CRANIAL ANATOMY OF KRYPTOBAATAR DASHZEVEGI (MAMMALIA

CRANIAL ANATOMY OF

KRYPTOBAATAR DASHZEVEGI

(MAMMALIA, MULTITUBERCULATA),

AND ITS BEARING ON THE EVOLUTION

OF MAMMALIAN CHARACTERS

JOHN R. WIBLE

Research Associate, Division of Vertebrate Zoology, American Museum ofNatural History; Section of Mammals, Carnegie Museum of Natural History,

5800 Baum Boulevard, Pittsburgh, PA 15206

GUILLERMO W. ROUGIER

Research Associate, Division of Paleontology, American Museum of NaturalHistory; Department of Anatomical Sciences and Neurobiology, School of

Medicine, University of Louisville, Louisville, KY 40292

BULLETIN OF THE AMERICAN MUSEUM OF NATURAL HISTORY

Number 247, 124 pages, 39 figures, 3 tables

Issued February 28, 2000

Price: $10.90 a copy

Copyright q American Museum of Natural History 2000 ISSN 0003-0090

2

CONTENTS

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Descriptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Premaxilla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16Nasal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19Lacrimal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Frontal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Maxilla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23Palatine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Pterygoid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30Sphenoid Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Petrosal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36Jugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Squamosal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Parietal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46Supraoccipital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46Exoccipital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47Basioccipital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Endocranium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Mandible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59

Vascular Reconstructions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Veins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Arteries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72Snout and Palate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Prenasal Process of Premaxilla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72Septomaxilla . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74Nasal Overhang and Anterior Nasal Notch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76Nasal Foramina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76Infraorbital Foramina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Palatine Foramina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Facial Process of Lacrimal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

Orbitotemporal Region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Orbital Mosaic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Orbital Foramina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80Postorbital Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Nasoparietal Contact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Jugal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

Basicranium and Lateral Braincase Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Pterygopalatine Ridges and Troughs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83Ectopterygoid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84Alisphenoid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Foramina for Branches of the Mandibular Nerve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86Venous System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87Arterial System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89Pterygoid Canal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92Petrosal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93Hypoglossal Foramen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

2000 3WIBLE AND ROUGIER: CRANIAL ANATOMY OF KRYPTOBAATAR DASHZEVEGI

Endocranium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96Cavum Epiptericum and Primary Braincase Wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96Hypophyseal Fossa, Tuberculum Sellae, and Jugum Sphenoidale . . . . . . . . . . . . . . . . 98Cribriform Plate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114Index of Anatomical Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

We affectionately dedicate this publication to Mike Novacek, for his years ofencouragement, support, and friendship.

4 NO. 247BULLETIN AMERICAN MUSEUM OF NATURAL HISTORY

ABSTRACT

The cranial anatomy of the Mongolian Late Cretaceous multituberculate Kryptobaatar dash-zevegi is described based on exquisitely preserved specimens collected from Ukhaa Tolgodand Tugrugeen Shireh in the Gobi Desert by joint expeditions of the American Museum ofNatural History and the Mongolian Academy of Sciences. Most sutural relationships are pre-served, enabling a bone-by-bone description of the skull and lower jaws exclusive of the nasalfossa and paranasal sinuses. A reconstruction of the principal components of the cranial ner-vous, arterial, and venous systems is facilitated by specimens with exposed endocranial sur-faces.

Comparisons with previously described multituberculates, other Mesozoic mammaliaforms,and extant mammals allow an assessment of major topics in the evolutionary morphology ofthe multituberculate and mammaliaform skull, as well as identification of Kryptobaatar as anappropriate model regarding most aspects of the multituberculate skull for future phylogeneticstudies. Elements previously unknown or poorly known in multituberculates are described.Included are a complete jugal on the internal surface of the zygoma; the orbital mosaic andforamina, including the optic foramen, the metoptic foramen, the transverse canal, and theforamen for the pituito-orbital vein; and the endocranium with an extensively ossified primarybraincase wall formed by the pilae metoptica and antotica. The latter pila is very robustcompared with its ossified remnants in non-mammalian cynodonts and monotremes, suggestingthat it is a derived multituberculate condition. The co-occurrence of the pilae metoptica andantotica in multituberculates is thus far unique among mammaliaforms, but agrees with themorphology expected to be primitive for Mammalia. This in turn implies an independent lossof an ossified pila metoptica in monotremes, marsupials, and Vincelestes or the loss of thepila metoptica in the ancestry of multituberculates and therians combined with the independentreacquisition of a neomorphic pila metoptica in multituberculates and eutherians. The absenceof several elements from the multituberculate skull, controversial in nature, is confirmed,including the prenasal process of the premaxilla, the septomaxilla, the ectopterygoid, and theorbital process of the palatine. Also confirmed is the presence of several controversial elementsin the multituberculate skull, including an alisphenoid with a reduced contribution to thebraincase and an anterior lamina expanded dorsal to the alisphenoid. Competing anatomicalhypotheses for several elements are addressed, including the function of the lateral pterygo-palatine trough as muscle attachment and not for the auditory tube, the homology of thepostorbital process on the parietal with that on the frontal, the identity of foramina in theanterior lamina as for mandibular nerve branches and not for the mandibular and maxillarynerves, and the function of the jugular fossa as primarily having housed a diverticulum of thecavum tympani and not large cranial nerve ganglia. The cranial arterial system in Kryptobaatargenerally resembled that restored for other multituberculates and for other mammaliaforms, inparticular the prototribosphenidan Vincelestes. Both Kryptobaatar and Vincelestes had a trans-promontorial internal carotid artery, a stapedial artery that ran through a bicrurate stapes, ramusinferior, ramus superior, and an arteria diploetica magna. The cranial venous system in Kryp-tobaatar resembled that described for other Mongolian Late Cretaceous multituberculates andfor monotremes with the major exits of the dural sinuses having been the prootic canal andthe foramen magnum.

A revised diagnosis of Kryptobaatar distinguishes it from other djadochtatherians (thegrouping that includes 10 of the 11 genera of Mongolian Late Cretaceous multituberculates)by a pterygoid canal either confluent with or barely separated from the carotid canal and aseparate hypoglossal foramen.

RESUMEN

Se describe la anatomıa craneana del multituberculado Kryptobaatar dashzevegi, del Cre-tacico Tardıo de Mongolia. Los especımenes, muy bien conservados, han sido colectados enlas localidades Ukhaa Tolgod y Tugrugeen Shireh, en el desierto del Gobi por expedicionesconjuntas del American Museum of Natural History y la Mongolian Academy of Sciences.La mayor parte de las suturas de Kryptobaatar estan conservadas permitiendo una descripcionhueso por hueso del craneo y mandıbulas con excepcion, de la fosa nasal y los senos para-


nasales. La reconstruccion de los principales nervios craneanos, sistema arterial y venoso quese presenta es facilitada por especımenes presentando la superficie endocranial expuesta.

La determinacion de los topicos mas importantes en la evolucion morfologica del craneode los multituberculados y otros mamaliafomes es posible a traves de la comparacion deKryptobaatar con multituberculados descritos previamente, otros mamaliaformes y mamıferosvivientes. Estas comparaciones permiten identificar a Kryptobaatar como un modelo apropia-do, en la mayorıa de los caracteres craneanos, para representar a los multituberculados enanalisis filogeneticos. Elementos previamente desconocidos, o mal representados en multitu-berculados, son descritos. Ellos incluyen: el jugal completo en la cara interna del arco cigo-matico, los elementos constituyentes de la orbita y sus foramenes, incluyendo el foramenoptico, metoptico, canal transverso y vena orbitopituitaria. Ademas de estos caracteres, sedescribe en detalle el endocranium. Este posee una pared primaria extensamente osificada,formada por las pilas metoptica y antotica. La ultima de ellas es muy robusta si se la comparacon los restos osificados de esta estructura en cinodontes no mamalianos y monotremas. Laocurrencia simultanea de un pila metoptica y antotica en multituberculados es hasta ahora unacaracterıstica unica de multituberculados, pero esta de acuerdo con la morfologıa que se pre-sume primitiva para Mammalia. Esto a su vez implica una perdida independiente de una pilametoptica osificada en monotremas, marsupiales y Vincelestes, o la perdida de la pila metopticaen el ancestro comun de multituberculados y terios, combinado con la readquisicion de unapila metotica neomorfica en euterios. Varios elementos del craneo de los multituberculadoscuya presencia ha sido discutida son aquı reconocidos como ausentes. Estos rasgos incluyenla ausencia de un proceso prenasal del premaxilar, septomaxila, ectopterigoides y la laminaorbital del palatino. Hemos confirmado tambien la presencia de varios elementos o rasgoscontrovertidos. Entre ellos se incluyen un aliesfenoides reducido y la presencia de una extensalamina anterior expandida dorsalmente al aliesfenoides.

Hipotesis alternativas para diversos caracteres son discutidas, incluyendo la funcion de lossurcos pterigopalatinos, aquı interpretados para insercion muscular y no para el tubo auditivo;la homologıa del proceso postorbitario en el parietal de los multitubercualdos de Mongoliacon aquel del frontal de otros multitubercualdos; la identidad de los foramenes en la laminaanterior, interpretadas como salidas de las ramas mandibulares del trigemino y no para lasramas mandibulares y maxilares de este nervio. Ademas, la gran fosa jugular de los multitu-berculados habrıa albergado un divertıculo de la cavidad auditiva y no grandes ganglios ex-tracraneanos. El patron general de las arterias craneans de Kryptobaatar es semejante a aquelrestaurado en otros multituberculados y algunos mamaliaformes, en particular el prototribos-fenido Vincelestes. Tanto Kryptobaatar como Vincelestes tienen una carotida interna de cursotranspromontorial, una arteria estapedial que atraviesa un estribo con dos cruras; las ramassuperior e inferior del sistema estapedial, y la arteria diploetica magna estan presentes. Elsistema venoso del craneo de Kryptobaatar, es similar a aquel descrito en otros multituber-culados del Cretacico Tardıo de Mongolia y en monotremas, en los cuales los senos duralesson drenados a traves del canal prootico y el foramen magnum.

Se presenta aquı una nueva diagnosis de Kryptobaatar, agregando los siguientes caracteresdiagnosticos: canal pterigoideo confluente o estrechamente asociado al canal carotideo y unforamen hipogloso separado del foramen jugular. Estos caracteres lo distingen de otros Dja-dochtatherians, el grupo que incluye con Kryptobaatar diez de los once generos de multitu-berculados del Cretacico Tardıo de Mongolia.

INTRODUCTION

Multituberculates are an extinct group ofearly mammals with a temporal range thatprobably spans from the Middle Jurassic(Freeman, 1976, 1979; McKenna and Bell,1997) to the Late Eocene (Kristhalka et al.,1982; Prothero and Swisher, 1992). A pur-ported multituberculate has been describedfrom the Late Triassic by Hahn et al. (1987),

a determination accepted by some authors(e.g., Godefroit, 1997). However, one of us(G.W.R.) has studied the specimen and be-lieves it to be a theroteinid (see Sigogneau-Russell et al., 1986). Multituberculates alsohave a very wide geographic distributionand, with recent discoveries in Morocco (Si-gogneau-Russell, 1991), are known from all


major land masses with the exception ofSouth America, Australia, and Antarctica. Ifgondwanatheres from the Late Cretaceous ofArgentina (Bonaparte, 1986; Krause et al.,1989; Kielan-Jaworowska and Bonaparte,1996), Madagascar (Krause and Grine, 1996;Krause et al., 1997), and India (Krause et al.,1997) are multituberculates as considered bythe above authors, but not in the most recentaccount by Pascual et al. (1999), then thegeographic distribution of Multituberculata iseven broader. Currently, more than 70 generaof multituberculates are recognized (McKen-na and Bell, 1997), but the overwhelmingmajority of these are known only from iso-lated teeth and fragmentary jaws (Simmons,1993).

