British Museum (Ratural History). 2 - TT ; - This 1s No.. De of 25 copies of the “ Descriptive Catalogue of the Marine Reptiles of the Oxford Clay,” Pt. 1, printed on special paper. PRESENTED The TCruste OF THE BRITISH MUSEUM. Sere = Nee a TUTE EF ad. ae 4443 sARSS Mt Nd g TV OVE ree! HOWIAS LLN Lal A DESCRIPTIVE CATALOGUE OF THE MARINE. REPTILES DEE Ox ORD, CLAY. BASED ON THE LEEDS COLLECTION IN THE BRITISH MUSEUM (NATURAL HISTORY), LONDON. PART I. BY CHARLES WILLIAM ANDREWS, D.Sc, PRS, yf) q LONDON: 214 Gb0 PRINTED BY ORDER OF THE TRUSTEES OF THE BRITISH MUSEUM. SOLD BY LONGMANS & Co., 39 PATERNOSTER ROW, E.C.; B. QUARITOH, 11 GRAFTON STREET, NEW BOND STREET, W.; DULAU & Co., Lrv., 37 SOHO SQUARE, W.; AND AT THE BRITISH MUSEUM (NATURAL HISTORY) CROMWELL ROAD, 8.W. 4910. (All rights reserved.) PRINTED BY TAYLOR AND FRANCIS, RED LION COURT, FLEET STREET. Pon Pee. JuRING the past twenty years the British Museum has gradually acquired the fine collection of Reptilian skeletons obtained by the Messrs. Leeds, of Eyebury, from the Oxford Clay in the neighbourhood of Peterborough. Most of the specimens represent marine Reptiles of the Orders Ichthyopterygia, Sauropterygia, and Crocoditia; and the associated sets of bones have been extricated from the rock with so much skill and care that they afford an unique opportunity for acquiring a good general knowledge of the Reptilian fauna existing in the Upper Jurassic sea. Dr. Charles W. Andrews has therefore been entrusted with the preparation of a Descriptive Catalogue of the collection, and it is hoped that his exhaustive work will form a useful basis for future researches in the same field. The separate bones of many of these reptiles have now been studied and described as thoroughly and satisfactorily as if they were from freshly-macerated skeletons; and it is only to be regretted that a considerable proportion of the specimens are too much distorted by crushing in the soft moist clay to allow of any exact measurements. The variations observed in the different individuals of some species are especially noteworthy ; and the growth-stages traceable in certain parts, such as the Elasmosaurian shoulder-girdle, are also of great interest. In accomplishing his task Dr. Andrews has been much assisted by Mr. Alfred N. Leeds, who made the greater part of the collection, and has given the Museum the benefit of his long experience. Part I. contains the account of the Ichthyosaurs and Plesiosaurs. Part II. will be devoted to the Pliosaurs and Crocodiles. A. SMITH WOODWARD. DEPARTMENT OF GEOLOGY, British Museum (Natura Hisrory), 13th July, 1910. a2 icNeRik-©O: DUG LON, Neary all the remains of the marine Reptilia of the Oxford Clay enumerated and described in this Catalogue, were collected from the numerous clay-pits near Peter- borough worked for the making of bricks, an industry that is extensively carried on in that neighbourhood. A few of the earlier specimens were discovered by Mr. Charles E. Leeds, M.A., but the greater part of the collection was made by his brother, Mr. Alfred N. Leeds, F.G.S., of Eyebury, who soon became associated with him. It is more than forty years since the collection was begun by Mr. Charles E. Leeds, and some of his first discoveries were described and figured by Phillips in his ‘Geology of Oxford and the Valley of the Thames,’ published in 1871. He left for New Zealand in 1887, but his brother has continued the work to the present day with the most astonishing results. Both in the number of species represented and in the perfect preser- vation of their remains, the Leeds Collection far surpasses any other single collection of Mesozoic Vertebrates, especially one in which all the specimens are from one horizon and from a restricted area. Not only marine forms, but remains of terrestrial reptiles, including several species of Dinosaurs, have been obtained. In nearly all cases the specimens have been collected with extreme care, usually by Mr. Leeds himself, the bones of the different parts of the skeleton being numbered and packed in separate parcels. Frequently, portions of the skeleton, such as the skull or limb-girdles, can only be extricated from the clay in fragments, but these haye been reunited with the greatest skill and patience by Mr. Leeds. The consequence of this care is that, in the case of some of the more nearly complete and uncrushed skeletons, it has been possible to mount the bones in their natural] relations vi MARINE REPTILES OF THE OXFORD CLAY. as easily as if they had been obtained by the maceration of a fresh carcass. A notable instance of this is the fine skeleton (R. 2860) of Cryptocleidus oxoniensis, which is figured on the Frontispiece and forms the basis of the restoration given in text-figure 94 on page 188. In this case, as in many others, the bones, which all belong to a single individual, are uncrushed and undistorted. Often, however, the skeletons have been subjected to great pressure, and have thus been extensively fractured and deformed. Unfortunately the skulls are especially liable to injury, and therefore any specimens approaching completeness are very rare. Occasionally the whole skeleton or portions of it are embedded in an intensely hard pyritous clay, and when this is the case all attempts at clearing away the matrix are usually hopeless. Another cause of the imperfection of many of the skeletons seems to have been the dismemberment of the carcasses by carnivorous reptiles, probably some of the Crocodiles and Pliosaurs whose remains are also common in the Oxford Clay. Bones are often found scored across by deep grooves, obviously cut by sharply pointed teeth. Moreover, the curious manner in which whole sections of the skeleton, as, for example, a limb, are sometimes wanting in otherwise nearly complete specimens, or, on the other hand, the occurrence of isolated paddles and other parts of the skeleton, seems to show that the dismember- ment occurred while the bones were stil] united by the soft tissues. A notable instance of this incompleteness is seen in the case of the portions of the skeleton of the giant Dinosaur, Cetiosaurus lcedsi, described and figured by Dr. A. 8S. Woodward (Proc. Zool. Soc. 1905). In this specimen, the skull, the whole of the vertebral column in front of the sacral region, the left fore limb and the right hind limb, as well as the ischia and pubes, are wanting, while the left hind limb is almost complete, even to the phalanges, also the right fore limb except the manus; the vertebral column of the tail, again, is represented by two series of successive and complete vertebre, an anterior series of about 27, and a posterior one of 10, while the intervening portion is entirely absent. Although careful search and extensive excavations have been made, none of the missing parts have been found, and such absence of whole sections of the body seems to be best explained by supposing the carcass to have been dismembered while the bones were still united. he horizon at which these reptilian bones occur is that characterised by the 5] presence of the “Ornatus” group of Ammonites, one of the species most commonly found in actual association with the bones being Cosmoceras gulielmit, INTRODUCTION. vil J. Sowerby (= Ammonites jason, Reinecke, sp., fide Oppel). Many other species of Cephalopods have been collected in the beds by Mr. Thurlow Leeds and others, the most important being Cardioceras lamberti, C. serratwm, Cosmoceras spinosuwin, C. duncant, C. ornatum, Peltoceras athleta, P. williamsoni, var., Aspidoceras per- armatum, Quenstedtoceras marie, and Belemnites owent. The horizon at which these forms occur is described by English stratigraphers as the Lower Oxford Clay *. By continental geologists strata of the same age would be called Upper, or Middle and Upper Callovian, but, as Mr. H. B. Woodward? has remarked in the memoir referred to below, “This seems a quite unwarranted stretching of a formation to suit local stratigraphy and in defiance of its original significance.” Accordingly the horizon in which the reptilian remains are found is here called the Lower Oxford Clay (Middle Oxfordian). The general succession of the beds of this age in Northamptonshire has been described by Prof. J. W. Judd { under divisions b-e as follows :— (f) Zone of Ammonites cordatus. (2) Clays with Ammonites of the group of the Ornati. Dark blue clays with nodules of pyrites and numerous pyritic Ammonites, including A. ornatus, A. duncani, A. bakerie, and A. athleta, and also Waldheimia impressa, Dug in brickyards about Whittlesey, at Thorney, and Hye Green. (d) Clays with Belemnites hastatus. Blue clays with many fossils found in Division c, but characterised by the abundance of B. hastatus. Dug at Werrington, Ramsey, and Eyebury. (c) Clays with Belemnites oweni. Dark blue clays and shales with B. oweni, often cf gigantic size. Gryphea dilatata occurs, but is more plentiful in the beds above. Saurians and fishes occur, and masses of lignite, sometimes converted into jet, are found. ~ Exposed in brickyards at Standground, Fletton, and Woodstone, near Peterborough, and at Connington, Luddington, and Great Gidding. (b) Clays with Nucula. Laminated blue shales with compressed Ammonites and Vucula nuda. Dug at Haddon, Holme, south of Peterborough, and at Hyebury to the north-east. (a) Zone of Ammonites calloviensis. * See ‘ Memoirs of Geol. Survey of the United Kingdom—The Jurassic Rocks of Britain,’ vol. v. “The Middle and Upper Oolitic Rocks of England,” by H. B. Woodward (1895) p. 8. + Tom. cit. p. 9. £ ‘Geology of Rutland’ (1875) p. 232. Vili MARINE REPTILES OF THE OXFORD CLAY. There seems to be a little uncertainty as to the conditions under which these beds were deposited. They are usually supposed to have been laid down in fairly deep water, but the presence of remains of land-reptiles, and perhaps the occurrence of large masses of lignite, may indicate that the coast was not far off. Probably they were mud-banks accumulating off the mouth of a large stream. In the Collection at the British Museum (Nat. Hist.) there are remains of the following Vertebrates collected from the series of clays just enumerated :— RepPTILia. Cetiosaurus leedsi, Hulke, sp. Pliosaurus ferox, Sauvage, sp. Omosaurus durobrivensis, Hulke. Simolestes voraz, Andrews. A large Stegosaurian. Peloneustes philarchus, Seeley. Camptosaurus leedsi, Lydekker. Metriorhynchus superciliosus, Deslongchamps. Sarcolestes leedsi, Lydekker. 5 brachyrhynchus, Deslongchamps. Rhamphorhynchus sp. 33 sp. Ophthalmosaurus icenicus, Seeley. Suchodus durobrivensis, Lydekker. DMurenosaurus leedsi, Seeley. Dacosaurus sp. i durobrivensis, Lydekker. Steneosaurus edwardsi, Deslongchamps. i platyclis, Seeley. | ‘ leedsi, Andrews. Picrocleidus beloclis, Seeley, sp. | 3 nasutus, Andrews. . sp. | 3 durobrivensis, Andrews. Tricleidus seeleyi, Andrews. | 3 obtusidens, Andrews. Cryptocleidus oxoniensis, Phillips, sp. | It is possible that the number of species of Crocodiles may be increased when the material is examined later in detail. It is a very remarkable circumstance that no trace of any Chelonian has been found. PISCES. Hybodus obtusus, Agassiz. | Heterostrophus sp. Asteracanthus ornatissimus, Agassiz, var. | Mesturus leedsi, A. 8. Woodward. jlettonensis, A. S. Woodward. Caturus sp. 1. Pachymylus leedsi, A. S. Woodward. | a eespas Brachymylus altidens, A. S. Woodward. Osteorachis leedsi, A. 8. Woodward. Ischyodus egertoni, Buckland, sp. Eurycormus egertoni, Egerton, sp. © heaumonti, Egerton. Hypsocormus leedsi, A. 8. Woodward. Lepidotus leedsi, A. 8S. Woodward. be tenuirostris, A. S. Woodward. | = latifrons, A. 8. Woodward. | Leedsia problematica, A. S. Woodward. | macrochirus, Egerton. Pholidophorus sp. INTRODUCTION. ix Most of the fish-remains are important as exhibiting osteological characters which cannot be seen so satisfactorily in crushed specimens preserved in hard rock. In this volume only the Ichthyosaurs and the Elasmosaurian Plesiosaurs are dealt with, the Pliosaurs and Crocodiles being reserved for the second volume. The Ichthyosaurs are represented by one genus, Ophthalmosaurus, of which only a single species, 0. tcenicus, Seeley, is here recognised, though the variability of the skeleton is so great that some might be inclined to consider several species to be present. It has, however, been found, from the examination of a very large number of more or less nearly complete skeletons, that the different forms pass into one another, so that no line between this and that can be drawn. Many of the apparent differences are due to the different extent to which ossification has proceeded in individuals of various ages, and others arise from the differing conditions of preservation (presence or absence of compression, &c.). Ophthalmosaurus, which is here regarded as congeneric with Baptanodon* from contemporary or approximately contemporary deposits of the United States, seems in many respects to represent the most highly specialised type of Ichthyosaurian as yet known. It first appears in the Oxford Clay, and it is by no means certain that the genus survived even in the period of the Kimmeridge Clay, although possibly the so-called Ichthyosaurus entheciodon, in which the teeth are very small, may be a related form. It is true that a small species of Ichthyosaur from the Cambridge Greensand has also been referred to the same genus under the name Ophthalmosaurus cantabrigiensis by Mr. Lydekker, but very little is known of this animal and the presence of facets for three bones on the distal end of the humerus does not seem sufficient evidence, since this character is not confined to Ophthalmosaurus. A portion of a lower jaw from the Upper Greensand of Warminster has also been referred to this genus on account of the small size of the teeth, but in this case also the evidence seems insufficient. It is certain that the Cretaceous Ichthyosaurs that are at all well known are not related to Ophthalmosaurus, this being shown by the great development of the teeth (in I. campylodon, Carter, from the Gault and later) or by the structure of the paddles (in I. platydactylus, Broili, from the Lower Greensand of Hanover). * For an exhaustive account of the American species, see C. W. Gilmore, ‘ Osteology of Baptanodon,” ‘Memoirs of the Carnegie Museum,’ vol, ii. (1905) p. 77. b xX MARINE REPTILES OF THE OXFORD CLAY. The Ichthyosaurs form a singularly homogeneous group, the earliest known forms being already highly specialised for aquatic life, though some traces of a terrestrial ancestry are retained. The first known member of the order is from the lower beds of the Muschelkalk (Middle ‘Trias) of Germany and Switzerland: this was first described by Quenstedt * under the name Jchthyosaurus atavus, and it has since been discussed in detail by Fraasf and Merriam under the name Mivrosaurus atavus. Other remains of Ichthyosaurs from the Middle and Upper beds of the Muschelkalk have been found in various European localities, but they are mostly fragmentary. From the Upper Trias of Northern Italy (the Bituminous shales of Besano, in Lombardy) excellently-preserved Ichthyosaurian skeletons are known and have been described by Bassani § under the name J. cornalianus. Baur || subsequently pointed out that this species was more primitive than the later forms in several respects, and proposed to place it in a separate genus Mrosaurus. The more important of the characters that seem to point to a terrestrial ancestry are, first, the elongation of the epipodial bones (radius and ulna, tibia and fibula) and their contraction in the middle to form more or less of a shaft; and, second, the differentiation of the teeth, those in the maxillary region being stout and blunt, those in the front of the jaws sharp and conical. In the later forms the epipodials are shortened up and show little or no trace of ever having possessed a shaft, and the teeth are sharp, conical, and numerous, the evolution of the Ichthyosaurs in this last respect, as in some others, showing an interesting case of parallelism with that of the Toothed Whaies. A more detailed account of these Italian Ichthyosaurs has been given by Repossi and by Merriam (op. cit.). Remains of ‘Triassic Ichthyosaurs are by no means confined to Europe: numerous forms have been described from Spitzbergen and the United States, and one * Tr’, A. Quenstedt, ‘ Petrefaktenkunde,’ Ist ed. (1852) p. 129. T E. Fraas, ‘ Die Ichthyosaurier der Siiddeutschen Trias- und Jura-Ablagerungen ’ (1891) p. 37. = J. C. Merriam, “Triassic Ichthyosauria, with Special Reference to the American Forms,” Mem. University of California, vol. i. no. 1 (1908) p. 90. § F. Bassani, “Sul fossili e sull’ eta degli Schisti bituminosi triasici di Besano,” Atti Soe. Ital. Sci. Nat. vol. 29 (1886) p. 20. || G. Baur, ‘* Ueber den Ursprung der Extremitiiten der Ichthyopterygia,” Berichte ueber der XX. Versam. des Oberrh. geol. Ver. vol. xx. (1887). 4] F. Repossi, “ Il Mixosauro degli strati triasici di Besano in Lombardia,” Atti Soc. Ital. Sci. yol. 41 (1902) p. 361, INTRODUCTION. xi species * probably of this age is known from New Zealand, showing that very early in its history the group had become cosmopolitan. ‘T'riassie Ichthyosaurs from Spitz- bergen were first described by Hulke f in 1875 under the names Ichthyosaurus nordenskioldi and I. polaris; these species have since been referred by Dames { to Mixosaurus and by Yakolew § and Merriam || to the American genera Cymbospondylus and Shastasaurus. Ina paper lately published Wiman § describes a quantity of new material, and finds that while J. nordenskioldi is referable to Mixosaurus, I. polaris is to be placed in a distinct genus, Pessosaurus. At the same time he describes a new genus, Pessopteryx, including several species. In Wiman’s paper (p. 151, fig. 3) there is an interesting restoration of Mixosaurus, showing the form of the tail-fin, which in many of these early forms was supported by elongated neural spines, and in some cases chevrons, while the sharp deflexion of the posterior part of the tail found in later forms was only slightly indicated. The most important series of Triassic Ichthyosaurs is from the United States, where remains referred to several genera have been described in detail by Merriam **. These are from the Middle Trias of Nevada and the Upper Trias of California. The principal genera are Cymbospondylus, Toretocnemus, Merriamia ft, Delphinosaurus, Shastasaurus, several species of some of these genera being known. Merriam has elaborately tabulated the characters distinguishing these ‘Triassic Ichthyosaurs from the recent types. Some of the more important differences between the ‘Triassic forms and Ophthalmosaurus are shown in the following table :— * J. Hector, ‘ On the Fossil Reptilia of New Zealand,” Trans. New Zealand Inst. vol. 6 (1874) p. 355. 7 J. W. Hulke, “ Memorandum on some Fossil Vertebrate Remains collected by the Swedish Expeditions to Spitzbergen in 1864 and 1868,” Bihang k. Svensk. Vet.-Akad. Handl. vol. i. (1873) no. 9. = W. Dames, “ Die Ichthyopterygier der Triasformation,” Sitzb. Akad. Wiss. Berlin, 1895, p. 1045. § N. Yakolew, “‘Neue Funde von Trias-Sauriern auf Spitzbergen,” Verh. Russ.-Kais. Min. Gesell. St. Petersb. ser. 2, vol. 40 (1902) p. 179. || J. C. Merriam, “ Triassic Ichthyosauria,’” Mem. Uniy. California, vol. i. no. 1 (1908). @ C. Wiman, “ Ichthyosaurier aus der Trias Spitzbergens,” Bull. Geol. Inst. Upsala, vol. x. (1910) p. 124, ** J.C. Merriam, “ Triassic Ichthyosauria, with Special Reference to the American Forms,” Mem. Uniy. California, vol. i. no. 1 (1908). This memoir contains a very exhaustive account up to the date of its publi- cation of all the known Triassic Ichthyosauria. ‘The only important paper on the subject published since, is that by Wiman referred to above. t+ The name Merriamia was substituted for the preoccupied Leptocheirus by Boulenger (Proc. Zool. Soc. vol. i. (1904) p. 425), On page 3 of the present yolume Leptochetrus is employed, the correction not haying been made when this part was printed. 62 Xl MARINE REPTILES OF THE OXFORD CLAY. Triassic Ichthyosauria. 1. The orbits are relatively small and the tem- poral bar behind them broad. . The maxilla relatively large and the pre- bo maxilla correspondingly smaller. 3. Teeth set in distinct sockets and those in the posterior part of the jaws often differing in form from those in front. 4, The anterior cervical vertebrae separate from one another. 5. The neural spines thick and sometimes circular in section. 6. The zygapophyses of opposite sides separate from one another. 7. The terminal portion of the caudal series of vertebre only slightly bent down, the caudal fin being comparatively small and in some cases supported by the elongated neural spines (and sometimes chevrons). on Hind limbs larger in proportion to the fore limbs than in the later forms, the reduction of the hind limb, however, already making considerable progress in some species, e.g. Mixosaurus nordenskioldi, 9. The epipodial bones elongated, with traces of a shaft. 10. The pelvic bones heavy, the ischium and pubis expanded and never fused with one another. Ophthalmosaurus. 1. The orbits very large and the temporal bar behind them greatly reduced in width. bo . The maxilla small and edentulous and the premaxilla relatively very large. 3. The teeth, when present, small and loosely fixed in a continuous groove, the anterior and posterior teeth of the same form (as in Jurassic Ichthyosaurs which possess maxillary teeth). 4, The two anterior cervical centra (atlas and axis) fused with one another. 5. The neural spines broad and strongly com- pressed laterally. , 6. The zygapophyses of opposite sides, in most of the vertebrae, in the same plane and united in the middle line. 7. The terminal portion of the caudal region of the vertebral column sharply bent down, and, notwithstanding its large size, the caudal fin not supported by the neural spines or chevrons, which are much reduced, nN Fore limbs much larger than the hind limbs. 9. The epipodial bones shortened, with no trace of a shaft. 10. Pelvic bones small, the ischium and pubis fused with one another at both ends. In the Jurassic Ichthyosaurs other than Ophthalmosaurus, the differences from the ‘Triassic types above enumerated are, of course, nearly equally well marked; but those numbered 1, 2, 8, and 10 are especially well illustrated by the Oxford Clay type. As to the origin of Ophthalmosaurus there is no certainty, but probably it was derived from one of the “latipinnate” group of Ichthyosaurs. There is nothing in the structure of the skull that is opposed to this suggestion, and the arrangement of the bones in the paddles seems rather to support it. It is unfortunate that no specimens of the paddles have been collected with the bones in an undisturbed INTRODUCTION. xill condition, but so far as the fore paddle is concerned it is believed that the specimen figured on PI. II. fig. 6 represents as nearly as possible the actual arrangement of the bones, every piece having been carefully numbered and a sketch-plan of their arrangement having been made before their removal from the matrix. The circumstance that the ossicles do not fit together in a close pavement, as in the paddles of Ichthyosaurus, but were surrounded by a considerable amount of cartilage, adds to the uncertainty as to the precise arrangement. It can be seen that the intermedium supported two digits, as in the typical latipinnate forms (e. g., Jchthyosaurus communis, I. intermedius); but the paddle differs from those of the earlier forms owing to the fact that its width has been still further increased by the great enlargement of the pisiform, which has acquired an articulation with the distal end of the humerus, in some cases almost as large as that possessed by the radius; the row of ossicles supported by the pisiform become enlarged and form a well-developed digit; a preaxial row of small sesamoid ossicles may also be present. There are traces of the widening of the paddle by the increased size of the pisiform and its digit even in the Triassic Mirosaurus, and probably some early Jurassic descendant of that genus is the ancestor of Ophthalmosaurus, though at present no species is known to which that position can be definitely assigned. ‘There seems to be no reason for the suggestion that Ophthalmosaurus is descended from Shastasaurus, for in that Triassic genus the digits have already undergone much reduction, there being probably, according to Merriam, only two large digits and one reduced digit in the manus. In the later forms of Ichthyosaurus the broadening of the fore paddle which occurs in Ophthalmosaurus may be effected in other ways; thus in Jchthyosaurus extremus, described by Boulenger * from an unknown locality and horizon, but now known to be almost certainly of Kimmeridgian age, the intermedium is thrust between the radius and ulna, and articulates with the humerus by a well-marked facet, so that that bone comes to resemble closely the humerus of Ophthalmosaurus, and if found isolated might be mistaken for it. The width of the paddle in this case is also added to by the presence of a row of sesamoid ossicles on both the preaxial and postaxial borders. Another method of widening is found in Jchthyosaurus platydactylus, described by Broili¢ from the Cretaceous (Aptian) of Hanover; in this species, although it is * Proc. Zool. Soc. vol. i. (1904) p. 424. _ T ‘Palewontographica,’ vol. 54 (1907-8) p. 139, pls. xii. & xiii. XIV MARINE REPTILES OF THE OXFORD CLAY. clearly a member of the longipinnate group, the fore paddle attains great width through the addition of at least two rows of supplementary ossicles on the radial side and one on the ulnar side. Ophthalmosaurus, with its powerful tail-fin, pointed head, and porpoise-like body, must have been a very swift and powerful swimmer, even for an Ichthyosaur, and probably lived in the open sea like most of the Toothed Whales of to-day ; like them, too, it was no doubt capable of diving and swimming at considerable depths, the structure of the auditory apparatus, in the opinion of Dollo*, being specially adapted for use under great pressures such as the animal would be subjected to at some distance beneath the surface. Although in its mode of life Ophthalmosaurus probably did not differ greatly from other members of the order, the reduction of the dentition indicates that its food probably differed from theirs, though of its nature nothing is known. All the Plesiosaurs described in the present yolume are members of the Family Elasmosauride, characterised especially by the structure of the shoulder-girdle, in which, in the adult, the scapule meet ina median symphysis, which is continuous posteriorly with the symphysis of the coracoids. The ingrowth of the scapule towards the middle line takes place beneath the clavicular arch, which thus comes to lie on the visceral surface of the ventral rami of the scapule, which usurp its functions, The consequence of this is, that the clavieles and interclavicles undergo reduction in varying ways. In some genera all the elements of the clavicular arch persist in a reduced form, in others the clavicles or interclavicle may dwindle away to mere vestiges. These varied conditions of the clavicular arch supply some of the chief characters employed in defining the different genera. If Professor Seeley’s f restoration of the shoulder-girdle of Eretmosaurus rugosus be correct, it would appear that the arrangement of the coracoids and scapule found in the Elasmosauride had already come into existence in the period of the Lower Lias; and, at any rate, it seems certain that it had done so in the Upper Lias, for Mr. D. M. 8. Watson { has lately described from beds of that age at Whitby, a shoulder-girdle of Ples‘osaurus homalo- spondylus (referred by him to a new genus Microcleidus), in which the form and * L. Dollo, “ L’audition chez les Ichthyosauriens,” Bull. Soc. Belge Géol. ete. vol. xxi. (1907) p. 157. T Quart. Journ. Geol. Soe. vol. xxx. (1874) p. 445. + Mem. & Proc. Manchester Lit. & Phil. Soc. yol. liv. (1909-10) no. 4, p. 4. INTRODUCTION. XV arrangement of all the elements are almost exactly as in Cryptocleidus. It should, however, be noted that neither Hretmosaurus rugosus nor Microcleidus homalospondylus