articulation, =... 2+. 2.2 2.5. s225--- --4---se ene 5.7 Wadst diameter Or distal articnlation,..-5~ses----2---.5-->------ s2s- ae 5.0 Measurements of Una, (Ichthyornis agilis, No. 1453.) Usp rraitn Git LET UR UNE pot ae Se te a ee, 5 i Petar He nrO rial CO eee ere ae ene eee ee te Se eee 9. PAINE LORIERCINC AMON Gress same a eee ek Sees ae flees eee 8. Measurements of Una. (Ichthyornis validus, No. 1740.) Tail G1 WE a So te eee RS See eee 68.5"™ Creaccn@orameten Gua protimal 6NO soe 222 hae ee waa wane oo PELVIC ARCIL OF ICHTHYORNIS. 16% They correspond with only three such processes in the type of Ichthy- ornis Behind these processes, the central portion of the sacrum is thickened below, and sends off a strong transverse process on each side, nearly at right angles with its axis. Behind this, are three pairs of trans- verse processes directed well backward. The second of these is preserved in the specimen, and formed a strong process for attachment with the ilium. The last pair of processes were short, and the independence of the vertebral centrum to which they were attached is indicated by a tuber- cular thickening at the point where the last sacral centrum has coalesced with the preceding. The posterior articular face of the last centrum is somewhat concave, and is nearly semi-oval in outline, being strongly flattened above for the neural canal. Measurements of Sacrum. (Ichthyornis dispar, No. 1450.) LS SUING Ly Gana 2 A oe SE EO ee ee a6." Vertical diameter of anterior end of first centrum, --------.------------------- 3.0 Transverse diameter of anterior end of first centrum,-_---.---.---------------- 3.2 Mean eanechree GlaMeter Of CONias\— =o 2 ooo So wean g ass ose neseceee 1 Vertical diameter of posterior end of last centrum, --------------------------- 1.2 Transverse diameter of posterior end of last centrum, ------------------------- 2.0 Greatest transyerse Giameteriol Bacram,--2-.2- .-_- -----<.-----=-4=2--<:---- 10.5 Measurements of Sacrum. (Apatornis celer, No. 1451.) eau Ge parkon preserved, —. 2.2 -- 222. == - a5 .- sews ~-- ~~~ 0---=- ry Mo Transverse diameter of last centrum at articulation,__-----..----------------- 2.5 Vertical diameter of last centrum at articulation, ----_.-_.-------------------- 1.8 Vertical diameter of sacrum, near proximal end, ----------------------------- 7.0 Tue Peryic Bones. (Plate XXXII.) The pelvic bones in Ichthyornis (Plate XXXII, figures 2 and 3) are strongly codssified with each other. The ilium is obtusely rounded in front, and its antero-superior border is separated a short distance from its fellow. The pre-acetabular part of the ilium is concave externally, and considerably longer than the posterior portion. The ischium is expanded medially, and extends further back than the ilium. It is not united with the ilium posteriorly, thus agreeing with the corresponding part in Hesper- ornis. The pubis has no distinct anterior process, but the post-pubic element is elongated, and its distal portion free. 164 ODONTORNITHES. Measurements of Pelvic Bones. (Ichthyornis victor, No. 1732.) Distance across pelvis between upper margins of acetabula, ------------------- 205 2 Extent of ilium, in front of acetabulum see 20. Greatest breadth of ilium, in front of acetabulum, .-------_-----..----------- 8.5 Least breadth of ilum,in frontiof acetabulumees-s- sees) see ae ee 5. Height of; ilinm, aboyerace tea bol urn: eee ee eee 4 Extent of ischium, behind) acetabulums.25-—- == ee oe eee 22.5 Vertical diameter! of shatt-ofcischium) = 2ee—2- ==) sae 3.5 Greatest vertical diameter of aschiumis on - == —= == a ee 5.8 Transverse diameter Of (Slant, Of tS C My Oris ee mecca ee en 1.5 Hxtentiof pubis, behind acetabulum) == ane ses] ase aa eee ee 26. Vertical diameter of shaft, =-22 522-29. --=s25-- 62 ee ee eee 1.5 Vertical diameter ot pu piss ati Cis tellers ee ee 2.5 The pelvic bones of oy ie {A agp Tks oy pei em ‘ ce oS 7 aa “i ® we > ae si~ uy as Gore Fas » ; ; ; > (odie ee nde, Te ee Cee FY + < - Ss 7 4 - Pie aug tinie™ pony ome ate wer eee a 78 = ys eae ah x Aree ake chee - a * eae ns 4 om ap Seager ae cif > Pua a een i. a CHAPTER XV. LEGS AND FEET OF ICHTHYORNIS AND APATORNIS. (Plates XXV, XXXIII and XXXIV.) Tue legs and feet in the two genera of Odontotorme are comparatively small, and present no peculiar features. The remains preserved agree most nearly with the corresponding parts in modern Carinate birds. Tue Femur. (Plates XXV and XXXIIL) The femur in IJchthyornis is a short and comparatively small bone. The shaft is slender, and nearly smooth, being destitute of the rugosites seen on the femur of Hesperornis. Both the articular faces present the ordinary avian type of articulation, and the shaft of the bone is hollow, with thin walls. The ratio of the fibular face to that for the tibia is about the same as in the Tern, and the general proportions of the bone, as to length, diameter of shaft, and size of articular faces, are nearly the same as in that bird. The femur of Jchthyornis, however, is more than half the length of the humerus, while in Sterna the femur is less than half as long as the humerus. The femur of Apatornis is proportionally larger than in Ichthyornis. The shaft is hollow, with thin walls, and the whole bone is nearly smooth, as in most modern birds. There is, however, just behind the outer condyle, on the inferior surface of the bone, a well marked pit for the attachment of a ligament, and just behind the inner condyle, on the same surface, is a low tubercle. The lateral surfaces of both condyles are also 171 172 ODONTORNITHES. slightly roughened, but not more so than in Colymbus. The precise length of the femur, and, of course, its length as compared with that of the tibia, cannot be determined; but the leg was evidently more strongly built than in Ichthyornis, while the wings appear to have been less powerful. The femur in the type specimen of IJchthyornis has the following dimensions : Measurements of Femur, (Ichthyornis dispar, No. 1450.) Tength'of femur; =< - 2-22 oa es ce ee eee ee ee 30.5"™™ Diameter‘of heads..22>. ) 52225. Ore eee Ae AE te OE eee ee 2.8 Greatest diameteriof proximaliend = 22 =5- ss ee ee ee 6.0 Diametersiofishatt:.4 42-7 +4545 4-. Nes 2 eee eS ee ee 3.0-2.5 Trankyerse diameter on. aistaliends) = 2 ee eae eae = ee eee 6.0 Antero-posterior diameter of inner condyle, ----------------------------- 4.0 Antero-posterior diameter of outer condyle,___+.-..-_-.-..--------=----- 4.5 Antero-posterior diameter at fibular articulation, .-.___-_..