Among Mesozoic multituberculates, themost notable exceptions to the usually in-complete morphological record for the groupare taxa collected from the continental LateCretaceous formations of Mongolia (Kielan-Jaworowska, 1970, 1971, 1974; Kermackand Kielan-Jaworowska, 1971; Kielan-Ja-worowska and Gambaryan, 1994; Dashzeveget al., 1995; Kielan-Jaworowska and Hurum,1997; Kielan-Jaworowska et al., 2000). Todate, 13 monospecific genera have beennamed from these formations, the majority ofthem by Zofia Kielan-Jaworowska, the leaderof the Polish–Mongolian Paleontological Ex-peditions of the late 1960s and early 1970s.Two of the 13 genera, Gobibaatar and Tu-grigbaatar, subsequently have been regardedas junior synonyms of a third, Kryptobaatar(see below). Another taxon, Buginbaatar(Kielan-Jaworowska and Sochava, 1969),has been said to be of Late Cretaceous orearly Paleocene age (Trofimov, 1975; Clem-ens and Kielan-Jaworowska, 1979). Howev-er, Martinson (1982), followed by Jerzykie-wicz and Russell (1991), considered the lo-cality where Buginbaatar was collected to beof Late Cretaceous age, representing ‘‘Ne-megt times.’’ The discovery of dinosaur egg-shells at this locality by recent Joint Pale-ontological Expeditions from the MongolianAcademy of Sciences and the American Mu-seum of Natural History supports a Late Cre-taceous age for Buginbaatar. Therefore, wecurrently recognize 11 genera of MongolianLate Cretaceous multituberculates from theNemegt Formation, the Djadokhta Forma-

tion, and the probably equivalent Barun Go-yot Formation (Novacek et al., 1996): Bugin-baatar Kielan-Jaworowska and Sochava,1969; Bulganbaatar Kielan-Jaworowska,1974; Catopsbaatar Kielan-Jaworowska,1994; Chulsanbaatar Kielan-Jaworowska,1974; Djadochtatherium Simpson, 1925;Kamptobaatar Kielan-Jaworowska, 1970;Kryptobaatar Kielan-Jaworowska, 1970; Ne-megtbaatar Kielan-Jaworowska, 1974; Nes-sovbaatar Kielan-Jaworowska and Hurum,1997; Sloanbaatar Kielan-Jaworowska,1971; and Tombaatar Rougier et al., 1997a.Regarding the state of preservation, 6 of the11 genera are described from well-preserved,nearly complete skulls and lower jaws (Kie-lan-Jaworowska, 1970, 1971, 1974; Kielan-Jaworowska et al., 1986; Kielan-Jaworowskaand Hurum, 1997), and 5 of these are alsorepresented by some postcranial elements(Kielan-Jaworowska, 1969; Kielan-Jawo-rowska and Gambaryan, 1994). The mostcompletely described taxa from the stand-point of anatomy are Nemegtbaatar gobien-sis and Chulsanbaatar vulgaris. They areknown from most aspects of cranial anatomy(Kielan-Jaworowska, 1974), including thebrain endocast (Kielan-Jaworowska, 1983,1986), the basicranium and the cranial vas-culature (Kielan-Jaworowska et al., 1984,1986; Hurum et al., 1995, 1996), the brain-case (Hurum, 1998a), the inner ear (Hurum,1998b), the nasal fossa and paranasal sinuses(Hurum, 1994), and the masticatory appara-tus (Gambaryan and Kielan-Jaworowska,1995); in addition, the postcranial skeleton ofboth is almost fully known (Kielan-Jawo-rowska and Gambaryan, 1994).

Although Nemegtbaatar and Chulsanbaa-tar are currently the most fully describedMongolian Late Cretaceous multitubercu-lates, Kryptobaatar dashzevegi, the subjectof this report, is the taxon represented by themost specimens. It is the most abundantmammal both in the traditional localities ofthe Djadokhta Formation (Kielan-Jaworow-ska, 1974) and in Ukhaa Tolgod (fig. 1), and,in fact, is represented by more skulls thanany other Mesozoic mammal. Ukhaa Tolgodis the locality showing the highest concen-tration of mammalian skulls and skeletonsfrom any Mesozoic site; it was discovered inJuly 1993 by the Joint Mongolian–American


Museum Expeditions (Dashzeveg et al.,1995). The Djadokhta Formation is thoughtto be of early Campanian age (Jerzykiewiczet al., 1993; Dashzeveg et al., 1995; Rougieret al., 1997a; Averianov, 1997). The age ofUkhaa Tolgod relative to Djadokhta and Ba-run Goyot is under study (Dingus et al., inprep.), but the faunal contents of these unitsare more uniform than previously thought(Novacek et al., 1996).

Kryptobaatar dashzevegi was named byKielan-Jaworowska (1970) based on an in-complete skull (missing the braincase) andlower jaws recovered in the Djadokhta For-mation, Bayn Dzak, Shabarakh Usu (the‘‘Flaming Cliffs’’; fig. 1). It was allocated tothe suborder Taeniolabidoidea, family Eucos-modontidae, as were most of the other mul-tituberculates from the Djadokhta and BarunGoyot Formations (Clemens and Kielan-Ja-worowska, 1979). Recently, in fact, Rougieret al. (1997a) have suggested that nine of theMongolian Late Cretaceous multituberculategenera are included in a monophyletic group-ing along with Pentacosmodon from theNorth American Paleocene (see fig. 38); ex-cluded was Buginbaatar, which was closerto the taeniolabidids, and Nessovbaatar,which had not yet been named. Kielan-Ja-worowska and Hurum (1997) proposed asomewhat modified monophyletic Mongo-lian assemblage, which they named the Dja-dochtatheria. Included were the same taxa inthe unnamed Mongolian clade of Rougier etal. (1997a) plus Nessovbaatar and Paraci-mexomys from the North American Creta-ceous. Kryptobaatar was assigned by Kielan-Jaworowska and Hurum (1997) to the Dja-dochtatheriidae, along with Djadochtathe-rium, Catopsbaatar, and Tombaatar.

Based on additional specimens from BaynDzak, some additions and corrections to theoriginal description of Kryptobaatar dash-zevegi subsequently were made by Kielan-Jaworowska and co-authors. For example,the slender palatal vacuities included in theoriginal diagnosis were found to be artifac-tual (Kielan-Jaworowska and Dashzeveg,1978). Comparisons were made by Kielan-Jaworowska and Dashzeveg (1978) with theskull of a new eucosmodontid, Tugrigbaatarsaichanensis, from Tugrugeen Shireh or Too-greeg (fig. 1), considered to be contempora-

neous with Bayn Dzak. These authors notedmany resemblances between Tugrigbaatarand Kryptobaatar, and subsequently the for-mer was declared a junior synonym of thelatter (Rougier et al., 1997a; Kielan-Jawo-rowska and Hurum, 1997). In 1980, Kielan-Jaworowska recognized Gobibaatar parvus(Kielan-Jaworowska, 1970), originally de-scribed as the only ptilodontoid from theLate Cretaceous of Mongolia, to be a juniorsynonym of K. dashzevegi. Further remarkson the masticatory apparatus of K. dashze-vegi have been published by Gambaryan andKielan-Jaworowska (1995), and other incom-plete skulls from Bayn Dzak have been fig-ured by Kielan-Jaworowska and Gambaryan(1994) and Kielan-Jaworowska and Hurum(1997). A revised diagnosis of K. dashzevegihas been offered by Kielan-Jaworowska andHurum (1997).

Regarding the postcranium, a partial skel-eton, including nearly complete hindlimbsand pelvis, an incomplete scapulocoracoid,and some damaged cervical, lumbar, sacral,and caudal vertebrae, was collected in asso-ciation with a skull of Kryptobaatar dash-zevegi at Bayn Dzak during the 1968 Polish–Mongolian expedition. Described in thisspecimen for the first time in multitubercu-lates were epipubic bones (Kielan-Jaworow-ska, 1969). Other details of the pelvis werereported by Kielan-Jaworowska (1979) andof the remaining postcranial elements byKielan-Jaworowska and Gambaryan (1994).Partial humeri, an ulna, and ribs were foundand described with the holotype skull of Tu-grigbaatar saichanensis by Kielan-Jawo-rowska and Dashzeveg (1978), and a com-plete humerus of Kryptobaatar was de-scribed by Kielan-Jaworowska (1998). Weadd that Kryptobaatar also provides the firstevidence of a tarsal spur in multituberculates.The bone that Kielan-Jaworowska and Gam-baryan (1994: fig. 2A) labeled as a possiblefragment of dentary between the distal endsof the tibia and fibula is a displaced tarsalspur (personal obs.) as occurs in monotremes(Griffiths, 1978), the gobiconodontid Gobi-conodon (Jenkins and Schaff, 1988), and thesymmetrodont Zhangheotherium (Hu et al.,1997).

The current report documents the anatomyof several well-preserved specimens of


Fig. 1. Map of Mongolia (A) with inset of south-central region (B) indicating the major fossillocalities visited by the joint expeditions of the Mongolian Academy of Sciences and the AmericanMuseum of Natural History. The specimens of Kryptobaatar dashzevegi described here come fromUkhaa Tolgod and Tugrugeen Shireh.


Kryptobaatar dashzevegi collected over thelast few years by the Joint Mongolian–Amer-ican Museum Expeditions (fig. 2). Two spec-imens have been figured elsewhere (Novaceket al., 1994; Dashzeveg et al., 1995: fig. 3a,misidentified as Chulsanbaatar; Rougier etal., 1996c: figs. 1–4; Monastersky, 1996;Rose and Sues, 1996; Novacek, 1997: fig. 2),but thus far only the anatomy of the middle-ear ossicles has been described in detail(Rougier et al., 1996c). Although K. dash-zevegi is known from more skulls than anyother Mesozoic mammal, its entire cranialanatomy had yet to be documented in detail,and therefore we present a bone-by-bone de-scription of the skull exclusive of the ear os-sicles and reconstruct the course of the majorcranial nerves and vessels. The dentition ofK. dashzevegi has been described by Kielan-Jaworowska (1970) and Kielan-Jaworowskaand Dashzeveg (1978), and will not be re-described here, although relevant dental mea-surements are presented in table 1.

Because postcranial and cranial remains(exclusive of teeth) are rarely recovered formultituberculates, the few well-preservedtaxa that have been described are central toour understanding of the morphology and re-lationships of the group. Included among themost thoroughly studied multituberculates, inaddition to several of the Mongolian LateCretaceous taxa (e.g., Nemegtbaatar gobien-sis, Chulsanbaatar vulgaris), is the taenio-labidid Lambdopsalis bulla from the Paleo-cene of China, whose anatomy has been thesubject of several reports (e.g., Miao, 1986;Kielan-Jaworowska and Qi, 1990), includingmonographic treatment of the skull (Miao,1988). Lambdopsalis is a highly specializedform, perhaps a burrower, with an expandedvestibular apparatus; it exhibits many apo-morphic cranial features compared with theMongolian Late Cretaceous taxa (Miao,1988; Luo, 1996; Kielan-Jaworowska andHurum, 1997). Despite its extreme speciali-zations, Lambdopsalis has served as a modelfor Multituberculata in some phylogeneticstudies (e.g., Wible, 1991; Meng, 1992;Meng and Wyss, 1995) because of the pau-city of material and of comprehensive, de-tailed descriptions of other taxa. Therefore, agoal of our report on Kryptobaatar dashze-vegi is to provide a detailed description of

the skull and lower jaws that can serve as anappropriate model for future comparativemorphological and phylogenetic studies onMultituberculata and on Mammaliaformes ingeneral, the latter being the clade comprisingthe common ancestor of the Liassic taxonMorganucodon and Mammalia plus all itsdescendants (Rowe, 1988).

MATERIALS AND METHODS

To date, more than 400 multituberculatespecimens have been collected by the JointMongolian–American Museum Expeditions.The majority are in the size range of Krypto-baatar dashzevegi, but only a handful havebeen prepared sufficiently to enable precisetaxonomic determination. Our descriptionsherein are based on the following specimensof K. dashzevegi, all of which are cataloguedin the Institute of Geology, Ulaan Baatar. Thefirst two specimens comprise the bulk of thedescriptions of external surfaces and themeasurements in tables 1 and 2; the remain-ing specimens reveal structures of the endo-cranium and/or selected external features.