fall within the limits of the family Elasmosauride, because their cervical ribs are double-headed, a condition not found in the true Elasmosaurs. The genus Elasmosaurus itself was founded by Cope * for the reception of several species of Plesiosaurians from various Cretaceous deposits in Kansas and Nebraska. The type species was described by Cope from an imperfect skeleton, consisting of the greater part of the vertebral column and the imperfect limb-girdles, the pectoral girdle showing the peculiar structure which has been regarded as of sufficient importance to justify the establishment of a distinct family for the reception of the genera in which it occurs. One peculiarity of H/asimosaurus proper is the enormous length of the neck, which consists of no less than 76 vertebre ; in Murenosaurus, the English genus in which the neck is longest, there are only 44 cervical vertebre. Of the numerous other genera of American Plesiosaurs many are very imperfectly known: some almost certainly belong to this family (e. g., Cimoliosaurus, in which the shoulder-girdle is not yet known). In the genera such as Dolichorhyncops aud Brachauchenius, which have been described in detail by Williston 7, the shoulder-girdle is not Elasmosaurian; these animals seem to approach rather the liosaurian type, though they differ from the typical Pliosauride in possessing single-headed cervical ribs, aud in some respects the structure of the skull seems to be very different. In their habits the Plesiosaurs probably differed widely from the Ichthyosaurs. In the first place, their mode of swimming was quite different, propulsion through the water being effected entirely, or almost entirely, by the oar-like paddles, the tail being short and, so far as is known, possessing no fin, or at most a very small one. This manner of swimming and the great length of the neck are characters preventing the supposition that these animals moved at a great pace beneath the water, and it is much more likely that they lived mainly at the surface and at no great distance from the shore. The shorter-necked Pliosaurs are more * Proce. Acad. Nat. Sci. Philadelphia, 1868, p. 68. + “North-American Plesiosaurs, Pt. I.,” Field Columbian Museum—Geology, vol. ii. no. 1 (1903); “North- American Plesiosaurs: Elasmosaurus, Cimoliosaurus, and Polycotylus,” Amer. Journ. Sei. [4] vol. xxi. (1906) p. 221; “ The Skull of Brachauchenius, &c.,” Proc. United States Nat. Mus. vol. xxxii, (1907) p. 477; ‘* North-American Plesiosaurs : Lrinacromerum,” Journ. Geol. vol. xvi. (1908) p. 715. XV1 MARINE REPTILES OF THE OXFORD CLAY. adapted for a pelagic life, but still to a much smaller degree than the whale-like Ophthalmosaurus. If these animals, as we suppose, lived near the shore, the conditions of life would be much more various than in the case of a truly pelagic animal, and this would probably account for the much greater variety of form found among them than among the IJchthyosaurs. One of the most remarkable circumstances about these Oxford Clay reptiles is the occurrence in a limited area of so large a number of closely related species and genera, in the case both of the Plesiosaurs and of the Crocodiles. This can be reasonably accounted for by supposing that the conditions under which the different forms lived presented considerable variety, some, for instance, living in shallow, some in deeper water, some perhaps in swamps, and some in rivers or river-estuaries. For the same reasons, although the Ichthyosaurs Ophthalmosaurus and Baptanodon are generically identical, it by no means follows that the American Plesiosaurs contemporary with them will, when better known, be found to be closely similar to the English species. Although in the Cretaceous period the Plesiosauria had spread over the whole world, being known not only from Europe and North America but also from Asia, South America, South Africa (a species discovered lately), Australia, and New Zealand, it is not certain how far the Elasmosauride spread, for in most cases too little is known about the skeleton of these foreign species to make it possible to determine to what family they belong. The Plesiosaurs were no doubt predaceous, their long sharp teeth being well adapted for the prehension of living prey, which would probably be swallowed whole. ‘The occurrence of numerous stones in the stomach of these animals, first observed by Mr. Thomas Codrington * in a Plesiosaur from the Upper Greensand of Wiltshire, and since noticed in the case of various English and American species, may indicate that the food was broken up in a muscular stomach by the aid of these stones, much as in the gizzard of a bird. No specimen of an Elasmosaur in which the stomach-stones are preserved has been collected from the Oxford Clay, but in the case of a Pliosaur, Peloneustes, Mr. Leeds has obtained a hard mass lying within the ribs, containing many stones of various sizes, from * ¢ Wiltshire Archeological Magazine,’ vol. ix. (1868) p. 170. INTRODUCTION. XVii that of a small hen’s egg downwards, and no doubt representing the fossilized contents of the stomach. The stones are of various kinds, including quartz, sandstone, and gneiss ; for the most part they are rather angular with the angles somewhat rounded off. The mass in which the stones are embedded consists mainly of angular grains of quartz-sand of various sizes, and mingled with these are numerous hooks from the arms of Cuttle-fishes, with black masses which show the characteristic structure of portions of the ink-bags of the same creatures. It is notable that the stomach does not seem to contain any of the hard ‘ guards” of the Belemnite shell, so that probably the animal either bit off the soft anterior portion of its prey or perhaps, as Mr. Crick has suggested, fed on some such form as Geoteuthis, in which the hard parts were not present. Although no doubt both the long-necked Elasmosaurs and the short-necked Pliosaurs could catch their prey on the surface, the former probably fed largely on animals living at the bottom, reaching down with their long necks much as do swans. At least, this manner of feeding would account for the tendency to increase the length of the neck in this group, for such increase would be of considerable advantage to the animals in widening their radius of action in the search for food. The longest-necked Plesiosaur at present known is Elasmosaurus itself, from the Cretaceous of Kansas: in this reptile, as already noted, there were no less than 76 cervical vertebre, the total length of the neck being about twenty-three feet, while the body was only about nine feet long. Some such explanation as that suggested above seems necessary to account for the otherwise apparently disproportionate development of this part of the body. ‘The comparative lack of flexibility of the neck, especially of the posterior portion, in some of these reptiles would not be any disadvantage, because the whole body would probably be tilted up much as it is in the case of birds feeding in a similar way. Any discussion as to the relationship of the Sauropterygia to the other reptiles is deferred till the Pliosaurs have been described. CHARLES W. ANDREWS. Department of Geology, July 1910. SYSTEMATIC SOCCOS Order ICHTHYOPTERYGIA . Family OPHTHALMOSAURID® . Genus Ophthalmosaurus . +5 icenicus Order SAUROPTERYGIA Suborder PLESIOSAURIA . Family ExasMosaURID& Genus Murenosaurus . < leedsi . 4 durobrivensis 35 platyclis . Genus Picrocleidus . ra beloclis . sp. Genus Tricleidus a seeleyi . Genus Cryptocleidus . 6 oxoniensis INDEX. 164 164 Fen ONS Pe Ca: Tue photograph represents the nearly complete skeleton of Cryptscleidus oxoniensis (R. 2860), as mounted in the Gallery of Fossil Reptiles. All the bones belong to a single individual and, with the exception of two or three ribs and neural spines, no restoration has been made on the side shown. ‘The total length of the skeleton is ileteet. EIST OF TELUSERATIONS IN THE, TEXT. Oe Fig, Page 1. Ophthalmosaurus : basioccipital and basisphenoid 3 2. 55 exoccipital andi supraoccipitall (9c susys wy sh te sy, 9) =) 7 3. pro-otic, opisthotic, and stapes . . Ga do pea wo, Ale 4, As reconstruction of the posterior region of the sea Cae ate en ae ay ee Ds = basis phenoidy (sade ere ttt cede es acter 1h ape eS 6. Ophthalmosaurus and Ichthyosaurus: basisphenoid . . . ..... =... .. OS4 7. Ophthalmosaurus: basisphenoid and parasphenoid . . . . . . . . es... . 1 8. ee UENO EMClTNoRoHOWeM ty =e 15 8 6 G8 6 o ne eo o go o 0 aff a: s quadmaters "<7 fen. SAUER ee retort esc eee Gea eg vera een 10. 5 quadrato-jugal and ee Epa a Ie OTS bar one pyr Al) 1 * lachrymal 21 Ie e nasal . 22 3s ss THA KAllayeles hEhe sey say Goh eee Vows Rika Tee, seus ph aes) Saree ie Meera ao 14. ap parietal soe a. a Sip eel oate Ona eo) fob co -fo co ae ee ily 33 part of upper region of skull 26 16. - postfrontal . 27 17. 9 pterygoid BE. weaver Po ac aic’ Ccummoresoune.. col, (cr Qu an, eee) 18. f palatines(#)s ce a) ited ty ca yey ice an UR eee rom cp ere ne 29 19: = VOICI: | slueeus So oe ale canes Rea ad eon lee Now tem ate a eee () 20. ss posterior portion of right ramus of ante BU fal oyna Se ee cee tae | Oe, 21. suran cular 2) sgn aid cod ace RE leet UP ee eae al ieee OO 22. i MOK ERE A Aad. GB oo Mr a Moen ace oi Roa en ree dtl Oe 23% restoration of skull and aut Seabee tn ah seats basins. ice A Oo. 24, Centra ob axisrand atlas) vertebiccey mwas aie umie) a mele nrkie ne nice rein om ON 25. , Centra‘ofanterloricervicall vertebrcciecmneaieeh ete a) et eee noo 26. 5 dorsallanducaudaliventebreeice sp eres Si ei eet ey eee ewe) AO) 27. 5 caudal’ vertebrac. 54 s..£.:\\e0ls Peet Plt ee ts, ura PA kee ee oe pal 28. a caudalevertebren ai) cat Aan : Pica hn Ta ace) eet TRE? 29: m left half of neural arch of anterior en ver Anes Han RO eres brs aaa AMD Z. neural arch of a dorsal vertebra . . . .. : SD oe tat ee <4: al. xs sections of vertebral centra and the upper end of a Gere TID seh a ee tO 32. a GONACOLM ct neck tin, “oy nak tne MaRS Pe EL haa gee et cs 8 toe cae ALY 33. Scapulaice anterior dorsal vertebra . 54. - middle dorsal vertebra . 55. 3 posterior dorsal vertebra 56. sacral vertebra and rib . Die middle caudal vertebra . 58. posterior caudal vertebrae 59. ss caudal vertebree and chevrons . 60. ventral and dorsal ribs 61, PP eab curtis and Nothosaurus : shoulder-girdle 62. Muranosaurus : shoulder-girdle 63. a fore and hind paddles 64. Nothosaurus : pelvis 65. Murenosaurus : pelvis es 66. é restoration of eee : 67. e durobrivensis: shoulder-girdle . 68. . platyclis : shoulder-girdle 69. oS 5 humerus and femur 70. Picrocleidus : shoulder-girdle Tale * sacral vertebre and ribs 72. Tricleidus: exoccipital-opisthotic . 73. ed basioccipital, basisphenoid, and parnephenora 74, - basis cranii and pterygoids 15. en left squamosal and quadrate 76. shoulder-girdle . Fini lett fore paddle Page Fig S: 78. Cryptocleidus : 79. 80. 81. 82. 83. 84. LIST OF ILLUSTRATIONS IN THE TEXT. atlas and axis ; anterior cervical vertebree posterior cervical vertebra . middle dorsal vertebra posterior dorsal vertebra. sacral vertebra and ribs . anterior caudal vertebra. posterior caudal vertebrze plastron of ventral ribs . shoulder-girdle . shoulder-girdle . immature shoulder-girdles . portions of fore paddles . portion of fore and hind paddles . immature pelvis hind paddle . restoration of skeleton Xxili Page 169 170 171 171 172 173 lhe 174 175 176 179 180 182 184 186 187 188 : be ; A fj 5) ae a ’ Vesela aa ar a de ei es eee A DESCRIPTIVE CATALOGUE OF THE M Awe NER Ee Pe Dalla. S OF THE OXFORD CLAY. Order ICHTHYOPTERYGIA. CarNivorous marine reptiles with a cetacean-like body and no visible neck ; a dorsal fn without skeletal supports, and a large vertical caudal fin, of which the lower lobe is strengthened by the downwardly-turned end of the vertebral column. A plastron of ventral ribs present, but no dermal armour. Skull large, with a more cr less elongated rostrum composed mainly of the pre- maxilla. External nares a little in front of the very large orbits, in which there is always a ring of ossified sclerotic plates. A large parietal foramen. Supratemporal fossa large, post-temporal fossa small, and lateral temporal fossa absent; a vacuity enclosed between the quadrate and quadrato-jugal. Postorbital and postfrontal bones distinct ; lachrymal and prefrontal also distinct. Quadrate immovable, being closely united below with the pterygoid, above with the squamosal and supratemporal, which remain separate. In the otic region, pro-otic and opisthotic elements distinct; stapes (where known) short and stout, extending from basioccipital to quadrate. In the palate, pterygoids extending forwards to meet the vomers and excluding the palatines from the middle line; an epipterygoid (columella cranii) present; a large parasphenoid extending forwards between the pterygoids; apparently no trausverse bone. ‘Teeth simply conical, usually with a vertical Labyrinthodont-like folding of their walls; restricted to the edge of the jaws; heterodont and set in distinct sockets in some early genera (e. g. Mixosaurus) ; homodont and set in continuous dental grooves in the typical and later genera. 2 MARINE REPTILES OF THE OXFORD CLAY. Vertebral centra amphiccelous and usually very short. A varying number of the anterior ribs with a double articulation with the vertebra. No sacrum. In the shoulder-girdle, distinct clavicles and an interclavicle, which is usually T-shaped (triangular in Leptocheirus, Merriam*). Pelvis, at least in the later forms, much reduced, and the ilium not united with the vertebral column. Limbs modified to form paddles, and the fore limb, as a rule, considerably the larger (not in the Triassic genus Toretocnemus, Merriam); digits varying in number from three to six and consisting of numerous phalanges. Range from Trias to Lower Chalk. Family OPHTHALMOSAURID. Specialised Ichthyosaurs in which the orbit is extremely large and the dentition reduced in the adult to a number of small teeth, which are loosely set in the anterior half of the jaws only. The humerus articulates distally with three bones, and the fore paddle is much larger than the posterior, which is greatly reduced; the pubes and ischium are fused together. Middle and Upper Jurassic of Europe and North America, with doubtful repre- sentatives in the Upper Greensand of England. Genus OPHTHALMOSAURUS, Seeley. [ Quart. Journ. Geol. Soc. vol. xxx. (1874) p. 699.] 1879. Sauranodon, O. C. Marsh, Amer. Journ. Sci. [3] vol. xvii. p. 85. (Name previously employed by Jourdan for a Rhynchocephalian f.) 1880. Baptanodon, O. C. Marsh, Amer. Journ. Sci. [3] vol. xix. p. 491. 1902. Microdontosaurus, C. W. Gilmore, Science, n. s. vol. xvi. p. 914. Orbit very large ; teeth small, confined to the anterior half of the jaws, at least in the adult. Clavicles uniting in a complex suture (sometimes fused) and embracing the anterior bar of the interclavicle; coracoid normally without posterior notch. Humerus with strong trochanteric ridge and articulating distally with three elements. Pelvis greatly reduced ; ischium and pubis normally fused with one another at the ends and enclosing between them a small foramen. Hind limb very small, femur articulating distally with two elements only. Middle and Upper Jurassic. This genus was established (loc. cit. supra) in 1874 by Prof. H. G. Seeley for the * «* New Ichthyosauria from the Upper Triassic of California,” Bull. Geol. Dept. Univ. California, vol. iii. (1903) p. 253. + Tom. cit. p. 259. + Quoted by Gervais in Comptes Rendus Ac. Sci. Paris, vol. Ixxili. (1871) p. 605, from Jourdan’s MS. OPHTHALMOSAURUS. 3 reception of the species O. icenicus. The type specimen was a shoulder-girdle from the Oxford Clay of Peterborough: with this there were also associated other parts of the skeleton of the same individual, including portions of the skull, mandible, numerous vertebre, portions of ribs and neural arches, and some paddle-bones (see Catalogue, p. 63, R. 2133). The feature regarded as specially characterising the genus was the union of the clavicles by suture in the middle line so as closely to embrace the anterior bar of the interclavicle. In the same paper a fore limb of another individual was described and its chief peculiarity, viz. the articulation of the distal end of the humerus with three elements, pointed ont. It is unfortunate that the type shoulder-girdle is greatly diseased and deformed, the right coracoid being an almost shapeless mass of bone, while the left has a deep posterior notch which is entirely wanting in all normal specimens. ‘The structure of the shoulder-girdle was further discussed by Prof. Seeley in his paper “ Further Observations on the Shoulder-Girdle and Clavicular Arch in the Ichthyosauria and Sauropterygia” *. Additional information concerning the genus has been given by Lydekker +, A. 8. Woodward {, and Bauer §. The last-mentioned writer has described the occipital and otic regions of the skull from material in the collection now under discussion, but, as will be shown below, his account of the arrangement of the otic elements is not quite accurate. The Ichthyosaur from the Upper Jurassic beds of the United States, originally described by Marsh under the name Sawranodon ||, afterwards emended to Baptanodon 4, has lately been the subject of several important papers by Knight **, and especially by Gilmore. Knight describes an imperfect skeleton and gives a number of reasons for regarding Baptanodon as distinct from Ophthalmosaurus, with which several writers have considered it as identical. Gilmore first recorded the occurrence of teeth in Baptanodon 7}, but in his first note he referred the specimen in which teeth were observed to a new genus Microdontosaurus, a name subsequently withdrawn. He has lately published the most detailed account {{ of the skeleton of this form that has yet appeared, and while pointing out that many of the differences between Baptanodon and Ophthalmosaurus referred to by Knight, do not really exist, and were partly the consequence of the bad state of preservation of the specimens * Proc. Roy. Soe. vol. liv. (1893) p. 149. + Catal. Foss. Rept. Brit. Mus. pt. ii. (1889) p. 8; also pt. iv. (1890) p. 268. + ‘Vertebrate Paleontology ’ (1898) p. 183. § “ Osteologische Notizen iiber Ichthyosaurier,” Anat. Anzeiger, vol. xviii. (1900) p. 581. || “A new Order of Extinct Reptiles (Sauranodonta) from the Jurassic Formation of the Rocky Mountains,’ Amer. Journ. Sci. [31 vol. xvii. (1879) p. 85. Also “ The Limbs of Sauranodon, with a Notice of a new Species,” loc. cit. [8] vol. xix. (1880) p. 169. | Amer. Journ. Sci. [3] vol. xix. (1880) p. 491. “ Notes on the Genus Baptanodon,” Amer. Journ. Sci. [4] vol. xvi. (1903) p. 76. tt Science, n. s. vol. xvi. (1902) p. 913, and vol. xvii. (1903) p. 750. $F *Osteology of Baptanodon,” Mem. Carnegie Museum, vol. ii. (1905) p. 77; also tom. cit. p. 325, B2 4 MARINE REPTILES OF THE OXFORD CLAY. examined, nevertheless he still considers that the American and English forms are generically different. Apparently the only reasons for this belief worth considering are: (1) that in Baptanodon the clavicles are fused in the middle line, instead of merely uniting in a close suture; (2) in the American forms the anterior cervical vertebre are uniformly biconcave at the ends of their centra, while in the English types the author states that only the middle portion of the centrum is cupped, the concave portion being surrounded by a flattened area; (5) Baptanodon is said to have an additional digit in the fore limb. With regard to these differences, taking them in reverse order, it may be said that the number of digits is by no means certain, and, in fact, Knight's figure of the fore paddle of Baptanodon, apparently the only one known in which the bones are in situ, lends no support to the view that six digits were present; and even if it did so, it is by no means impossible that an extra row of phalanges may not occa- sionally have been present in Ophthalmosaurus. As to the form of the vertebrae, it may be said that in many cases the anterior cervicals of Ophthalmosaurus are biconcave without any broad flattened area round the concavity. Finally, with regard to the fusion of the clavicles in Baptanodon, it would be somewhat remarkable if two bones so closely interlocking as the clavicles of Ophthalmosaurus did not, at least sometimes, fuse in old age, and, as a matter of fact, this seems to have actually happened in some specimens; in any case, the character does not appear to be of generic value. It seems, therefore, that the English and American species may be regarded as belonging to a single genus, which must be called Ophthalmosaurus, that name having the priority, a conclusion already arrived at by E. Fraas * and other writers. The case for this identity is further supported by the fact that the associated invertebrate fauna proves that the beds in which the remains occur were contemporary, and also that there are found in the American deposits remains of a Plesiosaur called by Marsh} Pantosaurus and clearly identical with Murenosaurus of the Oxford Clay of England. If further proof of the identity of the American and English genera is needed, it will be found in comparing the present account of the skeleton of the latter with the excellent and detailed account of the former given by Mr. C. W. Gilmore in the papers referred to above. Skull.—The skull is represented in the collection by a number of more or less nearly complete examples. The description of the bones of the back of the skull is founded mainly on a series of separate and uncrushed bones of a very large individual (R. 2162), while the account of the facial region is taken from a smaller skull (R. 2180), in which nearly all the bones are preserved separate from one another and only slightly crushed. Only a few bones of the skull of the type specimen are preserved (PI. I. figs. 11-15), and to these reference will be made below. * “Weitere Beitrage zur Fauna des Jura von Nordost-Groenland,” Meddelelser om Grgnland, vol. xxix. (1904) pt. i. p. 285. + “The Reptilia of the Baptanodon Beds,” Amer. Journ. Sci. [3] vol. 1. (1895) p. 406. OPHTHALMOSAURUS. 5 The description of the form and relations of the bones of the occipital and auditory regions of the skull is rendered difficult by the circumstance that a considerable amount of cartilage persisted throughout the animal’s life. ‘The consequence of this is, that not only does the form of the several elements vary considerably according to the degree to which ossification has proceeded, but also, owing to the persistence of the cartilage, actual surfaces of contact between the different bones are only found in a Text-fig. 1. Neel. it > ) p id ANY fit. foss. f =f a XS Basioccipital and basisphenoid of Ophthalmosaurus : A, basioccipital from above; B, basioccipital from behind; C, basioccipital and basisphenoid from side. (R. 2162, 2 nat. size.) cond., occipital condyle; ewo.f., facet for exoccipital ; n.c., neural canal; op.f., facet for opisthotic; pas., posterior part of parasphenoid; p.cl., posterior clinoid processes; pit.foss., pituitary fossa; pt.p.. pterygoid processes ; st,f., facet for stapes ; v.c.p., lower cylindrical processes of basisphenoid. few cases, and usually the separate elements have been displaced and scattered, so that their original position is difficult to make out. ‘The separate elements are first described and then a restoration of this part of the skull is attempted. The basioccipital (P1.1. figs. 13, 14; text-figs. 1 & 4) is a short and very massive bone. It forms the whole of the occipital condyle (cond.), which is sessile and forms less than 6 MARINE REPTILES OF THE OXFORD CLAY. a hemisphere. The condyle is usually about equally convex in all directions, but sometimes (e. g. in type specimen, PI. I. figs. 15, 14) it may be somewhat pinched in laterally towards its upper end; the outline is neatly circular, but it is flattened above for a short distance (n.c.), where it forms the lower border of the foramen magnum ; its surface is usually marked by a series of slight concentric ridges, and there is near its middle a small pit or dimple, probably marking the original position of the notochord. The upper surface of the bone (text-fig. 1, A)is occupied in the middle line by a smooth, slightly concave surface (n.¢c.), extending from the upper border of the condyle to the anterior edge ; this surface, which is the floor of the neural canal, is narrowed somewhat in the middle by the encroachment of the large roughened concave surfaces for union with the exoccipitals (evo.f.). In front of, anda little to the outer side of these surfaces there is in many specimens a slight prominence terminating in a smooth facet (op.f.), which appears to have supported the anterior portion of the opisthotic. On the sides of the bone in front of the condyle there is a smooth area slightly concave from before backwards, and in front of this a broad roughened surface (s¢,f.) looking outwards and a little downwards, with which the head of the stapes articulates. The ventral surface is also occupied by a smooth area which ends in front in a straight or slightly concave border, along which the bone is in contact with the basisphenoid. The anterior face slopes somewhat backwards and is entirely occupied by a coarsely roughened surface for cartilage, usually divided into two bosses by a slight median groove; it is clear that even in old individuals the basioccipital and basisphenoid were only in contact at most along their ventral edge, and were separated above by a thick wedge of cartilage (text-fig. 1, C). The evoccipitals (evo., text-fig. 2, A & B, also text-fig. 4) are short, stout, columnar bones which form the lower part of the lateral border of the foramen magnum. At their ventral end (oc.f.) they are considerably expanded, their base extending forwards in a long tongue-like process, and in consequence of this their surface for union with the basioccipital is very extensive. Their flattened posterior face seems to have sloped somewhat forwards; near the middle of its outer border it is perforated by a large foramen (XI1’), the inner opening of which lies at about the middle of the inner (cranial) surface. This is concave from above down- wards, and in addition to the large foramen just referred to, there is an oblique slit-like opening (XII) just anterior to it. The anterior border is strongly notched, the notch apparently forming the posterior border of the so-called foramen jugularis (jfor.). Judging from the arrangement of the nerve-exits of this part of the skull of Hatteria, as described by Osawa*, it seems probable that the two foramina (XII, XII’) perforating the exoccipital, transmitted two branches of the XII nerve, of which the posterior is the larger, and that the IX—XI nerves passed out through the jugular foramen, the hinder border of which is formed by the exoccipital as above described. ‘The outer face of the * Archiv f. mikroscop. Anatomie, vol. li. (1897) pp. 494-5. OPHTHALMOSAURUS. exoccipital is short and strongly concave from above downwards—in fact, forming merely a deep groove, in the middle of which is the outer opening of the anterior hypoglossal foramen (XII) above noticed, while anteriorly it passes into the border of the jugular notch. The upper end bears two facets, one roughly triangular and slightly convex for union with the supracccipital ; this surface looks directly upwards. The other surface looks outwards, upwards, and forwards, while above it is continuous with the supraoccipital surface; this facet, in some cases at least, was in contact with a corresponding surface on the upper posterior angle of the opisthotic (g. v.), but in other specimens the two elements were probably separated by a pad of cartilage. Text-fig. 2. Exoecipital and supraoccipital of Ophthalmosaurus: A, right exoccipital from outer side; B, ditto from inner side; C, supraoccipital from side; D, ditto from behind. (R. 2162, 2 nat. size.) a.v.c.., depression for anterior vertical semicircular canal; boc.f., facet for basioccipital; exo,f., facet for exoccipital ; for., foramen of supraoccipital; j,for., jugular foramen; op.f., facet for the opisthotic ; p- process of supraoccipital projecting into foramen magnum; pa.f., facet for the parietal; pro.f., facet for the prootic; p.v.c., depression for the posterior vertical semicircular canal; soc,f., facet for supraoccipital ; XII, XII’, foramina for the exit of the hypoglossal nerve. The supraoccipital (text-fig. 2, C & D, also text-fig. 4) is a large M-shaped bone forming the upper portion of the boundary of the foramen magnum, which is greatly narrowed in this region; the posterior surface of the bone, as a whole, is gently convex from side to side. The form of the edge bordering on the foramen magnum differs in different examples; in some it is simply fM-shaped, in others (e. g. that figured in text-fig. 2, D) there is a median process projecting down from the middle of the upper § MARINE REPTILES OF THE OXFORD CLAY. edge. ‘The opening is, moreover, usually more or less constricted by a pair of blunt processes (p.) situated near the lower end of the opening, and it seems most likely that the actual neural canal was only that portion of the opening below these processes. The lower ends of the arch are occupied by gently concave triangular surfaces (ez0.f.) for union with the exoccipitals. At the upper outer angle of the occipital surface there is a funnel-shaped depression, at the bottom of which is a large foramen (for.), which perforates the bone, passing into the cranial cavity by a large smoothly rounded aperture. The function of this opening, which does not seem to have been observed elsewhere, is doubtful ; possibly it transmitted a large blood-vessel. From its position it seems not unlikely that this opening may mark the line of separation between a primitively separate epiotic and the true supraoccipital; for, although Baur * has stated that no trace of a separate epiotic element has ever been observed in Reptilia, this seems to apply only to skulls in a comparatively advanced state of ossification, for a separate eplotic has been figured by Parker + in young embryos of several types (e. g. Tropidonotusand Lacerta). Another possible explanation of this opening is, that it may have given passage to a part of the enlarged upper end of the ductus endolympha- ticus of the ear, such as seems to occur in some Geckoes, in which, according to Wiedersheim {, a portion of the enlarged saccus endolymphaticus lies on the outside of the skull on and among the muscles of the neck, this external portion communicating with that inside the cranium bya duct passing through a foramen between the parietal and auditory capsule (or between the parietal and supraoccipital), which, it is suggested, inay be equivalent to the opening here noticed. Wiedersheim believes that the object of this enlarged saccus, which is more or less full of fine crystals, is to increase the sensitiveness to sound-vibrations §. If the conjecture that some such structure existed in the Ichthyosaurs is correct, it may be supposed to have compensated for the loss of sensitiveness to sound-vibration that must have resulted from the peculiar modification of the stapes, both as to its form, size, and relations to the surrounding bones. The lateral (epiotic) region of the supraoccipital projects forwards at right angles to the occipital portion of the bone; it is irregularly triangular in outline and bears on its outer surface a smoothly rounded triradiate depression, which marks the position of the inner wall of the anterior and posterior vertical semicircular canals (@.v.