--------------- 3.5 Tue Tip1a. (Plates XXV and XXXIITL) The tibia of Ichthyornis dispar is a slender bone of moderate length. The cnemial process is but little developed, and only slightly elevated above the articular face of the bone. The shaft was hollow, with rather thin walls, and was gently curved inward toward its distal end, rather more strongly than in the Tern. In section, the shaft was transversely oval. The fibular ridge arises just below the outer articular face, and is continued downwards until broken off by a fracture of the shaft, as seen in plate XXV, figs. 5-8. It is probable, however, that the ridge continues but a short distance farther on the shaft. The distal portion of the shaft preserved shows no trace of a fibular ridge. It is gently curved inward, and expands gradually to the articulation. Toward the distal end, the shaft becomes more flattened, and is excavated in front for the passage of tendons, which were not, however, kept in place by an osseous supraten- dinal bridge. The articular face has been somewhat injured by compres- sion in the type specimen (number 1450), but seems to have resembled in general form that seen in the Tern, although of comparatively less antero-posterior diameter. LEGS AND FEET OF ICHTHYORNIS 173 The tibia in Apatornis is somewhat larger and stronger than in Ichthyornis. The proximal end of the bone is unknown, and the upper part of the shaft is much crushed and mutilated in the specimen preserved. Distally, it is straighter than in Ichthyornis, and expands to about the same degree. As in that genus, there is no osseous supratendinal bridge, nor is there any apparent roughening for the attachment of the end of the fibula. The measurements given below indicate the size of the tibia in several specimens of the Odontotorme: Measurements of Tibia. (Ichthyornis dispar, No. 1450.) Groatesumuiamerenor proximal €nd,---------_--- 2- -- =~ 322 eo 5.0™ GrearesuuetiiGrO-NOSterion GlaMmeter, — = ~ sae ea a ee py) Vertical diameteriotdistal end): 2-2 52-0 See ase ie ri ee 1.4 CHAPTER XVI. THE RESTORATION OF ICHTHYORNIS. (Plates XXVI and XXXIV.) Tue preceding description of the remains of Jchthyornis, in connection with the illustrations given in Plates XXI to XXXIV, will make known most of the more important characters which separate these remarkable Mesozoic birds from all others, recent or extinct. The material on which the present restoration of two species of this genus are made consists of portions of more than seventy different individuals. None of these various skeletons were as complete, when found, as were several of Hesperornis, from the same formation. This was no doubt mainly due to the smaller size, and more fragile character, of the individual bones of Ichthyornis The restoration of Apatornis, evidently a near ally of Ichthyornis, has not been attempted, as the specimens at present available for this purpose are not sufficiently numerous or complete. In the restoration of Ichthyornis dispar, as given on Plate XXVI, only the remains of the type specimen are figured, as it was deemed important to show the exact material on which the genus was based. These portions are shaded in the restoration, and are of the natural size. The bones represented in outline are taken in part from other specimens of the same species, but mostly from the skeleton of the Royal Tern (Sterna regia, Gambel), which, after a careful comparison, seemed well adapted for completing, approximately, the restoration here presented. The position chosen is one which seemed on the whole to accord best with the remains at present known. 23 177 178 ODONTORNITHES. The restoration of Ichthyornis victor, on Plate XXXIV, is also of natural size, and is based on more complete material, including portions of several skeletons. The other species of the genus Ichthyornis did not differ widely from these in size, and this is true also of the known specimens of Apatornis. In considering the skeleton of Ichthyornis, the anatomist is at once confronted with a strange combination of characters. The wing-bones are conclusive proof that Ichthyornis was a highly specialized bird, with great powers of flight. The individual bones correspond closely with those of birds living to-day. The legs and feet, also, are much like those of some modern birds. With these portions alone before him, the comparative anatomist would unhesitatingly refer the remains to the class of Birds, and would naturally conclude that they belonged to the modem type. I, however, the skull should then be found with the wings and feet, very strong evidence would be required to convince him that they were parts of \\ one and the same bird. The jaws and teeth present reptilian characters wholly unknown in modern birds, while the base of the skull and the small brain point strongly in the same direction. The biconcave vertebre lead Ichthyornis still farther away from all known birds, recent and extinct, and, if found alone with the jaws and teeth, would force any anatomist to the conclusion that he had before him the remains of a reptile. The skeleton of Ichthyornis, as we know it to-day, can be interpreted only, in the light of modern science, by supposing that certain parts have become highly specialized in the direction of recent birds, while others have been derived, with but little change, from a reptilian, or even a more lowly, ancestry. In the wings, the most characteristic modern feature is the codssification of the metacarpal bones, a character universal among existing birds. In reptiles, however, and in the only known Jurassic bird, Archeopteryx, these bones are separate. The sternum of Ichthyornis is very similar to that of modern carinate birds. In the feet of Zchthyornis, also, the compound tarso-metatarsal is another modern feature, especially characteristic of recent birds. RESTORATION OF ICHTHYORNIS. 179 If, now, we consider the skull of Ichthyornis, we find the avian and the reptilian characters strangely blended. The teeth are evidently a strong reptilian feature, and, before the discovery of Ichthyornis, were entirely unknown in the class of Birds. Their method of implantation in distinct sockets is a specialized character in reptiles, and was not shared even by Hesperornis, the contemporary of Jchthyornis. The diminutive elongated brain, also, points back to the reptiles. Other features of the skull, for example, the single headed quadrate, are shared only by the most reptilian of birds. The union of the lower jaws in front, by ligament only, is characteristic of many reptiles, and is seen in Hesperornis, but is unknown in all other birds.'' The form of the skull and the obliteration of most of the cranial sutures are points of resemblance to many modern birds. The locomotive organs of Ichthyornis are so similar to those of typical birds that they present no such interesting mechanical problems as were suggested by the skeleton of Hesperornis. In the vertebra, however, we find a primitive form retained, and also have the key to the origin of one remarkable character