(1) PSS-MAE 101 (figs. 2, 3): Anteriorhalf of a fully articulated skeleton collectedfrom Ukhaa Tolgod, including a pristineskull with lower jaws attached, the anteriorportion of the vertebral column, and bothshoulder girdles and forelimbs. Some exter-nal surfaces of the skull are not accessible;matrix has not been removed from part of theleft ear region, the area of the foramen mag-num, and part of the palate in order to pre-serve the stylohyal, atlas and axis, and lowerjaws, respectively. The skull has been figuredelsewhere in different views (Novacek et al.,1994; Rougier et al. 1996c: fig. 2; Dashzeveget al., 1995: fig. 3a; Monastersky, 1996; No-vacek, 1997: fig. 2), and the forelimb hasbeen figured in dorsolateral view (Dashzeveget al., 1995: fig. 3a; Rose and Sues, 1996).

(2) PSS-MAE 113 (figs. 4, 5): Skull andthe left lower jaw collected in 1991 from Tu-grugeen Shireh. Missing are parts of the cra-nial vault, the left alisphenoid, part of thepterygoid, and the left zygomatic arch. Theskull has been figured elsewhere in ventralview, as have closeups of the fragmentaryleft malleus-ectotympanic complex and the


←

Fig. 2. Kryptobaatar dashzevegi PSS-MAE 101 in left lateral and two oblique dorsolateral views(from top to bottom), prior to removal of the skull and lower jaws from the block containing thepostcranium.

fragmentary right stapes (Rougier et al.,1996c: figs. 1, 3, 4).

(3) PSS-MAE 123 (fig. 25): Partial skullwith lower jaws collected in 1994 from Uk-haa Tolgod (Sugar Mountain). Missing arethe braincase roof, right braincase wall, oc-ciput, right zygomatic arch, and posteriorpart of the right lower jaw. Preparation hasexposed the inner surfaces of the skull baseand left braincase.

(4) PSS-MAE 124: Partial skull with low-er jaws collected from Ukhaa Tolgod. Miss-ing are the braincase roof and left zygomaticarch. Little matrix has been removed fromthe exterior of the skull, but the interior hasbeen partially prepared. Exposed is the innersurface of part of the sphenoid complex.Specimen is not figured.

(5) PSS-MAE 125 (fig. 26): Partial skullwith fragments of both lower jaws collectedfrom Ukhaa Tolgod. Missing are the left zy-gomatic arch, left braincase wall, and mostof the occiput. Preparation has exposed mostof the left orbital wall, the braincase floor,and the inner surface of the right braincasewall.

(6) PSS-MAE 127: Skull and lower jawscollected in 1993 from Ukhaa Tolgod. Matrixhas been removed only from selected exter-nal surfaces. Exposed is the tip of the ros-trum, right upper and lower jaws, and theposterior part of the roof of the braincase andzygomatic arch. Specimen is not figured.

In addition, three other indeterminateskulls resembling Kryptobaatar from theMongolian–American Museum Expeditionsare included here for comparative purposes,but are not figured.

(1) PSS-MAE 126: Skull without lowerjaws collected in 1994 from Gilbent Wash(locality 20) near Ukhaa Tolgod. Missing areboth zygomatic arches and exoccipitals, aswell as the tip of the rostrum. Also missingis part of the cranial vault, which exposes theposterior part of the brain endocast and vas-cular molds. Matrix has been removed fromthe right orbit, but not the from left.

(2) PSS-MAE 128: Partial skull withoutlower jaws collected in 1993 from UkhaaTolgod. Missing are the left braincase andzygomatic arch. Little matrix has been re-moved from the exterior of the skull, but theinterior has been prepared. Exposed is theinner surface of the right petrosal and ante-rior lamina, as well as parts of the sphenoidcomplex.

(3) PSS-MAE 134: Natural nasal endocastcollected in 1996 from Ukhaa Tolgod (Zo-fia’s Hill).

Kielan-Jaworowska and Hurum (1997)recognized two species of Kryptobaatar, K.dashzevegi and K. saichanensis, with themost substantive difference being the cuspformula (in the labial and lingual rows) ofthe second lower molar (3:2 in K. dashzevegiversus 4:2 in K. saichanensis). We have in-spected the single specimen of K. saichanen-sis (GI SPS 8-2 PST) and found three cuspsin the labial row on the left second lowermolar and three cusps plus an accessory cuspon the right side. Given that the other differ-ences identified by Kielan-Jaworowska andHurum (1997) are minor (e.g., the length ofthe mandibular ascending ramus), we rec-ognize K. saichanensis as a junior synonymof K. dashzevegi. Our specimens of Krypto-baatar also exhibit differences. For example,in PSS-MAE 101, the sphenopalatine fora-men lies between the frontal and maxilla,whereas in 113 it is wholly within the max-illa. Additionally, in PSS-MAE 101 the jugalis concealed in lateral view by the maxillaand squamosal, whereas in 113 a thin sliverof jugal is exposed along the dorsal edge ofthe zygomatic arch. PSS-MAE 113, collectedin Tugrugeen Shireh, is slightly larger than101 from Ukhaa Tolgod (see tables 1 and 2);this is in agreement with the purported largersize of K. saichanensis also from TugrugeenShireh compared to K. dashzevegi from BaynDzak (Kielan-Jaworowska and Dashzeveg,1978). We believe that these and a few otherdifferences in our specimens are minor andwe treat the specimens as conspecific.


Fig. 3. The skull and lower jaws of Kryptobaatar dashzevegi PSS-MAE 101 in (clockwise fromupper left) dorsal, ventral, anterior, posterior, right lateral, and left lateral views.


Fig. 4. The skull of Kryptobaatar dashzevegi PSS-MAE 113 in (clockwise from upper left) dorsal,ventral, anterior, posterior, right lateral, and left lateral views.


Fig. 5. The left lower jaw of Kryptobaatar dashzevegi PSS-MAE 113 in lateral and medial views.

Different authors (e.g., Kermack et al.,1981; Novacek, 1986; Miao, 1988) haveused various organizational schemes in de-scribing the skulls of fossil mammals. Herewe provide a bone-by-bone description of theexterior of the skull of Kryptobaatar. In ad-dition, the ventral endocranial surfaces visi-ble in several specimens are described as asingle anatomical unit. Also provided are re-constructions of the major cranial nerves, ar-teries, and veins. Following the descriptions,comparisons of individual regions of theskull are made with those of other multitu-berculates and mammaliaforms. Emphasis ison characters used in previous phylogeneticanalyses (e.g., Simmons, 1993; Wible et al.,1995; Rougier et al, 1996a, 1996c).

Over the years, cranial remains of multi-

tuberculates have been described by numer-ous authors (e.g., Broom, 1914; Simpson,1937; Hahn, 1969; Kielan-Jaworowska,1970; Miao, 1988), and the anatomical ter-minology used has not always been consis-tent. In general, our usage of terminologyagrees closely with that used by Kielan-Ja-worowska and co-authors (e.g., Kielan-Ja-worowska, 1970, 1974; Kielan-Jaworowskaet al., 1986; Gambaryan and Kielan-Jawo-rowska, 1995), but for the auditory regionand cranial vasculature we follow Wible(1987, 1990), Rougier et al. (1992, 1996a,1996c), Wible and Hopson (1993, 1995), andWible et al. (1995). In addition, for many ofthe endocranial structures not described pre-viously in multituberculates, we follow theterminology in the dog (Evans and Christen-


sen, 1979). Our usages for the cranial foram-ina and canals are specified in a glossary atthe end. We have also provided an index forthe major page citations for most anatomicalterms used. For the dentition, we use the ab-breviations ‘‘I, P, M’’ and ‘‘i, p, m’’ to referto upper and lower incisors, premolars, andmolars, respectively. Regarding the numera-tion of the upper premolars, we follow Sim-mons (1993) and designate the five premo-lars present in primitive multituberculates(e.g., the Late Jurassic paulchoffatiines,Hahn, 1993) as P0–P4. The first upper pre-molar having been lost, the remaining fourin Kryptobaatar are designated P1–4. How-ever, our usage of the term premolars doesnot necessarily imply strict homology withthe premolars of therians. There is no in-stance in which the multituberculate P4 isknown to have a deciduous precursor(Greenwald, 1987; Hahn and Hahn, 1999); adeciduous p4 has only been reported for oneform, Kuehneodon dietrichi (Hahn, 1978),but this was based on a single broken lowerjaw that contained no replacement teeth. Giv-en that there are no unequivocal examples ofreplacement of the upper and lower last pre-molars in multituberculates, it is possible thatthese teeth are molars, i.e., unreplaced per-manent teeth (see Clemens and Lillegraven,1986; Luckett, 1993).

Several authors (e.g., Bryant and Russell,1992; Witmer, 1995) have recently proposedexplicit methods for reconstructing soft tis-sues in fossils and for evaluating levels ofconfidence in those inferences. In formulat-ing hypotheses about soft-tissue reconstruc-tion here, we follow recent phylogeneticanalyses (Rowe, 1988, 1993; Wible et al.,1995; Rougier et al., 1996a, 1996c; Hu et al.,1997; Ji et al., 1999) that identify multitu-berculates as members of Mammalia (fig.39), the group including the last common an-cestor of monotremes and therians plus de-scendants (Rowe, 1988). We acknowledge,however, that the relationships of multituber-culates within Mammalia remain controver-sial (e.g., Rougier et al., 1996b; Sereno andMcKenna, 1996; Meng and Wyss, 1996),

with some authors (e.g., Meng and Wyss,1995) promoting a sister-group relationshipfor multituberculates and monotremes, andothers (e.g., Rougier et al., 1996a, 1996c; Huet al., 1997; Ji et al., 1999) allying multitu-berculates more closely with therians. Nev-ertheless, under the terminology proposed byWitmer (1995), the extant phylogeneticbracket (minimally, the first two extant out-groups) for multituberculates consists ofmonotremes and therians. Inferences that arebased on soft-tissue structures and osteolog-ical correlates occurring in both extant out-groups are considered more decisive thanthose occurring in only one.

Institutional AbbreviationsAMNH Department of Vertebrate Paleon-

tology, American Museum of Nat-ural History

FMNH Department of Geology, Field Mu-seum of Natural History, Chicago

GI Geological Institute, MongolianAcademy of Sciences, Ulaan Baa-tar

IVPP Institute of Vertebrate Paleontolo-gy and Paleoanthropology, Chi-nese Academy of Sciences, Beijing

MAE Mongolian–American MuseumExpedition

MCZ Museum of Comparative Zoology,Harvard University, Cambridge

PIN Institute of Paleontology, Acade-my of Sciences, Moscow

PSS(SPS)Paleontological and StratigraphicSection of the Geological Institute,Mongolian Academy of Sciences,Ulaan Baatar

USNM Department of Paleobiology, Na-tional Museum of Natural History,Smithsonian Institution, Washing-ton, D.C.

V.J. Museum of the Geological Ser-vice, Lisbon

YPM-PU Yale Peabody Museum, PrincetonUniversity Collection, New Haven

ZPAL Institute of Paleobiology, PolishAcademy of Sciences, Warsaw


DESCRIPTIONS

As accompaniment to the following de-scriptions, composite views of the skulls andlower jaws of PSS-MAE 101 and 113 areshown in figure 3 and 4–5, respectively. Ste-reophotographs of anterior, dorsal, lateral,ventral, and posterior views of these speci-mens are in figures 6–17. Enlargements ofparticular regions (i.e., orbit, basicranium,zygoma, endocranium, and mandible) are infigures 18–23, 25–26, 29, and 30. Recon-structions of the skull of Kryptobaatar dash-zevegi in various views are in figures 27, 31–35, 36A, and 37A, with vascular reconstruc-tions in figures 36B and 37B. Abbreviationsused in the figures are listed in table 3.

PREMAXILLA

The premaxilla is a large bone with facialand palatal components of near equivalentsize oriented approximately at right angles to

each other. It forms the walls of the anteriorpart of the nasal cavity and bears two inci-sors, a larger I2 and a smaller I3. The pre-maxilla is complete in PSS-MAE 101 (fig.6), but its anterior portion is missing in PSS-MAE 113 (fig. 7).