¢., p.v.¢.). These are bordered by a roughened edge of varying width for cartilage, the posterior edge being the broadest, the anterior of moderate width, the ventral very narrow and even wanting anteriorly and posteriorly where the channels for the canals reach the * Zovl. Anzeig. vol. xi. (1899) pp. 46-47. + Phil. Trans. 1879 (pt. ii.), p. 595, pl. 41. fig. 5; also 1878 (pt. i1.), p. 385, pl. 31. figs. 3, 4. 4 Morphologisches Jahrbuch, vol. i. (1876) pp. 495-534, pls. xvii—xix. § Mr. Boulenger has pointed out to me that the size of this lime-containing sac varies greatly even in different individuals of the same species, and it is often apparently absent; this would, of course, be against regarding this organ as having important anditory functions. OPHTHALMOSAURUS. 9 edge of the bone. Probably there was no actual contact either with the prootic or opisthotic, broad tracts of cartilage having intervened between the several otic elements. The upper border of the supraoccipital is gently convex from side to side and is also curved forwards laterally ; the edge is broad and occupied by a deep roughened groove (paf.), indicating that probably there was a pad of cartilage between this bone and the overlying parietal. The prootic (text-fig. 38, A, B) is a very small bone, oval in outline, with the outer surface gently convex in all directions, the convexity being most marked at one end of the long axis. Its inner face bears a triradiate smooth channel corresponding to the anterior vertical and horizontal semicircular canals; round these is a roughened border for cartilage, varying in width, the widest being at the most convex end of the long axis. The precise position of this bone cannot be made out, as it does not appear to have been in actual contact with either of the other otic elements, but was surrounded by cartilage. Probably its position has been most nearly determined by Bauer * in his figures given in his paper on the otic bones of this genus. The opisthotic (Pl. I. fig. 15; text-fig. 3, EK, F) is a large and solidly constructed bone, the position of which with regard to the surrounding elements is not easy to determine. The following account is founded on an examination of several sets of bones of the occipital region, and may be taken as representing fairly exactly the actual condition of things. It will be seen that the position here ascribed to the opisthotic approaches most nearly to that described by Gilmore and shown in his figure of the skull of Baptanodon fF, but differs in the relations with the exoccipitals; on the other hand, the descriptions given by Bauer in Ophthalmosaurus f, and by E. Fraas §, Cope ||, and Owen ¥ in Jchthyosaurus, differ considerably from the present account. ‘The bone consists of an inner greatly thickened region and an outer short stout process directed upwards and outwards, and terminating in a convex facet (sqg.f.), which fits closely into a corresponding surface on the inner face of the squamosal, in the angle between the process of that bone which joins the parietals and the downwardly-directed portion which embraces the upper end of the quadrate (see text-figs. 4 & 8, A, B). This outwardly directed bar of the bone bears both on its anterior and still more markedly on its posterior face strong ridges and roughened tuberosities (¢.) for union with muscles or tendons. ‘The expanded inner end of the bone is roughly trihedral; ventrally it bears a pair of rather narrow, roughly triangular surfaces (st,f. & st.f’.) (the posterior of * Anat. Anzeig. vol. xviii. (1900) pp. 586-7, figs. 17, 18. + Mem. Carnegie Museum, vol. il. (1905) p. 85, pl. xi. fig. 2. t Anat. Anzeiger, vol. xviil. (1900) p. 581, § ‘ Die Ichthyosaurier der Siiddeutschen Trias- und Jura-Ablagerungen,’ pl. ii. fig. 3. || Proc. Amer. Assoc. vol. xix. (1870) p. 199, fig. 2. 4 «Foss. Rept. Lias. Form.’ pt. tii, (Mon. Pal. Soc. 1881) pl. xxvi. fig. 1. 10 MARINE REPTILES OF THE OXFORD CLAY. which is concave, the anterior convex) for union with corresponding facets on the upper surface of the stapes. In both elements these facets are separated by a deep eroove (g.), so that when the two bones are in apposition a channel is enclosed Text-fig. 3. Prootic, opisthotic, and stapes of Ophthalmosaurus: A, inner face of prootic; B, outer face of prootic ; C, right stapes from above; D, ditto from front; E, right opisthotic from below; F, ditto from above. (A, B, R. 2161; C-F, R. 2162: all 2 nat. size.) boc,f., facet for basioccipital ; ewo.f., facet for exoccipital; g., grooves in stapes and opisthotic, forming a foramen when the two are articulated with one another; /.c., channel for horizontal semicircular canal; op.f., op.f'., the two facets on the stapes for the opisthotic; p.v.c., channel for posterior vertical semicircular canal; q.f., facet for quadrate; 7., ridge on shaft of stapes; sq.f., facet for squamosal; sé.f., st.f'., the two facets of the opisthotic for the stapes; ¢., tuberosities on shaft of opisthotic. between them, running from within outwards and probably transmitting one of the nerves (?the ninth) issuing from the jugular foramen. Above the posterior of the two facets just referred to, there is a large, roughly semicircular surface (boc/.), by which the bone unites with the basioccipital. Above this again, and separated OPHTHALMOSAURUS. 11 from it by a notch, is a pointed process, bearing in some cases a small facet (exo.f.) for articulation with the outer upper facet of the exoccipital ; in cases where the facet was wanting, no doubt the actual junction of the two bones was prevented by the intervention of a pad of cartilage. The junction of the opisthotic with the exoccipital above and the basioccipital below, encloses between it and the former of these two bones an oval vacuity, through which must have run some of the nerves and vessels passing into the jugular foramen which, as above described, notches the front of the exoccipital. The upper and anterior face of the inner end of the opisthotic formed the posterior, outer and lower portion of the auditory capsule, and is deeply impressed by two grooves meeting towards the inner end and marking the position of part of the posterior vertical (p.v.c.) and of the horizontal (/.c.) semicircular canals. Round these the bone forms a roughened border of varying width for cartilage, and probably it had no contact with either the supraoccipital or epiotic elements. The chief peculiarities of this bone (é. g., its extensive union with the basioccipital and stapes) all seem to tend to increasing the rigidity of the squamoso-quadrate complex. The stapes (Pl. I. fig. 12; text-figs. 3, C, D, & 4), asin other Ichthyosauria, has under- gone a most peculiar modification. Instead of being, as in most Reptiles, a slender rod of bone connecting the tympanic dram with the inner ear, it seems to have lost its auditory function and has become a stout bar of bone, lying between the basicccipital and quadrate and helping to add to the rigidity of the latter. The proximal end is greatly enlarged and forms a massive head, the inner face of which is occupied by a surface (boc,f.), slightly convex from above downwards, by which it articulates with the corresponding facet on the basioccipital. Above this surface on the upper side of the head are two obliquely elongated facets (op.f. & op.f’.) separated by a groove ; these, as above described, articulate with the corresponding surfaces of the opisthotic, the groove forming the lower half of the channel (g.) running between the two elements. The posterior face of the head is flattened and a little roughened ; the anterior face is produced ventrally into a blunt triangular prominence, from which a ridge (7.) runs obliquely upwards on to the narrow outer process of the bone. This latter terminates in a blunt point, which bears a facet, looking upwards and forwards, for union with the corresponding pit on the inner face of the quadrate. In addition to its inner and outer unions with the basioccipital and quadrate respectively, the lower angle of this bone is wedged into the angle formed by a ventral shelf-like process of the portion of the pterygoid which is united to the inner face of the quadrate (see pterygoid below, also text-fig. 4). ‘This union added still further to the rigidity of the occiput. The arrangement of the various bones described above is figured in text-fig. 4. In this the form of the foramen magnum is well shown, as also is the remarkable manner in which the quadrate is supported by the stapes, opisthotic (indirectly), squamosal, and pterygoid. c2 12 MARINE REPTILES OF THE OXFORD CLAY. The basisphenoid (text-figs. 1, 5, 6, 7) is a stout and very massive bone. Its posterior and most of its upper surfaces are greatly roughened and were obviously thickly covered by cartilage in life. In this region the bone is divided into two prominent convex bosses, separated by a deep median groove running from the middle of the upper anterior edge to the lower posterior border. The posterior faces of these prominences are slightly flattened, or even slightly concave, and were directed towards Text-fig. 4. LAS. Sa \ it Wy h \ Lp are y R aly Sy My Reconstruction of the posterior region of skull of Ophthalmosaurvs from behind. (About 3 nat. size.) art., articular surface of quadrate; boc., basioccipital ; cond., occipital condyle; ewo., exoccipital; for., foramen in supraoccipital ; formay., foramen magnum; op., opisthotic; op.f., facet for opisthotic ; p., process of supraoccipital projecting into foramen magnum; pa.f., facet for parietal; p.b.sq., parietal branch of the squamosal ; p.e.a., postero-external angle of the squamosal ; pt., pterygoid; pt.foss , post- temporal fossa ; q., quadrate ; .b.sy., quadrate branch of the squamosal; soc., supraoccipital; sé., stapes; t.b.sq., temporal branch of squamosal; XII’, foramen for posterior branch of the hypoglossal nerve. the corresponding anterior surfaces cf the basioccipital, though probably separated from them by a thick pad of cartilage. The presence of the deep median groove above noticed may be a trace of the original ossification of this bone from two lateral centres. The anterior border of the upper surface is raised into a pair of blunt processes separated by a slight notch: these are the posterior clinoid processes (p.cl.). Beneath them the anterior face of the bone is at first vertical and then slopes slightly backwards, forming the posterior wall of the very large internal carotid foramen (i.¢f.), OPHTHALMOSAURUS. 13 the anterior wall of which is formed by the upper surface of two prominent bosses, separated by a deep notch and each terminating in a surface for cartilage: these are the processes called by Siebenrock *, in his account of the skull of Hatteria, the lower cylindrical processes (v.c.p.) ; they mark the beginning of the presphenoidal region of the basis crani?, which remained unossified. Beneath them is the posterior end of the parasphenoid (pas.), which is adherent to the ventral face of this bone, as described below. The hollow bounded by the upper surface of the lower cylindrical processes below, and by the posterior clinoid process behind, is the pituitary fossa (pit.fos.), into which, as already noted, the very wide carotid canal opens and passes downwards Text-fig. 5. Basisphenoid of Ophthalmosaurus: A, from below ; B, from the front. (R. 2162, } nat. size.) é.c., foramen for a branch of the carotid artery; i.c.f., internal carotid foramen; pas., parasphenoid ; p.cl., posterior clinoid processes ; pit.foss., pituitary fossa; pt.f., facet for pterygoid; pt.p., pterygoid processes ; v.c.p., lower cylindrical processes of basisphenoid. and backwards, its single ventral aperture being situated about the middle of the ventral face of the bone: this lower opening is usually more or less asymmetrical in form. In some of the Liassic Ichthyosaurs, as pointed out by Cuvier + and more fully by Maggi f, there are two posterior openings, separated, in some cases at least, by the posterior end of the parasphenoid (see text-fig. 6, C); this condition seems to be the most primitive, since in /Hatteria the paired openings are situated on either side of the posterior end of the parasphenoid. In Ophthalmosaurus the condition of the posterior end of the parasphenoid (see below) is very different in different individuals * “ Zur Osteologie des Hatteria-Kopfes,” Sitzungsber. k. Akad. Wiss. Wien., math.-naturw. Cl., vol. cii. (1898) p. 250 ; also translated in Ann. Mag. Nat. Hist. [6] vol. xiii. (1894) p. 297. + * Ossements fossiles,’ vol. v. pt. 2 (1824) p. 460, pl. xxix. figs. 12 & 13. + “Il Canale Craneo-faringeo negli Ittiosauri ete.,” Rendiconti R. Istit. Lombardo, [2] vol. xxxi. (1898) p. 761. 14 MARINE REPTILES OF THE OXFORD CLAY. (see text-figs. 5, A; 6, A, B), but in no case does it divide the opening of the carotid canal. On either side of the pituitary fossa the basisphenoid is produced outwards into the two prominent pterygoid processes (pt.p.), the upper surfaces of which look upwards and forwards; at their base there is a small perforation (0.c.) leading into the carotid canal and no doubt transmitting a small branch of the internal carotid. ‘The outer ends of the processes are truncated by the facets for the pterygoids looking almost directly outwards and sloping downwards from behind forwards (see fig. 1, C). The ventral surface of the pterygoid processes is roughened for union with the inner plate of the pterygoids, which overlapped and closely united with this part of the ventral face of the basisphenoid, extending back even to the posterior portion of the body, as is shown by the presence of roughened surfaces for union. When the basisphenoid is looked at from below (text-fig. 5, A) its posterior border is seen to be nearly a semicircle, while the anterior border is nearly straight, except for the slight projections formed by the ends of the lower cylindrical processes (v.c.p.) Text-fig. 6. Basisphenoid of Ophthalmosaurus and Ichthyosaurus: A and B, basisphenoids of Ophthalmosaurus from below ; C, basisphenoid of Ichéhyosaurus from below. (A, R. 2164; B, R. 2161; C, R. 2063: 2 nat. size.) i.c.f., internal carotid foramina; pas., parasphenoid; pi,f., facets for pterygoids; pt.p., pterygoid processes ; v.c.p., lower cylindrical processes. and by the anterior angles of the pterygoid processes (pt.p.). As already mentioned, the middle of the ventral face is pierced by the internal carotid foramen (7.¢.f.). In front of this the adherent posterior end of the parasphenoid is seen. In different individuals the degree to which the parasphenoid overlaps varies (see text-fig. 6). OPHTHALMOSAURUS. 15 Text-fig. 7. Basisphenoid and parasphenoid of Ophthalmosaurus. The basisphenoid is much erushed from above downwards: A, from above; B, from side. (R. 2180, 2 nat. size.) p.cl., posterior clinoid processes; pit.foss. § ¢.f., pituitary fossa and carotia foramen; prs.g., presphenoidal sroove ; pt.f., facet for pterygoids ; pt.p., pterygoid processes of basisphenoid ; v.c.p., lower cylindrical processes of basisphenoid. 16 MARINE REPTILES OF THE OXFORD CLAY. In one case (fig. 6, A) it extends back round the sides of the carotid foramen (é.c.f.), forming the anterior and part of the lateral border of that opening; in most (text- fig. 5, A) it extends just to the anterior border of the foramen, while in one instance it only extends about halfway to it (text-fig. 6, B). In front of the basisphenoid the parasphenoid (text-fig. 7) extends forwards as a long pointed rostrum, the length of which is about three times that of the basisphenoid in the mid-ventral line. Posteriorly this rostrum is transversely oval in section, but soon becomes triangular, the upper surface being concave from side to side, so that it forms a shallow groove (prs.g.) which received the lower edge of the presphenoidal bar of cartilage, which probably, as in Hatteria, formed the ventral edge of the interorbital septum: this groove occupies about the posterior half of the free portion of the bone. In front of this the rostrum becomes more compressed laterally, and on its sides are long surfaces (pt,f-) slightly concave and ridged, apparently for union with the inner side of the anterior limbs of the pterygoids, between which its anterior half extended. The squamosal (text-fig. 8, A, B.) isa bone of complex shape. It occupies the postero- superior angle of the skull, where it forms a prominent boss (p.e.a.). From this angle there runs forwards a broad plate, the somewhat thickened upper border of which forms the posterior half of the outer border (0.b.) of the supratemporal fossa (s.t,foss.) ; the ventral and anterior edges of this plate are thin and no doubt united with the supratemporal and postfrontal bones, as they are shown to do in Gilmore's figure * of the skull of Baptanodon, and as is the case in the earlier Ichthyosaurs. From the posterior angle again there runs inwards and a little forwards a very stout bar of bone, making an acute angle with that just described and forming the outer part of the posterior border of the supratemporal fossa. At its inner end this process widens out considerably from above downwards and terminates abruptly in a deeply hollowed, somewhat diamond-shaped cavity (pa.f.) for the reception of the outer end of the squamosal process of the parietal. On the posterior face of this process of the squamosal there is a small shelf-like projection, making nearly a right angle with the quadrate region about to be described: in this angle is the facet for the reception of the outer end of the opisthotic (op,f.). Beneath the posterior angle (p.e.a.) there isa broad plate of bone continuous in front and behind with the processes already described: this is the quadrate region of the squamosal (¢.f.); it consists of an outer and an inner plate separated by a deep narrow fossa, into which the upper end of the quadrate is firmly fixed. The inner plate extends down the inner face of the quadrate, to which it is closely adherent, and at its lower end it overlaps the ascending quadrate plate of the pterygoid; by this arrangement the quadrate, apart from its other supports, the stapes and quadrato-jugal, is held rigidly by the squamosal above and the pterygoid below. ‘The above description agrees in the main with that given by Gilmore in his account of the skull of Baptanodon, but it cannot be said * “Osteology of Baptanodon (Marsh),” Mem. Carnegie Museum, vol. ii. (1905) pl. viii. OPHTHALMOSAURUS. 17 Text-fig. 8. Squamosal and postorbital of Ophthalmosaurus: A, left squamosal from above; B, ditto from below (the dotted line represents the restored outline of the quadrate border); C, right postorbital. (A, B, R. 2146; C, R. 2180: 2 nat. size.) o.b., outer branch of squaimosal; op,f., facet for opisthotie; 0.7., orbital rim on postorbital ; pa.f., surface for union with the parietal ; p.c.a., postero-external angle of squamosal ; q,f., surface for union with upper end of quadrate ; s.t.foss., supratemporal fossa. 18 MARINE REPTILES OF THE OXFORD CLAY. that the inferior plate filled all the space between the opisthotic, stapes, and quadrate, nor does its lower end seem to extend so far down as to pass under the upper lateral margin of the stapes. It seems probable that in the skull figured by Gilmore the squamosal and quadrate have been somewhat dislocated from their natural positions with regard to one another. The supratemporal is missing or crushed beyond recognition in most cases, but in Right quadrate of Ophthalmosaurus: A, outer side; B, inner side; C, articular end. (R. 2133: 2 nat. size.) a.b., anterior border; a.i.a., antero-internal angle; art., articular surface for mandible; n., neck of bone ; p.e.a., postero-external angle; pt. surface of union with pterygoid; q.j.f., facet for articulation with quadrato-jugal ; sq., surface for union with the squamosal; st.f., facet for outer end of stapes. (The outlines of the surfaces for the squamosal and pterygoid are shown by dotted lines.) one (R. 2740) it can be seen that it was roughly triangular, overlapping the post- orbital anteriorly, and sending forwards along it a process to meet the postfrontal. The relations with the bones behind and above cannot be made out. The quadrate (P1. 1. fig. 10; text-fig. 9) isa large, broadly sickle- or rather ear-shaped bone, consisting of a comparatively thin upper portion and a greatly thickened articular OPHTHALMOSAURUS. 19 region. The form of the articulation (text-fig. 9, C) varies somewhat, according to the degree to which the ossification of the cartilage has proceeded; in the most fully ossified specimens the surface is strongly convex from before back. In the transverse direction there is a double curve, the inner portion being convex, especially towards the middle of its length, while the outer is concave, especially in front. The prominent postero-external angle of this region forms a well-marked projection (q.j.f.), with the upper end of which the quadrato-jugal articulates by a concave surface (qf, text-fig. 10, A-C). aes, een alleo R. 2181 (Leeds Coll. 66). Imperfect skull and skeleton of a very young individual. In this specimen the anterior portions of the jaws bear a number of small, very closely crowded and rather backwardly directed teeth (PI. I. figs. 7,8): in the anterior part of the right side of the mandible there are ten on a space of 5-5 cm. The pubis and ischium show no more sign of separation than in older individuals, and the same may be said of the atlas and axis vertebrae. The portions of the skull preserved include : basioccipital, K MARINE REPTILES OF THE OXFORD CLAY. stapes (1), opisthotics, squamosal, jugals, lachrymal, nasals, quadrates, pterygoid : also a considerable part of the mandible, including both articular bones, and about forty teeth ; hyoid bones. Of the skeleton there are: atlas and axis and sixty-five other vertebrae, a few neural arches, numerous fragments of ribs, right clavicle, coracoids, scapulee, humeri, radius, ulna, ilia, parts of ischio-pubes, femora. The dimensions (in centimetres) of some of the bones are :— iBasroccipital.:= widthvot condyles G9. 4) .) 4%) Sake nore Q@uadrate:sextremellengthy 9. 2s 28 i ve ee ee ened ee OLS WwidthVatsneciin sr) Memucmeee eucuiel Suen) SleS 3 length of articulation . . . . . . (approx.) 3°5 Articularsbone: Wengthe eisai. 2) ge ec ets eee oe eS Hyoid bone: length Fe geet Sea ee Oytal Man bok comer EP width of middle of shaft . . ...... O7 Atlasiandéaxiss lengthie (Gee | es Ge ye age alc sreatestiwidth) 4% i. 6 ks ee 6 ee 846 Middlesprecaudal:) width :9. 5 . 1. 5 2 & 3 wus « 40 Anterior caudalimlengthiye: a uuaeciue lit sien ciees (cn gelliccl Wilby Micrnc ms ciara lemukoge Cat.ce | MSamere dere oracoidi:" sreatests width) bo oD Scapula: length : 21 width of blade . 5:0 ss distal expansion 14-4 Humerus: length ras 156 width of shaft at narrowest 6-9 » distalend . 12:9 68 MARINE REPTILES OF THE OXFORD CLAY. INsrADey NS sayAdal < koe dt Halo aces op cent e OM a Woweendss 6.) JGHS) widthiok sharbrat marrowesbys yi. | ij Voeiber py) iene toa6 PeECIStAlTED Gc.) « caceuee (mis leewesp are) (area OLS Hschio-pubistwlengbhies ere. ca. «ell cco comets elites Seta wNlOtc R, 2853 (Leeds Coll. 85). Fragments of skull, including pterygoids, sclerotic plates, six centra of vertebrae, both scapule, fragment of coracoid and clavicles, interclavicle, both fore paddles complete (the left is figured in text-fig. 37, and also by A. S. Woodward in ‘Qutlines of Vertebrate Palwontology,’ p. 182, fig. 113, B), ilia (text-fig. 38), both ischio-pubes (text-fig. 39, A—C), both femora and numerous paddle-bones. The dimensions (in centimetres) of some of the bones are :— Scapula: greatest length . width of blade at narrowest bo AN wo He Soko EF Oo GQ », distal expansion eae doy comerc Right humerus: length . . . .. . . . ~ (approx.) width of shaft at narrowest Arte as esp rGistalucnd, te: sg cg) pos wee se elo Right radius: width Eaagolts at lengthy en eemete noe e ks ie eres Se cm( 2D PLOX) mamOLo Miphtmlnast widths ss. eee alts ee ey ee ee (Apprexs) 76,0 Lene Gh Wapam nyo 8 oobiee Wart eta eta tay Meme) reo ees Richtantermediumis width! vasa ees ciakes sel owl oe Me gael as TAO lenotihtes | semen sane isl sno (LD DEOX:) mlOsO Ilium (text-fig. 38): length in straight line. . . . . . . 120 ereatest widthyic. 2 4% Veaus es. glee er ue ee thee aos width otlowernend: @ ews el ae Gut issues tote Ischio-pubis (text-fig. 39, A-C): length. . . . . .. . 159 width of proximalend . . . . 40 distallven divagev Gc, j- secees oe Ors) ” Memumcolenp thy 22 (owe we Us ist as Mase fst eehie) Seem” asp steam clelue width of shaft at narrowest Tes cto oe Ae POE = Gistal:errd, 4 se, Seen. thee bounces Bat eye Oto) Mitiayylencthw vce ace earch son ee fey uc pal owe: Okey ee eng ae widthe sawacse: iso use Gsy wey ep ke. HURTS sel dot Sas) Rem Fibula: length . 