in recent birds, which has hitherto remained unexplained. While all existing birds, and all of the extinct forms so far as known, including Hesperornis, have the peculiar saddle-shaped vertebre, those of Ichthyornis, and its near ally Apatornis, are biconcave. This form is seen in a few recent and in many extinct Reptiles, and in the Amphibians; but it is especially characteristic of Fishes, from which class it was undoubtedly inherited by the superior groups. This character alone indicates unmistakably a great antiquity for the class of birds. The saddle-shaped vertebree are certainly one of the most marked features in the skeleton of modern birds. This form is so peculiar and so constant that it has been considered by many anatomists as the best distinctive character for the class. In no other group of animals known do we find the true saddle-shaped articulation of the centra seen in the vertebrie of birds.? ‘In the Pelicans and Curlews, the rami unite late by ankylosis. * An approach to this form is shown in the cervical vertebrae of the Kangaroos. 180 ODONTORNITHES. Not only do the presacral vertebrae of all existing birds exhibit this structure, but the many extinct species now known from the whole series of Tertiary deposits have the same articulation. If we knew only these fossil forms, in addition to the existing species of birds, the origin of this peculiar vertebral articulation would perhaps remain a mystery. Most fortunately, however, one of the vertebrae of Ichthyornis throws much light on this point, and virtually explains the difficulty. We have seen that Hesperornis has the same kind of vertebre as modern birds, while those of Jchthyornis are biconcave. The marked contrast between the shape of the vertebral articulation in these two genera is seen in the figures below, which show a characteristic cervical vertebra in each form. In the vertebra of Ichthyornis shown in figures 32 and 33, it will be seen that the articulation of the centrum is cup-shaped ; while, in the corresponding vertebra of Hesperornis, the ends of the centrum are saddle-shaped, as in ordinary birds. Thus the distinction between the two types in this part of the skeleton is as wide as between Ichthyornis and any living bird. Fig. 32. Ficure 32.—Twelfth cervical yertebra of Jchthyornis dispar, Marsh; front view; twice natural size. Figure 33.—The same vertebra; seen from the left side. Figure 34.—Third cervical vertebra of Ichthyornis victor, Marsh; front view; twice natural size. Fig. 36. Figure 35.—Thirteenth cervical vertebra of Hesperornis regalis, Marsh; front view; natural size. Figure 36.—The same yertebra; posterior view. a. anterior articulation; d. diapophysis; p. parapophysis* J: lateral foramen; ne. neural canal; s, neural spine; 2. pre-zygapophysis; 2’. post-zygapophysis. RESTORATION OF ICHTHYORNIS. 181 To the evolutionist, who believes that birds are all closely connected genetically, this difference in structure, at first sight, offers a most serious difficulty ; since hitherto we have had no hint of a transformation from the one form to the other. In the third cervical vertebra of Ichthyornis, however, we catch nature in the act, as it were, of forming a new type; by modifying one form of vertebra into another. Following this clew, the connection between these widely divergent types of structure soon becomes apparent; and the development of the modern style of avian vertebra from the fish-like, biconcave form finds a ready solution. In the anterior articulation of this vertebra of Ichthyornis (figure 34), the surface looks downward and forward, being inclined at an angle of nearly 60° with the axis of the centrum. In vertical section, it is moderately convex, while transversely it is strongly concave; thus presenting a near approach to the saddle-like articulation. None of the other known vertebree of Ichthyornis possess this character _ This highly specialized feature occurs at the first bend of the neck, and greatly facilitates motion in a vertical plane. If, now, we consider for a moment that the dominant motion in the neck of a modern bird is in a vertical plane, we see at once that anything that tends to facilitate this motion would be an advantage, and that the motion itself would tend directly to produce this modification. With biconcave vertebra, the flexure in any direction is dependent on the elasticity of the fibrous tissue that connects them, as the edges of the cups do not slide over each other. An increasing movement in the neck of Ichthyornis in a vertical plane would tend to deflect the upper and lower margins of the circular cup, and to produce a vertical constriction, and at the same time to leave the lateral margins projecting; and this is precisely what we have in the third vertebra of this genus. This modification of the vertebree would naturally appear first where the neck had most motion, viz: in the anterior cervicals, and gradually would be extended down the neck; and, on to the sacrum, if the same flexure were continued. 182 ODONTORNITHES. Behind the axis, or where the vertical motion prevails, we find in modern birds no exception to the saddle articulation of the vertebra in the whole cervical series, In the dorsal vertebrae, this cause would be less efficient, since the ribs and neural spines tend to restrict vertical motion, and hence to arrest this modification. This region, then, as might be expected, offers strong confirmatory evidence of the correctness of the above explanation; for here occur, among existing birds, the only true exceptions known in the presacral series to the characteristic saddle-shaped articulation. In Strigops and a few other land birds; in the Penguins, the Terns, and some other aquatic birds, one or more vertebree in the dorsal region are without the saddle-shaped articulation, and are either opisthocclian, or imperfectly biconcave. In such instances, we can usually, if not always, detect evidence of an arrest of vertical flexure. This may lock together the posterior dorsals by their neural spines, as in Strigops, leaving the power of lateral flexure; or several vertebree may be codssified, as in Accipiter, and some other Raptores, in which a stiff back is a positive advantage. In the codssified sacral series of many birds, one or more of the anterior vertebrae have the saddle-shaped articulation. This, however, is no valid objection to the above explanation, since these vertebre are really dorsals, and have gradually coalesced with the true sacral vertebre. In the caudal vertebrae of recent birds we have, in a measure, the original biconcave structure preserved, for here the motion in every direc- tion was much restricted. The caudal vertebre of these birds, even in the most aberrant forms, are essentially the same, and in the fossil species the articulations at least appear to follow the general rule. In Pavo and Geoccocyx, the caudal vertebrae exhibit a tendency to a proccelian union. Some other forms also show unimportant modifications of the normal type of caudal articulation, but nothing to suggest a real objection to the explanation now proposed of the origin of the saddle-shaped vertebra, characteristic of Birds. In bringing together the above facts, and others suggested by them, the classification and development of the various forms of vertebree appear to be somewhat as follows: RESTORATION OF ICHTITYORNIS. 