In lateral view (figs. 10–13, 33), the facialcomponent comprises approximately one-third of the preorbital skull length. It formsthe anterolateral wall of the nasal cavity andthe ventral half of the margin of the externalnares. Facets for the septomaxilla on the ex-ternal narial aperture are lacking. The ante-rior margin of the facial component is ver-tical, flush superiorly with the anterior tip ofthe nasal and inferiorly with the anterior edgeof the I2 alveolus; consequently, the rim ofthe external nares is circular in outline andopens directly anteriorly without any lateralexposure. The facial component contacts themaxilla posteriorly and the nasal dorsally.The premaxillary–maxillary suture is verticalnear the palatal margin, roughly halfway be-


tween I2 and P1. Near the nasal, the suturecurves posteriorly, and a slender tongue ofthe premaxilla, the posterodorsal process, ex-tends between the maxilla and nasal to thelevel of I3. In rostral view in PSS-MAE 101

(figs. 6, 31), the anterior part of the facialcomponent, which is complete on both sides,is very narrow dorsoventrally and shows nosign of a prenasal process, which, if present,would have formed an internarial bar. Con-


Fig. 6. Stereophotograph of the skull and lower jaws of Kryptobaatar dashzevegi PSS-MAE 101 inanterior view, with accompanying line drawing. Gray pattern represents matrix; parallel lines denotebreakage. Abbreviations: con (mandibular) condyle; fr frontal; gl glenoid fossa; iof infraorbital foramen;mf mental foramen; mx maxilla; na nasal; pmx premaxilla; sq squamosal.

tained within the anterior part of the facialcomponent is the alveolus of I2, and extend-ing posteriorly from the anterior part to thepremaxillary–maxillary suture is a small hor-izontal furrow, perhaps representing an at-tachment for facial musculature (‘‘muf’’ infig. 14). The most probable occupant of sucha furrow was the musculus incisivus superio-

ris, which raised the upper lip (Evans andChristensen, 1979).

In ventral view (figs. 14, 15, 34), the pre-maxilla occupies roughly one-third of thepalate, with the most conspicuous featuresbeing the alveolus of I2 anteriorly, and thealveolus of I3 and the incisive foramen pos-teriorly. The lateral margin is thickened to


Fig. 7. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in anterior view.

form a sharp, ventrally projecting ridge. Thecentral portion of the palatal component, be-tween the alveoli of I2 and I3, is very longand concave ventrally. Subdividing this partof the palatal component into two longitu-dinal depressions is a cord-shaped ridge, thethickening of the premaxilla (‘‘tpmx’’ in fig.14) of Kielan-Jaworowska et al. (1986). Ofthe two depressions divided by the thicken-ing of the premaxilla, the medial is the deep-er; it lies on the midline and meets its com-plement of the opposite side. The alveolus ofI2 is large and although close to the midline,is separated from the I2 alveolus of the op-posite side by considerable bone. The leftand right I2s are directed ventromedially to-ward one another and come into contact ven-trally, leaving a triangular space betweenthem and the edge of the premaxilla visiblein rostral view (figs. 6, 31). A similar ar-rangement with contact between the left andright I2s has been illustrated for Sloanbaatarfrom the Mongolian Late Cretaceous (Kie-lan-Jaworowska, 1971: fig. 9) and for Tae-niolabis taoensis from the North AmericanPaleocene (Granger and Simpson, 1929: figs.5, 6). The portion of the premaxilla in frontof the alveolus of I2 in Kryptobaatar slopesposteroventrally from the narial aperture tothe alveolus. The smaller alveolus for I3 liesroughly halfway between the median sutureand the labial margin. The I3 alveolus isformed entirely by the premaxilla and is sep-arated from the premaxillary–maxillary su-ture by bone approximating the diameter ofthe alveolus. Medial to the I3 alveolus and

of equivalent size is the incisive foramen(‘‘inf’’ in fig. 14). This aperture is oval andlies between the premaxilla and maxilla (fig.34). Separating the left and right incisive fo-ramina and meeting on the midline are thebroad palatal processes of the premaxillae.

NASAL

The nasals are elongate bones that formnearly the entire skull roof in the preorbitalarea. They are severely damaged in PSS-MAE 101, 113, 123, and 127, but the pres-ervation is sufficient to restore their suturalrelationships.

In dorsal view (figs. 8, 9, 32), the nasal onits lateral margin contacts the premaxilla andmaxilla and on its posterior margin, the lac-rimal and frontal. The sutural relationships ofthe nasal are as follows. Where the nasalcontacts the premaxilla, the lateral margin ofthe nasal is subparallel to the medial margin.However, where the nasal contacts the max-illa, the nasal is strongly expanded laterally.The lateral three-quarters of the nasal’s pos-terior suture are roughly transverse, but inthe medial quarter there is a slender tongueof frontal that projects into a short, broad U-shaped notch between the two nasals. Thefrontal is overlapped slightly by the nasal, asis evident on the right side of PSS-MAE 113where the posterolateral part of the nasal ismissing and the frontal is exposed (fig. 9).Nasal foramina (Simpson, 1937) are presenton the dorsal surface (‘‘naf’’ in figs. 8, 10).However, because of damage, the number


Fig. 8. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 101 in dorsal view,with accompanying line drawing. Gray pattern represents matrix; parallel lines denote breakage. Ab-breviations: al anterior lamina; ano nasal notch; con (mandibular) condyle; cor coronoid process; exocexoccipital; fdac foramen of dorsal ascending canal; fr frontal; frt foramen for ramus temporalis; jujugal; juf jugal facet; lac lacrimal; mx maxilla; na nasal; naf nasal foramen; pa parietal; pmx premaxilla;sgf supraglenoid foramen; sq squamosal; sup supraoccipital.

and position of the foramina cannot be fullyascertained. In the anterior third of the nasalsin PSS-MAE 101, there are two foramina onthe right side and three on the left, and theyare not arranged symmetrically between thetwo sides (fig. 8). On the basis of more com-pletely preserved nasals, Kielan-Jaworowskaand Hurum (1997) reported two pairs of na-sal foramina for Kryptobaatar dashzevegi.The anterior edge of the nasal is not fullypreserved in any MAE specimens, butenough is present to report the presence of anotch there (‘‘ano’’ in figs. 8, 12), especiallyin PSS-MAE 127, which we term the ante-rior nasal notch following Lillegraven andKrusat (1991). Additionally, enough is pre-sent to preclude a substantial nasal overhangof the external narial aperture and to excludethe presence of an internarial bar, as is alsoindicated by the morphology of the premax-illa.

LACRIMAL

The lacrimal occupies the anterior marginof the orbit and has orbital and facial pro-cesses, with the latter being the larger of thetwo. A complete lacrimal is not preserved ineither PSS-MAE 101 or 113; the most com-plete is on the left side of the former (figs.8, 18), which is missing only the anterior-most part.

The orbital exposure is wedge-shaped anddoes not extend ventrally far from the orbitalrim, contributing mostly to the overhangingorbital roof (figs. 16, 18, 33). It contacts themaxilla laterally and inferiorly, and the fron-tal medially. The facial exposure (figs. 8, 32)is transversely convex and subrectangular inshape (see Kielan-Jaworowska and Hurum,1997: figs. 2B, C). It contacts the maxillaanterolaterally, the nasal anteromedially, andthe frontal posteromedially. The part of thelacrimal forming the orbital rim is thickened,


Fig. 8. Continued.

and on its ventrolateral aspect bears a singlesmall lacrimal foramen, which is not subdi-vided (‘‘lacf’’ in fig. 16). The foramen ishigh in the orbit, surrounded by strong crests,and opens near the contact between the lac-rimal and the facial process of the maxilla.

FRONTAL

The frontal has a horizontal component inthe skull roof and a vertical component in theorbit. Both components are well preserved inPSS-MAE 113, but the horizontal is dam-aged in PSS-MAE 101 (fig. 8).

The horizontal components of the frontals(figs. 8, 9, 32) are flat and meet on the dorsalmidline at a slightly irregular suture. Theyroof the posterior part of the nasal cavity andthe anterior part of the cranial cavity. Ante-riorly, the frontals contact the lacrimals andsend a tongue-shaped process anteromediallyinto the preorbital area between the nasals.Posteriorly, the frontals narrow, extending tothe level of the postorbital processes on theparietals and contacting the parietals largelythrough a broad U-shaped suture; lateral tothe arms of the U is a small prong of the


Fig. 9. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in dorsal view.

frontal that projects posterolaterally, formingthe medial rim of the supraorbital notch(‘‘son’’ in figs. 10, 12). Along their suture,the frontal and parietal do not overlap. Thelateral border of the horizontal component ofthe frontal forms a thickened ridge contrib-uting to the orbital rim, the supraorbital crestof Miao (1988). In the orbital margin, thefrontal is widest anteriorly where it contactsthe lacrimal and tapers posteromedially, be-ing narrowest where it contacts the anteriorprong of the parietal in front of the postor-bital process. At the frontal–parietal contactin the orbital margin, the frontal is indentedby the supraorbital notch (see below).

In the orbit (figs. 10–13, 18, 19, 33), thevertical component of the frontal forms thedorsal third of the orbital mosaic, with itsventralmost projection extending to justabove the sphenopalatine foramen. The fron-tal contacts the lacrimal and the maxilla an-teroventrally, the orbitosphenoid posteroven-trally, the anterior lamina posteriorly, and theparietal posterodorsally. Within the orbit, ata level just in front of the supraorbital notch,

there is a low, raised area directed dorsoven-trally in the same position as is the well-de-fined, high orbital ridge of Kielan-Jaworow-ska et al. (1986) that occurs in some otherMongolian Late Cretaceous multitubercu-lates (e.g., Nemegtbaatar). Interestingly, theorbital ridge is present in a specimen ofKryptobaatar from Tugrugeen Shireh,GISPS 8-2 PST (Kielan-Jaworowska, per-sonal commun.), originally identified as Tu-grigbaatar saichanensis by Kielan-Jawo-rowska and Trofimov (1980). Anterior to thisraised area, the frontal is fairly flat in Krypto-baatar, PSS-MAE 101 and 113, whereassome other Mongolian Late Cretaceous taxa(e.g., Nemegtbaatar) have a small, roundedfossa, the orbitonasal fossa of Kielan-Jawo-rowska (1971). This part of the frontal boneis the posterodorsal end of a pocketlike struc-ture (orbital pocket or theca orbitalis) occur-ring in many multituberculates that repre-sents the origin for the pars anterior of themedial masseter muscle (Gambaryan andKielan-Jaworowska, 1995). As noted bythese authors, the orbital pocket is roofed


dorsally and laterally by the frontal, lacrimal,and maxilla in Kryptobaatar. Because the or-bital pocket is open ventrally, its roof is vis-ible in ventral view (figs. 14, 15).

Either in the frontal or between it and oth-er bones are several nervous and vascular fo-ramina and grooves (figs. 12, 18, 19, 36A).At the posterodorsal limit of the orbital plateof the frontal, beneath the postorbital pro-cess, is a well-developed, anteriorly facingforamen (‘‘otc’’ in figs. 12, 36A). This is theanterior opening of the orbitotemporal canalof Rougier et al. (1992) for the ramus su-praorbitalis of the stapedial artery and ac-companying veins (postorbital foramen ofKielan-Jaworowska et al., 1986). The bound-aries of this aperture are the frontal ventro-medially, the anterior lamina ventrolaterally,and the parietal dorsally. In the right side ofPSS-MAE 101 in which the postorbital pro-cess is missing, the frontal can be seen tocontribute to the floor of the anteriormostpart of the orbitotemporal canal. Runningforward from the anterior opening of the or-bitotemporal canal is a shallow longitudinalsulcus in the frontal. This sulcus turns dor-sally ventral to the supraorbital notch andthen continues onto the notch. Where the sul-cus turns dorsally, a small foramen opens an-teromedially into the frontal (‘‘fdv’’ in figs.18, 36A); this opening may have transmittedthe frontal diploic vein, as does a similarlyplaced foramen in the frontal of the dog(Evans and Christensen, 1979). The frontaldiploic vein has been described in only ahandful of placentals and functions either asan emissary or diploic vein between the su-perior sagittal sinus or veins of the frontalparanasal air sinus and the ophthalmic vein(Thewissen, 1989). Given that the supraor-bital notch is subdivided by a faint ridge, ittransmitted at least two structures from theorbit to the skull roof: one ran anteriorly ina sulcus on the dorsal surface of the frontal,and the other ran posteriorly into a foramenbetween the parietal and frontal. The likelyoccupants of the notch, sulcus, and foramenare branches of the frontal artery, vein, andnerve derived from the ramus supraorbitalis,orbital veins, and ophthalmic nerve, respec-tively. The final aperture associated with thefrontal is the ethmoidal foramen for the eth-moidal vessels and nerve (‘‘ef’’ in figs. 12,

18, 19, 36A). It lies within the orbit on thesuture between the frontal and orbitosphe-noid, at the latter bone’s anterodorsal margin.Leading to the ethmoidal foramen from be-low is a broad sulcus whose anterior borderis formed by a raised ridge along the frontal-orbitosphenoid contact.