2. . . . < 3) SARS ae, SPs 2 awl bits Was? mote cee Met Ss orks ES AAU REM etree em ez) R, 2149 (Leeds Coll. 70). Portions of a skull and mandible, including basisphenoid, stapes, opis- thotics, one pro-otic, squamosal, parietal, both articulars, one surangular, and other fragments ; atlas and axis, thirty other precaudal and twenty-five caudal vertebre ; the clavicles, scapulee, coracoids, left humerus, and twenty-three paddle-bones ; one ilium and both imperfect ischio-pubic bones. This skeleton does not present any very notable features : the vertebre are for the most part crushed ; the clavicles show clearly the surfaces by which they unite with the cross-bar of the interclavicle ; the ischio-pubiec bones are rather shorter and stouter than in most specimens. : OPHTHALMOSAURUS ICENICUS. 69 Some dimensions (in centimetres) of parts of this skeleton are :— =I or or bo Basisphenoid: length . : sreavestpwidthien 0. os) supers width width of occipital condyle Basisphenoid: greatest length . », Width Stapes: greatest length width of inner end . Atlas and axis: length width height Humerus: length (approx.) 1 QO COTM COM Crt 00m C1) heat Om HAI © Oo oO Tho oO 8-2 6: o 72 MARINE REPTILES OF THE OXFORD CLAY. R. 2132 (Leeds Coll. 61). Includes the quadrates (imperfect), some fragments of skull-bones and sclerotic plates, twenty-six precaudal and thirty-one caudal vertebral centra, all well preserved and undistorted, the coracoids, imperfect scapule, humeri, radius, intermedium, pisiform, and twenty-three other paddle-bones, a femur, and numerous fragments of ribs. The length of the humerus is 15 em., that of the femur 11 cm. The width of the narrowest part of the shaft in the former is 7 em., in the latter 3°9 cm. R. 2150 (Leeds Coll. 77). Parts of skull, including basioccipital and imperfect quadrates, numerous loose teeth, seventy-five greatly-crushed vertebree, three caudal neural arches, imperfect coracoid, scapulee, humerus, and twenty-nine paddle-bones. The length of the humerus is 14:4 cm., the width of its distal end 12°2 cm. The length of the scapula is 18-2 em., the width of its distal expansion 10°8 em. Some of the teeth are 1°9 em. long, with a maximum diameter of 1 cm. at about the middle of the root. R, 2153. Fused atlas and axis, twelve other precaudal and three caudal vertebra, eight odd paddle-bones. The dimensions (in centimetres) of some of the bones are :— Atlassand axis: length-oficentra . 4. 4 = 2 « |). «! = wort depthvotcentiras ayes -iisce ete scien wea my. esi mOnl WACO RCInHLe 4 (Oh Go He Gee on a OHO Largest caudal vertebra: Jength of centrum . ... . . 387 depth of centrum. . . .. . . 102 width ofcentrum. .... . . 95 R. 2150a. Atlas and axis, figured in text-fig. 24, A, B. The length of the combined centra is 4-5 em., the greatest height 8°6 cm., the greatest width 8 cm. R. 2152. A basioccipital, atlas and axis (text-fig. 24, E, F), twenty-one other precaudal and fifteen caudal centra. The caudal centra have been strongly compressed longitudinally, so that they are not much more than half their original length, but’ are not otherwise distorted. Also the coracoids, left scapula, part of left humerus, and six paddle-bones. The dimensions (in centimetres) of some of these bones are :— Basioccipital: greatest length . . . . . . . (approx.) 66 ay, WIdth tn jo clack ee eae eo” widthvof condyle “3 uses) seme teat ces One Atlasiandiaxis: lengths i" 3 sw 6 ery eee eaten Si0 WAGED cbt Gibb, aca ie: ns, hed es tee oe ae OO height is est in team ee get eee EOS @oracoidi:\ width” "7. 3 50) 4 = a) ee) (approx), 21-0 Scapulajmleng thie ay vy tcc) 4 ee) ee tree meee lc width:of lower expansion . . =. . = - « » «= 4146 R. 2143 (leeds Coll. 82). The left quadrate, right scapula, the centra of three posterior precaudal and twenty-two anterior caudal vertebree of a large individual. The centrum of vertebra which appears to be the first caudal is 10-7 cm. high, 10-7 cm. wide, 4°2 em. OPHTHALMOSAURUS ICENICUS. 43 long, approximately. The quadrate is massive and very extensively ossified ; its extreme length is 13°5 em., the width of the neck 6°8 em. The scapula is very fully ossified, and the deflected antero-inferior angle forms a definite and prominent process (?acromium). The length is 23 cm., the width of the proximal expansion 14°8 cm. R. 2173 (Leeds Coll. 56). A caudal vertebra, a left humerus with radius, ulna, and seventeen other paddle-bones. The humerus has the prominences on the anterior angles of its distal end strongly developed on both its upper and lower surfaces ; the pisiform facet is very sinall ; its length is 18°5 cm., the width of its distal end 15-4 em. R. 2174. Atlas and axis with twenty-seven other precaudal vertebree and a femur of a young individual. The original line of separation between the fused atlas and axis centra is clearly marked, even on the floor of the neural canal, where in most cases it is early obliterated. The dimensions of the combined atlas and axis centra are: height 5°3 cm., width 6°0 cm., length at neural canal 2°?7 cm. The femur is only 4:7 cm. long, its distal end being 3:2 cm. wide. R. 2163. Bones of the back of the skull of a small individual, including the basioccipital, basi- sphenoid, stapes, one opisthotic, and quadrates. R. 2135 (Leeds Coll. 71). A series of vertebrae, including the atlas, axis, and thirty other precaudal centra, together with twenty-four caudals from various parts of the tail. Also portions of the coracoids, the right humerus, both radii, an intermedium, a pisiform, and fifteen other paddle-bones. Some of these last are very thin and seem to have belonged to the edge of the paddle ; their surfaces also are broken up by grooves into a number of irregular areas with smooth surfaces, which may indicate that they were merely covered with some sort of horny epidermal structure. The atlas and axis, with the four succeeding vertebra, are shown in text-fig. 25. Some dimensions (in centimetres) of these specimens are :— Atvlistandvaxis'lengthvobicentra eam eunh get tee eet mnect widtheoticentrae te cee- seers cet mea mies ae We OT depth of:centran i Garena ise tt ae) sere mo 6 Phird cervicali length: Ww" ahh. hee ease on eon mega Mean. S will Ghiscy Sed oe ese Pa eA oa eke ieamtee ese) 11656 Ge pth woke tac iat tee Me ema Cater taee cet oi! a gine Wourthiceryicallslenothy 2) a pera erat Meneses nensTi s m2:6 Wadthr: HOS ee ane nr ewe OSS depth 70th aie cre eet vege Mice eae tne eersce an OO) Total length of the six anterior vertebra (text-fig. 25) . . . 17:0 iFinmerusisvien g thos to) a.) SU ehewre: aoe aoe eet a, ail! 927, R. 2188. Portion of basioccipital. About eighty-six vertebral centra, mostly crushed and broken, with the exception of the posterior caudals, about fifty-four in number. Also a femur and sixteen paddle-bones. The chief peculiarity about this skeleton is, that the basi- occipital seems to have been formed by two centra, an anterior and a posterior, the L 74 MARINE REPTILES OF THE OXFORD CLAY. posterior being proccelous, the convex hinder surface forming the occipital condyle, the anterior concave face closely resembling the face of a vertebral centrum of ordinary type. R. 2148 (Leeds Coll. 69). Fragments of mandible, three anterior cervical centra, and fifteen caudal centra, including some from the bend of the tail and some of the small terminal bones, the coracoids, and a femur of a young individual. The dimensions (in centimetres) of some of the bones are :— Cervical vertebral centrum: Jength . . ...... =. «20 height yeaauen gi Od Vir Go Sty Fe hae, GA widtht ays fynkate tomas ween) Menem OL Caudal centrum from bend of tail: length (at top) . . . . 16d 3714 (ab bottom) ysek 7) lO) height | 2. ee) eh feo ner ps ec WACEI Wie. Wes Ge) valy) es eee, ep esO : Climotiloirdisdt 3G Go a eo) Ba ee a a a GHD breadth. wf cio be soso ns ee ae, 1620 Womursn lone th tees sitee. io une) es seins ces ain nee ae ms. R. 2137 (Leeds Coll. 63). Eleven posterior precaudal vertebral centra, three neural arches, many ribs, the complete shoulder-girdle (coracoids, scapule, clavicles, interclavicle), one complete ischio-pubic bone and the proximal end of the other. All these bones, which are those of a large individual, are uncrushed and exceptionally well ossified. The shoulder-girdle has been figured by Prof. H. G. Seeley in Proc. Roy. Soe. vol. liv. (1893) p. 151, fig. 1 ; see also text-figs. 82-35. The dimensions (in centimetres) of some of the bones are :— Dorsal vertebral centrum: length. . . .... 2.2.2. «4 eile htretac joa. een tee ton eA. Width. 3.02 Ye al ge Fe EOD INeurallarchi= total height eg see vee pes) gy ea ease UGG VACHE GG ss Glue oo 5 a oe tt!) Coracoid (text-fig. 32): greatest length . . . . . . . . 23:0 yyy “WAGth? Gace ce) (oly ey eer cutee ee ses length of symphysial surface .(approx.) 14:5 depth of symphysial surface .(approx.) 7:1 length of glenoid surface . .(approx.) 10-0 Scapula (text-fig. 33, A & B): extreme length. . . . . . 240 width of blade at narrowest . 4:4 3 expanded lower end . 16:0 length of glenoid surface . . 6:0 width of glenoid surface . . 4:0 Claviele (text-fig. 54): length along outside of curve. . . . 37°5 s2. dnvstraichtzlines. em. a ay to4b greatest width (in middle). . . . .) 57 OPHTHALMOSAURUS ICENIOUS. 75 Tnterclaviele (text-fig. 54): greatest length in middle line . . 13:8 length of cross-bar. . (approx.) 23-0 Ischio=pubisislenothaycwcmtci soit f= ‘ ein ee orf widthietuproximalvendiy. Veen sepenit melon en ec Ord . distalyendigy aay jus, ee age Sere yet ay hed R. e 2147 (Leeds Coll. 68). Centra of fifteen precaudal vertebrae, nine neural arches, and parts of the clavicular arch including the median portions of both clavicles, the right one closely united with the corresponding portion of the interclavicle. Both clavicles seem to be somewhat deformed ; that on the right side may have been fractured and afterwards mended during the animal’s life. R. 2141. Twenty-two precaudal, twenty-nine anterior caudal, and eight posterior caudal centra of a small individual. Some of the posterior caudals are from near the end of the tail and are very small. The dimensions of one are: length 1 em., width 1°25 cm., height 1°3 cm. R. 2157 (Leeds Coll. 29). Centra of twelve precaudal and nineteen caudal vertebrae of an old individual. R. 2139 (Leeds Coll. 73). Centra of eleven vertebree, six neural arches, one scapula, one coracoid. R. 2164. Basioccipital, basisphenoid (text-fig. 6, A), stapes. The parasphenoid extends round the sides of the carotid foramen. The dimensions (in centimetres) of these bones are :— Basioccipital:oreatestilenpthyeie/.) wt 2) vel 0) een Gel proc WAd ble emir ieee tise nee cement eet Og, Basisphenoid: greatest length . . . . .... 2... #70 fo: We DRAG tH ry 0) aC ei at MUNA nae ReaneO:() Stapeslengthe sie pcs esos are asm tare ee re eee ae mn Widthvorinneniends es, “avig ise shee eens mesecee 4-9 R 2134. Right fore paddle. Co-type described and figured by Seeley in Quart. Journ. Geol. Soc. vol. xxx. (1874) p. 703, pl. xlvi. fig. 3. The dimensions (in centimetres) of this specimen are :-— Flumerus\ lengthy <2 ocn” ytd role ont mie ee eee ence wal 726 widthiat proximal ender 20-0 < yee eee ey so: 7, a middlerotishattiy aie ela een veneers yen, SO) “3 distaliendis; (suis. pi enews uemr nc Radiuspalengthtes, pi, ‘intr | vsekeceeg ye ONE re man tan eee 7-6 WIG es intiatice st 24 ba Neal ae See i pace NMS 2 estakon'7 eC) Wintavmlena ths jas. ys. dhe iia, (oon ae ames eke hee ahs en wid bhty 2 es E55.) ate yoo bch ee a a eae (eat r oG 76 MARINE REPTILES OF THE OXFORD CLAY. R. 2175. A series of vertebral centra and a right coracoid of a young individual. The vertebrae include the atlas and axis (text-fig. 24, C & D), the line of junction of which is shown by a sharply-defined groove which is only interrupted at the floor of the neural canal ; the axis has distinct diapophysial and parapophysial facets ; the position of the inter- vertebral wedge-bones is shown by the facets with which they united. There are twenty-nine other precaudal vertebre and twenty-eight caudals. The small coracoid presents no remarkable characters. The dimensions (in centimetres) of some of the bones are :— Atlas and axis: combined length depth of centra width of centra , Anterior caudal vertebra: length of centrum depth of centrum width of centrum Coracoid: greatest width . tan meses te) cute os pe plenp thw ry. ican cou ec Mee aren Se ey eles gr to hm ty wredvdswonp o so Ro) R. 2169 (? Leeds Coll. 80). Twenty-four caudal centra; of these, one from the bend of the tail and another further back are figured (text-fig. 28). Also three neural arches. The dimensions (in centimetres) of the first figured centrum and arch are :— Centrum: length at neuralcanal. . . . . . - . . ~~. 21 - ventralced@ee pi 0 ya ig eee ome eli SWC LLG eases poe aire ue aetrs gn ini cute eu Tue eon 3.277 Nita aaa aoe oP ce wee eee Soyo Ores op ro) woaesel R. 3533. Portions of a young skull and skeleton, including basioccipital, exoccipital, supraoccipital, stapes, opisthotics, quadrates, pterygoids, squamosals, maxillze, nasals, vomers (text- fig. 19), and other bones of the skull and mandible. Sixty vertebral centra, some neural arches and ribs. R. 3535. Bones of skull and skeleton of a large individual, including parietals, frontals, and squamosals united (text-fig. 15), basisphenoid with parasphenoid, quadrates, stapes, opisthotic, portions of pterygoid, two vertebral centra; also the clavicular arch. Numerous sclerotic plates. R. 3534. Eighteen centra of caudal vertebrie, mostly with the neural arches and ribs associated with them (text-fig. 27). Also imperfect shoulder-girdle, fore paddles, imperfect pelvis, hind paddles. Nore.—In the explanation of fig. 40 (p. 59), for “Left ischio-pubis of Ophthalmosaurus: A, from outer side; B, proximal end; ©, from inner side,” read “ Right ischio-pubis of Ophthalmosaurus : A, from inner side; B, proximal end; C, from outer side.” MURZENOSAURUS, =—T =I Order SAUROPTERYGIA. Carnivorous aquatic reptiles in which the skull has only one temporal arcade and a fixed quadrate; the pterygoids extending forwards to meet the vomers (tin all) ; the external nares situated some distance behind the end of the snout; a pineal foramen present. ‘The teeth are thecodont, and sometimes a few are considerably enlarged; in the later forms they are confined to the edges of the jaws. The tail is relatively short, swimming having been effected mainly by the limbs, which become paddle-like; the hind pair never greatly reduced. Dorsal ribs with a single head ; a plastron of ventral ribs. Suborder PLESIOSAURIA. Clavicular arch undergoing reduction and tending to become situated on the visceral side of a ventral extension of the scapule, which in the later types replace it functionally ; coracoids large. ‘The ilium is directed backwards and articulates with the ischium only; the pubis is a broad plate of bone, and the ischia also are generally much expanded, The limbs form oar-like paddles. ‘The plastron consists of a median and several lateral series of overlapping ventral ribs. Family ELASMOSAURID. Head relatively small; neck long, in some cases excessively so. Cervical ribs with single head. Scapule meeting in the middle line, where they join the corresponding median anterior prolongations of the coracoids, at least in fully adult individuals. Clavicles and interclavicles may both be present, but one or both are usually greatly reduced. Epipodial bones much modified, being shortened up so as to resemble mesopodials. Middle Jurassic to Cretaceous of Europe, North America, and perhaps New Zealand. ’ Genus MURENOSAURUS, Seeley. [ Quart. Journ. Geol. Soc. vol. xxx. (1874) p. 197.] Skull short and broad, of relatively small size. About 24 teeth on each side in the upper jaw, five being situated in the premaxilla; of the maxillary teeth, the third, fourth, and fifth are enlarged. Mandible with a short symphysis and bearing 78 MARINE REPTILES OF THE OXFORD CLAY. about 20 teeth on each side. Neck consisting of about 44 vertebra, the centra of which in the anterior part are about as long as broad. In the shoulder-girdle there is a well-developed interclavicle, while the clavicles are generally greatly reduced, in some cases being mere films of bone adherent to the visceral face of the interclavicle ; in some cases probably they are wanting entirely. Coracoids not greatly produced outwards and backwards into postero-lateral processes. Fore limb a little larger than the hind limb, to which it is very similar in form, the humerus not being greatly expanded at its distal end even in the adult. Middle Jurassic. The specimen upon which Prof. H. G. Seeley founded Murwnosaurus leedsi, the type species of the genus, is included in the Leeds Collection (No. 25, R. 2421) ; it consists of portions of the skull and mandible, 79 vertebre, some of the caudals having been lost, numerous ribs, coracoids, scapule, pubes, ischia, ilia, and both the fore and hind paddles. The following account of the skeleton in this genus is founded so far as possible on this specimen, but many other nearly complete skeletons of the same or closely similar species have been employed to supplement the description. Skull (Pl. II1.; Pl. VI. figs. 1-2; text-figs. 43-47).—The skull is small in proportion to the size of the animal, and is roughly triangular in outline, the muzzle being bluntly pointed. The upper surface of the anterior portion was probably gently convex from side to side, while in the parietal region there is a high, sharp, sagittal crest, the posterior end of which is the highest point of the skull, from which it slopes gradually down to the tip of the snout. The following account of the individual bones is founded on several more or less nearly complete examples—the best (R. 2678) belonging to the skeleton which is the type specimen of JM. platyclis. Portions of the skull of MZ. durobrivensis (R. 2861), as well as of the type of I. leedsi, are also figured and described. The basioccipital (b.0c., Pl. III. figs. 1, la; text-figs. 45, 44) bears the whole of the nearly hemispherical occipital condyle (oc.c.), the border of which forms a sharp rim, sometimes separated by a short neck from another parallel rim (text-fig. 44) ; the upper border of the condyle in some specimens is a little flattened beneath the neural canal; there is no pit marking the original position of the notochord, such as has been described as occurring on the occipital condyle of Ophtha/mosaurus (see supra, p. 6). The upper surface of the bone (text-fig. 43, B) bears a pair of large elongated oval facets (ero.f.), for union with the ventral end of the exoccipitals and opisthotics; the space between these facets (n.c.), forming the floor of the brain-case, is narrow and concave from side to side behind, while in front it widens out and bears in its middle line a strong longitudinal ridge which in some specimens is paired. The anterior face of the bone (text-fig. 43, C) is nearly vertically truncated by the surface (ds.f.) for union with the basisphenoid. Antero- MURZNOSAURUS. 79 laterally the body of the bone bears a pair of stout processes which project outwards and downwards, their ends being truncated by large oval facets (pt.f.) which look outwards and forwards and articulate with the pterygoids, which in these animals extend very far back, their posterior ends, by which they unite with the quadrates, being actually behind the occipital condyle. Between these processes the ventral surface of the bone is concave from side to side; in front of this it is flattened and greatly roughened up to its anterior edge. In some cases the posterior surface of the lateral processes bears a roughened surface for the attachment of muscles. Text-fig. 43. \ / Wh aw En ik i width The dimensions (in centimetres) of this specimen are :— Skull (text-fig. 44): Length from posterior border of pineal foramen to tip of snout. Sir ea wo dues rE ea ROG 155. mc Length of tasioceipital « Gio Ge oy ro, Goin | (Eyoordere)) Width of basioccipital at lateral processes. Q Vertical diameter of occipital condyle . Width of articulation of quadrate Mandible : Length . CN. Deh) IG of aot C so * GEENA NSE Go cll sa. oo oc 10 (Gi oxors)) 16:0 (app.) 315 28°3 52 63 Greater part of a skull and skeleton of an old individual. exoccipital, other cervicals (in these the sutures between the centra and the neural arches and ribs are obliterated, but in most cases the ribs have been broken off), nineteen dorsals (much distorted), and twenty-four caudals (in these the arches, ribs, and, in many cases, chevrons are fused with the centra) ; numerous. ribs, imperfect shoulder-girdle (in this ossification is complete, the coraco-scapular symphysis being continuous ; the clavicular arch is not preserved) ; fore paddles ; ilia; ischium ; part of pubis; hind paddles. MURAZNOSAURUS LEEDSI. 123 Atlasand Third Fifth Tenth Fifteenth Twentieth Twenty- Posterior Posterior Vertebree: Cervicals... fourth axis, cervical, cervical. cervical. cervical. cervical, caer cervical, cervical. Length of centrum in mid-ventral line . 6-1 31 4-0 4-6 5°5 6-0 6-1 57 5°77 Posterior width of centrumis.. <>< 3°6 4:0 40 4:8 ? 5:8 5°7 (etl 8:3 Posterior height of COntrUMy ie seca age thas 2:9 3:2 3°7 ? 5:0 5:2 6-0 6-1 Height to top of neural spine. . . 5S:8(app.) 6:9 81 9-7 12:2(app.) ? 2 ? 17°8 The total length of the neck was about 241 em. (about 8 ft.). The dorsal vertebree are too much crushed for measurement, but the dimensions of an anterior and middle caudal centrum are :—Length on mid-ventral line 3°4, 3°5 ; posterior width of centrum 7:4, 6°6; posterior height of centrum 4°7, 4°3. Shoulder-girdle : width at hinder end of glenoid cavity (exaggerated by crushing). . . . . +. 43:0 long diameter of coraco- leg foramen . 13:7 Humerus: greatest length . . . . Masih ines LOOre width of shaft (widened by cane Sh RNS esp ee CRB) » distal end (ditto) Gcdint oles ID Relvis:——llinmuslenptlinue ) vcr cee se teres Wier Pettit Meme ee tty Oso WACTHOLMIPPErie DCs rein coMlvae ofl cole sce on sh Ore ‘ alowenmend; o : eae eee Ischium: width of upper end (cr ‘abn a Gute iee: aoh eM) 3 TECH at gy iar ae oe isenite enn Tee, MemursMlenp thy saeco Wy eaele er euh etree cpt othe meer eutn ed c4 Widthvotshatbogien We) Aventyes see smtvicmethal senses Ore, ipa pClstalvexpansloniey! she este cee oe LOSS R. 2423 (Leeds Coll. 22). Incomplete skull and skeleton, the bones being much crushed and distorted. The following parts are preserved :—Basioccipital, basisphenoid, supra- occipital, part of squamosal, the articular portions of the mandible, atlas, axis, and about forty other cervical vertebree wanting ribs and arches, nineteen dorsals and nineteen caudals, crushed neural arches, fragments of scapule and ischium, humeri, femora, epipodials, and other bones of paddles. Length of the basioccipital . . . . 35 Width of occipital condyle from side to wide) 2:5 basioccipital at Jateral processes 45 Length of united basioccipital and basisphenoid 6-9 axis and atlas 53 9 The other vertebree are too much distorted to give measurements of any value. The centra of the posterior cervicals appear to have been rather shorter than in the type specimen. 9 al 124 MARINE REPTILES OF THE OXFORD CLAY. R. 2424 (Leeds Coll. 23). Imperfect skeleton, the skull and mandible being entirely absent. The vertebral column is represented by thirty-four cervicals, two pectorals, twenty-three dorsals (including two sacrals), and eighteen caudals. In some cases tlie centra are much crushed and in all are separated trom the neural arches, cervical and caudal ribs, ete. Five complete neural arches (four dorsal and one caudal), with portions of several others, are preserved. Several cervical ribs and several more or less nearly complete dorsal ribs are present. The right humerus with the radius and ulna, two ischia, part of a pubis, both femora, with some other bones of the hind paddle are preserved. In this specimen ossification is not so far advanced as in the type, although it is already larger, so that probably this individual might have attained a considerably greater size than the type. In the form of the limb-bones, so far as known, the two are closely similar, but in the cervical vertebra, especially in the posterior members of the series, there are some differences, the most striking being that the width of the centrum is somewhat greater in proportion to the height. It must be noted, however, that in the type specimen many of the vertebrae have been distorted by pressure, and this may account for the dissimilarity. Associated with this skeleton and in exacily similar condition of preservation is a clavicle (figured by Seeley in Proc. Roy. Soe. vol. li. (1892) p. 141, fig. 8) of an irregular triradiate form, its irregularity of outline probably indicating that the clavicles were undergoing reduction. With the exception of a portion of the opposite clavicle, this bone is the only element of the shoulder-girdle preserved in this specimen, and no such bone has been found with any other Afwrenosaurus-skeleton, so far as Lam aware, so it is just possible that it may actually belong to a species of Cryptocleidus. If, on the other hand, this is not so, and the bone is actually that of J. leedsi, it indicates that in some cases the clavicles undergo much less reduction than usual, and that this species must, in the structure of its clavicular arch, have been much like MM. platyclis. The approximate dimensions (in centimetres) of this specimen are :— wr : Twenty- Thirty- Forty- Vertebre: Cervicals*......... Fifth Hans Sixteenth sixth fifth second cervical, cervical. cervical, = : - cervical. cervical. cervical, Length of centrum in mid- Memtralwlinekere 10). ict ie 3-4 4:2 ol 5-2 5:3 52 Width at pesteriorend . . 3 4:3 vis 56 63 67 Height at posteriorend . . 2-6 ot 4-1 4:8 48 4:8 Caudals..30.5 35022203 ccdvesocestespueasetes Anterior, Posterior. Length of centrum on mid-ventral line . . . . 3-6 3-4 Wadth:at posterior-end. 5 5 3 2%) as es os ot ol Height at posteriorend. .:. . . 5s « « « « 4-0 4-0 EInmerusslenoth. ~ 4: woh, e, ) Se aen oko te et eID Widthiof heads fr. ud anette wen Wakes) a, Wee greatest width of proximalend . . . . .. . 9:0 least antero-posterior diameter of shaft . .. . . 7:3 width: of distallenduss, “a enumerEemicuiee mer cow Gy ge: O22 * The numbers of the cervicals in the series are only approximate. The dorsals are all too much erushed for measurement. MURZNOSAURUS LEEDSI. — bo or bo Femur: length . . width of head . rie cog eA Mca aR Ea har ogese greatest width at proximal end (with trochanter) . “I om Ot or or @ least antero-posterior diameter of shaft 46 width of distal end . 12:9 Pubis: greatest width . . . 2. 5 . 2°9 Han MEANS Ate ie 2-4 length of acetabular surface . . . . 1. ss « «FO "3 surface tormischinm %.' sev 1s ye seals 3°5 Ischium : greatest length from acetabulum to middle line (approx.) 18-0 length of acetabular surface. . . . . (approx.) 4:5 5 ilidefsuriacey yu". fades) (approx), 22a ae pubicisuriace® Gv. sy (Approxs), woro width at narrowest point. Bu veal R. 2864 (Leeds Coll. 34). Imperfect skull and skeleton of an individual smaller than the type specimen. The portions preserved are :—Basioccipital, basisphenoid with part of parasphenoid, portions of exoccipitals, of frontals, and of premaxillee, vomer and part of the pterygoid, quadrate; greater part of mandible ; several detached teeth ; forty- five cervical and pectoral vertebrae, twenty dorsal (and ? sacral) vertebrae and twenty caudals with some caudal ribs and chevrons; portions of the limb-girdles, the fure and hind paddles. The fore and hind paddles (text-fig. 63) are the most nearly complete yet found, practically all the bones on one side having been collected and retained in their natural relative positions. The caudal vertebre (text-fig. 59) are in excellent preservation, many of them still retaining their union with the neural arches and caudal ribs and in one case with the chevrons; of these last several are preserved separated trom the centra (text-fig. 59, F, G, H). The dimensions (in centimetres) of this specimen are :— Skull: Basioccipital : transverse diameter of the occipital condyle . 2-2 greatest length on middle line 2 » Width at lateral processes . . . . 41 Mandible : Length from anterior end of symphysis to angle (approx.) 