1835 (1.) Biconcave vertebre (Fishes and Amphibians); the primitive type ; a weak articulation, admitting free, but limited motion. From this form, have been directly derived the other varieties, namely : (2.) Plane vertebre (Mammals); affording a stronger joint, with motion still restricted. (3.) Cup-and-ball vertebre (Reptiles); a strong and flexible joint, well fitted for general motion, and evidently produced by it. The vertebrae are proccelian when lateral motion is dominant (Serpents); opisthoccelian with varied motion (Dinosaur cervicals). (4.) Saddle vertebre (Birds); the highest type; a very strong and free articulation, especially adapted to motion in a vertical plane, and mainly due originally to its predominance. In considering the mode of life, and habits of Ichthyornis, many important suggestions may be derived from its structure, as well as from the localities where the remains are found. The sharp cutting teeth of Ichthyornis prove, beyond a doubt, that it was carnivorous; its great powers of flight, long jaws, and its recurved teeth suggest, moreover, that it captured its prey alive. Its food was probably fishes, as their remains are found in great abundance mingled with those of Ichthyornis. These fossils occur in the bed of the old Cretaceous ocean in which Hesperornis swam. Both of these birds were clearly aquatic in habit, as shown by various points in their structure, already described, and the conditions under which their remains were deposited. In many respects, Ichthyornis probably resembled the modern Terns in its mode of life. The powerful wings and small feet suggest similar habits in flight, and rest. That Ichthyornis was provided with feathers is proved beyond question by the tubercles for the attachment of quills on the forearm. Beside Ichthyornis and its allies, the only other denizens of the air at present known to have then inhabited the same region were the toothless Pterodactyles. Ichthyornis doubtless competed with these huge dragons for the fishes in the tropical ocean, about which they lived. Wj : re . i OG ong ‘* me “Vea belies “Te ee LP asd he ae <0 cca pe ade ys at ne, 7 r as — dui) ; —~ a gi MIA ae | " La a diode wo Sige gL aim r fe (2 pemeees hie ; ao j , ~=As - ¢L. tee viens Ve ; — Tes weet (ori! an Oe ligt ial oake, a iv. eo eee eG te iP Oa aj a ay 1 a hfe ¢ Ro i hh “1 “ S05 Tora EC a Ok ty ues weer cir Wiet 0F8 Ride riot Jae ee et dt bet Tale Pate foes" at wrap ear eas ee ee ‘oun Vaal aes iets iP t Ani os) a iw it a iti 3 Pia ei: 5 : ran’ ; , a a a . A r o - - ui Ps ee i iy fe : F ALLY aia bu Oana be ; k Rind dF ee j ' ila? .. ae \ fry ib None Me iL Si vy OY LENS Ve ee td elt ie age in Sik naar | myer age wk wines -_¢ pe Led | g@ "mr* rd e 2 ear Pie oe.) ae ‘ CHAPTER. XVII. CONCLUSION. Havine now described the more important characters in the structure, so far as known, of the two groups of Cretaceous Odontornithes, or Birds with teeth, it remains to consider what relation these birds bear to each other, and to allied members of the class; and, also, to inquire if the facts presented throw any light on the profounder question as to the origin of Birds. In comparing Hesperornis and Ichthyornis, as the types of their respective orders, the Odontolee and Odontotorme, the contrast in their principal characters is as striking as it is unexpected. Hesperornis had teeth implanted in a continuous groove, a low, generalized character; with, however, the strongly differentiated saddle-shaped vertebrae. Ichthy- ornis, on the other hand, had the primitive biconcave vertebrae, and yet the highly specialized feature of teeth in distinct sockets. Better examples than these could hardly be found to illustrate one fact brought out by modern science, that an animal may attain great development in one set of characters, and at the same time retain other low features of the ancestral type. This is a fundamental principle of evolution. : The more superficial characters of the absence of wings and th strong swimming legs and feet of Hesperornis are in striking contrast, also, with the powerful wings and diminutive legs and feet of Ichthyornis. These and other characters, already mentioned, separate the two birds so widely that a more detailed comparison seems here unnecessary. ; 24 185 186 ODON'VORNITHES, It would be highly desirable to carefully compare both Ichthyornis and Hesperornis with Archeopteryx, the still older Mesozoic bird. This unfortunately cannot be done at present, as the two skeletons of Archeopteryx, now known, have not yet been fully described, nor even prepared for examination by removal of the matrix. That Archeopteryx belongs to the Odontornithes, the writer fully satisfied himself by a personal examination of the well known specimen in the British Museum. This examination was made in 1878, several years after the writer had become familiar with the American forms of toothed birds. The teeth seen on the same slab with this specimen of Archeopteryx, and referred to it by Evans, although imperfectly preserved, agree so closely with the teeth of Hesperornis, that the writer identified them at once as those of Birds, and not of Fishes. It has since been announced that the specimen of Arche@opteryz, more recently found in Germany, also possessed teeth, although only two of small size were detected. The separate metacarpal bones, and especially the elongated tail, of Archeopteryx, moreover, remove it widely from the known American genera of Odontornithes. It will probably be found, however, that Archeopteryx possessed biconcave vertebra, somewhat like those of Ichthyornis. The other Mesozoic birds now known from the deposits of this country, and the few discovered in Europe, may, some or all of them, have had teeth, but their remains are too fragmentary to determine this point, or even their near affinities. It is an interesting fact that the Cretaceous birds at present known, some twenty species or more, were all apparently aquatic forms, which are of course most likely to be preserved in marine deposits, while the Jurassic Archeopteryx, the only one from that formation, was a true land bird. The Birds found in more recent formations all belong apparently to modern types, and hence present few points for profitable comparison with the Odontornithes. The existing birds with reptilian characters are nearly all confined to the Ratite, or Ostrich tribe. These are evidently the remnants of a very numerous group, once widely extended over different CONCLUSION. 