MAXILLA

The maxilla is an enormous bone, extend-ing from the snout and palate deep into theorbitotemporal fossa. In lateral view, itslength is more than half that of the entireskull (fig. 33). It bears six cheekteeth, fourpremolars (P1–4) and two molars (M1–2),and it has four distinct processes: facial, zy-gomatic, orbital, and palatal.

The facial process (figs. 10–13, 33) oc-cupies the preorbital region behind the pre-maxilla. It is incomplete in both PSS-MAE101 and 113, but its outer margin is well pre-served in the former. The facial process istall and strongly convex laterally; dorsally, itbends medially such that its contact with thenasal and lacrimal tends toward the horizon-tal (fig. 8). As described by Kielan-Jawo-rowska (1970), the infraorbital canal opensto the rostrum between the embrasures ofP1–P2 or slightly posterior to them (‘‘iof’’ infig. 10). The single infraorbital foramen isdorsoventrally compressed, and its roof isvisible in ventral view (figs. 14, 15). Be-tween the posterior margin of the infraorbitalforamen and the root of the zygomatic archis a small, horizontal shelf, visible in ventralview, that extends laterally from P2, P3, andthe anterior half of P4 (figs. 14, 15). Thisshelf merges laterally with the zygomaticprocess of the maxilla and likely providedadditional attachment area for the anteriorpart of the superficial masseter (Gambaryanand Kielan-Jaworowska, 1995).

The zygomatic process is preserved onboth sides in PSS-MAE 101 (figs. 10, 12, 22)and on the left in PSS-MAE 113 (figs. 13,23). It is a robust, posterolaterally trendinglamina of bone that contacts the squamosalposteriorly and the feeble jugal medially. Itis the main constituent of the zygomatic archand, following the characterization of Kie-lan-Jaworowska (1970), is described as con-fluent with the snout, that is, there is no flare


Fig. 10. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 101 in right lateralview. Gray pattern represents matrix; parallel lines denote breakage. Abbreviations: al anterior lamina;ali alisphenoid; fbu foramen buccinatorium; fr frontal; iof infraorbital foramen; lac lacrimal; mc mas-seteric crest; mx maxilla; naf nasal foramen; ocon occipital condyle; ompf orbital opening of minorpalatine foramen; opf optic foramen; or orbitosphenoid; pmx premaxilla; son supraorbital notch; spfsphenopalatine foramen; sq squamosal; tg temporal groove.

in the zygomatic arch (figs. 8, 14). As notedby Kielan-Jaworowska (1970), the posterioredge of the root of the zygomatic processoriginates opposite the posterior half of P4.The ventral margin of the zygomatic processis fairly straight and horizontal, and posteri-orly it stops short of the glenoid fossa; thedorsal margin does not extend posteriorly asfar as the ventral margin, producing a diag-onal suture between the maxilla and the

squamosal (figs. 10, 12, 33). On the lateralsurface of the zygomatic process is a distinct,elongate, arcuate depression extending fromthe squamosal suture forward to the root ofthe zygomatic arch and being bordered dor-sally by a crest. This depression and crest,the anterior zygomatic ridge (‘‘azr’’ in fig.14), provided attachment area for the anteriorpart of the superficial masseter muscle (Gam-baryan and Kielan-Jaworowska, 1995). The


Fig. 11. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in right lateralview.

anterior origin of this crest is on the ventralface of the shelf extending between the den-tal arcade, the posterior edge of the infraor-bital foramen, and the root of the zygomaticprocess. The anterior origin is marked by ablunt, rounded process and a conspicuouskidney-shaped depression behind that. Onthe left zygomatic arch in PSS-MAE 101, thejugal is lost, exposing an elongate, shallowfacet on the medial surface of the maxilla forthis element (‘‘juf’’ in figs. 8, 22).

The orbital process is complete in PSS-MAE 101 (fig. 18), but it is broken in PSS-MAE 113 (fig. 19). It is very extensive andrepresents the main constituent of the orbitalmosaic. In the anterior part of the orbit (figs.18, 19), the maxilla forms the wall and baseof the roof, which is completed by the frontaland lacrimal; there is no floor to the orbit,which emphasizes the vertical nature of themaxilla here. In the posterior part of the orbit(figs. 10, 12), the maxilla is confined to theinferior one-fourth of the wall and forms thefloor to the sphenorbital recess, the spacemedial to the lateral rim of the sphenorbital

fissure (‘‘sphf’’ in fig. 36A), which is the ap-erture that transmitted nerves and vesselsfrom the cavum epiptericum into the orbit.The posteromedial extent of the maxilla can-not be confidently identified, because it isdeep within the sphenorbital recess. That re-cess is completed by the frontal and orbito-sphenoid dorsomedially and by the alisphe-noid posterolaterally. In the anteroventralcorner of the orbital process of the maxilla,immediately above the root of the zygomaticarch, a circular opening into the infraorbitalcanal is present (‘‘oiof’’ in fig. 18). Prepa-ration of the infraorbital canal on the rightside of PSS-MAE 113 exposed two openingsleading medially into the maxilla (either tothe maxillary sinus or nasal cavity), whichmay represent entrances into the alveolar ca-nals, which transmitted alveolar branches ofthe infraorbital nerves and vessels (Rougieret al., 1997a). The presence of alveolar ca-nals is confirmed by the study of high-reso-lution CT scans of PSS-MAE 101. In the or-bital process at the level of the anterior partof M2 is a large, circular sphenopalatine fo-


Fig. 12. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 101 in left lateralview, with accompanying line drawing. Gray pattern represents matrix; parallel lines denote breakage.Abbreviations: aar atlas arch; al anterior lamina; ano nasal notch; ef ethmoidal foramen; fr frontal; izrintermediate zygomatic ridge; maf masseteric fossa; mafo masseteric fovea; mf mental foramen; mxmaxilla; na nasal; ocon occipital condyle; opf optic foramen; or orbitosphenoid; otc orbitotemporalcanal; pa parietal; pmx premaxilla; pop postorbital process (broken); ppr paroccipital process; ptcposttemporal canal; son supraorbital notch; spf sphenopalatine foramen; sq squamosal; sth stylohyal.

ramen, the enclosing elements of which dif-fer between the two specimens. In PSS-MAE113 the foramen is entirely within the max-illa and the frontal approaches but does notcontribute to the rim (‘‘spf’’ in fig. 19). Onthe other hand, in PSS-MAE 101, the pos-terodorsal margin of the foramen is formedby the frontal (fig. 18). The sphenopalatine

foramen transmitted the sphenopalatinenerve and vessels, which went to the nasalcavity, and the major palatine nerve and ves-sels, which went to the hard palate. The dor-sal aperture into the palatine canal, whichtransmitted the major palatine nerve and ves-sels to the palate, cannot be seen. Posteriorto the sphenopalatine foramen, a sulcus, the


Fig. 13. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in right lateralview.

sphenopalatine groove of Kielan-Jaworow-ska et al. (1986), extends posteriorly alongthe maxillary–frontal suture toward the sphe-norbital fissure (fig. 19). Occupying thisgroove were the nerves and vessels destinedfor the sphenopalatine foramen. Near theposterior limit of the orbital process of themaxilla is a foramen that runs anteroventrallyinto the maxilla (‘‘ompf’’ in figs. 10, 36A);we believe this is the dorsal opening into theminor palatine canal that transmitted the mi-nor palatine nerves and vessels to the hardand soft palate.

The palatal processes, the largest elementsof the hard palate (fig. 34), are fully acces-sible in PSS-MAE 113, where they are com-plete but distorted (fig. 15); in PSS-MAE101, the lower jaws obscure all but the an-terior- and posteriormost parts of the palatalprocesses (fig. 14). When restored to theirnatural positions, the palatal processes aremoderately concave ventrally. As noted byKielan-Jaworowska and Dashzeveg (1978),palatal vacuities are absent. The intermaxil-lary suture extends from the premaxilla to

the level of the anterior half of M1, where itmeets the transverse suture with the pala-tines. The palatines have an essentiallysquare-shaped exposure at the rear of thehard palate. Only a small splinter of the max-illa is interposed between the palatine and themedial alveolar margin. Despite distortion inPSS-MAE 113 (fig. 15), two medium-sizedforamina (one on each side) are preservedthat notch the transverse part of the maxil-lary–palatine suture (see fig. 34); these arethe major palatine foramina, which transmit-ted the major palatine nerves and vessels.Where the longitudinal part of the maxillary–palatine suture meets the alisphenoid, lateralto the postpalatine torus, there is a slitlikeforamen, called here the minor palatine fo-ramen (‘‘mpf’’ in fig. 14), that leads poste-rodorsally into a canal, the minor palatine ca-nal. The alveolar portion of the maxilla pro-jects ventrally below the level of the palateand extends posteriorly beyond the posteriormargin of the hard palate to contact the ali-sphenoid.

A small triangular portion of the orbital


Fig. 14. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 101 in ventral view,with accompanying line drawing. Gray pattern represents matrix. Abbreviations: ali alisphenoid; axaxis; azr anterior zygomatic ridge; bo basioccipital; cp crista parotica; fhy fragment of hyoid arch; foiforamen ovale inferium; fv fenestra vestibuli; gl glenoid fossa; i3a alveolus for third upper incisor; infincisive foramen; iof infraorbital foramen; jf jugular fossa; mpf minor palatine foramen; msy mandibularsymphysis; muf muscular facet; mx maxilla; oiof orbital aperture of infraorbital canal; P3 third upperpremolar; pal palatine; pat postpalatine torus; pef perilymphatic foramen; pmx premaxilla; ptc posttem-poral canal; ptca pterygoid canal; rvnf recess for vascular and nervous foramina (prootic canal, ventralascending canal, canal for ramus inferior, and facial canal); sth stylohyal; tpmx thickenings of premax-illa; ttf tensor tympani fossa; vo vomer.

process of the maxilla is exposed in the lat-eral wall and roof of the choana. It is limitedanteroventrally by an oblique contact withthe palatine, medially by the pterygoid, andposteroventrally by the alisphenoid.

PALATINE

The palatine lacks any orbital exposureand therefore can only be seen in ventralview, where it is fully exposed in PSS-MAE113 (fig. 15), but is partially covered by ma-trix left on the palate to support the lowerjaws in PSS-MAE 101 (fig. 14). The absenceof the orbital exposure has been confirmedin PSS-MAE 101 via the study of high-res-olution CT sections through the skull. Thepalatine contacts the maxilla on the palate,the alisphenoid at the ventrolateral edge ofthe choana, and the maxilla (and likely the

pterygoid) inside the choana. A contact withthe vomer within the choana is not preservedand was probably absent.

The horizontal portion of the palatine issubrectangular (fig. 34). As seen in PSS-MAE 113 (fig. 15), the anterior limit of thepalatine, at the level of the P4/M1 embrasure,is the transverse suture with the maxilla,which encloses the major palatine foramen.The lateral limit is the longitudinal suturewith the maxilla, which encloses the minorpalatine foramen, and the posterior limit isformed by the strong postpalatine torus. Par-allel to the sides and close to the suture withthe maxilla are four to five small foramina inthe palatine in the position of accessory pal-atine foramina (fig. 34). No accessory pala-tine foramina are visible in the exposed areaof the left palatine in PSS-MAE 101 (fig.


Fig. 14. Continued.

14); that portion of the right palatine is cov-ered by matrix. In the midline, the palatineprojects ventrally, forming a strong crestalong the indistinguishable interpalatine su-ture. This crest is higher and thicker poste-riorly and at the border of the choanae merg-es into the postpalatine torus (‘‘pat’’ in fig.14). The torus is massive, thick anteropos-teriorly, projects strongly ventrally, and ex-tends laterally to the suture with the maxillaand the alisphenoid. As preserved, the torusextends ventrally to the level of the occlusalsurface of M2. The lateral part of the torus,at the level of the middle of M2, is continued

posteriorly by a wedge-shaped process thatis covered laterally by the alisphenoid. Thisprocess marks the ventrolateral border of thechoana.