25°0 si BOL, SVMPOYSISe oe epee One meen cL Era ea OSE » of postarticular region . . . . . . (approx.) . 3:6 Vartebrme Cecvicals * Atlas and Fifth Tenth Fifteenth axis. cervical, cervical. cervical, Length of centrum-. .... .- 45 2:2 3:5 3-9 Width of hinder end of centrum. 2°5 Sl 40 4-4 Height of hinder end of centrum . 21 2:5 31 o4 * The numbers of the vertebre in the series are approximate. Behind the fifteenth the cervical vertebre are too much distorted for measurements to be of any value; the same is the case with the dorsals. 126 MARINE REPTILES OF THE OXFORD CLAY. Caudals. Length of centrum in mid-yentral line. 31 31 2:7 2:3 21 2:0 Width of hinder end of centrum .. . 4-7 41 37 3:2 SI) 7) Height of hinder end of centrum . . . 4-1] 39 Gril 2 24 2-0 » totopofneuralspine ... . z 7:8 6-1 48 4:3 Width between outer ends of caudal ribs. 146 13:8 iis 8:2 7:0 Fore paddle (text-fig. 63, B): total length 72:0 shes Eade Go - 5 a yo -d os Gu eee Bec 25°2 greatest width at upper end (with tuberosity) 9-1 least antero-posterior width.of shaft 6°3 greatest width at lower end 14-1 Radius: greatest length 72 » Width Qe Ulna: greatest length . 4-9 » width . 61 The lengths of the successive phalanges of the longest (fourth) digit are :—4:1, 4:0, Bion O10, FO OnLy 270s) 2205.250;, 056, dea 3, dsl 8: Hind paddle (text-fig. 63, A): totallength . . . . . . . 705 emursw@leng th eamesuer een eee cae oe ue) Mm teem OLS greatest width as upperend . ........ V4 least antero-posterior width of shaft . . . . . . 5:0 greatest width at lowerend . ...... . =. 12:2 Mibiatnereatest lenothys, 2: cw cier gist yo. ue) 6) eu Meee re GOL oy Gee \itehd rename oe wate Sareau naa eeMrey noe tae bay © LeHta Pibula:’ greatestlength ~. 5 aes eh G2 we we ee) Aa 3 WAG. are) Meee etal ncle en Uceetee Ret me tact tan ROTO The lengths of the successive phalanges in the longest (fourth) digit are:—4:1, 3°9, 4:0, 3°6, 3-4, 2°8, 26, 2°3, 2°0, 1°7, 1:4, 1-1, °9. R. 3704. Shoulder-girdle of an old individual, probably of this species. In this specimen (text-fig. 62) the ossification of the bones seems to have proceeded further than in any other; this is especially notable in the posterior region of the coracoids, which are produced backwards with well-marked postero-lateral processes and less developed median processes not seen in younger examples. The scapule and coracoids meet in the mid-ventral line, and the interclavicle (Pl. VI. figs. 6, 6 a) is preserved in an almost perfect state. The middle of its anterior border is marked by a shallow concave notch (a.n.), the edge of which is smooth ; on either side of this the convex anterior and lateral edges of the bone are thin and with numerous gmall indentations. Posteriorly there seems to have been a short pointed process (p.p), the continuation backwards of a median ridge, seen on the ventral face of the bone ; this ridge begins anteriorly as a broad very slightly convex surface, and narrows backwards to this posterior point. The ventral surface seems to have been gently convex from side to side, the dorsal surface concave in the same direction. Crossing the dorsal face of the bone from side to side, at abeut two-thirds of its length from the anterior end, is a broad convexity, behind which the bone is slightly concave from before backwards. MURANOSAURUS DUROBRIVENSIS. Dye The dimensions (in centimetres) of this shoulder-girdle are :— Total length of the whole on middle line . 75:0 Scapula: greatest length . Beate: es! od bb nape 33°6 length from anterior angle to posterior end of youteal TAMUS! 2 seeius 5 Gee Gio e 24:6 length on a straight Tees poe posterior angle of ventral ramus to tip of dorsalramus . . . 26°6 approx. width of glenoid surface from before back WCU Siateet a reterstays oD ae OTe, approx. width of Slenaial aatincs fom abate down SUENSE Sa! Vig hp. Ate glad Wad Ngai Ryo! Lo. Uiguemon Roane] Coracoids: greatest length . .. . 43-9 width of united bones at meaee sels of pieced CaN by eeman ominous pics he) st oe) cee 20:8 width of united bones at narrowest. . . . . 28:2 a 5 between the poster eier nal anclesjy meme Sane eae (AD PLOK)an 479 antero- nestoviondininetee of coraco-scapularforamen. 13:2 transverse diameter of ditto. . . . . (approx.) 94 Interclavicle (Pl. VI. fig. 6): length . . . . . (approx.) 11°9 Widthis? soa auaerntss, (pprox)p «ds R, 2443a. Left femur probably of this species, figured in Phillips, ‘Geology of Oxford, ete., p- 317, text-fig. exxi., as belonging to Pliosaurus grandis. Murenosaurus durobrivensis, Lydekker, sp. [Plate V.; text-figs. 43, 45, 49-57, 60, 65, 67.] 1889. Cimoliosaurus durobrivensis, Lydekker, Catal. Foss. Rept. Brit. Mus. pt. ii. p. vill. 1895. Murenosaurus plicatus, Andrews, Ann. Mag. Nat. Hist. [6] vol. xvi. p. 429 (pars). Type Specimen.—An imperfect skeleton, including thirty cervical vertebre, fifteen dorsals, and some caudals, wanting ribs and arches; shoulder-girdle wanting the clavicular arch ; imperfect fore paddle; pelvis including both iha and ischia and the pubis; imperfect hind paddles (R. 2428, Leeds Coll. 28). The pectoral and pelvic girdles have been described and figured in Ann. Mag. Nat. Hist. [6] vol. xvi. (1895) p. 429 (see also Pl. V.; text-figs. 43, 45, 49-07, 60, 69, 67). This species was established by Mr. Lydekker for the reception of an Oxford Clay Plesiosaur which, while closely resembling his Cimoliosaurus plicatus, differs in wanting the median bony bar uniting the coracoids and scapule, and in possessing cervical vertebre with relatively shorter centra, especially in the posterior portion of the neck. The absence of the median bony bar in the shoulder-girdle is probably merely an age-character, but the shortness of the cervical vertebre is a well-defined peculiarity and is accompanied by others which distinguish this species from the other members of 128 MARINE REPTILES OF THE OXFORD CLAY. the genus. One of these peculiarities is, that ossification does not become complete (as indicated by the fusion of the cervical and caudal ribs and of the neural arches, Text-fig. 67. Cor. 4A jy ae Z TG D . fi’ = LE \( { 4 If SS —— = SSS==- Shoulder-girdle of Murenosaurus durobrivensis : A, from above; B, from right side. (Type specimen, R. 2428, 1 nat. size.) cor., coracoid ; d.p., dorsal ramus of scapula; gl.c., glenoid eavity ; scap., scapula ; v.p., ventral ramus of scapula. MURZNOSAURUS DUROBRIVENSIS. 129 and the condition of the proximal ends of the humerus and femur) until a much greater size has been attained than in J. leedsi, and all parts of the skeleton are more massive than in that form. Another point is, that the expansion of the distal end of the humerus is greater than in I, leedsi. As already noticed, the cervical vertebre are shorter than in M. leedsi; and this difference is most marked in the posterior part of the neck, where the centra in the present species are shorter in comparison with those of the anterior region. ‘The plications on the edges of the centra, just outside the articular faces, are particularly well marked in this species, certainly much more so than in MM. platyclis, where they seem to be replaced by irregular rugosities. It was the existence of these plications on a cervical vertebra that caused Phillips to name one of the Oxford Clay Plesiosaurs Plesiosaurus plicatus: this no doubt is a Murenosaurus, but it is not possible on the available evidence to be sure whether it is identical with the present species or with AM, leedsi or is a distinct form. In the pectoral girdle in this species the median union of the scapule with one another does not take place till a very large size has been attained, and the median junction of the coracoids and scapule was still later. ‘The interclavicle is a small oval bone, with anterior and posterior rounded notches in the middle line; it is thickened in the middle, especially towards the anterior border, but laterally it thins out to a sharp edge; no trace of clavicles has been observed. R. 2428 (Leeds Coll. 28). Portion of a skeleton, including twenty-nine cervicals (Pl. V. figs. 1-3), two pectorals, fourteen dorsals (PI. V. figs. 4, 5), and some caudals (PI. V. figs. 6, 7); coracoids and scapule (text-fig. 67) ; imperfect fore paddles ; ilia, ischia, and one pubis (text-fig. 65) ; incomplete hind paddles. Type specimen referred to by Lydekker in Catal. Foss, Rept. Brit. Mus. pt. ii. (1889) p. viii. The pectoral and pelvic girdles have been described and figured in Ann. Mag. Nat. Hist. [6] vol. xvi. (1895) as those of of M. plicatus, Phillips, sp. In this specimen the neural arches are wanting in all but two or three of the vertebree, and in the cervical region the ribs are also missing, fusion with the centrum not having taken place in either case. The ventral rami of the scapul have not yet reached the middle line and are widely separated from the coracoids, so that there is no doubt that although this was a large individual it had not nearly attained its full size. The dimensions (in centimetres) of some of the bones of this skeleton are as follow :— Vertebral centra, Anterior Middle Posterior First dorsal Caudals Caudals (Pl. V. figs. 1-7.) cervicals. cervicals. cervicals, crushed. anterior, posterior. Length in mid-ventral line . 2:7 4:0 49 50 5:3 5:2 yes) S55 OFT 4:2 28 2-5 Width of posterior articular surtacanm eas. ( Wh Bi) (4:00 15-7 6:0! 7AM TOM TON G5 | 4-6. Sz Height of posterior articular SULLACe) eaten a Meee ed POs oy 13st 4:6 5:2 Delos 6-2 60 6-4 4-9 3:5) onl Scapula (text-fig. 67): greatest lencth . . . . . . . . 274 length from anterior angle to posterior end of ventralramus . . . . . 168 1350 MARINE REPTILES OF THE OXFORD CLAY. Seapula (text-fig. 67): length in straight line from posterior angle of ventral ramus to summit of dorsal ramus . . = aeons isquen mea width of glenoid autiace from above downwards. . . . 2 4:8 width of glenoid surface en iiefore backwards... . c Be a 3 Coracoids (text-fig. 67): greatest antero-posterior length Bg OL width of united bones at posterior angle of glenoid cavity . . . . 335 width of united bones at narrowest . 26:9 » between postero-externalangles 36-8 Greatest length of united coracoids and scapule . . . . . 60:0 Humerus (Pl. V.figs.11,1l@,116):lenpth . . ... . 315 greatest width of upper a 10°6 width of shaft at narrowest 7:8 - distalend . . . 194 Pelvis (Pl. V. figs. 8, 8a, 86,9, 9a; text-fig. 65). thinmsWenoth ven tel oes kiv (e onue lM ey en MLO Sl greatest width of upperend ....... . 56 width of shaft . . . . a Cewen nomacivin ee ee sere greatest width of lower aa Pah ord: Woe sa ns cen, RY pubis: ereatestlensth. 9%. 6 2 ss 2 ee 2 1s a0 Bo Willd New ‘ol at ce ou ee) oa BGS. Gas mezice: length of acetabular surface . . . .... =. 80 = symphysialbordem =. is ee = on, LOie Ischium’:widthofupperend’ . 9. 2. 5 . © . = .,. 970 an MOCK ~o) 5. ey Bes ta Gtee ee Rs car “OID 3 lowerexpansion . ...... . 202 length of acetabular surface. . . . . . . . 38 Femur (Pl. V. figs. 12,12@,126): length . . . . . . . 284 width of upperend . . . 80 2 middle shaft (at narrowest) . 5-4 3 lower end 15:8 Tibia (PI. V. figs. 12, 12a): length ei width 9-0 Fibula (Pl. V. figs. 12,12): length . 6-5 width v2 R. 2863 (Leeds Coll. 29). Imperfect skeleton, including fragments of skull and mandible ; forty- two cervical vertebra (text-figs. 49-51), two pectorals (text-fig. 52), twenty-one dorsals (text-figs. 52-55), and eight sacrals (text-fig. 56) and caudals (text-fig. 57) ; numerous ribs (cervical, dorsal (text-fig. 60), and caudal), ventral ribs (text-fig. 60); coracoids, scapule and interclavicle (Pl. V. fig. 10), humerus and some other bones of fore paddle ; ilia, ischia, pubes ; femora and some other bones of hind paddle. In this specimen the neural arches with the neural spines are preserved united with MURZNOSAURUS DUROBRIVENSIS. 131 the centra in the cervical and dorsal regions, although the neuro-central sutures are still open ; in the sacral and caudal regions the arches are lost in all cases but one (text- fig. 57). In the pectoral girdle the scapulee met in median symphysis, but did not yet extend back in the middle line to meet the coracoids, the interval between them being, however, only about 4.cm. The heads of the humeri and femora are showing signs of becoming rounded owing to the extension of ossification into the originally cartilaginous ends: in the type specimen they are nearly flat. The whole skeleton shows very well the massiveness characteristic of this species. The dimensions (in centimetres) of this specimen are :— Skull too fragmentary to measure. Mandible: width of articular surface. . . ... =... «26 length of postarticular region . . . . ... 38 9 Vertebre: Cervicals............ eee Third. Fifth, Eleventh, Twentieth, Tay fe ages cervical). Length of centrum in mid-ventral line . 5:0 31 3:3 4:3 52 53 51 Width of posterior face of centrum . . 3:0 31 3°5 42 5:2 66 ETE Height of posterior face of centrum . . 2:6 27 3:0 3-7 4-8(app.) 5-7 6-0 Height to top of neural arch . . . . .. Ae Mi 10-0 15°3 18:0 Pectorals and aaa ae Soot First RRLerCE Anterior Post. Post. (?last) Sacral Anterior Middle sacrals and caudals (figured), dorsal. (figured). dorsal. dorsal. dorsal. (figured). caudal. -(fgured). Length of centrum in mid- ventral line. . . . 56 a7 56 5°9 5:6 54 AT 34 2-9 Width of posterior face of centrum . ba | oo T fon) a oO va lor} 69 6°3 6:7 63 a4 Height of posterior face of centrum . 6°53 6:2 oo 6:5 6-0 5:8 51 47 41 Height to top of Leet eae ae Ar 20°0 22:0 20°0 3-3 Width between ends of trans- verse processeS. « . see ate 145 18-4 14°6 10°6(app.) .- Coracoid): greatestilengthy 2) 0.) (4) )4) se aes BU tay) ete) width of united bones at hinder angle of glenoid Cavity s a1 Bone Ns 40-4 width at narrowest te the te nnited hanes (approx. ) 28-4 » between angles of postero-external processes. 34:0 Scapulas ereatestilenathy = 7. jee Gcveachireniienpetr Benen sais 2 eo length from anterior angle to posterior end of VentralsTamuUs! Winch muci somone Gace ve LO: length from end of median coracoid process 6 tip of dorsal tamus).) 205). 3 Met aoiecuites poco width of glenoid surface (ators dowaracas) eee amcor te 7 » (before backwards) . . . ° 65 Interclavicle (Pl. V. fig. 10): greatest length . . . . . . 85 length onmid-line . . . . . 66 width (so far as preserved) . . 9:8 77) bo 132 MARINE REPTILES OF THE OXFORD CLAY. Hnmeruskwlengthisseseykis ile rs) te greatest width at upper end . diameter of head. width of shaft at narrowest dorsal end . Ilium : length at St inen be greatest width at upper end width of shaft . : greatest width of lower end Pubis: greatest length width length of acetabular surface F symphysial border Ischium: width of upper end es shaft 3 lower expansion length of acetabular surface (approx.) 8-4 20:0 6-4 oO 6°7 26°5 30°5 73 (approx.) 20-0 9:3 61 21:2 (approx.) 55 R.2861 (Leeds Coll. 18). Imperfect skeleton of a large individual probably of this species, though the plications round the edges of the cervical centra are less marked than usual. The parts preserved are an imperfect and much broken skull (basioccipital, exoccipital, opisthotic, and supraoccipital, figured in text-figs. 43, 45); a nearly complete mandible; atlas, axis, and twenty-seven other cervicals, mostly with the fused arches and ribs broken away; nine dorsal and two caudal vertebree ; about twenty dorsal and cervical ribs, some portions of ventral ribs ; portions of the acetabular region of the pelvis with the con- stituent bones united; left femur, both tibiz and fibulee, tarsals, and sixty-three metapodial bones and phalanges. Some dimensions (in centimetres) of this specimen are :— Skull (text-figs. 43, 45): Length in mid-dorsal line from posterior end of the parietals to the tip of the snout . Length from pineal foramen to tip of snout . Transverse diameter of occipital condyle . Vertical diameter of occipital condyle . Length of basioccipital . Greatest width at lateral processes . Mandible: length : » of symphysis . width of articular surface for quadrate Wertebraty) ic.» distal end. Ulna: greatest length . » width Pisiform: width . length Pubis: greatest length ss awadth: ss depth of symphysis length of surface for ischium es acetabular surface A least width of neck 4 . 2... 8, Hind limb : Femur (Pl. VIIL. fig. 4): length . greatest width at nrosifial aa ane WVidtavOnbeady st stoic te) ule) tee View nelinen tee » shaft at narrowest » distal expansion ~ . . 3 % Tibia (Pl. VILL. fig. 4): length. . . . width . Fibula (Pl. VII. fig. 4): length width . 25:0 23-0 21°6 78 6-9 43 11:3 36 56 35 54 Y2 165 164 MARINE REPTILES OF THE OXFORD CLAY. Genus CRYPTOCLEIDUS, Seeley. [Proc. Roy. Soe. vol. li. (1892) p. 145, as a subgenus of Murcenosaurus. | Plesiosaurs in which the skull is relatively small, about a quarter the length of the neck in the adult. Neck consisting of about 32 cervical vertebre, the centra of which are short, with oval articular faces which are deeply concave, at least in old individuals. There are two or three pectoral vertebree and 21 or 22 dorsals. There seem to have been three or four sacrals, the ribs of which are thickened and converge towards their outer ends ; the number of caudals is uncertain, but probably was about thirty ; the posterior caudals diminish in size rapidly. Cervical ribs with single heads, not much flattened, and in some cases with a fairly prominent anterior angle. The ventral ribs forming a strong plastron, each transverse row consisting of a median and three pairs of lateral elements. Shoulder-girdle of the true Elasmosaurian type, the scapule having an extensive median symphysis continuous with that of the coracoids. Coracoids with prominent postero-lateral processes in the adult. The clavicular arch consisting of two triangular clavicles meeting in median symphysis: in some cases with a rudimentary interclavicle interposed between them posteriorly (text-fig. 88); the clavicular arch, except for its extreme anterior edge, lying entirely on the visceral surface of the ventral bar of the scapule in the adult. The humerus greatly expanded distally; radius large, with an elongated anterior border, so that the axis of the expanded portion of the paddle makes a slight angle with that of the humerus. In the pelvis the pubis short in proportion to its width ; the femur not greatly expanded distally. Only a single species of this genus is recognised from the Oxford Clay of Peterborough. Cryptocleidus oxoniensis, Phillips, sp. [Plates IX. & X.; text-figs. 78-94. ] 1871. Plestosaurus oxoniensis, Phillips, Geology of Oxford, etc. p. 307. 1871. Plestosaurus eurymerus, Phillips, op. cit. p. 315. 1888. Plesiosaurus ovoniensis, Lydekker, Geol. Mag. [3] vol. v. p. 352. 1889. Cimoliosaurus eurymerus, Lydekker, Catal. Foss. Rept. Brit. Mus. pt. ii. p. 205. 1889. Cimoliosaurus oxoniensis, Lydekker, tom. cit. p. 209. 1892. Plesiosaurus durobrivensis, Seeley, Proc. Roy. Soc. vol. li. pp. 132-134 (for young). 1892. Cimoliosaurus eumerus, Seeley, Proc. Roy. Soc. vol. li. p. 145. 1892. Cryptocleidus platymerus, Seeley, tom. cit. p. 145. 1895. Cryptoclecdus oxoniensis, Andrews, Ann. Mag. Nat. Hist. [6] vol. xv. p. 335. Type Specimen.—Cervical, dorsal, and caudal vertebre described and figured by Phillips in the ‘ Geology of Oxford, etc., pp. 307-309, figs. cxiii-cxv. Phillips also CRYPTOCLEIDUS OXONIENSIS. 165 ascribes to this species a shoulder-girdle (figured and described as a pelvis—op. cit. p- 310, fig. cxvi.) and a paddle (op. cit. p. 512, fig. exvii.), both of which probably belong to a species of Murwnosaurus. Phillips, in his original account of the cervical vertebree which must be regarded as the type specimens, simply states that they are biconcave with narrow tumid inter- foraminal space, while his figures and measurements show that the centra are short ; the vertebre figured, however, are not anterior examples as Phillips supposed, but from some little distance back in the neck. The dorsals and caudals described are not definitely stated to have been associated with the cervicals, and, as above noted, the shoulder-girdle and paddle are those of a Murenosaur, consequently the determi- nation of the species must rest entirely on the cervical vertebree. Comparison of these with the corresponding vertebre of the commonest type of Plesiosaur from the Oxford Clay of Peterborough, shows such similarity of form that both are clearly of the same species, and therefore the name Cryptocleidus oxoniensis is applied to them. Lydekker employed the name Cimoliosaurus oxoniensis for the smaller individuals of this type, while he called the larger C. eurymerus, a name which had been given by Phillips to a large broad paddle from the Oxford Clay of Bedford associated with vertebra similar to those of C. oxoniensis. The series of specimens in the Leeds Collection tends to show that, as Lydekker himself suggested, these two forms are probably only a single species, and that. the difference in size and form are merely the result of increased age. Since, however, the ossification of the shoulder-girdle is completed while the individual is considerably smaller in some cases than in others, it is possible that the difference may be a sexual one, as I have already suggested in a paper on the development of the shoulder-girdle (Ann. Mag. Nat. Hist. [6] vol. xv. (1895) p. 333). The differences in the form of the limb-bones which led Professor Seeley to establish his species Cryptocleidus platymerus for a specimen (Leeds Coll., R. 2412) in this collection, are probably due entirely to the advanced state of ossification that had been reached in this case. One of the most important points about this species is, that remains of individuals of all ages are common in the Peterborough deposits, and are easily distinguished from the other forms. From these specimens it has been possible to make out the history of the development of several parts of the skeleton, notably that of the shoulder-girdle (see Ann. Mag. Nat. Hist. [6] vol. xv. (1895) p. 533). The description of the skeleton given below is founded mainly on the almost perfect adult skeleton (R. 2860) which has been mounted in the Gallery of Fossil Reotiles ; reference will also be made to other specimens, especially to the mounted skeleton of a young individual (R. 2417), a brief account of which has been published in the Geological Magazine (1895), p. 241. Skull (Pi. [X.).—The skull, so far as known, is very similar in its structure to 166 MARINE REPTILES OF THE OXFORD CLAY. that of Murewnosaurus, and will be described mainly by pointing out such differences as occur. Unfortunately the material available for description is very imperfect, all the specimens being much crushed and wanting many important parts. The basioccipital (b.0c., Pl. IX. figs. 1, 2, 4, 5) is closely similar to that of Murwno- saurus, as will be seen from the figures. The occipital condyle is perhaps a little broader in proportion to its height, and its articular surface is continued quite up to the facet for the exoccipital, there being no trace of a neck to the condyle such as is seen in most specimens of the basioccipital of Murenosaurus; the basipterygoid (lateral) processes (pt.p.) are a little more rounded in section than in that genus. The dbasisphenoid (b.sp., Pl. IX. figs. 2, 5) does not differ in any important respects from that of Murenosaurus or Tricleidus. The facet marked f. in text-fig. 73, A, is here very well developed, and probably received the anterior lower angle of the pro-otic. ‘The ventral face is extensively overlapped by the parasphenoid in the usual way. In the young skull (Pl. LX. fig. 5) the basisphenoid shows evidence of its original ossification from two centres, since it is deeply notched posteriorly in the middle line and there is also a large vacuity beneath the pituitary fossa. This opening corresponds to the original space between the trabecule, through which the pituitary body was connected with the pharynx; possibly, even in the young, it was partly closed by the posterior end of the parasphenoid, and it corresponds in position to the pit occurring in some specimens of the basisphenoid of Murenosaurus (see Pl. III. fig. 1). The structure of the evoccipital-opisthotic (ex.op., Pl. IX. figs. 1, 1a, 4, 4 a) differs yery little from that seen in A/urenosaurus ; the paroccipital process is rather shorter, and in the oldest specimen available fur comparison (R. 2860) the fosse for the ampulla of the posterior vertical semicircular canal and for the canals themselves are larger and more open (see Pl. IX. fig. 1). In the young specimen (R. 2417) the exoccipital and opisthotic are united above the jugular foramen (jwg.), their line of union being still clearly visible ; but below that opening they are still separate (Pl. IX. fig. 4), each terminating in a distinct facet for union with the basioccipital. In this specimen also the posterior face of each exoccipital bears a small facet (f:) somewhat resembling a zygapophysis; this probably indicates that a well-developed pro-atlas was present, since the form of the neural arch of the atlas is such that there certainly was no point of contact between it and the exoccipitals. The supraoccipital (soc.) is a high arch curving forwards, the occipital surface being concave from above downwards and continuous with the posterior face of the postero-lateral processes of the parietals (PL IX. fig. 1a@). In front the supraoccipital bears a facet looking downwards and forwards for union with the pro-otic, and the inner side of this surface is deeply channelled for the reception of the upper portion of the posterior vertical semicircular canal (p.v.c.). Between the pro-otic facet and the junction with the parietal the border of the supraoccipital is sharp and concave, and clearly did not unite with any other element. CRYPTOCLEIDUS OXONIENSIS. 167 The parietals ( par.) widen out posteriorly into lateral processes, on which are the sutural surfaces overlapped by the upper ends of the dorsal processes of the squamosals (sqg.1), which in this genus do not appear to have met in the middle line, so that the actual vertex of the skull is formed by the parietals alone. In front of the lateral process the united parietals rise into a high sagittal crest, widening out again at the pineal foramen, of which they form, at least the greater part of and possibly all, the margin. From an examination of this region in the young skull it seems probable that the parietals completely surround the pineal foramen at its inner (cranial) end, while on the outer surface of the skull a portion of its anterior margin may be formed by the overlapping frontals. In front of the pineal foramen the frontals widen out considerably and unite at their outer ends with the postfrontals; in front of this they form the roof of the orbit, but their relations with the bones further forwards is unknown, all the specimens being very imperfect in the rostral region. The squamosal is of the usual triradiate form (PI. IX. fig. 3); the slender dorsal rami (sqg.!) run up to the lateral processes of the parietals, with which they unite, over- lapping them on the upper and probably also on the lower surface, but not meeting one another in the middle line. Judging from the appearance of this portion of the squamosal in the young specimen, it seems possible that it may have originally ossified from a distinct centre and may therefore represent a supratemporal. The ventral ramus unites closely with the quadrate, down the outer side of which it sends a long process. The anterior (zygomatic) ramus (sqg.) is thin and broad; its upper border is strongly convex, the lower concave; anteriorly it bears a sutural surface for union with the jugal and presumably also with the postorbital. The guadrate (q.) is a large bone; its anterior face is concave from side to side, its posterior face convex or flat. The upper end is embraced by the squamosal, while at its lower end it bears the broad articular surface for the mandible. This surface is convex from before backwards, and is imperfectly divided into a larger outer anda smaller inner convexity. There is no trace of any division of this bone into two elements. The premaxille seem to have borne six teeth each ; the muzzle was probably rather more pointed than in Murenosaurus. The mandible (Pl. LX. figs. 6, 7) is somewhat more slenderly built than in AWureno- saurus, and the symphysis (sym.) seems to have been somewhat shorter, otherwise the structure is similar. ‘There is no trace of any division between the articular and surangular, even in the youngest specimens. The postarticular region is relatively smaller and especially shorter than in J/urenosaurus. ‘There are 25 or 26 teeth on either side. The teeth (PI. IX. fig. 7) are long, slender, and very sharply pointed; the enamel is smooth, except for a few fine ridges confined to the lower part of the inner side cf the crown. In Murenosaurus (P1. III. figs. 4-6), on the other hand, the whole of the crown except the tip is covered with fine longitudinal ridges; on the anterior and 168 MARINE REPTILES OF THE OXFORD CLAY. the posterior side there is a main ridge, more prominent than the rest and continuous almost to the top of the crown, forming in some cases a very slightly marked keel. The rest of the surface (Pl. III. fig. 4 a) is covered with fine ridges of varying length, running in a generally longitudinal direction and often anastomosing. The crown is circular or nearly circular in section throughout its length ; in Cryptocleidus the crown is sometimes slightly compressed, so that in section it is oval. The roots of the teeth in both genera are very long and circular in section; their surface is smooth and the pulp-cavity is large. Vertebral Column.