187 parts of the earth; and it is to the fossil forms of these birds that we must look eventually for the intermediate types between them and the less specialized Mesozoic birds. For the present, at least, it seems advisable to regard the Odontornithes as a sub-class, and to separate them into three orders, according to the characters given below. These orders are all well marked, but evidently are not of equal rank. Archeopteryr is clearly separated much more widely from both Ichthyornis and Hesperornis than are these two genera from each other. The free metacarpals and long tail of Archeopteryx are significant characters. Gegenbaur and Morse have shown, however, that young birds of existing species have the metacarpals separate, and this is true for all these birds up to a certain age. Hence this character is of less importance than the presence of true teeth, since in no recent birds, young or old, have these been found. The length of tail is perhaps a char- acter of more value, but even this is a variable feature in modern birds. Sub-class ODONTORNITHES (or Aves Dentar#), Marsh. Order Opontotc#, Marsh. |OponrorormM#, Marsh.| Saurur#, Haeckel. Genus Hesperornis, Marsh. | Ichthyornis, Marsh. Archaeopteryx, von Meyer. Teeth in grooves. Teeth in sockets. Teeth in ———? Lower jaws separate. Lower jaws separate. | Lower jaws ———-? Vertebree saddle-shaped. | Vertebrae biconcave. | Vertebrae ————? Wings rudimentary. Wings large. Wings small. Metacarpals wanting. Metacarpals ankylosed, Metacarpals separate. Sternum without keel. Sternum with keel. | Sternum ——-—? Tail short. Tail short. Tail longer than body. That the three oldest known birds should differ so widely from each other points unmistakably to a great antiquity for the class. Archeopteryz, Hesperornis, and Ichthyornis, are all true birds, but the reptilian characters they possess are convergent toward a more generalized type. No Triassic birds are known, and hence we have no light on this stage of the develop- ment of the class. They will doubtless be found, however, and, if we may 188 ODONTORNITHES. judge from Jurassic Mammals and Reptiles, the next classes above and below Birds, the avian forms of that period would still be birds, although with even stronger reptilian features. For the primal forms of the bird- type, we must evidently look to the Paleozoic; and in the rich land fauna of our American Permian we may yet hope to find the remains of both Birds and Mammals. The genera Archeopteryx, Hesperornis, and Ichthyornis, each possessed certain generalized characters not shared by the others. These characters were undoubtedly united in some earlier form, and this fact gives us a hint as to what the more primitive forms must have been, and suggests the prominent features of the ancestral type. In the generalized form to which we must look back for the ancestral type of the class of Birds, we should therefore expect to find the following characters : (1.) Teeth, in grooves. (2.) Vertebrze biconcave. (3.) Metacarpal and carpal bones free. (4.) Sternum without a keel. (5.) Sacrum composed of two vertebree. (6.) Bones of the pelvis separate. (7.) Tail longer than the body. (8.) Metatarsal and tarsal bones free. (9.) Four or more toes, directed forward. (10.) Feathers rudimentary or imperfect. These various characters may indeed have been combined in an animal that was more reptile than bird; but such a form would be on the road toward the Birds, rather than on the ancestral line of either Dinosaurs or Pterodactyles, as feathers were not a character of these groups. With this exception, all of the characters named belong to the generalized Sauropsid, from which both birds and the known Dinosaurs may well have descended. An essential character in this ancestral type would be a free quadrate bone, since this is a universal feature in Birds, and only partially retained in the Dinosaurs now known. CONCLUSION, 189 The Birds would appear to have branched off by a single stem, which gradually lost its reptilian characters as it assumed the ornithie type, and in the existing Ratitae we have the survivors of this direct line. The lineal descendants of this primal stock doubtless early attained feathers and warm hlood, but, as already shown (p. 114), never acquired the power of flight, The volant birds doubtless separated early from the main avian stem, probably in the ‘Triassic, since, in the formation above, we have Archeopteryx, with imperfect powers of flight. This power of flight probably originated among the small arboreal forms of reptilian birds. How this may have commenced, we have an indication in the flight of Galeopithecus, the flying squirrels (Pteromys), the flying lizard (Draco), and in the flying tree-frog (Rhacophorus). In the early arboreal birds, which jumped from branch to branch, even rudi- mentary feathers on the fore limbs would be an advantage, as they would tend to lengthen a downward leap, er break the force of a fall. As the feathers increased, the body would become warmer, and the blood more active. With still more feathers, would come increased power of flight, as we see in young birds of to-day. A greater activity would result in a more perfect circulation. A true bird would doubtless require warm blood, but would not necessarily be hot-blooded, like the birds now living. The short wings and clumsy tail of Archeopteryx were quite sufficient for short flights from tree to tree, and if the body were essentially naked, as now supposed, we have in this Jurassic form an interesting stage in the development of birds before full plumage was attained. Whether Archeopteryx was on the true Carinate line cannot at present be determined, and this is also true of Ichthyornis ; but the biconcave vertebrz of the latter evidently suggest that this form was an early offshoot. It is probable that Hesperornis came off from the main Struthious stem, and has left no descendants. These three ancient birds, so widely different from each other, and from all modern birds, prove beyond question the marvelous diversity of the avian type in Mesozoic time; and also give promise of a rich reward to the explorer who successfully works out the life-history of allied forms, recorded in ages more remote. : be “ied ove ak : ie ; i a 4452 ia ied |e tp ti ci. at ath! retary ; 7 Licee wi. ; 'S br teal Sar e: ay ). Gat oe ee eo 7 bob ba , Bas 7 aeet®. a: \e : g Vag i Th : i ; 7 Fae F ib wie fe “= vied ae FF a 4 Set We, ‘ : BS dose § Pte Ae oe Mme ad _ i pe . 7 ~=7 m¢ ij : , Pe | =r ie ‘ail Qabeogaep ta ei oid stil a eae 7 ‘ 4k , ng (ae Jeq & rer @: 7 (lp ‘ aie wer J Dineite rid off'- >) Cab mee)