The palatine forms the floor and a smallpart of the ventrolateral wall of the choana.Behind the palatine contribution to the ven-trolateral choanal wall is the maxilla. Thesetwo bones are separated by an oblique suturethat runs anterodorsally inside the choanaabove the minor palatine foramen. In themidline between the choanae, the palatinehas a strong crest that projects dorsally, par-tially subdividing the air passageway.


Fig. 15. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in ventral view.

PTERYGOID

The pterygoid lies on the skull base, ex-tending from within the choana to the ante-rior pole of the ear region (fig. 34). It is pre-served in both PSS-MAE 101 (fig. 14) and113 (fig. 15), but is most complete on theright side of the former. The following de-scription is based mostly on PSS-MAE 101,where the pterygoid appears to be essentiallycomplete, although parts of its sutures are notvery clear. The medial suture with the pre-sphenoid and basisphenoid is distinct onPSS-MAE 113.

The pterygoid is elongate and underliesthe presphenoid and basisphenoid, which arefused with the alisphenoid and orbitosphe-noid to form the sphenoid complex. The pter-ygoid contacts the alisphenoid laterally, thebasisphenoid posteriorly, and the presphe-noid, vomer, and basisphenoid medially. Itmay reach as far posteriorly as the anteriorpole of the promontorium of the petrosal andepitympanic recess, but the sutures are un-clear. What is visible of the pterygoid shows

no contact across the midline, with the baseof the vomer separating them. The ptery-goids converge rostrally inside the choanae,but the full anterior extent cannot be ascer-tained because the interior of the nasal cavitycannot be accessed.

Each pterygoid bears a tall, longitudinal,ventrally directed crest, the pterygopalatineridge (‘‘ptr’’ in fig. 37A) of Barghusen(1986), that delimits a medial and lateralpterygopalatine trough (‘‘mpt’’ and ‘‘lpt’’ infig. 37A). As there is also a midline crestventral to the presphenoid, likely formed bythe vomer, four troughs extending within thechoanae into the rear of the nasal cavity arepresent on the mesocranium. The two lateraltroughs expand both anteriorly in the nasalcavity and posteriorly on the mesocranium,and their lateral wall likely included a con-tribution from the alisphenoid. However, thesuture between the alisphenoid and pterygoidcannot be established. The two medialtroughs are deeper and become only slightlywider posteriorly. They are confluent behind


the vomer where they are roofed by the sphe-noid complex. The pterygopalatine ridge be-comes taller posteriorly and ends at the levelof the anterior portion of the epitympanic re-cess in a rounded, posteroventrally directedprocess. A similar process was described forKamptobaatar by Kielan-Jaworowska(1971). As noted by her, this process is rem-iniscent of the pterygoid hamulus in modernmammals.

In PSS-MAE 113 (fig. 15), the pterygoidsare broken posteriorly on both sides, expos-ing a groove for the internal carotid artery inthe petrosal. The pterygoid likely formed thefloor to an enclosed carotid canal (see Petro-sal).

SPHENOID COMPLEX

The individual components of the sphe-noid complex—presphenoid, basisphenoid,orbitosphenoid, and alisphenoid—are not de-limited by sutures in any specimens ofKryptobaatar. Although these elementsclearly form a unity in the adult skull, theyare described here as separate entities thatcorrespond with similar structures in livingmammals for which the individual ossifica-tion centers are known. As sutures betweenthe sphenoid and what we interpret as thevomer are lacking or cannot be identified, wedescribe the latter bone together with the pre-sphenoid and basisphenoid.

PRESPHENOID/BASISPHENOID/VOMER

We describe this portion of the sphenoidcomplex and vomer as the elements lying onthe midline in ventral view between the pos-terior edge of the choanae and the basioccip-ital (fig. 34). Based on the morphology inliving mammals, the presphenoid comprisesthe anterior part of this complex, the basi-sphenoid the posterior part, with the vomerlying ventral to both. These three elementsare most fully preserved in PSS-MAE 101(fig. 14); however, additional details areshown in PSS-MAE 113, in which the mid-line crest (vomer) and parts of the pterygoidsare not preserved (fig. 15).

The presphenoid and basisphenoid formthe midline floor of the braincase in the me-socranium; their ventral surfaces are essen-tially flat. Laterally, they contact the ptery-

goids through a mostly longitudinal suturethat diverges laterally toward the back of theskull; therefore, the basisphenoid is widerthan the presphenoid. As mentioned above inrelation to the pterygoids, the midline crestin the roof of the choanae is likely formedexclusively by the vomer. Its anterior termi-nus is obscured by the sediment still presentinside the nasal cavity in both PSS-MAE 101and 113. The midline crest becomes tallerposteriorly, and at its terminus, approximate-ly at the level of the anterior margin of theepitympanic recess, the ventral and posterioredges of the crest become quite massive.This enlargement makes the posterior portionof the crest a robust and prominent, ventrallyprojecting process; it does not, however, ex-tend as far ventrally as the pterygopalatineridges. Behind the midline crest, the basi-sphenoid exposure is wedge-shaped andslightly convex ventrally. Extending posteri-orly from each of the pterygopalatine ridgesto the suture between the basisphenoid andthe basioccipital, there is a small, blunt, lon-gitudinal crest. This crest marks the lateralboundary of the basisphenoid, but thebone(s) forming the crest as well as thoseimmediately lateral to it cannot be confi-dently ascertained. Candidates include thepterygoid and the petrosal. In Kamptobaatar(ZPAL MgM-I/33), this crest is formed bythe pterygoid.

In PSS-MAE 113 (fig. 15), the back partof the right pterygoid is missing, exposingthe overlying basisphenoid. A large antero-medially directed foramen on the thick dor-solateral surface of the basisphenoid is foundat the end of a vascular groove traceableback to the anterior pole of the promonto-rium. We interpret this foramen and grooveas for the internal carotid artery. As statedabove, the pterygoid likely formed the floorto an enclosed carotid canal and obscured thecarotid foramen in the basisphenoid.

ORBITOSPHENOID

The orbitosphenoid is the most complicat-ed portion of the sphenoid complex. It formspart of the posteromedial wall of the orbito-temporal fossa and is continuous posteriorlywith the ossified primary wall of the brain-case, the pilae antotica and metoptica (see


Fig. 16. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 101 in posterior view,with accompanying line drawing. Gray pattern represents matrix; parallel lines denote breakage. Ab-breviations: aar atlas arch; ax axis; con (mandibular) condyle; cor coronoid process; exoc exoccipital;lac lacrimal; lacf lacrimal foramen; mx maxilla; ocon occipital condyle; pc pterygoid crest; pet petrosal;ptc posttemporal canal; sq squamosal; sup supraoccipital.

Endocranium). Additionally, the orbitosphe-noid wholly or partially encloses the exits forthe nerves and vessels that reached the orbitfrom the cranial cavity and constitutes theposteroventral edge of the ethmoidal foramen(figs. 33, 36A). The orbitosphenoid is a del-icate element and is only partially preservedin both PSS-MAE 101 (figs. 10, 12, 18) and

113 (figs. 11, 13, 19). However, the speci-mens showing the endocranial surface of theskull provide additional information that en-ables a fairly complete reconstruction of theorbitosphenoid’s external morphology.

As is visible in lateral view (figs. 10–13,33, 36A), the orbitosphenoid contacts thefrontal anteriorly and dorsally, the maxilla


Fig. 17. Stereophotograph of the skull of Kryptobaatar dashzevegi PSS-MAE 113 in posterior view.

ventrally, and the anterior lamina posterolat-erally. The orbitosphenoid is a laminar ele-ment with two major components, an obliquedorsal portion forming the floor of the cranialcavity and a subvertical ventral portion con-necting the braincase to the skull base.Where these two portions meet is a broadgroove leading into the ethmoidal foramenfrom behind and below. It transmitted nervesand vessels to the ethmoidal foramen and be-comes deeper and broader dorsally as it ap-proaches the foramen.

The dorsal portion of the orbitosphenoidhas a slightly arched contact with the frontaldorsally and extends from the ethmoidal fo-ramen in front to the anterior lamina behind.Ventral to its contact with the frontal, thedorsal portion of the orbitosphenoid bulgesanterolaterally, reflecting the outer contour ofthe brain. Posteriorly, at the level of the ros-tral edge of the hypophyseal (pituitary) fossa,the dorsal portion flares laterally, contribut-ing at least partially to the dorsomedial edgeof the sphenorbital fissure. This aperture iscompleted laterally by the anterior lamina.

The ventral portion of the orbitosphenoidextends ventrally from the ethmoidal fora-men in front and converges on the midlineto contact the ventral portion of the orbito-sphenoid of the opposite side. Together, inPSS-MAE 101, these elements form a mid-line structure, a broad, short crest that ven-trally contacts the maxilla as well as the pre-sphenoid and basisphenoid portions of thesphenoid complex. Anterodorsally, this crest

is notched on both sides. Upon comparisonwith the endocranial specimens, it is clearthat the notches on the left and right sidesconstitute the anteroventral portions of theoptic foramen (‘‘opf’’ in figs. 10, 12), whichwould have been completely enclosed withinthe orbitosphenoid (figs. 33, 36A). On theright side of PSS-MAE 101 dorsomedial tothe sphenorbital fissure is a small, circularforamen anterolaterally directed. A similaraperture seems to be present on the right sideof PSS-MAE 113, but is not as clear. Aftercomparison with the specimens showing theendocranial surfaces, this aperture is inter-preted as the metoptic foramen for the ocu-lomotor nerve (‘‘mef’’ in fig. 36A).

ALISPHENOID

The alisphenoid is a small, laminar com-ponent of the sphenoid complex with contri-butions to the orbitotemporal region and themesocranium. It is preserved on both sidesof PSS-MAE 101 (fig. 14) and on the leftside of 113 (fig. 15).

Despite its small size, the alisphenoidtouches a number of bones on the skull base(figs. 14, 34, 37A) and in the orbitotemporalfossa (fig. 10). Its contacts are on the palatethe maxilla and palatine anteriorly, on thechoana the pterygoid medially, on the me-socranium the petrosal posteriorly and me-dially, and in the orbitotemporal fossa the an-terior lamina posterodorsally, the remaining


Fig. 18. Stereophotograph of the left orbitotemporal region of the skull of Kryptobaatar dashzevegiPSS-MAE 101 in oblique dorsolateral view, with accompanying line drawing. Gray pattern representsmatrix; parallel lines denote breakage. Abbreviations: al anterior lamina; ali alisphenoid; fdv foramenfor frontal diploic vein; fr frontal; frt foramen for ramus temporalis; lac lacrimal; man mandible; mxmaxilla; na nasal; oiof orbital aperture of infraorbital canal; or orbitosphenoid; pa parietal; sgf supra-glenoid foramen; spf sphenopalatine foramen; sq squamosal.

Fig. 18. Continued.

portions of the sphenoid complex medially,and the maxilla anteriorly.

The alisphenoid has the shape of a portionof a Mobius strip, so that its medial marginin the lateral wall of the choana through acontinuous line becomes the lateral edge ofthe epitympanic recess in the ear region.Based on this shape, the alisphenoid can bedivided into anterior and posterior portions.The anterior portion is exposed only on itsventral surface; the posterior portion is ex-posed ventrally and dorsally. The anteriorportion (figs. 14, 34, 37A) contacts the pal-atine and maxilla in the wall of the choana.Lateral to this contact, there is a deeply re-cessed, semilunar area on the alisphenoidthat probably housed the medial pterygoidmuscle (Gambaryan and Kielan-Jaworow-ska, 1995: fig. 8B). The anterolateral part ofthis attachment area for the medial pterygoidis completed by the maxilla. From the an-teromedial edge of this recess, a slender pro-cess of alisphenoid extends forward approx-imately to the level of M2 and contributes tothe minor palatine foramen (fig. 14). Medi-ally, the anterior portion of the alisphenoidmeets the pterygoid in the lateral pterygo-palatine trough, but as already stated, the su-


Fig. 19. Stereophotograph of the left orbitotemporal region of the skull of Kryptobaatar dashzevegiPSS-MAE 113 in oblique dorsolateral view, with accompanying line drawing. Gray pattern representsmatrix; parallel lines denote breakage. Abbreviations: al anterior lamina; ef ethmoidal foramen; fr fron-tal; ju jugal; lac lacrimal; mx maxilla; pa parietal; pop postorbital process (broken); spf sphenopalatineforamen; sq squamosal.