—The atlas and avis (text-fig. 78) are very similar in structure to those of Murenosaurus, but differ in some details. Asin the case of the other cervicals, the centra of these vertebrae, particularly that of the axis, are shorter than in Mureno- saurus. In the formation of the cup for the occipital condyle the bases of the lateral pieces of the neural arch of the atlas (at.a.) take a somewhat greater share than in the other genus, but they are still separated by a considerable interval from the sub- vertebral wedge-bone (a.w.d.) which constitutes the lower fourth of the cup. Asin Murenosaurus, the lateral pieces of the neural arch do not unite above, but run back and articulate with the anterior zygapophyses of the axis; from the presence of a pair of peculiar facets on the exoccipitals in the young specimen (Pl. IX. fig. 4a, f.), it seems probable that a pro-atlas was present, but whether single or paired there is no means of ascertaining. In the axis the neural arch (az.a.) is lower than in Mureno- saurus, and there is not so well-developed a neural spine; on either side of the base of the neural spine there is a strong ridge, not present in the other genus. The posterior zygapophyses (p.z.) are very large, and project a long way behind the level of the posterior face of the centrum ; their facets are flat and look outwards and down- wards. The posterior face of the centrum is concave in the middle with a raised and convex outer portion ; it is considerably wider than high. ‘The facet (r.1f.) for the rib of the atlas is much larger than in M/urenosaurus, and its antero-inferior portion is borne by the subvertebral wedge-bone. The rib of the axis (7.?) is larger than that of the atlas; the facets for these two ribs are confluent. ‘The subvertebral wedge-bone (a.w.b.) bears a strongly developed hypapophysial ridge (hy.7.), which forms a distinct anterior prominence not seen in Murenosaurus; the ridge only extends backwards on to the anterior portion of the centrum of the axis. In the young specimen (text- fig. 78, A, B) in which the constituent elements of the atlas-axis are still separable, there is no trace of the presence of a second subvertebral wedge-bone. In addition to the atlas and axis there are thirty other cervical vertebre. Of these the centra are much shorter than in Murenosaurus, and their articular ends are strongly concave in the middle with rounded margins; in outline they are transversely oval, the upper border beneath the neural canal being a little concave. ‘The ventral face of the centra is gently concave from before backwards and also from side to side, with the exception of the longitudinal ridge between the pair of nutritive foramina. Above the rib- CRYPTOCLEIDUS OXONIENSIS. 169 facets the sides of the centra are flat or slightly concave. The neural arches occupy nearly the whole length of the centra. ‘They bear very strongly developed anterior and posterior zygapophyses, the facets of which are nearly flat in the anterior part of the neck, but further back become somewhat concave and convex respectively from side to side. In the anterior cervical vertebre the anterior and posterior zygapophyses are connected by a ridge which disappears on the posterior part of the neck. The Text-fig. 78. ii) | \ f 3 : cf \fl Atlas and axis of Cryptocleidus oxoniensis: A, from right side; B, from front of a young specimen (R. 2417, nat. size); OC, from right side; and D, from front of an older specimen (R. 2860, nat. size). at.a., arch of atlas; a.w.b., anterior wedge-bone ; aw.a., arch of axis; hy.r., hypapophysial ridge ; 7.7, rib of axis; r.lf., facet for rib of atlas; od., odontoid (centrum of atlas); p.z., posterior zygapophysis of axis. 170 MARINE REPTILES OF THE OXFORD CLAY. neural arches, like the centra, are much shorter from before back than in J/ureno- saurus, and the neural spines seem never to have attained the height found in the posterior cervicals of some species of that genus. ‘The facets for the cervical ribs are about as deep as long. In the anterior part of the neck they occupy nearly the whole length of the centrum, but further back are separated from the anterior border by a short interval. ‘The cervical ribs are somewhat compressed from above downwards, and in some individuals in which ossification is very far advanced Text-fig. 79. Anterior cervical vertebre of Cryptocleidus owoniensis: A and C, from front ; B and D, from left side. (R. 2412, 3 nat. size.) a.z., anterior zygapophysis; c.7., cervical rib; 7.c., neural canal ; 7.8. neural spine; p.z,, posterior zygapopbysis: (e. g., R. 2862) they have a well-marked anterior angle; this is wanting in the nearly adult mounted specimen (R. 2860), in which the neural arches are fused with the centra throughout the column except in the posterior caudals, while the cervical and caudal ribs are free throughout. In the young skeleton (R. 2417) none of the arches and ribs are fused with the centra, and the cartilage-covered surfaces for the neura CRYPTOCLEIDUS OXONIENSIS. 171 Text-fig. 80. Posterior cervical vertebra of Cryptocleidus owoniensis: A, from front; B, from left side. (R. 2412, 4 nat. size.) a.z., anterior zygapophysis ; ¢.7., cervical rib; 7.¢., neural canal ; n.s., neural spine ; p.z., posterior zygapophysis. Text-fig. 81. Middle dorsal vertebra of Cryptocleidus owoniensis: A, from front; B, from left side. (R. 2418, 3 nat. size.) a.z., anterior zygapophysis; 7.c., neural canal; n.s., neural spine ; p.z., posterior zygapophysis ; rf., facet for rib; t.p., transverse process. zZ2 172 MARINE REPTILES OF THE OXFORD CLAY. pedicles are continuous with those for the cervical ribs. The neural spines are short and thickened ; their summits were evidently cartilaginous. ‘There are two pectoral vertebrae, in which the articulation for the rib passes from the centrum to the arch ; their centra assume a more circular outline, passing into the form of the dorsal centra (text-figs. 81, 82), the vertical diameter of which is approximately equal to the transverse. At the same time the articular ends are less concave and the rounded rim disappears ; the nutritive foramina pass on to the sides of the centra. There are 22 dorsals, the neural arches of which bear transverse processes ; these increase in length and rise on the arch in the first seven or eight dorsals, while in the posterior five or six, on the other hand, they descend and shorten (text-fig. 82). In the middle of the back Text-tig. 82. 71.8. | iol il Nl Posterior dorsal vertebra of Cryptocleidus owoniensis: A, from front; B, from left side. (R. 2418, 3 nat. size.) a.z., anterior zygapophysis ; .c., neural canal; n.s., neural spine ; p.z., posterior zygapophysis ; r.f., facet for rib ; t.p., transverse process. they are moderately long and curved, the concavity being downwards. The neural spines on the dorsal region are shorter and narrower than in Murenosaurus; in the young animal they are stout, but much shorter than in the hinder part of the neck. The zygapophyses in the anterior part of the neck are larger, and look more directly upwards and downwards than in the posterior portion. ‘There seem to have been four sacral vertebre (text-fig. 83), each bearing a pair of stout ribs (s.r.), which articulate partly on the arch and partly on the centrum and are eularged distally (s.f.); whether or not they actually joined the ilium is uncertain, but from the presence of deep CRYPTOCLEIDUS OXONIENSIS. 173 oO. roughened pits on the inner face of the upper end of the ilia in very old individuals, it seems possible that they did so, at least in advanced life. ‘Che centra in the sacrals Text-fig. 83, Sacral vertebra of Cryptocleidus oxoniensis: A, from front; B, from below. (R. 2412, 4 nat. size.) a.c., neural canal; s.f., facet for ilium ; s.r., sacral rib. Text-fig. 84. q | We I its } i} WS Anterior caudal vertebra ef Cryptocleidus oxoniensis: A, from behind ; B, from left side. (R, 2412, 3 nat. size.) c.f., facet for chevron-bone ; c.r., caudal rib ; #.c., neural canal ; 2.s., neural spine ; p.z., posterior zygapophysis. become slightly depressed, and pass posteriorly into the transversely oval centra of the anterior caudals (text-fig. 84). In the sacral region the neural spines begin to decrease 174 MARINE REfTILES OF THE OXFORD CLAY. in height, the decrease continuing to the end of the tail. The zygapophyses are well developed, somewhat concave (or convex) from side to side, and looking more inwards (or outwards) than further forwards. In the caudal region (text-figs. 84, 85) the centra, -as already mentioned, are wider than deep; laterally, at least in the young, they bear prominent facets for union with the caudal ribs, which in the adult become joined to the centra. These ribs are compressed from above downwards, and a little behind the middle of the tail some are considerably expanded towards their outer ends; in all there is a tendency to curve backwards. ‘The facets for the chevrons commence on the second or third caudal; at first they are confined to the hinder part of the centrum, where they form a pair of projections truncated by a flat nearly circular surface looking downwards and backwards. Further back (text-fig. 85) these facets are on both the Text-fig. 85. Posterior caudal vertebrae of Cryptocleidus owoniensis: A, end of caudal series from left side ; B and C, posterior caudal vertebrae from below. (R. 3705, 4 nat. size.) c.f., facet for cheyron-bone ; ¢.7., caudal rib; 7.a., neural arch. anterior and posterior edges of the centra, the heads of the chevrons articulating between the successive centra; the posterior facet is the larger. The neural arches on the caudal region decrease in height gradually from before backwards; the zyga- pophyses disappear about the middle of the tail; the arches of the caudal vertebre are the last to fuse with the centra, and the fusion takes place from before backwards. The chevrons are not well known ; in the young they are little rounded rods of bone with a slightly expanded vertebral extremity. The cervical, sacral, and caudal ribs have already been referred to. The dorsal ribs are thickened and oval in section towards their articular ends, which terminate in a Qt CRYPTOCLEIDUS OXONIENSIS. 17 nearly flat oval facet for the transverse process, External to this proximal thickening they are somewhat compressed from before backwards, and their upper edge bears a sharp ridge which terminates externally in a slightly backwardly deflected crest. External to this again the ribs are nearly circular in section; they become very little thinner towards their lower end, which terminates in a flat or shghtly concave surface probably tipped with cartilage in life. The ventral ribs (text-fig. 86) form a close plastron consisting of eight or nine transverse rows of bones, each consisting of a median element (1) and three lateral pairs (2, 3, 4), besides two posterior rows in which the median element is wanting. Text-fig. 86. Ventral view of plastron of ventral ribs of Cryptocleidus owoniensis. (R. 2862, 1 nat. size.) pu., pubis; p.7., forked end of posterior ventral rib; pu.sym., symphysis of pubis ; 1, 2, 3, 4, median and three lateral ribs of a transverse series. The median bone of a row is oval or circular in section in its middle portion, which is often much thickened; towards the outer ends the bone thins down toa point, and its anterior face at either end bears a flattened facet for union with the inner end of the first lateral rib. This is pointed at both ends and forms a very open S-shaped curve; its inner end bears on its posterior face a surface for the overlap upon the median bone, while its outer end has on its anterior surface a facet for the reception of the inner end of the second lateral rib. This latter is similar to the first, and the third differs only in wanting the outer facet, the outer half being rounded or oval in section and terminating in a point. ‘There is no trace of any connection with the true ribs. The two hinder rows differ from the others m not possessing a 176 MARINE REPTILES OF THE OXFORD CLAY. median element. In the anterior of the two the inner ends of the inner pair of lateral bones turn sharply forwards, and their extremities are closely adherent to the median bone of the row in front. The inner ends of the inner elements of the hindmost row are forked; what other lateral elements were present in this row is not known. This posterior row (p.7.) in the specimen figured seems to have actually underlain the anterior end of the pubis (pw.), but this may be the consequence of displacement Text-fig. 87. Adult shoulder-girdle of Cryptocleidus oxoniensis, from above. (R. 2616, about } nat. size.) el., clavicle ; cor., coracoid ; gl., glenoid cavity ; sc., scapula. resulting from the flattening out of the carcass. The whole ventral surface of the body seems to have been very strongly protected, by the expanded scapule and coracoids in front, the plastron of ventral ribs in the middle, and the plate-like pubes and ischia behind. Shoulder-girdle (Pl. X. figs. 1, 2; text-figs. 87-89)—This has been described in some detail in the Annals and Magazine of Natural History, [6] vol. xv. (1895) p. 333, and the following account is largely founded on the one there given. CRYPTOCLEIDUS OXONIENSIS. 177 The scapule (sc., Pl. X. figs. 1, 1a, 16, 1¢; text-figs. 87-89) are, as usual in the group, triradiate bones, consisting of a backwardly directed bar carrying the articular surfaces for the coracoid and humerus, an upwardly directed process (d.sc.), and a ventral ramus (v.sc.), which in the adult extends forwards and inwards to the middle line, where it unites in symphysis with its fellow of the opposite side. The following description of the scapula is based mainly on the adult shoulder-girdle (R. 2616) figured on Pl. X. and in text-fig. 87 :— The posterior bar is triangular in section; its inner edge, forming the outer border of the coraco-scapular foramen, is sharp, thickening a little as it approaches the coracoidal surface ; its upper outer border is rounded and passes above into the hinder border of the dorsal ramus or blade. The lower outer border rises into a rough ridge about 2 cm. from the glenoid surface and then runs forwards and outwards, forming on the outer face of the bone the boundary between ventral and lateral regions of the outer and ventral surfaces; anteriorly it terminates in a strong outwardly directed tubercle having a smooth facet on its summit. The glenoid surface and that for union with the coracoid are at right angles with ore another, the line of junction being slightly concave and about 6°5 cm. in length. The form of the glenoid surface is that of half a rather irregular oval, measuring about 6°5 cm. from the middle of its line of union with the glenoid surface of the coracoid to the top of the curve. The surface for union with the coracoid is an isosceles triangle, the sides of which are slightly convex and measure 8°2 cm. in length: the base is the line of union with the glenoid surface. This latter is nearly smooth, while the coracoidal surface is greatly roughened by the presence of irregular pits and ridges. The dorsal ramus of the scapula (d.sc.) is compressed from within outwards and is between 4 and 5 cm. wide at its summit, which is occupied by a rough depressed surface to which, in life, probably a small suprascapular cartilage was attached. ‘The anterior border of this ramus, especially on its lower portion, is greatly roughened, probably for the attachment of muscles above, and at its lower end for union with the roughened facet on the outer angle of the clavicle. The ventral ramus (v.sc.) is the largest and most important part of the scapula, at least in the adult; its anterior border is a continuation of the anterior edge of the dorsal ramus; it is at first rounded and concave as far as the prominent tubercle referred to above, then it becomes relatively sharp and thin, running inwards and forwards to the middle line, where by a sharp curve it passes into the median border with which it makes an angle of about 45°. The posterior border of this region of the scapula forms the anterior and half the inner edge of the coraco-scapular foramen ; it increases in thickness from without inwards and then backwards to the point of union with the anterior prolongation of the coracoid (Pl. X. fig. 1c), the surface for union with which is nearly semicircular and at right angles to the median symphysis. This latter (sym.), occupying the inner face of the thickened posterior prolongation of the 2A 178 MARINE REPTILES OF THE OXFORD CLAY. ventral ramus, forms a nearly rectangular surface about 7-8 cm. long and 5 cm. deep ; its antero-dorsal angle is rounded, while its antero-ventral angle is prolonged forwards as the inner edge of the thin anterior region of the scapula. The symphysial surface is deeply pitted and grooved by channels, which seem to have communicated with the exterior by a foramen situated on the upper surface at the middle of the symphysis. The outer surface of the ventral ramus of the scapula is nearly flat, but the visceral (upper) surface is divided into the raised and thickened symphysial region behind and the thin depressed anterior area which supports the clavicle, the two being separated by the ridge against which the hinder edge of the clavicle rests. In this thin anterior region the two scapule do not actually meet in the median line, but are separated by a narrow V-shaped interval, covered by the inner borders of the overlying clavicles; it is possible that this interval was filled by cartilage and that in advanced age the scapular symphysis was prolonged to the extreme anterior end. The clavicles (cl., Pl. X. figs. 1, 2; text-figs. 87-89) are in the form of scalene triangles. ‘The outer and posterior borders meet in an acute angle and the posterior border is sometimes concave. The lower surface of the outer angle bears a roughened facet (s.sc.), which fits against a corresponding rugosity on the anterior edge of the scapula and no doubt united closely with it. This union of the outer ends of the clavicles with the anterior borders of the scapul is a point of considerable interest, since it is probably a remnant of the original condition seen in the Nothosauride (text- fig. 61 B, p. 108), in which the clavicular arch stretches from one scapula to the other, the ventral plates of these bones being still widely separated, as, indeed, they are in the young shoulder-girdle of Cryptocleidus (see below). ‘The outer border is nearly straight and is thin and sharp, and is usually turned a little downwards so as to fit closely against the anterior edge of the scapula. The inner (median) borders (sym.) of the clavicles are thickened and meet in the middle line at least in the anterior third. Behind this there is a notch on each clavicle, which in a former paper was described as showing either that a blood-vessel passed between the two bones at this point or that a rudimentary interclavicle was present. ‘The latter explanation seems to be the correct one, though it is by no means sure that this element always ossified. In the specimen figured on Pl. X. fig. 2 the notch is well developed, but the clavicles behind it seem to have met in the middle line, so that the interclavicle, if present at all, must have been very small. In the specimen shown in text-fig. 88, on the other hand, there is a distinct interclavicle, the anterior slightly forked end of which fits into the notches in the clavicles, while its thin posterior prolongation lies between them in the middle line for the posterior two-thirds of their length. The occurrence of this rudimentary interclavicle in Cryptocleidus, like the existence of rudimentary clavicles in Murenosaurus, shows that both these genera were probably derived from some form in which, as in Tricleidus, both clavicles and interclavicles were well developed. The coracoids (cor., Pl. X. fig. 1; text-figs. 87-89) are very large, and, so far as concerns CRYPTOCLEIDUS OXONIENSIS. 179 their anterior region, massive bones. ‘Their form will be best understood by reference to the figures. ‘The anterior median prolongations of the coracoids towards the scapule are almost semicircular in section, the diameter of the semicircle being represented by the symphysial face. This portion of the bones forms a marked projection below the rest of the ventral face (Pl. X. figs. 14, 1c). Behind the scapular processes the concave anterior border of each bone is thin and sharp, forming the hinder boundary of Text-fig. 88. Adult shoulder-girdle of (?) Cryptocleidus owoniensis, showing the rudimentary interclavicle : A, from above ; B, clavicles and interclavicle from below (outer surface). (R. 3538, about } nat. size.) cl., clavicle ; cor., coracoid ; gl., glenoid cayity ; 7.cl., interclavicle ; sc., scapula. the coraco-scapular foramen. External to this the bone is greatly thickened and bears the facets for the scapula and the glenoid surface, these making an angle of about 135° with one another: the scapular facet is triangular and is greatly roughened by pits and ridges ; the glenoid facet is half an oval, the short diameter being the line of junction 2a 2 180 MARINE REPTILES OF THE OXFORD CLAY. with the scapular facet, its surface is gently concave and nearly smooth. Internal to these facets the bone is greatly thickened to the median symphysis, the form of which is shown on Pl. X. fig. 1¢. Behind this the bone thins greatly, especially towards the median line, and the straight symphysis is only interrupted at about a third of its length from the hinder end by a small foramen ( for.) which is present in several specimens examined and probably transmitted a blood-vessel. The lateral borders, which are deeply concave, are somewhat thickened and terminate posteriorly in the postero- lateral processes (p.é.p), which in adult individuals in this genus are strongly developed, and project outwards and backwards; they terminate in an oval concave surface, which was no doubt capped with cartilage during the life of the animal, and this cartilage Text-fig. 89. Immature shoulder-girdles of Cryptocleidus owoniensis: A, a very young specimen with clavicles restored (R. 2416); B, an older example. (About + nat. size.) cl., clavicle ; cor., coracoid; gl., glenoid cavity ; sc., scapula. was continuous with the fringe of that substance which continued backwards the some- what irregular posterior borders of the bones. The visceral surface of the united coracoids is concave from side to side, and, though slightly convex from before backwards in the region of the symphysial thickening, is strongly concave in that direction in its posterior two-thirds, The above description refers to the shoulder-girdle of old individuals in which ossification is in an advanced condition. Fortunately the Leeds Collection contains shoulder-girdles of individuals of various ages, so that it has been possible to give an CRYPTOCLEIDUS OXONIENSIS. 181 account * of the development of this part of the skeleton, and a short description of the growth-changes is appended. In the young shoulder-girdle (text-fig. 89,A) the scapula is already triradiate in form, but the dorsal and, more particularly, the ventral rami are very imperfectly developed. ‘The two bones did not meet in the middle line, and there is yet no trace of the extension backwards of the ventral rami to meet the anterior prolongations of the coracoids. The clavicles articulate by their outer ends with the anterior border of the scapule and meet in a median symphysis, the structure at this stage being essentially the same as in the primitive Sauropterygia. The ventral surface of the clavicles was exposed, the scapula, or at least their ossified portion, not yet extending beneath them. In the successively later stages the ventral rami of the scapule grow inwards and forwards beneath the clavicles and at the same time are gradually prolonged backwards in the middle line towards the gradually developing anterior median prolongations of the coracoids (text-fig. 89, B). Finally, the condition described above is attained, the scapule extending almost completely beneath the clavicles and meeting in a median symphysis, which through the backward continuation of the bone becomes continuous with the symphysis of the coracoids, the coraco-scapular foramina being completely separated from one another (text-fig. 87). In the coracoids the chief growth-changes that take place are the prolongation forwards in the middle line to join the scapule, and the formation of the prominent postero-lateral processes (Pl. X. figs. 1 a, 16, p.e.p.; text-figs. 87,88). As has already been pointed out, the prolongation inwards of the scapule beneath the clavicular arch causes the latter to become functionally unimportant or useless, and consequently in the family Elasmosauride it is extremely variable in form and is met with in all stages of reduction. Fore Limb.