2 Cervical and Dorsal Vertebre of HESPERORNIS REGALIS Marsh. All Figures Natural Size. Page. Fic, 1.—Seventeenth or last cervical vertebra; lateral view, seen from the left. Fypepephyas imperiect, ---3-—2- <= 23-3 === nn yes oe a ee la —Superior viey. 14 —Anterior view. le —Inferior view. Fic. 2.—Eighteenth vertebra; first dorsal, lateral view, seen from the left. Neural spine : SVC UOTE Cg eel a la —Superior view. Right diapophysis removed, _ 2b —Anterior view. 2¢e —Inferior view. Left diapophysis removed, Fic. 3.—Nineteenth vertebra; second dorsal, lateral view, seen from the left,.---.---------.. 3a —Superior view. 36 —Anterior view. 3c —Inferior view Fic. 4.—Twentieth vertebra; third dorsal, lateral view, seen from the left,.._-. -..-._--._-. .- 48 4a —Superior view. 4b —Anterior view. t 4¢ —Inferior view. Fic. 5.—Twenty-first vertebra; fourth dorsal, lateral view, seen from the leit, ----..-. -.------ 507 5a —Superior view. 56 —Anterior view. 5c —Inferior view. Fic. 6.—Twenty-second vertebra; fifth dorsal, lateral view, seen from the left,_--...---------- 53 6a —Superior view. 6 —Anterior view. 6c —Inferior view. Fic. 7.—Twenty-third vertebra; sixth dorsal, lateral view, seen from the leit, ....-------.-.-- 55 7a —Superior view. 7b —Anterior view. Neural spine somewhat too short. , Te —Inferior view. ‘ \ 4 . - A\e \> 4 > - F h, 4 “2159: 34 DS B Fic. Fic. Fic. Fic. Itc. Fia. Fic. PLAWH,. Va. ODONTORNITHES. Bones of Scapular Arch and Sternum of HESPERORNIS REGALIS Marsh. All Figures Natural Size. 1.—-Sternum ; lateral view, seen from the left. Outline restored in part from figures on Plate. Vilj22.- 22... 5-02 24 sobe debe ea weed ese ee ee ee 2°.—Sternum ; ‘inferior views. .52--= =.22.42. 6-22 6262 20 5- oar eee ee ee 3.—Left clavicle; lateral view, seen from the right, -........---.-.----+-.------------- 3a—Inferior end, showing face for cartilaginous union with its fellow. (Compare Plate VIII, Fig. 8.) 4-—Right coracoid ; external surface, -.<.-—- -- a2 a> oor en nn ses oe ee en 5.—Right coracoid ; lateral view, seen from the right, -----------.----.-------.-------- 6:—hightcoracoid's ‘inner'!suriace,-.- 42. -- oa a2 oe eee ee 7.—Left scapula; external surface. (See also Plate VIII, Figs. 5, 6 and 7,)--.---.---. oe Page. 60 60 61 = HESPERORNIS REGALIS, Marsh PLATE VII. PEAT H var ODONTORNITHES. Sternum, Coracoid and Clavicle of HErSPERORNIS CRASSIPES Marsh. All Figures Natural Size. , 1.—Sternum; lateral view, Seen from the left, ---..---5---- 2-5 S22 . 2.—Sternum; anterior view;-..--=-<=-22- 526 2-22 - ens se ae eee a ee ee . 3.—Sternum ; inferior view, showing entire absence of keel, ----.-.----- ---------------- pds RNG COTACOIG ss sIMTAer ENEOG et re » Det Glavaole’s (seen\fromn DOV Gey. cre ence we eee HESPER RNIS CRASSIPES, Marsh. VIII. PLATE Aseria ODONTORNITHES. Humerus, Scapula and Shoulder-girdle of HESPERORNIS REGALIS Marsh. All Figures Natural Size. Fic. 1.—Right humerus; inner surface,-------------------------------------------------- Fie. 2.—Right humerus; radial side, . .....--.--.------- -=--+-------------------=-== <==5 2a —Proximal end. Fic. 3.—Right humerus; ulnar side,_.----. -..-------.---. ----------+---- ---- ---- -------- & —Distal end, showing entire absence of articulation. Fic. 4.—Right humerus; external surface, .------.--- ----~--- ---- ---= --== -=-- -=- - «2 =~ == == Hie. 5.—left scapula; anterior view, ----—--—— ~~ Fic. 6.—Left scapula; inner surface, -- -- -- -- ---- ~~ 0 - <3 -- n n 6a —Proximal end. Fic. 7%,—Left scapula; posterior view, --- ---- ---- --------e-ee-an= Semcon ans. = oe eee Fic. 8.—Sternum and scapular arches, restored ; front view,.--- .--------------------------- st —Sternum. cl —Clavicle. ce —Coracoid. s —Scapula. hk —Humerus. Note.—In figure 8, portions of the scapular arch are diminished in size by the perspective. This is especially noticeable in the humerus. 62 62 ~ RE( RORNIS HESPE Fic. Fic. Fic. Vie. Fic. Fic. Fie. Fia. Fic. Fic. Fic. Fic. Fic. Fic. Fic. Ke. Fic. Vic. Fic. Vic. PLATE £2 ODONTORNITHES. Ribs and Uncinate processes of IIESPERORNIS REGALIS Marsh. All Figures Watural Size. 1.—First articulated rib; from the left side, exterior view, ------------ ------------ ---- 2,—Second articulated rib; “ ss s ee ee en es ha ee es SS 3.—Third articulated rib; “ § SS Oe 3 oe ee ee ee u—Articular surface for first uncinate process. 4,—Fourth, or first true dorsal, rib; from the left side, exterior view, -. ---- ---- -------- 5.—Fifth rib; from the left side, exterior view,--------- -.- -- -2-- === -=-- --=- .--=--<= 6.—Sixth rib; “ £ is eae dic ut soeneer = soe ee Soar 7.—Seventh rib; “ a es EEE Se ee a ee ae on, 8.—Eighth rib; “ « ee eR o's eee eee u—Articular surface for last uncinate process. 9.—Ninth rib; from the left side, exterior view, .-..--- ---------.--. ---------------- 10.—First uncinate process; from left side, articulating with third rib at the point MOTE 24; @XTOLLOL, VIC Ws) sores oe ae ee ea ae! 11.—Second uncinate process; from left side, exterior view, --._.------- .--- .------- ---- 12.—Third uncinate process; “ Ny ss SSG Anan 3 oe ee Ree as x 13.—Fourth uncinate process; “ a < $6. te Soest as pe aslos — 14.—Fifth uncinate process ; as ss s CON ee ee Ee Bs Saree = 15.—Sixth uncinate process ; se S ss A aie ee erie oe ces 16,—First sternal rib; articulating with first true dorsal rib, from left side, exterior view, -- 17.—Second sternal rib; from left side, external view, .-------- ---- Peer a aes 18.—Third sternal rib; ir a & SC ae Se ee a Re IE we re 19.—Fourth sternal rib; “ BS cs ON SE Ge ge an ee 20.—Fifth sternal rib ; si i: - SE ne ee ee eee p = < «bec 8 & G&&ee RE eee PLATE 1x Sencha SP sicand, lith New Haven i 2G ”) ic 0) is (io >) ai 4 4 qj a ne mn al A ng O ng fy Oy icp) 4 ie) PLATE XIT. Fic. Fic. Fie. Fie. Vic. Ira, Fa. Iia. Fic. Via. PLAT Ha 41a- ODONTORNITHES. Caudal vertebre of HrESPERORNIS REGALIS Marsh. All Figures Natural Size. Page. 1.—First free caudal vertebra; lateral view, seen from the left. Neural spine restored, -. 78 1@ —Superior view. 16 — <== === 3a—Proximal end of femur. . 4.—Left femur ; fibular side, ... 2 - == - 52-8 32 = ee ee 88 88 PLATEH XIV. PLAS, _Sceye ODONTORNITHES. Tibia of HrsprrorNis REGALIS Marsh. Natural Size. Page. Kray 1. Left tibia; anterior wiew;e> ce osc ee ee ae Se ee ee ee Fic. 2.—Left tibia; exterior surface, showing fibular ridge, -.....-.-------.---------------- 90 2a—Proximal end. IBxGiao;— ete tibial: posterior View, seeo ese a ee ee eee ea eet 8a—Distal end. co. nN ATX ALY 1d ma) PLATE XV. = 24 PLATE Xe ODONTORNITHES. Patella and Fibula of HrEsPERORNIS REGALIS Marsh. All Figures Natural Size. Page Wig. 1:—Left patellas, anterior View), 222. 