Fig. 19. Continued.

ture between these two bones is unclear (fig.37A).

The posterior portion of the alisphenoidcan be divided into two parts: one exposedventrally and the other dorsally. The ventralexposure constitutes the anterior pole of theepitympanic recess (fig. 37A). It also formsthe anterior part of the tall crest demarcatingthe lateral margin of the epitympanic recess;the bulk of this crest is formed by the ante-rior lamina. Extending medially from the an-terior pole of the epitympanic recess is atongue of alisphenoid that forms the antero-lateral edge of the posterior aperture into thecarotid canal (see Petrosal). The dorsal ex-posure of the posterior portion of the ali-sphenoid lies behind the maxilla and in frontof the anterior lamina (fig. 10). This part ofthe alisphenoid forms the ventrolateral edgeof the sphenorbital recess (figs. 10, 36A), thelateral limit of which is demarcated by a crestthat continues anteroventrally in a ridgealong the alisphenoid–maxillary suture. In-ferior to this ridge is a concave surface onthe dorsal exposure of the alisphenoid that iscontinuous with the surface described in ven-tral view as for the medial pterygoid muscle.


This concave surface likely represented anadditional attachment area for pterygoidmusculature, including the lateral pterygoid(Gambaryan and Kielan-Jaworowska, 1995:fig. 8B), which we term the pterygoid fossaof the alisphenoid.

Close to or within the suture between thedorsal exposure of the alisphenoid and theanterior lamina is a small, anteriorly directedforamen (‘‘fbu’’ in figs. 10, 36A). We thinkthat this foramen communicated with the ca-vum epiptericum, based on an as yet undes-cribed skull of a new species of MongolianLate Cretaceous multituberculate (PSS-MAE126). Among living mammals, comparablysituated foramina (i.e., within the attachmentarea of the pterygoid musculature and con-necting to the cavum epiptericum) are re-ported in some rodents and transmit branchesof the mandibular nerve (Hill, 1935; Wahlert,1974). In fact, four apertures in or along thealisphenoid accommodating branches of themandibular nerve are known to occur in ro-dents: the foramen ovale, the foramen ovaleaccessorius, the masticatory foramen, and thebuccinator foramen (Wahlert, 1974). Of thepossible occupants of the anteriorly directedopening in the pterygoid fossa of Kryptobaa-tar, we think the most plausible was the buc-cal nerve, which is the most medial branchof the anterior division of the mandibularnerve and runs forward dorsal to the lateralpterygoid muscle in the dog (Evans andChristensen, 1979). Consequently, we iden-tify this opening as the foramen buccinato-rium (figs. 10, 36A).

PETROSAL

The petrosal is the most complex elementof the basicranium; it houses the organs ofhearing and balance, and contributes to thebraincase wall on the lateral surface, floor,and occiput. In living therians, the petrosal isgenerally conceived as comprising two dif-ferent regions: the pars cochlearis (housingthe cochlea) and the pars canalicularis (hous-ing the vestibule and semicircular canals). Inmultituberculates, in addition to these two re-gions, an anterior lamina forms part of thelateral wall of the braincase and an extensiveepitympanic recess. Among living mammals,an anterior lamina contributing to similar ar-

eas of the skull is found only in monotremes.The monotreme anterior lamina forms as thelamina obturans, an intramembranous ossifi-cation in the sphenoobturator membrane, thatfuses with the endochondral petrosal properin subsequent ontogeny (Kuhn, 1971; Pres-ley, 1981; Zeller, 1989). Given that themonotreme lamina obturans is the only mod-el for the anterior lamina in extinct taxa, itseems likely that the element in multituber-culates and other mammaliaforms also formsintramembranously.

We describe the petrosal of Kryptobaatarin three views—ventral, lateral, and occipi-tal—reserving the endocranial surface of thepetrosal for a description of that space as asingle unit (see Endocranium below). InPSS-MAE 101, both petrosals are preservedin situ, but the left is partially hidden in ven-tral view by the stylohyal and matrix left tosupport the stylohyal (figs. 14, 20). In PSS-MAE 113, the right petrosal is preserved inlife position, but the left is displaced poste-roventromedially and the anterior end of itsepitympanic recess is missing (figs. 15, 21).

Ventral View (figs. 14, 20, 21, 34, 37A):In this view, the petrosal contacts mediallythe basisphenoid and basioccipital, postero-medially the exoccipital, laterally the squa-mosal, and anteriorly the alisphenoid andpterygoid. The most conspicuous feature ofthe petrosal is the promontorium (‘‘pr’’ in fig.37A), the tympanic surface of the cochlearhousing, which is elongated, ventrally flat-tened, and anteromedially directed. In bothPSS-MAE 101 and 113, the ventral surfaceof the promontorium is distinctly marked bygrooves, presumably for the internal carotidand stapedial arteries, and it bears flangesalong its medial aspect. However, these sur-face features do not mask the essential fin-gerlike contour of the promontorium, whichin turn likely reflects the shape of the en-closed cochlear duct, known to be rodlike inother Late Cretaceous and Paleocene multi-tuberculates (Miao, 1988; Luo and Ketten,1991; Fox and Meng, 1997; Hurum, 1998b).

The grooves on the promontorium (fig.37A) exhibit a Y-shaped pattern. The shortstem of the Y, the medial groove, is orientedin a near transverse plane at a level posteriorto the basisphenoid–basioccipital suture; it isinterpreted as having housed the internal ca-


Fig. 20. Stereophotograph of the basicranium of Kryptobaatar dashzevegi PSS-MAE 101.

rotid artery. The anteriorly trending arm wasfor the rostral continuation of the internal ca-rotid (‘‘gica’’ in fig. 37A), and the postero-laterally directed arm was for the internal ca-rotid’s main extracranial branch, the stapedialartery (‘‘gpsa’’ in fig. 37A). Of these threegrooves, the widest is the main stem of theinternal carotid, and the longest is the onefor the stapedial artery, which is subequal indiameter to the one for the rostral continua-tion of the internal carotid. This Y-shapedpattern is very rostrally positioned on thepromontorium, which is unusual among liv-ing mammals (Wible, 1987) but is found insome other multituberculates (e.g., Kampto-baatar, ZPAL MgM-I/33).

From the point of the origin of the stape-dial artery, the internal carotid groove ex-tends anterolaterally close to the medial bor-der of the epitympanic recess. On the rightside of PSS-MAE 101 (fig. 20), the carotidgroove leads anteriorly into a foramen situ-ated between the lateral aspect of the anteriorpole of the promontorium and the medialmargin of the epitympanic recess. This fo-

ramen is formed jointly by the alisphenoidand petrosal and is called here the posterioraperture of the carotid canal (‘‘pacc’’ in fig.37A). The lateral position of this aperture de-termined a long course for the artery to thecarotid foramen in the hypophyseal fossa,which is known from the specimens showingthe endocranial surface. The exact course ofthe internal carotid artery from the posterioraperture of the carotid canal to the endocra-nium is not entirely clear, as it was hiddenwithin bone, but the following reconstructionprovides the best fit with the evidence avail-able. On the right side of PSS-MAE 113 (fig.21), the back part of the pterygoid is missingand the rostral continuation of the carotidgroove can be followed as it curves mediallyon the anterior pole of the promontorium.From the promontorium, the carotid grooveruns anteromedially to a foramen in the dor-solateral part of the basisphenoid that pre-sumably represents the ventral aperture ofthe carotid foramen in the hypophyseal fossa.With the pterygoid in place, as on the rightside of PSS-MAE 101 (fig. 20), the carotid


Fig. 21. Stereophotograph of the basicranium of Kryptobaatar dashzevegi PSS-MAE 113.

groove on the anterior pole of the promon-torium was presumably enclosed in a canal,which left no visible trace on the ventral ba-sicranial surface. The carotid canal appearsto be have been formed by the petrosal, ali-sphenoid, and pterygoid. The length of thecarotid canal is a consequence of the lateralposition of its posterior aperture and the verydorsal placement of the ventral aperture ofthe carotid foramen, which in turn implies avery thick basisphenoid from the floor of thehypophyseal fossa to the ventral surface ofthe skull base.

The absence of the pterygoid on the rightside of PSS-MAE 113 exposes anothergroove in the alisphenoid that notches thedorsal surface of the ventral aperture of thecarotid canal and leads anterolaterally towardthe orbit (fig. 21). It is uncertain whether thissecond groove opened in the posterior aper-ture of the carotid canal or in a separate fo-ramen anteromedial to it. The exact positionof the orbital opening of this groove is un-certain, but was likely in the floor ventral tothe foramen buccinatorium. This anterolat-erally directed groove was likely enclosed ina canal by the pterygoid that is subequal tothat for the internal carotid artery. In fact, onthe right side of PSS-MAE 101, in which thepterygoid is in place, there is a ridge runningforward from the posterior aperture of thecarotid canal that likely marks the positionof the enclosed canal (fig. 20). We interpretthis canal as the pterygoid (Vidian) canal

(‘‘ptca’’ in fig. 37A). Its usual occupant inmodern mammals is the nerve of the ptery-goid canal, which is formed by the greaterand deep petrosal nerves, carrying parasym-pathetic and sympathetic fibers, respectively(Evans and Christensen, 1979; Williams etal., 1989). However, in some instances, thereis also an accompanying artery off the inter-nal carotid (McDowell, 1958; MacPhee,1981). The relatively large size of the pre-served groove on the alisphenoid in PSS-MAE 113 suggests that the pterygoid canalcontained both an artery and nerve. The en-trance of the deep petrosal nerve, a branchof the internal carotid nerve, into the ptery-goid canal must have been from the carotidgroove, but the entrance of the greater petro-sal nerve is unclear. Given that there is nolikely tympanic aperture for the greater pe-trosal nerve, we speculate that it must haveentered the pterygoid canal through the su-ture (gap) between the petrosal and alisphe-noid directly from the cavum epiptericumand/or cavum supracochleare.

From its origin, the groove for the stape-dial artery runs posterolaterally on the pro-montorium toward the fenestra vestibuli oroval window (‘‘fv’’ in figs. 14, 37A). Thestapedial groove exhibits a slightly differentrelationship to the fenestra vestibuli in PSS-MAE 101 and 113. In the former, the groovenotches the rim of the fenestra vestibuli,whereas in the latter it extends slightly pos-terior to the fenestra, with the crest forming


the ventral edge of the stapedial groove ex-tending posteroventral to the ventral marginof the oval window.

On the medial aspect of the ventral pro-montorial surface, near the contact with thebasisphenoid, is a fan-shaped flange trendinganteroposteriorly. Running transversely onthe surface of the flange is the groove for themain stem of the internal carotid describedabove. Extending rostrally from the sulcus onthe flange is a low crest, which we call herethe rostral tympanic process of the petrosal.It may mark the contact with either a mem-brane, cartilage, or bone contributing to thefloor of the tympanic cavity, as it does inextant mammals (Novacek, 1977; MacPhee,1981). However, the composition of thisfloor, whether membranous, cartilaginous, orbony, cannot be determined. To date, there isno evidence for a bony bulla in any non-therian mammal. On the right side of PSS-MAE 101, the rostral tympanic process ofthe petrosal is continuous with a low crestthat extends rostrally to the base of the ptery-gopalatine ridge (fig. 20). At least the pos-terior part of this crest is petrosal.

At the posterior end of the promontorium(fig. 37A) are two apertures, the fenestra ves-tibuli and the perilymphatic foramen, sepa-rated by a narrow bridge of bone, the cristainterfenestralis, that reaches posteriorly tothe base of the paroccipital process. The peri-lymphatic foramen (‘‘pef’’ in figs. 14, 37A)is roughly circular, and the subequal fenestravestibuli has an average stapedial ratio (seeSegall, 1970) of 1.39 in PSS-MAE 113. Thefenestra vestibuli is oriented in a near verticalplane and faces anterolaterally with a slightventral component. Immediately anterior tothe fenestra vestibuli is a deeply excavatedpocket (‘‘ttf’’ in figs. 14, 37A), the fossa forthe tensor tympani muscle (fossa muscularismajor of Kielan-Jaworowska et al., 1986).The lateral margin of the tensor tympani fos-sa is formed by the flaring medial edge ofthe epitympanic recess.