—The fore paddle (text-figs. 90, 91 A) is chiefly remarkable for the great expansion of the distal end of the humerus in the adult: in the young this charac- teristic is not seen and only develops with advancing ossification (text-figs. 90, A-C). The head of the humerus is strongly convex in full-grown individuals, and its roughened surface shows that it was capped with cartilage. The tuberosity (¢w.) is strongly developed and forms a quadrate prominence on the postero-superior surface at the upper end of the bone; its upper cartilage-covered surface is continuous with that of the head of the bone, or in individuals of advanced age separated from it by a slight concavity. The anterior border of the tuberosity is continued down as a strong ridge on the upper part of the shaft, while the posterior border forms a prominent angle continuous below with the posterior border of the shaft. The shaft is oval in section and increases gradually in width towards the distal expansion ; its postaxial border * “On the Development of the Shoulder-girdle of a Plesiosaur (Cryptocleidus ovoniensis, Phillips, sp-),” Ann, Mag. Nat. Hist. [6] vol. xv. 1895, p. 3388, 182 MARINE REPTILES OF THE OXFORD CLAY. bears a roughened surface for muscle-attachment, and its ventral surface towards its upper end is likewise roughened (text-fig. 90, C, m.r.). The great distal expansion bears two large facets for union with the radius and ulna: of these surfaces that for the radius is the larger and is slightly concave, while that for the ulna is usually straight or very gently concave; the two make a very obtuse angle with one another. In some cases there may have been a small postaxial accessory ossicle articulating with the humerus. The radius (r., text-figs. 90, 91) is very large and of peculiar form: this shape being already marked at a very early age (text-fig. 90, A, B) is of considerable value in determining as belonging to this genus very young individuals in which most of the Text-fig. 90. Proximal portions of fore paddles of Cryptocleidus owoniensis: A, left fore paddle of very young individual (R. 2417, 4 nat. size); B, left fore paddle of older individual (R. 2416, 1 nat. size); CO, right fore paddle (ventral face) in which ossification is complete (R. 2412, 1 nat. size). In A the carpals are not in place. a.0., accessory ossicle; h., head of humerus; hum., humerus; int., intermedium ; mc.V., fifth metacarpal ; m.r., ridges for muscle-insertion ; r., radius; rad., radiale; w., ulna; uln., ulnare; tu., tuberosity of humerus. other characteristics are not yet developed. Inthe adult the humeral border is slightly convex ; its outer (preaxial) border is greatly elongated, convex above and concave below. The ulnar border is short and straight or slightly concave, while the distal border is also slightly concave and articulated with the radiale, with, in some cases, a short CRYPTOCLEIDUS OXONIENSIS, 183 surface of contact with the intermedium. In one case (a.0., text-fig. 91, A) it articulated externally by a short facet with a peculiar phalange-like bone lying on the preaxial side of the radiale (?a prepollex). The wlna is much smaller than the radius; its humeral facet is gently convex, its inner (radial) surface is likewise convex, while distally it bears two flat facets making an obtuse angle with one another, for the intermedium and ulnare. The proximal row of carpals consists essentially of the radiale, intermedium, and ulnare, but in many specimens there is a tendency to develop accessory ossicles on the preaxial or postaxial border or on both, and there is considerable variation in the form and manner of development of these accessory ossicles. In some cases it seems as if the increased width of the radiale and ulnare, consequent upon the expansion of the paddle, led to a tendency to ossify from more than one centre, and this may bring about the total or partial separation of the preaxial portion of the radiale and the postaxial part of the ulnare (text-fig. 90, C); often there is a want of symmetry in the paddles of opposite sides. In some cases it seems as if instead of a separation of the ulnare into two elements there has been a fusion with an originally distinct element (text-fig. 91, A, a.o.) corresponding to the bone called the pisiform in some of the paddles described above (e. g., Tricleidus, text-fig. 77, p. 160). The intermedium articulates mainly with the ulna, the facet for the radius being small and perhaps in some cases absent. The distal carpals are three in number, the first (preaxial) articulating with the radiale (and anterior accessory ossicle if present), the second with the radiale and intermedium, the third (postaxial) with the intermedium and ulnare, while its postaxial border may have a facet at its proximal end for contact with the fifth metacarpal (text- fig. 91, A), which, as usual, articulates directly with the ulnare. ‘The first metacarpal, which is flattened like the carpals, articulates only with the first distal carpal ; the second metacarpal, which is cylindrical, with the second carpal; the third has two facets, one for the second the other for the third carpal, with which also the fourth metacarpal articulates ; the fifth metacarpal has already been referred to. The phalanges are cylindrical and somewhat constricted in the middle; the faces by which they articulate with one another are nearly flat. ‘The number of phalanges in the different digits is not known. Pelvis (Pl. X. figs. 3, 3a, 535; text-fig. 92)—The structure of the pelvis in this species has been described in some detail in Geol. Mag. [4] vol. iii. p. 145. It is composed of the usual three pairs of bones. The pubis (pu., Pl. X. figs. 3, 3a, 3d) is relatively much wider from side to side and shorter from before backwards than is the case with the pubis of Murenosaurus (cf. text-fig. 65, p. 116). The anterior convex border is thin, except at its outer angle, where it is thickened and produced into a short process (@.¢.a), separated by a notch from the median part of the border. This antero-external process of the pubis is also present in Murenosaurus, though less 184 MARINE REPTILES OF THE OXIFORD CLAY. prominent ; it has been suggested that it may be homologous with the lateral process of the Chelonian pubis. The symphysial surface (Pl. X. fig. 3 b) is deeper and at the same time shorter than in Murenosaurus; its form will be best understood from the figure. The curvature of the symphysial border shows that even when ossification was far advanced, as in the specimen figured, the actual contact of the two bones, if present at all, must have been short, they being separated anteriorly and posteriorly with wedges of cartilage: the Text-fig. 91. Proximal portions of fore and hind paddles of Cryptocleidus oxoniensis: A, right fore paddle (upper surface) ; B, left hind paddle (upper surface). (R. 2860, about 4 nat. size.) a.0., accessory ossicle; 7., fibula; fem., femur ; 7id., fibulare; h., head of humerus and femur ; hum., humerus ; int., intermedium ; me. V., fifth metacarpal ; mt. V., fifth metatarsal; 7., radius ; rad., radiale; t., tibia ; tib., tibiale ; t., trochanter of femur ; tu., tuberosity of humerus; u., ulna; wln., ulnare. anterior cartilage was probably small and continuous with the cartilage bordering the front of the pubis; the posterior cartilage, on the other hand, was thick and probably was continuous with that uniting the symphysial surfaces of the ischia, thus completely separating the foramina obturatoria from one another. CRYPTOCCLEIDUS OXONIENSIS. 185 The outer and posterior borders are both concave, thickening towards the massive articular region. This bears two facets, that for the ischium being nearly semicircular, and the diameter of the semicircle forming the line of division between it and the acetabular surface. This latter makes an angle of about 145° with the ischial surface and is slightly concave; there is no contact between the pubis and ilium. The isehiwm (isc.) is of the usual hatchet-head form. The anterior portion of its symphysial border is thickened and bears a deep symphysial surface (PI. X. fig. 5 4), behind which it thins rapidly, thickening again a little towards its posterior angle. The ischial symphysial surface was separated in front by a pad of cartilage, which, as already mentioned, was probably continuous with that between the pubes; probably there was also a small posterior cartilage. ‘The neck of the ischium is comparatively narrow and depressed in section; towards the articular surfaces the bone becomes greatly thickened; there are three facets—one, looking almost directly forwards and semicircular in outline, for the pubis, a median one rectangular in outline and slightly concave forming the middle and greatest part of the acetabulum, and a posterior one looking backwards and outwards but only a little upwards, for the ilium, which slopes backwards much more than in Murenosaurus; the surfaces for the pubis and ilium are roughened for cartilage, the acetabular surface is smooth. The dium (il.) is a stout slightly curved rod of bone. Its lower end is greatly thickened and bears a nearly flat oblique oval surface, the inner two-thirds of which unite with the iliac facet of the ischium; the remaining third, making a slight angle with the rest, forms the posterior wall of the acetabulum. ‘This surface was covered with cartilage, which also extended up on to an angular projection marked e. in the figure (Pl. X. fig. 5 a). The middle part or shaft of the ilium is contracted and oval in section; it is curved, the concavity being anterior, and on the middle of its posterior border there is a small angular prominence probably for the attachment of muscles. The upper part of the bone is compressed and somewhat like the blade of an oar ; the anterior border of this expanded region is thin and sharp, the posterior thick and rounded. ‘The inner face of the upper end of the ilium is flat and, except in very old individuals, shows little or no trace of any union with the sacral ribs, to which probably it was attached loosely by ligaments. In the pelvis, as a whole, it will be noted that the ilium is greatly inclined backwards, but this does not necessarily represent its exact inclination to the vertebral column, because probably the whole pelvis was inclined downwards and forwards. The pubes and ischia of opposite sides do not make a distinct angle with one another, their median portions being almost on the same plane, and the visceral surface of the ventral portion of the pelvis is only slightly concave from side to side. From before backwards the line of the symphysis is convex on the visceral side, its highest point being only about 85 cm. below a straight line joining the middle points of the acetabular cavities. 2B 186 MARINE REPTILES OF THE OXFORD CLAY. The pelvis of a young individual (R. 2417) is shown in text-fig. 92. None of the bones are completely ossified, but the immaturity of the pubis is most striking, there being no trace of the antero-external angle or of the articular surfaces; these must have been still cartilaginous. The ilium is less expanded at the ends, while the ischium, though resembling that of the adult more than is the case with the other elements, has its articular surfaces rounded and not sharply defined. Hind Limb.—The femur (text-figs. 91 B, 93, fem.) is not greatly expanded at its distal end like the humerus. Its proximal end in fully ossified specimens bears a convex head, oval in outline, the surface of which is roughened and was covered with cartilage. The trochanter is large and prominent; its cartilage-covered upper end is continuous with that of the head; anteriorly and posteriorly it is marked off from the Text-fig. 92. Immature pelvis of Cryptocleidus owoniensis, from above. (R. 2417, 4 nat. size.) acet., acetabulum ; 2J., ilium ; isc., ischium ; 02¢,f., obturator foramen ; pu., pubis. upper part of the shaft by longitudinal grooves, of which the posterior is the more strongly marked; its outer surface is raised into ridges and roughened for muscle- attachment. ‘The shaft is oval in section and bears on its ventral face and posterior border strong rugosities for muscle-attachment. The distal end, as already noted, is much less expanded than is the case with the humerus; it articulates only with the tibia and fibula, the facet for the first being slightly concave, that for the latter nearly flat. The ¢zéra (text-figs. 91 B, 93, ¢.) has a slightly convex femoral border; its preaxial edge is also convex ; its inner (postaxial) border for union with the fibula is sometimes CRYPTOCLEIDUS OXONIENSIS. 187 notched. Distally it bears a long facet for the tibiale and a shorter one, making an angle of about 130° with the last, for the anterior part of the intermedium. The fibula (text-figs. 91 B, 93, f.) has a long straight femoral edge, short and somewhat convex preaxial and postaxial borders, and distally two subequal facets for the inter- Text-fig. 93. Left hind paddle of Cryptocleidus ovoniensis: A, upper surface ; B, proximal end of femur. (R. 3708, 4+ nat. size.) Ff. fibula; fem., femur; jid., fibulare; %., head of femur ; iné., intermedium ; ¢., tibia; ¢d., tibiale ; tr., trochanter ; I.-V., the five digits. medium and fibulare, making an angle of about 120° with one another. ‘The former of these facets is slightly concave. The tibiale (tib.) articulates proximally solely with 2B 2 Coats quu Sb anoqy) ‘sesuarnovo snprajaozdhia7 JO UOJOTOYS {4 JO WOLWLOySoL OUBULLUBATEIP-1ULag wegen t CaS, Se TT nen ined oa S iS \ Q Sia SRR Wp ms “PG SY-IXOT, u CRYPTOCLEIDUS OXONIENSIS. 189 the tibia; distally it carries the first distal carpal and has a short surface for the preaxial part of the second. In sume specimens (see text-fig. 91, B) there is a small accessory ossicle (@.o.) articulating with the preaxial border of the tibiale near its proximal end. The presence of this accessory ossicle may indicate that although the hind paddle is little expanded in comparison with the fore paddle, there is nevertheless a tendency towards such an enlargement. ‘The intermedium (int.) has a short facet for the tibia and a longer one for the fibula, with which also the fibulare ( 77d.) articulates ; the postaxial border of the last-mentioned bone is thin and convex. Of the three distal carpals, the first articulates with the tibiale alone, the second with the tibiale and intermedium, the third with the intermedium and fibulare, while its postaxial border is in contact with the fifth metatarsal, which, as usual, articulates directly with the fibulare. Ail the metatarsals are flattened, and the same is the case in a decreasing degree with the first two or three rows of phalanges. The more distal phalanges are more cylindrical and constricted in the middle till near the ends of the digits, where they become more flattened again; the terminal phalanges are mere nodules of bone. The articular surfaces of the phalanges are somewhat convex and the articulations between the successive phalanges of one digit usually alternate more or less regularly with those of the adjacent digits. In one nearly complete hind paddle (R. 5705, text-fig. 93) the numbers of the phalanges in the digits from the first to the fifth arero, Oy la, Wass A semi-diagrammatic restoration of the complete skeleton of Cryptocleidus oxoniensis is given in text-fig. 94. This drawing is made almost entirely from the skeleton of the adult (R. 2860) mounted in the Gallery of Fossil Reptiles, British Museum (see Frontispiece), and used as a basis for the description of the skeleton given above. ‘The chief points of difference from Murenosaurus shown in this diagram are the relatively larger head, shorter neck, and more expanded fore paddles ; the important differences in the shoulder-girdle and pelvis cannot be shown in a side view. R. 2860 (Leeds Coll. 14). An almost complete skeleton of a nearly adult individual (Frontispiece). The skull is broken and incomplete, the parts preserved being :—basioccipital (Pl. IX. figs. 1, La), exoccipital, opisthotic (Pl. TX. figs. 1, 1 a), supraoccipital (Pl. LX. figs. 1, 12), basisphenoid (PI. LX. fig. 2), parietals, frontals, quadrate and squamosal (PI. LX. fig. 3), premaxille (part), portions of pterygoids. The mandible is nearly complete. The vertebral column consists of the atlas and axis (text-fig. 78, C & D) and thirty other cervicals, two or three pectorals, twenty-one or twenty-two dorsals, three or four sacrals and twenty-two caudals, the distal portion of the tail being wanting ; most of the vertebrae have their arches and ribs preserved: in the cervical and caudal regions the ribs in most cases have not yet fused with the centra. There are (as mounted) six rows of ab- dominal ribs, each consisting of a median element and three lateral pairs. The shoulder- girdle is complete; the fore paddles (text-fig. 91, A) want some of the phalanges ; the pelvis is complete ; the hind paddles (text-fig. 91, B) want the proximal end of the right femur, some tarsals, and phalanges. This specimen, which is mounted in the Gallery of 190 MARINE REPTILES OF THE OXFORD CLAY. Fossil Reptiles, is probably the most nearly complete skeleton of an Hlasmosaurian Plesiosaur known : the description of the skeleton given above and the restoration in text-figure 94 are founded mainly on it. The dimensions (in centimetres) of this specimen are :— Total length of the skeleton as mounted . . about (11 ft.) 335-0 Skull (Pl. IX. figs. 1-3): Length of basioccipital BSA cg ans Onn, 5 35 Width of basioccipital at pterygoid processes . . . . . 4:6 Transverse diameter of occipital condyle . . . ie 2-4 Height from bottom of basioccipital to vertex of skull oh 86 Length of paroccipital processes . . . . . (approx.) 2:6 Width of articular surface of quadrate . . . . ... 31 Vitielice lpia Fourth Tenth Fourteenth Twentieth Twenty-Afth Thirtieth First Fifth Tenth Fifteenth First First ‘Uwelfth (figured). cervical. cervical. cervical. cervical. cervical, cervical. pectoral. dorsal. dorsal. dorsal. sacral. caudal. caudal. Length of centrum in mid - ventral line, creas 42 2:3 27 oo 33 34 35 35 40 42 42 Soo OU Width of neeieioe end of centrum, 2°8 2°8 34 39 45 5:0 es ars or ric rae ty 61 47 Height of posterior end of centrum. 2°1 2°2 26 3:2 36 40 ate ee a es 50) ares ee 37 Height to top of neural spine . S'1 61 76 $9 10°83 11:0 He ar LD mes el Oe O mao 89 Owing to the mounting of this skeleton, complete measurements of the vertebree cannot be taken. Shoulder-girdle : Greatest length of combined scapula and coracoid . . . . 52:1 - ss SCAPULAZ Bromus SN si eeee Te oc 25:0 a A Coracoid {5 \.. &i ges 2 i ts Uapnios: ) 39:0 Width of united coracoids at posterior angle of glenoid cavities . . . : : 349 Width between outer *Netide of ike nestor: -exter fal arales of COTaCOIdS a mtegae uel as Sl OMmeai na a or cm. Use) Width of united cotacoide:s at narrowest pomt . . . . . 267 Length of clavicle (median border). . . . ... =. ~. 1210 Width of clavicle (median border to outer angle) . . . . 10:6 Fore limb : amerusi:vlenothec. sails eye ey Moench Monnens aime oS: (iameter, of head saan co wei get ae eens el CS) greatest width of upperend . .... . . 109 width of shaft at narrowesb . . . ... . 70 width-of distalexpansion . . . . . .. . 21:4 Radius: length of preaxial border . . . . . . . . . 105 width of surface for humerus. . . . . . . . 109 Ulna: greatest length SOL OO oo ean ge He SoU) Co COMMIS kok oOo Go oh CRYPTOCLEIDUS OXONIENSIS. 191 Pelvis : Miamyplen sth hw mvpnteh meee oil cht 17-2 greatest width at upper end 7:3 53 55 lower end 56 Pubis: greatest length . Be ICD” Geena, esi: ss width (from antero-external angle to sym- physialiborden)) Gy sy sasceuey cu ne eee cUco Wwidthvotanticularshead@ males: tals ves BlO.0) width between the antero-external angles of the two PUbeSame eee se eer. et hy eee te kr OOLO) Tschrumensreatestawidui ee set, a)! eo fee eae LOol! widthioMarhicularsheadumet ssi t beeen Utere aGcO length ofexpanded purtion. . . .. . . . 18-4 Widthvo tne cles Ruste gies) fiw, pamncmiic) (awa eH Os Hind limb; Bemur: silent hts, yesh ape We Maso e RM NOM Pico oS) rete O diameter of head. . . Sal cae UP aio coh ono peeich pe sO greatest width of upperend . . . . ... . 96 width of shaft at narrowest . . . . . . . . 58 widthio£ distal expansion 2) 0) 1° 7h apes dG 'O Mibiaiorcatestalenethin.. a. celta css Galea aN cr Oa % WAG reeech s Glin p ohana hg Rey ony MeO Hibulsemereabesh Len ot hie yee yess tecc ier aieey een pee eOoL os Wahine eth ry, ee ar A ey Sten Geyirowe Geo R. 2862 (Leeds Coll. 27). Imperfect skeleton of a large adult individual. The parts preserved are :— basioccipital, part of basisphenoid, atlas, axis and twenty-eight other cervical vertebra, about twenty-three pectorals and dorsals and thirty sacrals and caudals ; scapulee, coracoids, clavicles, fore paddles wanting only a carpal and some distal phalanges, portions of ilia, ischia and pubes, hind paddles wanting some distal phalanges, ventral ribs (text- fig. 86) embedded in matrix and showing their relations to one another and to the pelvis. In a short interval between the ventral ribs and the pubes there is a peculiar wrinkled surface which may represent a portion of the abdominal wall. This skeleton is that of a large and old individual, in which ossification is very far advanced. This is shown by the fact that all the neural arches, and, in the cervical and caudal regions, the ribs, are fused with the centra. The ossification of the limb-bones also is very far advanced, the heads of the humeri and femora being strongly convex, while the upper end of the tuberosity in the humerus and of the trochanter in the femur are much more clearly defined than in most specimens. In the shoulder-girdle all the sutures between the coracoids and scapule are obliterated, and the clavicles, which are very closely adherent to the visceral face of the scapule, are fused with one ancther in the middle line. The specimen is also interesting as showing the exact arrangement of the elements of the ventral buckler (text-fig. 86). The general relations of the posterier rows cf ventral! ribs to the peivis can also be made out, though some displacement may have taken place in the course of the flattening out that the carcass has undergone. In the present 192 MARINE REPTILES OF THE OXFORD CLAY. condition it appears that the posterior row of ventral ribs lay below the anterior portion of the pubes. It seems probable that in life they were just in front of them, and that the plastron with the shoulder-girdle and pelvis formed a complete bony armour on the ventral surface of the body. The form and arrangement of the ventral ribs with regard to one another in this specimen have been described above. In the fore paddles the ulnz are very wide, and the postaxial portion of that on the left side, which is bent a little downwards, seems to be on the point of separating off from the main body as an accessory ossicle, traces of the line of division being clearly visible. On the right side this part of the bone is already separate, though it remains in very close contact with the remainder. Of course, these separate, or imperfectly separate, portions of the ulne may be interpreted also as accessory ossicles on the way to fusion with the main bone. The dimensions (in centimetres) of this specimen are :— Basioceipital:lenpth @2 =. y. . 2 = s. ees es e(approx:)) | 53:8 width at pterygoid processes . . . . . - . OF transverse diameter of condyle . . . ... 26 SVerlobrac Musreececcseeee essa. eeceaneee eee ene Fe pate Length of centrum in mid-ventral line . . . 5:0 32 Width of posterior end of centrum. . . . . dr 4-0 Height 55 FF _ Sa sys aemetec iD 31+ The remainder of the cervicals and the other vertebre are too much crushed to supply measurements of any value. Humeruseslengbh) fo. a vas Meen ee ss) if Wels ise epee ar yey Noell lonsidiameteromhcadw sa cu teu dete eae Op LOsO short Py; a Ca me CCT Cn eee oeaere Pree m4 LOK, greatest width at upperend . . ... .. ... 12:8 width of shaft at narrowest point . . . . . . 76 width of distalexpansion . . . . . . .. . 248 Radius: length of preaxial border. . . . . . . . . . 183 5 humeral sborder wa) ia ue eles oe een ele S Wings Nengthy.. oc ce se" icc iseiis lem Ss cea eh werkt ee Ae eee mie width with postaxial portion . . .... .. . 133 »» Without postaxial portion . . . . . . . . 12:0 The pelvic bones are too imperfect to be measured. Femur: length eM ROPER ot ake oF te 313 diameter of head (exaggerated by crushing). . . . 95 greatest width at upperend ..... .... . . 102 width of shaft at narrowest . . . ... .. . 68 a distalsexpansiOne aman mercer eee OZ shibia:, sreatestilengthi |e) (20's ees a as Sen cece GIS os WAdGH sy. eh, Ws! Re eee ey eee 0 :() Hibula)-tgreatest length a.m as tees eee rae eee nee = WIC bhi © SORIA inate mee ys Meeps tai tece anO-() Length in mid-ventral line of plastron so far as preserved. . . 24:5 ‘Waidthi(exaggerated\bytcrushing)) cs. 2s) se see wes) ee See CRYPTOCLEIDUS OXONIENSIS. 195 R. 3730 (Leeds Coll. 144). Imperfect skeleton, including skull (imperfect posteriorly, Pl. LX. fig. 7), atlas, axis and fifteen other cervical vertebree, the centra of the posterior caudal vertebrze still united with one another ; six ribs (cervical, dorsal, and sacral), left clavicle, radius, ulna, and the greater part of one fore paddle, together with a number of odd bones of the other ; distal halves of both femora, tibie, fibulxe, and the greater part of the other bones of both hind paddles. The skull (Pl. LX. fig. 7) is very much broken and is incomplete posteriorly ; the upper and lower jaws are crushed together and it can be seen that the long sharp teeth of the upper and lower series alternated throughout: there are about 24-25 lower teeth on each side. In the cervical region the sutures between the centra and the neural arches and ribs are stillopen. The terminal caudals do not seem to have borne ribs, but the neural arches and chevrons were present on all but the last, or perhaps two last vertebrze, although in this specimen they are for the most part represented only by the facets for their attachment. The skull is too much crushed to give any reliable measurements. The dimensions (in centimetres) of other parts of this specimen are :— Vier Le bree saccsncsscverseacceuscsetearccactss eee aoe tH ea Length of centrum in mid-ventral line . . . .) 42 26app. 3:5 3°5 app. Width of posterior face of centrum. . . . . . 3:0 31 45 5:3 Height of posterior face of centrum . . . . . By 2°3 a5 4-1 app. The length of the united centra of the ten posterior caudal WELLE DES ey a Lal ail voice nse ah ayia sia vem Ay SP A oe enna Te che) Clavicle : length of outer border Tee SY feo ney ee se s symphysial border. . . . . . (app.) 18-1 Radius); length of preaxial borders; iain a 2 = 3 LL % humeral: borderwineusancurdie heels y a) a Hose Femur: width of distal expansion . 2.) §. 2 4 5 2) 6207 Tibia: length. F HE MACE AT Poi crLerMitce fen stn ae logo) WIG es, tt, ey steno a UaReweasn hunt Ee, cuore eae qe Ogee Fibula: length 5-4 width 86 B. 2412 (Leeds Coll. 31). A great part of the skeleton of a large adult individual. The parts preserved are :—eighteen cervical vertebrae, mostly with fused arches and ribs (text- figs. 79, 80), two pectorals, a dorsal, two sacrals (text-fig. 83), and twenty-two caudals (text-fig. 84); ribs, abdominal ribs; shoulder-girdle somewhat imperfect, especially posteriorly, and with only portions of the clavicles preserved; right fore paddle (text- fig. 90, C), left ulna, incomplete pelvis wanting one ilium, hind paddles wanting some tarsals and phalanges. This specimen is the type upon which Professor Seeley founded the species Cryptocleidus platymerus : he figured the shoulder-girdle, clavicles, and fore paddle in Proc. Roy. Soe. vol. li. (1892) pp. 145-148, text-figs. 13-15, the genus (or, as Professor Seeley in some places calls it, subgenus) Cryptocletdus being founded for its reception. The shoulder-girdle differs from those of R. 2616 and R. 2860 in being a little narrower and more lightly built, although the size and degree of ossification of the 2C 194 MARINE REPTILES OF THE OXFORD CLAY. humerus show that the animal was advanced in age ; the differences, however, do not scem to be sufficiently great to warrant separation as a distinct species. Length of centrum in mid- ventral line . The dimensions (in centimetres) of this specimen are :— Width of posterior face of centrum . Height of posterior face of teri Posterior Pec- First Anterior Vertebres .....-rs..-..04 Bes seein ieee Aoveal! Sacrals. caudal. FARR Cacti te aaa fig’. fiz’. A DAE Oey BUSH) aerehi ois) 48 4147 3:9 36 3-5 . 38338 42 46 5 Gor roel TO eOrtG:9 ca ee 252:512:9 «3:3 8Brb) F452: WAST 157 (5:8) 10:0) b1o: A496: 254 centrum . Height to top of neural spine . (Sits S2} 27) Distance between the outer ends of the sacral ribs 21°2 cm. Shoulder-girdle (approximate only) : Scapula: greatest length 27:2 length of median border : 19:0 Coracoid: length from middle of glenoid cee to tip of postero-external process ; . 35-4 width of the united bones at the fades nals of the glenoid cavity oe ae 38°'8 width of the combined bones at narrowest . 30°3 Humerus: length 362 width of head . ye ees 9-0 7 upper end with tuberosity . 12°6 43 shaft at narrowest 7:8 a distal expansion . 27-0 Radius: length of preaxial border . 145 5 humeral border . ys eee Poo 3 ulnar border . (approx.) 4:5 Ulna: length . 65 width . 14-4 Pelvis : Hium: length . Tomer oy ISRO) width of upper aa (approx.) 4:9 x lower end (erushed) (approx.) 7°6 Pubis: width from anterior angle of acetabulum to symphysis : 26°7 width of articular head . Cette: 125 Ischium: width from acetabular surface to symphysis . 24:3 width of neck th Femur: length 33°7 greatest width ae Head atic ee EZ width of upper end with trochanter (approx.) 10:8 if shaft at narrowest . 7:0 » distal eypansion . 20°7 Middle Posterior caudal. caudal. 3:1 3:0 2-8 56 4:7 AOVa Sl CRYPTOCLEIDUS OXONIENSIS. 195 Tibia: length of preaxial border. 