2-2 --a5 -2-- 55> oo eae ee oe ne eee ae - Wie. 2.—Left patella; lateral view, seen from theileft, = .--2 22-2222 cso -2 -ne- a= = eee ee Fic. 3.—Left patella; lateral view, seen from the right,.....-----. ------.----.---. .------ 93. Bia, 4.—Left fibula; anterior view, 2.22. <2522e--se= 2 5 =o eo ee ee Ic. 5.—Left fibula; exterior view,,= 2--== ms n ite a < a) A Ps ASE eon ODONTORNITHES. Femur and Tibia of IcHTHYORNIS DISPAR Marsh. All Figures twice Natural Size. Page. Kura, 1.—Left femurs ‘anterior, or superior, View, =2.- -62-.25- ----2- =o steep oe 171 la —Distal end. Kixe.2;— Left femurs inner, or tibial, surface, 2 = 222./5.0 3 -S ace soe- ss cee eae eee Frg..8:— ett femur; posterion view; y-.~-=.=2- 5626... 28 el a aoe Pee eee 3a —Proximal end. Fira. -4.—Heft femur; outer, or fibular; surfacos- 2) 2-228. 2. SP, ee Fic, 5.—Right tibia; proximal portion, anterior view,.---. -------------------- ---- -------- 172 5’ —Distal half; approximately in position, anterior view. 5a—Proximal end. Fic. 6.—Right tibia; proximal portion, exterior view, _--..-.-.------------- ------------- 12 6’ —Distal half; approximately in position, exterior view. Fic. 7.—Right tibia; proximal portion, posterior view, ----------~- = =-- -----se--=-<--sess—== 172 7’ —Distal half; approximately in position, posterior view. 7’a—Distal end. Fie. 8.—Right tibia; proximal portion, inner view, -.-.-.-.--~----. ----------------s=----. 178 8’ —Distal half; approximately in position, inner view. PAA) 2AxkSnN ODONTORNITHES. Restoration of ICHTHYORNIS DISPAR Marsh. Natural Size. Page. In this restoration, the shaded portions represent the parts preserved of the type specimen, (No, 1450, Yale College Museum). The portions in outline are taken in part from other specimens but mainly from the skeleton of a Tern, (Sterna regia Gambel.) -------- ---- --------------»----- 17 “OS. tetas okies a yey Aiba = ? A “ 2 st S m Fic. Fic. Fic. Fic. PLA VY iF ODONTORNITHES. Cervical and dorsal Vertebrz of IcuTHYORNIS vicroR Marsh. All Figures twice Natural Size, Page. 1.—Atlas and axis, or first and second cervical vertebre; lateral view, seen from the left,. 127 la—Superior view. 1—Anterior view 1c—Inferior view. 1d—Posterior view. 2.—Third (?) cervical vertebra; lateral view, seen from the left, ......-.---------------- 2a—Superior view. 2b—Anterior view. 2ce—TInferior view. 2d—Posterior view. 3.—Posterior cervical vertebra; lateral view, seen from the leit,...--------------------- 38a—Superior view. 8b—Anterior view. 8c—Inferior view. 38d—Posterior view. 4,—Anterior dorsal vertebra; lateral view, seen from the left, .-.~----------------------- 4a—Superior view. 4b—Anterior view. 4c—lInferior view. 4d—Posterior view. . 5.—Posterior dorsal vertebra; lateral view, seen from the left, 5a—Superior view. 5b—Anterior view. 5c—Inferior view. 5d—Posterior view. x om — ms a 3 < ia) Ay PLA TH XA at ODONTORNITHES. Sacrum of APpATORNIS, and Caudal Vertebre of IcHTHYORNIS. All Figures twice Natural Size. Page. Fic. 1.—Sacrum of Apatornis celer Marsh; lateral view, seen from the left, anterior end imperfect, 162 1a—Superior view. : 10 —Inferior view. le—Posterior view. Fic. 2.—T'irst of seven consecutive caudal vertebrae of Jchthyornis victor Marsh; lateral view, seen Trom ‘the left,=..:- 52 2 42s: 22222 soe Saas ee a eee eee 165 2a—Superior view. 2b—Anterior view. 2e—Inferior view. 2d—Posterior view. Fic. 3.—Median caudal vertebra; lateral view, seen from the left, _------------------------- 166 3a—Superior view. 3—Anterior view. 3¢e—Inferior view. 8d—Posterior view. Tic. 4.—Median caudal vertebra; lateral view, seen from the left, ...----.- ---------------- 166 4a—Superior view. 4b—Anterior view. 4ce—Inferior view. 4d—Posterior view. Vic. 5.—Posterior caudal vertebra; lateral view, seen from the left,_.... ..------------------ 167 5a—Superior view. 5b—Anterior view. 5¢e —Inferior view. 5¢—Posterior view. Tic. 6,—Posterior caudal vertebra; lateral view, seen from the left, ---.-.-.----.------------ 167 G6a—Superior view. ¢6+—Anterior view. 6c —Inferior view. 6d—Posterior view. Vic. 7.—Posterior united caudal vertebre; lateral view, seen from the left, ..........-.------ 167 Ta—Superior view. 7b—Anterior view. 7c—Inferior view. 7d—Posterior view. Norn,—The caudal vertobre figured on Plate XXVIII are consecutive, and were found with the pelvis represented on Plate XXX, figures 2 and 3, Ic. 1.—Left scapula of Apatornis celer Marsh; exterior view, .---.----------------------- la —Proximal end. Fic. 2.—Left scapula of Apatornis celer ; anterior, or superior, view, ----------------------- Fic. 8.—lLeft scapula of Apatornis'celer; inner view, ----<- =-- --2 == ==s5 a2 eee eae eee Fic. 4.—Left scapula of Apatornis celer ; posterior, or inferior, view, -------------.--------- Fic. 5.—Right coracoid of Apatornis celer; anterior view, --------.----------------------- 5a@ —Sternal end. Fic. 6.—Right coracoid of Apatornis celer ; posterior view, ------.----------.---------«--- Fic. 7.—Upper portion of clavicle of Apatornis celer ; exterior view, -.-----------.--------- 7a— Posterior view. 7b— Inner view. 7c— Anterior view. Fie. 8.—Sternum of Apatornis celer; inferior view, ------------------------------------=- ; Fic. 9.—Left scapula of Zchthyornis victor Marsh ; exterior view, ---. -----.----------------- 9a—Proximal end. Fig. 10.—Left scapula of Zchthyornis victor ; anterior, or exterior, view, --------.-------.---- Fic. 11.—Left scapula of Zchthyornis victor ; inner view, .--. ------------------«-------~---- Itc, 12.—Left scapula of Zchthyornis victor ; posterior, or inferior, view, ----------~--------- Fic, 13,—Left coracoid of Ichthyornis victor ; posterior view, ----------------------------+- Fic. 14.—Left coracoid of Jchthyornis victor ; exterior view, --- ---- -- mylene ee Fic. 15,.—Left coracoid of Zchthyornis victor ; anterior view,-------.---------------:------- 15a—Scapular end. Fie, 16.—Left coracoid of Ichthyornis victor ; inner view, -.--.--~.----.-------------------- PLAGE. Seis ODONTORNITHES. Sternum and Scapular Arch of APATORNIS and JcHTHYORNIS. All Figures twice Natural Size. 147 141 144 F Berger, del 1-8, APATORNIS ‘ i. Z “ < | fy PLATE X= ODONTORNITHES. Wing bones of APATORNIS and IcHTHYORNIS. All Figures twice Natural Size. Page. 2) Fic. 1.—Left radius of Apatornis celer Marsh ; proximal portion, inner view, ----------.--.-- 152 Fic. 2.—Left radius of Apatornis celer ; proximal portion, inferior view, -----------------_-- 152 Fic. 3.—Left radius of Apatornis celer ; proximal portion, exterior, or ulnar, surface,......... 152 3a —Proximal articular surface. Fic. 4.—Left radius of Apatornis celer ; proximal portion, superior view, ------------------- 152 Fic. 5.