78 x demoralaborderie | seeene woes vets 6 6, 102 Hibulayroreatestilensthe wee.) ty ca eset yee OST 3 WIth este onen spies ao foe! ce bee: Ceeeem fmeeluielD R. 2616 (Leeds Coll. 25*). Portions of a skeleton of a large adult individual. In this specimen the bones are quite uncrushed and not distorted : some are scored by deep scratches apparently made by the teeth of some predaceous reptile. The parts preserved are some cervical vertebree, mostly with the neural arches, which are just becoming fused to the centra ; one sacral and eight caudal vertebrae, the neural arches in most cases missing, the suture having remained open; some cervical ribs which had not yet united with the centra, and a number of abdominal ribs of great size and massiveness ; nearly perfect shoulder-girdle and pelvis, and a few odd paddle-bones. The shoulder-girdle has been described and figured in Ann. Mag. Nat. Hist. [6] vol. xv. (1895) pp. 335-340, figs. 1, 2; also figured on Pl. X. figs. 1, 1a,10,1¢. The pelvis described and figured in the Geol. Mag. [4] vol. iii. (1896) pp. 145-148 ; also figured on Pl. X. figs. 3, 3.4, 3b. The dimensions (in centimetres) of some parts of this skeleton are :— WWertebricgussscese ses paeeuce Sacral. ae bas L Length of centrum in mid- — A _ ventral line. . . 37 33 3:9 37 3°8 35 Width of posterior ead of centrum™=. 5)" Sse oro Ong 7:2 app. (ik 65 Height of posterior aiid of COMET UMA pike Mite 42 45 54 5°3 app. a3 5:0 Height to top of peur Spine eta May seu os PE Sh 6O}appl an 15:2 14:6 Shoulder-girdle: preatest length . . . ... =... . (G65 Clavicle: length of anterior border. . . . . . . . . 180 3 symphysial borders) a ss ee ep Scapula: greatest length . . . . 6. Dinka upoye length of median (eymiphyeial) Borden RLM Fre L358 length in straight line from median border to top Of dorsalvramusey, i) cf sree, omit as et length of glenoid surface. . . . . (approx.) 6-2 : Surface tor/coracoldy ean a) ses se Se GLO antero-posterior diameter of coraco-scapular Opening.) J. Ai co we eS 12-1 lateral diameter of coraco-scapular opening Gan = 11:0 Coracoid,:slenothie |) se) es caer a -6 NG NE o BRET width of united soricotdls at posterior anes of Glemord cayribyau. ur ttt cule we Cpe og o25 * The left scapula and clavicle were acquired by the Museum in 1892 and registered as R 1966, the remainder of the skeleton was received in 1895, 2¢ 196 MARINE REPTILES OF THE OXFORD CLAY. Coracoid: width of the united bones at narrowest. . . . 33°0 width between the outer ends of the postero-lateral PEOCESSES 3s Sh ce esc) 2 Johar COnicitsy MaRS Relyis;:;oreatestslengthe mien grs sits \ftenieme ce vey ver eet tOZO width between the antero-external angles of the pubes. 66-6 Ml iumy Mle mathew st Geer reten Wh mL eh, vo es, Sep esl ase are: atone Os0) Mid Chrofupperends Wriere no yeb.) Wee. 2 ele Oz Py lowersendneniene te cise ls) eke SRA Seg 282 () Pubis':"sreatest lengthen .uetottn ake) Je te ee E2568 » Width (from antero-external angle to sym- physialgborden) sean ciiseks geal auee BOA width of acetabular surface. . . . . (approx.) 87 depth of acetabular surface . 59 Ischium: greatest length of median expansion . . . . . 21:9 width from acetabular surface to symphysis. . . 223 widthiof articular heads = <3 2 j%:, 9 @ = 10:7 5 TOC leRPatgens hiss epee Hees wicca tera tom MOTTA antero-posterior diameter of obturator foramen . 10:2 R. 2417 (Leeds Coll. 36). A nearly complete skeleton of a very young individual. The parts preserved are :—skull (imperfect and much broken), the occipital portion and the basis cranii figured Pl. LX. figs. 4, 4a, 5; mandible (Pl. IX. fig. 6) which carried 20-22 teeth on each side, atlas and axis (text-fig. 78, A, B), 30-31 other cervical vertebrae, 2-3 pectorals, 21-22 dorsals, 3-4 sacrals, and about 21 caudals ; neural arches, cervical, dorsal, and caudal ribs, ventral ribs, chevrons; pectoral girdle and fore paddles (imperfect) (text-fig. 90, A), pelvis (text-fig. 92), and hind paddles (imperfect). This specimen was described and figured in the Geol. Mag. [4] vol. ii. p. 241, pl. ix. The ossification of the vertebral column is imperfect, the elements making up the atlas-axis complex being all free from one another, the neural arches nowhere fused with the centra, and all the cervical and caudal ribs free ; the cartilage-covered surfaces for union with the arches and cervical ribs are continuous with one another. The ossification of the posterior caudal vertebrz seems even less advanced than it is further forwards in the column. In the shoulder-girdle the ventral ramus of the scapula is little developed, and probably did not extend at all beneath the clavicle ; in the fore paddles there is little indication of the great distal expansion of the humerus character- istic of the adult, but the radius is already large and generally similar in form to that of the adult. The dimensions (in centimetres) of this specimen are :— Total length of the skeleton as mounted . . (about 6 feet) 184-0 Skull (Pl. IX. figs. 4, 4a, 5): length from occipital condyle to tip of snout. . . . 19:3 Fu OE NNN wishes ole on wiidao oo aH) width of basioccipital at pterygoid processes . . . . 3:0 transverse diameter of occipital condyle. . . . . . I9 Mandablesjlength) “7. ee gs) yee eee ee ene (@pprox.)a 23:0 5) QOL symphysis.) 21 auakew ton plone (@DPEOXs))saeoy CRYPTOCLEIDUS OXONIENSIS. 179 Atlas and axis (text-fig. 78, A, B): esr ea) 615 0.10 Gh po Vorto, aio en so, Ge on oll ts) widthtofsposterionaceonaxsy sii... 20 seele us Se yeh del oy Ded height to top of neural spine of axis. . . . . .... 388 Fifth Tenth Twentieth Twenty-fifth Thirtieth Anterior Posterior Anterior Middle NERA aeencte: cervical. cervical. cervical. cervical. cervical. dorsal. dorsal. caudal. caudal. Length of centrum in mid-ventral line. . 1:6 Ley 2-0 DEAL 2-1 2-4 2-4 1:9 16 Width of posterior face of centrum. . 24 2:3 3°6 3°9 4-2 4-4 4-1 4:2 34 Height of posterior face of centrum. . 1:7 2:0 25 3:0 (app.) 3:1 3°6 3°3 3°0 if Height to top of neuralspine ... . 42 4:8 61 76 74 76 6:6 6:2 4-2 Width between the ends of the trans- verse processes . : .. de Ae he oe 8:3 6-8 Shoulder-girdle : ClavicleMengthvofanteriorborder" sa) Fe. 10a a Symophiysialilborder! sy bi) ee, sa 027) Scapular: ereatest lengtht se et, SET eee nee wate (0) widthsotarticulamend a veren ens Melee a ee ces e4.9 Coracoid: greatest length. . . . . poe psd ketal width of each at hinder angle of Honoiil cavity . 11:0 umerusisslencthys tees) 211. Sas dst oat anche dm mene IHL) greatest width at upper ond vt sevuibectos otis i hale) bu neatss) width of shaft (at narrowest). . . . . ... 49 3 loweriendl (rests HA coh ee 1 A Batata fy UTED TI Radish lens the ayer ato ek cnn eae Unease ree el 20) WI EHU™ cy rieruce Coase oan Anon eta in onan ee ee aL Pelvis (text-fig. 92) : Tium : length. 10:3 width of upper etd 3°5 * middle of shaft. 2:2 a distal end 35 Pubis: length . 14:2 (app.) width . : 18-0 Ischium : width of articular Head 45 a neck : 2°9 » symphysial expansion . ries 85 1obes HEE eg en Fo oo 8 Go aly AHO (Gyn) WAS M CLE THO So aoe od ool ibe Gg Oa ae > Shaft (at narrowest). ° . 2) “ing. ve 74:3 lowered}? mainte: peti yaeds, Metre LOS ” R. 2416 (Leeds Coll. 37). Portions of the skeleton of an immature individual in which ossification is very incomplete. The parts preserved are:—the cantra of twenty-seven cervical 198 MARINE REPTILES OF THE OXFORD CLAY. vertebree, two pectorals, fourteen dorsals, and three caudals ; a few of the neural arches and cervical ribs, in all cases free from the centra; the scapule and coracoids (figured in Ann. Mag. Nat. Hist. [6] vol. xv. (1895) p. 341, fig. 3, B, with the clavicle restored, see also text-fig. 89, A), humeri, radii, ulne (text-fig. 90, B), pubes, ischia, and left femur. The dimensions (in centimetres) of this specimen are :— Vertebra............ UO? Middle Posterior pectoral, Dorsals. Caudal. ceryicals, cervicals. cervicals. Length in mid-ventral line. 1°38 19 2:3 2:4 26 2:7 3:2 34 33 2-4 Width of posterior face of Centres genes) AION p22) Kore 37 43 4:7 4:8 4:9 4:9 4:5 Height of posterior face of Cent numep ce tmtee eixe age) le 227 29 3:3 3°5 3°9 44° 4:5 3:3 Shoulder-girdle (text-fig. 89, A): Se MbeyawereMEN Seo 5 4g 0 a o ooo o NullbHy) Wwidthiot articular head) |. \e) 20s ae OI Pooh ) Hie Geer nalte: aa) loc oNeonam prlbe 6, © CHB) length in a straight line from the median angle to tiprotdorsalsramtusin Weay-tepaye) ee is) LOD Coracoid: length (from anterior internal angle to postero- external prOCess) im tau w bo bw co (approx. ) [IN DEX. [The asterisk denotes a figure on that page. | Baptanodon, 2, 3, 4, 9, 16, 31, 33, 35, 48._ Dolichorhyneops, 114. discus, 46. ELasMosauRID#, 77. Cimoliosaurus durobrivensis, 127. : Hatteria, 13, 16, 82. eumerus, 164. eurymerus, 164, 165. plicatus, 120, 127. snowil, 92. Cryptocleidus, 164. oxoniensis, 164. skull, 165. IcurHyoprTeryata, 1. Ichthyosaurus, 9, 14 *. acutirostris, 24. extremus, 54. zetlandicus, 29. basioccipital, 166. Leptocheirus, 2. basisphenoid, 166. exoccipital-opisthotic, 166. Microdontosaurus, 2, 3. supraoccipital, 166. Mixosaurus, 1, parietal, 167. cornalianus, 56. frontal, 167. Murenosaurus, 4, 77. squamosal, 167. skull, 78, 85 *, 88 *. quadrate, 167. basioccipital, 78, 79 premaxilla, 167. -| basisphenoid, 79, S0*. mandible, 167. parasphenoid, 80 *, $1 teeth, 167, 168. exoccipital, 80 *, Biess vertebral column, 168, 169*, 170 *, 171*, opisthotic, $1, 83 *. - 2 Wisin As prootic, 84. ribs, 174. parietal, 84. ventral buckler, 175 *. frontal, 86. shoulder-girdle, 176 *, 179 *, 180*. | postfrontal, 86. fore paddle, 181, 182 *, 184 *. postorbital, 87. pelvis, 183, 186 *. squamosal, 86. hind paddle, 184 *, 186, 187 *. jugal, 87. restored skeleton, 188. maxilla, 87. —— platymerus, 164, 165. prefrontal, $7. lo 9 Mureenosaurus (covd.). premaxilla, 88. yomer, 88. palatine, 89. pterygoid, 89. mandible, 89, 90 *. vertebral column, 92 *, 94*, 95 *, 96%, 98 *, 99 *, 100*, 101 *, 102*, 103%, 104 *, ribs, 105, 106*. shoulder-girdle, 106, 108 *, 109 *. fore paddle, 111, 112*. pelvic girdle, 115, 116 *. hind paddle, 112 *, 117. restored skeleton, 118 *. beloclis, 140, 144. durobrivensis, 127. leedsi, 120. platyclis, 134. —— plicatus, 120, 127. Nothosaurus, 107. shoulder-girdle, 108 *. pelvic girdle, 114 *. OPHTHALMOSAURIDA, 2. Ophthalmosaurus, 2. skull, 4-35 *. basioccipital, 5 *. exoccipital, 6, 7 *. supraoccipital, 7 *. prootic, 9, 10 *. opisthotic, 9, 10 *. stapes, 10*, 11. basisphenoid, 12, 13 *, 14 *, 15 *. parasphenoid, 15 *, 16. squamosal, 16, 17 * supratemporal, 18. quadrate, 18 *. quadrato-jugal, 19, 20*. jugal, 20 *. postorbital, 17 *, 21. lachrymal, 21 *, 22. nasal, 22 *, maxilla, 23 *, premaxilla, 24. parietal, 24, 25 *, 26 *. 204 INDEX. Ophthalmosaurus (cont. ). frontal, 26 *, 27. postfrontal, 27 *. prefrontal, 28. pterygoid, 28, 29 *. palatine, 29 *. vomer, 30 *. sclerotic ring, 31. mandible, 31, 32 *, 33 *, 34 *, 35 *. dentition, 36. hyoid arch, 36. vertebral column, 36, 37*, 39 *, 40%, 41*, 42 *, 43 *, 44*, 45 *, ribs, 45 *, shoulder-girdle, 46, 47*, 48*, 50 *, ol. fore paddle, 49, 52 *, 55 *. pelvic girdle, 56,57 *,.58 *, 59 *. hind paddle, 58, 60 *. restored skeleton, 62 *. icenicus, 61. Pantosaurus, 4. Picrocleidus, 139. — beloclis, 140. skull, 141. basioccipital, 141. quadrate, 141. vertebra, 141. shoulder-girdle, 140 *, 142. fore paddle, 143. femur, 144. —— sp., 146. PLESIOSAUBIA, 77. Plesiosaurus durobrivensis, 164, eurymerus, 164. leedsi, 120. oxoniensis, 164. —— plicatus, 120. (?) rostratus, shoulder-girdle, 108 ~*, Polycotylus, 91,113. Sauranodon, 2, 3. SavROPTERYGIA, 77. Shastasaurus alexandra, 47. osmonti, 47. _ Sphenodon, 156. Thaumatosaurus areuatus, 107. Toretocnemus, 2. Tricleidus, 91, 113, 149. seeleyi, 149. skull, 149. basioccipital, 150, 151 *, 153 *. exoccipital, 150 *. opisthotic, 150 *, basisphenoid, 150, 151 *. parasphenoid, 151 *, 153 *, parietal, 152. frontal, 152. maxilla, 152. premaxilla, 152. INDEX. 205 Tricleidus seeleyi (cont.). vomer, 152. pterygoid, 153 *, 154. palatine, 154, quadrate, 154, 155 *, squamosal, 155 *, 156. mandible, 156. teeth, 156. vertebral column, 156. shoulder-girdle, 157, 158 * fore paddle, 160 *. pubis, 161. hind paddle, 161. Trinacromerumn, 91. PRINTED BY TAYLOR AND FRANCIS, RED LION COURD, FLELT STREER, * Ver ee ities P tts | 42 WF ying een ah y | | , ooreumnanre elm? | ‘ ‘ < eC a Behe Vihar i | be Nia —- ’ ‘ : ca 03 ‘ % 4 4 = : = 2 oA Bie ty) ed « v; a es . \y fe = * u “ 4 : é Seth yo ew Te 7 Ve yay a) Nia ea Dag, | 7 k | af P A ~o . 2 ' & AN “4 ‘ ~ ‘< ij ; 7 y ‘ bee eet Lae do iia ar ‘ean w Fig. bo (Si) 13. 14. PLATE I. . Ophthalmosaurus icenicus, Seeley; transverse section of root of tooth : magnified twelve diameters. [R. 3013. ] . Ditto; transverse section of crown of tooth: magnified twelve diameters. [R. 3013. . Ditto; transverse section across the base of the crown of tooth: magnified twelve diameters. [R. 3018. | . Ditto; teeth: nat. size. [R. 3013. | . Ditto; anterior ends of the rami of the mandible of a young individual with teeth in situ: nat. size. [R. 2181. | . Ditto; three sclerotic plates: one-half nat. size. [R. 2740. | . Ditto; section across junction of two sclerotic plates showing the inter- locking suture: nat. size. [R. 2740. ] . Ditto; left quadrate, outer face : two-thirds nat. size. [Type specimen, R, 2133. | . Ditto; left stapes, from front: two-thirds nat. size. [Type specimen, R, 2133. | Ditto; basioccipital, from above: two-thirds nat. size. [Type specimen, R, 2138. | Ditto ; basioccipital, from behind: two-thirds nat. size. [Type specimen, R. 2183. | . Ditto; left opisthotic from above: two-thirds nat. size. [Type specimen, R 2133. } Page i) art., articular surface of quadrate. | _p-C-y pulp-cavity of tooth. boc.f., facet for basioccipital. | p.e.a., postero-external angle of quadrate. c., cement. p.v.c., channel for posterior vertical semicircular cond., occipital condyle. canal. d., dentine. q-f., facet for quadrate. e., enamel. qj. facet for quadrato-jugal. exo.f., facet for exoccipital. h.c., channel for horizontal semicircular canal. 7., ridge tor muscle-attachment. s., suture between sclerotic plates. n.c., floor of neural canal. sq.f., facet for squamosal. op.f., facet for opisthotic. st.f., facet for stapes. T CATAL. MARINE REPT. OXFORD CLAY. GM. Woodward deLetlith. OPHTHALMOSAURUS. Oh Es ey PLATE II. Fig. 1. Ophthalmosaurus icenicus, Seeley ; skull from above; the basioccipital is drawn from a second specimen: one-fourth nat. size. [R. 3702. ] 2, 2 a. Ditto; left prefrontal from above (2) and from below (2a): one-third nat. size. [ Leeds Coll. ] 3, 3 a. Ditto; postorbital, from outer side (3) and from inner side (3 @): one- third nat. size. [Leeds Coll. ] 4. Ditto ; left fore paddle from above: one-fourth nat. size. [R. 3702. ] Page 49 ‘These specimens were received after the descriptions were written, and therefore are not specially referred to in the text. art., articular bone. p., pisiform. b.oc., basioccipital. par., parietal. fr., frontal. p-for., pineal foramen. h., humerus. p.mx., premaxille. int., intermedium. js, Jugal. j.s., Surface for jugal. po.f-, postfrontal. po.f.s., surface for postfrontal. prf., prefrontal. 1., lachrymal. | g-, quadrate. Ls., surface for lachrymal. r., Tadius. me., maxilla. rad., radiale. n., nasal. H st., stapes. nar., external nares. | u., ulna. n.s., Surface overlapped by nasal. | uln., ulnare. I J , CATAL.MARINE REPT. OXFORD CLAY. PLATE I G.M. Woodward del.et lith. West, Newman imp. OPHTHALMOSAURUS. PLATE III. Fig. Page 1, la. Murenosaurus leedsi, Seeley ; basioccipital, exoccipital, and basi- sphenoid from below (1) and from side (1 a): half nat. size . . . 78-82 2, 2a. Ditto; anterior part of skull from below (2) and from above (2a): half Wate SIZE. 2 Fg We, WORN A os geste AE | Eee ete oR aD ccd artes Oe OG) 3, 3a. Ditto; mandible from above (3) and from below (3 a): half nat. size . 89-92 4, 4a. Ditto; tooth, nat. size (4), and portion of crown, five times nat. size(4a). 167 5,6; itto:;s teeth: matissize o.3 5 5 86 te oe. Bi ee ee GT All the specimens figured in this Plate are parts of the type skeleton, R, 2421. art., articular bone. oc.c., occipital condyle. boc., basioccipital. pa., parietal. bs., basisphenoid. | pas., parasphenoid. dent., dentary. pinf., pineal foramen. ewo., exoccipital. pmex., premaxilla. fr., frontal. ptf., facets for pterygoids. me., maxilla. spl., splenial. CATAL. MARINE REPT. OXFORD CLAY. PLATE Ii. A.H. Searle del.et lith. West, Newman imp. MURANOSAURUS LEEDSI. ren vpn ih Hi ie Ay cai ; moe) ae PN a ie a x one. re iy ue ; ar he i uy Wa ih RAIE ile: ; ei ee b Vesa ASAD inn ‘ea + ey son BG; } Cas : con co ihe ise mn ie d Hae ag ae ae es r PLATE IV. Fig. : Page 1. Murenosaurus leedsi, Seeley ; atlas and axis vertebra from left side: one- third nat. size 92 a. Ditto; anterior cervical vertebra from right side (2) and from behind (2 a): - one-third tate size.< (1 Te iyo ea. ala ws Lok Nie a gr er io eZ 3a. Ditto; middle cervical vertebra from right side (3) and from behind (3 a): one-third nat.siZe occ). ek ue eI eal i eee ee ee 4a. Ditto; posterior cervical vertebra from right side (4) and from behind/(4@): one-third mat. ‘size: 2). 93. . Ye et 5a. Ditto; anterior dorsal vertebra from right side (5) and from front (5a): One-third tat, S126.) ahi a ek es pe eee 6a. Ditto; anterior caudal vertebra from behind (6) and from below (6 a): one-third nat.size,s.c.0 4 goals le Gee ye em 7. Ditto; left fore paddle from above: one-fifth nat. size . . . . . . Ill 8, 8a. Ditto; left iium from outer (8) and inner (8 a) sides: one-thirdnat. size. 115 9. Ditto; left ischium, upper (visceral) surface: one-third nat. size . . . 117 10. Ditto; right hind paddle from above: one-fifth nat. size. . . . . . IT All the specimens figured in this Plate are parts of the type specimen, R. 2421. acet., acetabular surface. } is.f., facet for ischium. at., atlas vertebra. | n.sp., neural spine. ax., axis vertebra. pu.f., facet for pubis. a.z., anterior zygapophyses. /p.2., posterior zygapophyses. ch., chevron. | r., ribs (in fig. 7, radius). ch.f., facet for chevron. | rad., radiale. e.r., caudal rib, | 7f., facet for rib. cr.2., erista ilii. | sf., sacral facet. f., fibula. | sym., symphbysial surface of ischium. fem., femur. t., tibia. jib., fibulare. tib., tibiale. h., humerus. | t.p., transverse process. il,f., facet for ilium. u., ulna. int., intermedium. uln., ulnare. CATAL. MARINE REPT. OXFORD CLAY. PIATEY OVE A.H. Searle del.et lith. West, Newman imp. MURA NOSAURUS LEEDSI. Fig. We tk ae 8, 8a, 88. 9, 9a. 10. IES ema ra 12, 12 a, 12 b. PLATE V. Murenosaurus durobrivensis, Lydekker ; centrum of anterior cervical vertebra from behind (1), from right side (1a), from below (10): one-third nat. size . Ditto ; centrum of middle cervical vertebra from behind (2), from left side (2 a), from below (2 6): one-third nat. size . Ditto; centrum of posterior cervical vertebra from behind (3), from left side (3 a), from below (36): one-third nat. size . Ditto; pectoral vertebra from right side: one-third nat. size . . Ditto; centrum of dorsal vertebra from end: one-third nat. size . . Ditto; centrum of anterior caudal vertebra from behind (6), from left side (6 a), from below (6 4): one-third nat. size . Ditto; centrum of posterior caudal vertebra from behind (7), from below (7 @): one-third nat. size Ditto; right ilium, inner face (8), outer face (8 a), acetabular end (84): one-third nat. size Ditto; right ischium, inner (visceral) face (9), acetabular end (9 a): one-third nat. size Ditto; ventral face of interclavicle: one-third nat. size . Ditto; right humerus, ventral side (11), dorsal side (11a), proximal end (114): one-fifth nat. size Ditto; right femur, tibia, and fibula, vential side (12), dorsal side (12 a), proximal end of femur (124): one-fifth nat. size 130 All the specimens figured in this Plate, with the exception of the interclavicle (fig. 10), are parts of the type skeleton, R. 2428. The interclavicle (fig. 10) is from R. 2868. acet., acetabular surface of ilium and ischium. m.r., ridges for muscle-attachment on the ventral a.f., facet for pedicle of neural arch. faces of the humerus and femur. a.n., anterior notch of interclavicle. p.n., posterior notch of interclavicle. c., cartilage-covered crest of acetabular end of | pu,f., facet for pubis. ilium. ra.f., facet for radius. c.f, facet for chevron-bone. | r.j., facet for rib. sf., surface to which outer ends of sacral ribs cr.i., crista ilii, fs fibula. were probably united. h., head of humerus and femur. | sym., symphysial surface of ischium. il.f., facet for ilium. | t., tibia. is.f., facet for ischium. | t.p., transverse process. L.p., tuberosity of humerus. t., trochanter of femur. ‘ uf., facet for ulna. CATAL.MARINE REPT. OXFORD CLAY. PLATE V. £ 1 1a, 3} raf. ra.f. G.M. Woodward del. et lith. West, Newman imp. MURANOSAURUS DUROBRIVENSIS. ay uae ia co, wo rs Wee ey y PLATE VI. 135-6 126 Fig. 1. Murenosaurus platyclis, Seeley ; skull from above: one-third nat. size. [Type specimen, R. 2678. | 2. Ditto; mandible from above: one-third nat. size. [Type specimen, R, 2678. | 3. Ditto; anterior portion of shoulder-girdle from above : one-fourth nat. size. [Type specimen, R. 2678. | 4, Ditto; posterior cervical vertebra from right side: one-third nat. size. [Type specimen, R. 2678 | 5, 54,56. Ditto; anterior cervical vertebra from below (5), from left side (5a), and from front (5 2): one-third nat. size. [Type specimen, R. 2678. | 6, 6a. Murenosaurus leedsi, Seeley ; interclavicle from below (6) and from above (6a): one-fourth nat. size. [R. 3704. | a.n., anterior notch ef interclavicle. | par., parietal. a.z., anterior zygapophysis. | par.p., paroccipital process. boc., basioccipital. p-J., pineal foramen. cl., clavicle. | pme., premaxilla. cor., coracoid. po.f., postfrontal. d., dentary bone. p.orb., postorbital. d.sc., dorsal ramus of scapula. | p-p-, posterior process of interclavicle. F. irontal. | prf., prefrontal. gl., glenoid cavity. p.z., posterior zygapophyses. i.cl., interclavicle. #., cervical rib. i.s.f., interscapular fenestra. s.ang. & art., united surangular and articular. je Jugal. spl., splenial. me., maxilla. | sq., squamosal. nar., external nares. | sym., mandibular symphysis. n.sp., neural spine. , v.sc., ventral ramus of scapula. CATAL. MARINE REPT. OXFORD CLAY. GM Wocdward del.et lith. West, Newmen imp. 1-5, MURANOSAURUS PLATYCLIS 6 is LEEDS 2 : ah Hid a ay PLATE VII. Tig. 1. Picrocleidus beloclis, Seeley, sp.; posterior portion of right ramus of mandible: one-half nat. size. [Type specimen, R, 1965. | 2, 2a, 2b. Ditto; shoulder-girdle (2) and proximal portion of left fore paddle (26) from above: one-fourth nat. size; interclavicle (2a) from below: one-half nat. size. (‘Type specimen, R, 1965. | 3. Ditto; posterior cervical vertebree from right side: one-half nat. size. [Type specimen, R. 1965. ] 4, Ditto; posterior cervical vertebra from front: one-half nat. size. [Type specimen, R. 1965. ] 5,5 a, 56. Ditto; cervical vertebre from right side (5); front views of the vertebre marked a and 6 (5a, 5 8): one-half nat. size. [R. 3698. ] 6, 6a. Ditto; caudal vertebre from right side (6) and from front (6a): one-half nat. size. [R. 3698. | a,b, vertebre in figure 5 shown from front in 2.cl,, interclavicle. figs. 5 a, 5b. l.p., tuberosity of humerus. ang., angular bone. n.Sp., neural spine. a.p., anterior process of cervical rib. /p.z., posterior zygapophyses. a.z., anterior zygapophysis. 7., cervical ribs ; in fig. 2 6, radius. c.f., facets for chevrons. Page 141 141 141 141 r., r°., ribs of atlas and axis respectively. cor., coracoids. r.f., facets for ribs. d., dentary bone. | s.ang. & art., fused surangular and articular bones. d.sc., dorsal ramus of scapula. | u., ulna. gl., glenoid cavity. | v.sc., Ventral ramus of scapula. h., bead of humerus. CATAL. MARINE REPT. OXFORD CLAY. TCL & als wY S al. Ss ang. & arc. West, Newman imp. GMWoodward delet lith PLO IO) CHL IH MID) SIS} 1H IO) (CALLS; 4A, 4a. ~ he Th he 8, 8a, 8b, 8¢, 8d. PLATE VIII. Tricleidus seeleyi, Andrews; right half of the mandible, from outer side (1) and from inner side (1 @): one-half MeEvtsSIZE | OeuaeMee ca, . Ditto; lower tooth: nat. size. . Ditto; clavicular arch, from above (the interclavicle has been broken longitudinally, but the crack is not drawn) : one-half nat. size Ditto; left femur and proximal part of paddle, from below (4) and showing upper end of femur (4@): one- third nat. size . Ditto; dorsal vertebra, from front (5) and from right side (5 a): one-third nat. size . Ditto; dorsal vertebra, from front (6) and from right side (6a): one-third nat. size Ditto ; pectoral vertebra, from front (7) and from right side (7 a): one-third nat. size Ditto; series of cervical vertebre from right side (8), the vertebre marked a, 6, c, d being shown from the front in figs. 8a, 84, 8c, 8d: one-third nat. size. All the above figures are from the type specimen, R. 3539. a, b, c, d, in fig. 8 mark the vertebre figured from n.sp., neural spine. the front in ang., angular bone. at.a., neural arch of axv.a., neural arch of cl., clavicles. d., dentary bone. fy fibula. h., head of femur. i.cl., interclavicle. int., intermedium. m.r., ridges for the figs. 8a, 8b, 8c, 8d. | r., cervical ribs. 1 atlas. rf., facet for rib. 158 156-7 7., r°., ribs of atlas and axis respectively. axis. s.ang.§ art., fused surangular and articular bones. spl., splenial. sym., Symphysis of mandible. t., tibia. tib., tibiale. t.p., transverse process. tr., trochanter of femur. attachment of muscles. TAR IOMS SNCIMTOIUL 4a"[9p PxeMpoom'M'D ‘dure weurme yy “98a ' ' ' (ba 2lD p bun's THA HL Td AV'TIO GHOAXO Ld ANISVW 'TVLVO PLATE IX. Page 1, 1a. Cryptocleidus oxoniensis, Phillips, sp.; occipital region of skull from front (1), from behind (1 a): 2. Ditto; basis cranii from below: two-thirds nat. size. two-thirds nat. size. [R. 2860.] 166 [R. 2860.] 166 3. Ditto; right squamosal and quadrate : two-thirds nat. size. [R. 2860.] 167 4, 4a. Ditto; occipital region of skull of a young individual, from front (4) and from behind (4 a): two-thirds nat. size. [R. 2417] 166 5. Ditto; basis cranii of a young individual from below: two-thirds nat. size. 6. Ditto; mandible from above: two-thirds nat. size. sl two-thirds nat. size. a., ampulla of posterior vertical semicircular canal. art., articular surface for quadrate. boc., basioccipital. bsp., basisphenoid. bsp.f., facet for union with the basisphenoid. d., dentary bone. ew.op., united exoccipital and opisthotie bones. Ff. facet on hinder face of exoccipital, possibly for pro-atlas. for., median opening in basisphenoid of young individual. h.c., channel for horizontal semicircular canal. i.c.f., internal carotid foramen. jug-, jagular foramen. [R. 2417] 166 [R. 2417.] 167 . Ditto; crushed anterior portion of skull and mandible from below: [R. 8730.] 166-7 1.., lateral process of parietal. pal., palatine. pa'., parietal. par.p., paroccipital process. pas., parasphenoid. ptp., pterygoid (lateral) processes of basioccipital. p.v.c., channel for posterior vertical semicircular canal. q-, quadrate. soc., Supraoccipital. spl., splenial. sq, zygomatic process of squamosal. sq.', parietal process of squamosal. sym., symphysis of mandible. XII., foramen for twelfth nerve. CATAL. MARINE REPT. OXFORD CLAY. G.M Woodward del.et lith. West, Newman imp CRYPTOCLEIDUS OXONIENSIS. PLATE X. Big. Page 1, la, 1b, 1e. Cryptocleidus oxoniensis, Phillips, sp.; shoulder-girdle from above (1), from front (1a), from left side (14), and showing sym- physial surfaces (1c): one-sixth nat. size. [R. 2616.] 176 2,24. Ditto; left clavicle from upper (visceral) side (2) and from ventral (scapular) side (2 @): one-sixth nat. size. [R. 2616.] 178 3, 3a, 3b. Ditto; pelvis from above (3), from right side (3a), and showing symphysial surfaces (3 6): one-sixth nat. size. [R, 2616.] 183 acet., acetabulum. il., ilium. @.e.d., antero-external angle of pubis. isch., ischium. c., ridge on ilium above acetabular surface. | obt.f., obturator foramen. cl., clavicle. p-¢.p., postero-external process of coracoid. cor., coricoid. pu., pubis. cr.i., the crest of the ilium. sc., scapula. d.sc., dorsal ramus of scapule. s.sc., Surface on clavicle for union with scapula. for., foramen between coracoids. sym., median symphysis between various bones. qgl., glenoid cavity. v.sc., Ventral ramus of scapula. Ges & ) ; CATAL. MARINE REPT. OXFORD CLAY. G.M Woodward del.et lith CRY] PTOCUETII DUS OXONIENSIS. West, Newm. LATE A 1 L ¥ an imp. a ROT Maat sininl iit | re ei ; aa a LQ ii v_ 1A cecrpiv ctloge ofthe ma