—Left radial carpal of Apatornis celer ; anterior view, ----------------------------- 153 5a —Inner view. 56 —Posterior view. 5e —Lxterior view. 5d —Superior view. 5e —Inferior view. Fic. 6.—Left ulnar carpal of Apatornis celer ; anterior view, .-.---.-----------------..----- 154 6a —Exterior view. 6b —Posterior view 6c —Inner view. 6d —Inferior view. 6e —Superior view. Fic. 7—Right humerus of Jchthyornis victor Marsh; immer, or posterior, view, --...-.-...... 149 Fic. 8.—Right humerus of Zchthyornis tener Marsh; distal portion. Inner, or posterior, viey;, 149 8a@ —Ulnar surface. sb —Outer or anterior view. Se —Radial surface. 8d —Distal end. Fic. 9.—Left radius of Jehthyornis victor Marsh; proximal portion, inner view, -- ---- -- ---- 151 9a —Inferior view. 9b —Exterior, or ulnar, surface. ‘ 9¢ —Superior view. 9d —Proximal end. Fria. 10.—Left radius of Ichthyornis victor ; distal portion, inner, or anterior, view, -------- -- 151 10a—Inferior view. 10)—Exterior, or ulnar; surface. 10e—Superior view. 10¢—Distal end. Fic. 11.—Left ulna of Jchthyornis validus ; exterior view, -..---..--.-----------.---------- 162 es Fic. 12.—Left ulna of Ichthyornis validus ; superior view-.---. .--- .--. ----.--- ------.---- 152 12a—Proximal end. Fic. 13.—Left ulna of Jchthyornis validus ; inner, or radial, surface, ................-....... 152 Fig. 14.—Left ulna of Zchthyornis validus ; inferior view,.-..--.---.--.-.---.--------.------- 152 14a—Distal end. 10 MORNIS 'T 6, APA SE OKT PLAT 2 xa ODONTORNITHES. Wing bones of APATORNIS and IcHTHYORNIS. All Figures twice Natural Size. Fic. 1.—Right metacarpal of Apatornis celer Marsh; exterior view,-..-..-.---------------- Fic. Fic. Fic. Fic. Fic. Fic. Fic. Fic. Fic. Fic. Fic. Vic. Fic. Fic. Fic. Via. Ite. Fie. 1a —Distal end. 2.—Right metacarpal of Apatornis celer ; superior, or ulnar, surface, __.__._..-..--.----- 3.—Right metacarpal of Apatornis celer ; inner view, ---.---------------------------- 38a@ —Proximal end. 4.—Night metacarpal of Apatornis celer ; inferior, or radial, surface, ._...._..---------- 5.—Proximal phalanx of second digit of Apatornis celer ; from the right wing, inferior, or radial surface, 222 5-2 422552 30.cn5.20 eee oe eee ee 5a@ —Distal end. 6.—Proximal phalanx of second digit of Apatornis celer ; exterior view,_.-.---.------ - 7.—Proximal phalanx of second digit of Apatornis celer ; superior view,--------.------- 7a —Proximal end. 8.—Proximal phalanx of second digit of Apatornis celer; inner view, -..--------------- 9.—Second phalanx ot second digit of Apatornis celer ; from the right wing, inner view, - 9a —Distal end; with outline of proximal end. 10.—Second phalanx of second digit of Apatornis celer ; inferior, or radial, surface, -----.- 11.—Second phalanx of second digit of Apatornis celer ; exterior view,-.--.--.---------- 11a—Proximal end. 12.—Right metacarpal of Iehthyornis victor Marsh; exterior view, ....-.---. .---------- 12a—Distal end. 13.—Right metacarpal of Iehthyornis victor ; superior, or ulnar, surface,_...--.--------- 14.—Right metacarpal of Ichthyornis victor ; inner view,..----.--------.-------------- 14a—Proximal end. 15.—Right metacarpal of Ichthyornis victor ; inferior, or radial, surface, ...--.---- ae 16.—Proximal phalanx of second digit of Ichthyornis victor ; from the right wine, inferior, or radial, surficg,. soon 5-0 sae nee ee oe aan ee rr’ 16a—Distal end. 17.—Proximal phalanx of second digit of Zchthyornis victor ; exterior view, .......------ 18.—Proximal phalanx of second digit of Jchthyornis victor ; superior, or ulnar, surface, 18a—Proximal end, 19.—Proximal phalanx of second digit of Zchthyornis victor ; inner view, ....-----.-.-- ee 4) _— PLATE XXXTI. PLAT eis ODONTORNITHES. Pelvis of APATORNIS and ICHTHYORNIS. Twice Natural Size. Page. Fic. 1.—Pelvis of Apatornis celer Marsh; lateral view, seen from the right, ..--...---------.- 164 il —Ilium. és —Ischiur~. pb—Pubis. af—Acetabular foramen. a —Ischio-sciatie interval. 6 —Obturator interval. Fic. 2.—Pelvis of Ichthyornis victor Marsh; superior view, - .----..-.---------------------- Aer |S: él —Tlium. is —Ischium. pb—Pubis. s —Sacrum. Wie. 8.—Pelvis of Ichthyornis victor ; inferior view,----------------. ---------------------- 163 s —Sacrum restored in part from sacrum of Jehthyornis dispar, (Pl. XXII, fig. 7.) r s . 4 rs ’ * ; 4 Bee hee ah H 4 ) 7 5 oe ae Nemes HK XXXIItl. PLAT « PLATE Arr ODONTORNITHES. Leg bones of APATORNIS and IcHTHYORNIS. All Figures twice Natural Size, Page. Fic. 1.—Right tibia of Apatornis celer Marsh; anterior view. Proximal portion wanting, ---. 173 Fic. 2.—Right tibia of Apatornis celer ; exterior view, -----. -.--. -------- ---------------- 118 Fic. 38.—Richt tibia of Apatornis celer ; posterior view, ---- ---- -------------------- ------ 173 8a —Distal end. Fic. 4,—Right tibia of Apatornis celer ; inner view, -- ------------------------------------ 173 Fic. 5.—Right fibula of Apatornis celer ; posterior view, -- -------------------------.-.--- 174 Fic. 6.—Right tibia of Zchthyornis victor Marsh; anterior view, ---------. .---------------- 172 Fic. 7.—Right tibia of Ichthyornis victor ; posterior view, ----------.----------------.---- 172 Fie. 8.—Phalanx of Zchthyornis victor ; superior view, ----------------------------------- 176 8a —Proximal end. 8b —Lateral view. 8¢e —lInferior view. 8d —Lateral view. 8e —Distal end; with outline of proximal end, Fic. 9.—Right tarso-metatarsal of ZJehthyornis victor ; anterior View,. ---.---.-------------- 174 9a —Distal end. Fic. 10.—Right tarso-metatarsal of Ichthyornis victor ; exterior view,--.-------------------- 174 Fic. 11.—Right tarso-metatarsal of Ichthyornis victor ; posterior view,----.-------.---.----- 174 lla—Proximalend. , Fic. 12.—Right tarso-metatarsal of Zehthyornis victor ; inner View, -----.-------------- eRe So fy: * PLATE XXXII P Berger, del u ‘ « - 7 « 7 7 P - * . Th -3 - ¢ t 4 ‘ . . es 7 , . . 2 . eo o . . 5 z sd ' ’ ui + + PEATE XXXIV. PLAT HEH Seki ODONTORNITHES. Restoration of IcHTHYORNIS VICTOR Marsh. Page. Natural Size. In the restoration of this species, the shaded portions represent the parts preserved of the type specimen, (No. 1452, Yale College Museum), as well as of other individuals from the same region, and the same geological horizon, The portions in outline are taken in part from the type of Ichthyornis dispar, (Plate XXVI), and completed from the skeleton of a Tern, (Sterna regia Gambel) .........- 177 «le ¥ TE A I ee oe te - + oe , a * ‘ ri. pee a } F 5g. OO, Soe * ° . ¢ en OC Te a ~ ® ~”%, A ey "4 ’ 7 . — ty ‘o® > » «4 s a - “> by . ° ae J . a Pie yo 3 e : ; A , . co ‘ - ‘ we > s : 4 9 « * "a t t. ‘ baer 5 “~—* ‘ i ° "on The | . _ P ee y ‘ » = e ° ; +3 4 : bon i. ak if c “& . » * . ’ La f ¥ ‘ a" ° UNIVERSITY OF ILLINOIS-URBANA i HNN 0112 02721952