The perilymphatic foramen and environsare best shown on the right side of PSS-MAE101 but are preserved on both sides in PSS-MAE 113. The foramen is posteriorly di-rected and only its ventral margin is well de-limited; its roof lacks a definitive edge andis formed by the petrosal’s contribution to the

very conspicuous jugular fossa (‘‘jf’’ in figs.14, 37A), the large depression around thejugular foramen (‘‘jfo’’ in fig. 37A). A sulcusfor the cochlear aqueduct, which is presenton the petrosal in some other multitubercu-lates (Rougier et al., 1996c; Fox and Meng,1997), is lacking. We term this aperture inKryptobaatar a perilymphatic foramen, be-cause observations on the endocranium ofPSS-MAE 123 have failed to reveal a coch-lear aqueduct (cochlear canaliculus), a bonycanal that transported the perilymphatic duct.Consequently, the only possible channel forthe perilymphatic duct from the inner ear tothe jugular foramen was via the perilym-phatic foramen.

Immediately behind the perilymphatic fo-ramen and forming part of its posterodorsaledge is the prominent ventral bulge housingthe posterior ampulla of the semicircular ca-nals. Portions of the lateral and posteriorsemicircular canals can be traced posteriorlyfrom the posterior ampulla. These, alongwith the crista interfenestralis, delimit twodeep pits on either side of the posterior am-pulla. The larger pit is the one lateral to theampulla, which is encircled by the posteriorextension of the crista interfenestralis and theposterior extension of the lateral semicircularcanal. The small medial pit is subtriangularand is placed between the posterior exten-sions of the lateral and posterior semicircularcanals. These pits likely housed the expandedmiddle-ear cavity.

Positioned medial and very near to theperilymphatic foramen is the jugular foramen(fig. 37A), which is about half the size of theformer aperture. The jugular foramen lies onthe suture between the petrosal and exocci-pital; the basioccipital appears to be exclud-ed, judging by a suture visible on the rightside of PSS-MAE 101 (see Exoccipital).Also visible on the same side of this speci-men is a smaller foramen just posterolateralto the jugular foramen completely encircledby the petrosal; the function of this openingis unknown and its presence in PSS-MAE113 cannot be verified.

Directly in front of the anteromedial edgeof the perilymphatic foramen is a niche onthe promontorium, which is deeply recessed,more so in PSS-MAE 113 than in 101. Thisniche represents the anteromedial extent of


the greatly expanded jugular fossa, the largedepression around the jugular foramen (figs.14, 37A). A robust, long laminar processprojects posteromedially from the medialside of the promontorium and forms a hori-zontal shelf that partially floors this part ofthe jugular fossa. This shelf is well devel-oped in PSS-MAE 113 (fig. 21), but barelyhas any horizontal contribution in PSS-MAE101 (fig. 20).

As stated above, the crista interfenestralisreaches posteriorly to contact the paroccipitalprocess (‘‘ppr’’ in fig. 37A). Lateral to thecrista interfenestralis and in front of the par-occipital process (and hidden by that struc-ture in the figures) is a broad depression inthe tympanic roof, the fossa for the stapediusmuscle. The limits of this depression are notconspicuous, but its ventral extension on thelateral face of the crista interfenestralis isvery distinct. Nevertheless, this depression isat least twice the surface area of the fenestravestibuli. The well-developed, triangular par-occipital process (fig. 37A) is preserved onboth sides of PSS-MAE 101 (fig. 20) and onthe right side of 113 (fig. 21). It is not avertical structure, but is slanted somewhatanteroventrally. From the blunt apex of theparoccipital process two crests arise, oneleading medially and the other laterally (fig.37A). The medial one, the caudal tympanicprocess of the petrosal (‘‘ctpp’’ in fig. 37A),is very short and marks the posteroventrallimit of the conspicuous jugular fossa. Thethree bones encircling the jugular fossa (pe-trosal, exoccipital, and basioccipital) providelaminae that wall and partially floor thisspace. As one of these laminae, the caudaltympanic process of the petrosal is coplanarwith the laminar projections from the otherbones.

The lateral crest arising from the parocci-pital process is the crista parotica (‘‘cp’’ infig. 37A). It is L-shaped with transverse andlongitudinal components. The shorter trans-verse component is anteroposteriorly thickand smooth and runs from the tip of the par-occipital process to a scooped area housingthe external acoustic meatus on the squa-mosal-petrosal suture. The posterior part ofthe longitudinal component is not well indi-vidualized, and sutures indicate that its lat-eral surface contacted the squamosal; it was

here that the external acoustic meatus likelyentered the middle ear. The anterior part ofthe longitudinal component, beginning at thelevel of the posterior margin of the fenestravestibuli, is more prominent. It runs antero-medially as a tall subvertical crest that limitsthe medial margin of the epitympanic recessand continues forward as the medial edge ofthe infolded lateral flange.

Slightly posterolateral to the level of thefenestra vestibuli, the crista parotica showsthe attachment of a well-developed tympa-nohyal (‘‘th’’ in fig. 37A). In ventral view,this pronglike element is posteriorly and onlyslightly medially directed; in medial view, ithas a distinctly triangular outline. Thetympanohyal appears completely preservedon the left side of PSS-MAE 113 (figs. 21)and has a hooklike profile for cradling thehyomandibular branch of the facial nerve.This nerve left the middle ear via a stylo-mastoid notch immediately posterior to thetympanohyal.

The epitympanic recess (‘‘er’’ in fig. 37A),following Kielan-Jaworowska et al. (1986),is the fossa above the dorsal margin of thetympanic membrane that accommodated thebody of the malleus and incus and housedthe crus breve of the incus (fossa incudis). InKryptobaatar, the epitympanic recess is adeeply excavated, elongated, ellipsoidal fos-sa trending roughly parallel to the crista pa-rotica (figs. 20, 21, 37A). It has contributionsfrom three bones: the petrosal forms the bulkwith the alisphenoid at the anteriormost end.Additionally, the squamosal forms the pos-terolateral edge of the epitympanic recesswhere the fossa incudis is located. The me-dial wall of the fossa incudis is delimited bythe prominent crista parotica and its rostralcontinuation, the lateral flange. This sharp,ventrally projecting crest forms a barrier be-tween the fossa incudis and the fenestra ves-tibuli, and it severely constrained the likelypositions of the stapes and incus. As in othermultituberculates (Rougier et al., 1996a,1996c), the lateral flange in Kryptobaatar isinfolded such that its anterior end contactsthe cochlear housing; in other Mesozoic taxa,such as Morganucodon and Vincelestes, thelateral flange runs parallel to and is separatedfrom the promontorium by a depression


called the lateral trough (Wible and Hopson,1993).

In the anterior part of the epitympanic re-cess are several foramina interpreted forbranches of the mandibular nerve (‘‘foi’’ and‘‘fma’’ in fig. 37A); Kryptobaatar, as in allknown multituberculates and some rodents(Hill, 1935; Wahlert, 1974), has multiple ap-ertures for this division of the trigeminalnerve, one of which, the buccinator foramen,was described already (see Alisphenoid). InPSS-MAE 101 (fig. 20), two foramina for themandibular nerve are situated in the epitym-panic recess: the foramen ovale inferiumopens into the recess and faces anteroven-trally, whereas the foramen masticatoriumnotches the lateral rim of the recess and facesventrolaterally. The foramen ovale inferiumis centrally placed in a deep fossa and dom-inates the anterior pole of the epitympanicrecess; it is traversed by a broad sulcus,which anteriorly falls just short of the ali-sphenoid’s contribution to the recess. The fo-ramen masticatorium is strongly ellipticalwith the anteroposterior axis longer than thedorsoventral and is placed lateral and slightlyposterior to the foramen ovale inferium. Asimilar arrangement is present in PSS-MAE113 (fig. 21) with the exception that the fo-ramen masticatorium is subdivided into twoforamina on the right side. The division isaccomplished by a narrow bar of bone con-tinuous with the lateral edge of the epitym-panic recess. This delicate bar was complete,but was broken prior to illustration here andin Rougier et al. (1996c: fig. 3). The left sideof PSS-MAE 113 is damaged, but only oneforamen masticatorium seems to have beenpresent. Variation in the number of exits forthe mandibular nerve on different sides ofthe same skull is known in other multituber-culates (e.g., Kamptobaatar, ZPAL MgM-I/33).

Medial and dorsal to the crista parotica,and anterior to the tympanohyal, is a re-cessed area (‘‘rvnf’’ in figs. 14, 37A). On theleft side of PSS-MAE 113, three foraminaare visible in this recess. These foramina leadinto canals presumably within the petrosal,the exact course of which could not be de-termined without damaging the availablespecimens. Our reconstruction of the occu-pants of these foramina and canals is based

on comparison with isolated petrosals of oth-er Late Cretaceous multituberculates (seeKielan-Jaworowska et al., 1986; Luo, 1989;Wible and Hopson, 1995). Of the three fo-ramina in the recess in PSS-MAE 113, theposterior two share a common space and arehigher than the anterior one. The posterior-most foramen opens into a posterolaterallydirected channel and likely transmitted oneof the two end branches of the stapedial ar-tery, the ramus superior, into the ventral as-cending canal (the pterygoparoccipital fora-men). The similarly sized middle foramen isposteroventrally directed and likely transmit-ted the prootic sinus; therefore, it representsthe ventral aperture of the prootic canal. Thedorsal aperture of the prootic canal will bedescribed with the endocranial surfaces. Themorphology of these two foramina, the ven-tral apertures of the ventral ascending andprootic canals, accords well with that report-ed in some other multituberculates (Wibleand Hopson, 1995: fig. 7A). The anterior-most foramen in the recess in PSS-MAE 113,the largest, is anteromedially directed, hori-zontal, and elliptical in outline. We considerit to be a joint aperture leading to two canals,one for the hyomandibular branch of the fa-cial nerve (the secondary facial foramen) andthe second for the other end branch of thestapedial artery, the ramus inferior, an ar-rangement different from that in describedptilodontoid and taeniolabidoid multituber-culates (Luo, 1989; Wible and Hopson,1995). Supporting this interpretation is a dis-tinct sulcus leading posteriorly from this ap-erture to the dorsal rim of the fenestra ves-tibuli that resembles the sulcus for the facialnerve in other multituberculates (Wible andHopson, 1995: figs. 7A, 8A). Moreover, asecond, well-developed foramen at the ante-rior edge of the petrosal, lateral to the prom-ontorium, likely held the rostral continuationof a vessel that initially ran with the facialnerve. That this vessel was the ramus inferioris supported by the large size of the groovefor the stapedial artery on the promontoriumand the small size of the foramen for the ra-mus superior, implying that the other primaryramus of the stapedial artery was present.However, we cannot rule out that a vein, thepost-trigeminal vein, accompanied the ramusinferior in its passage through the petrosal.


The positions of the foramina transmittingthe superior and inferior rami and the sta-pedial groove on the promontorium in bothPSS-MAE 101 and 113 suggest that the sta-pedial artery ran across the fenestra vestibuliand through the presumed bicrurate stapes(Rougier et al., 1996c).

Lateral View (figs. 10–13, 33, 36A): Themost conspicuous feature of the petrosal inlateral view is the anterior lamina. It contactsdorsally the parietal, anteriorly the frontal,orbitosphenoid, and alisphenoid, and poste-riorly it is overlapped by the squamosal; ven-trally, it is continuous with the tympanic sur-face of the petrosal through the lateral flange.The anterior lamina formed wholly or par-tially a number of passageways for nervesand vessels leaving the braincase for the or-bitotemporal fossa and also provided a majorarea of attachment for several muscles ofmastication.

Dorsally, the anterior lamina overlaps theparietal as shown on the left side of PSS-MAE 101. From back to front, the suture be-tween the anterior lamina and parietal runsslightly dorsally from the triple junction ofparietal, squamosal, and anterior lamina tothe anterior opening of the orbitotemporalcanal between the parietal, frontal, and an-terior lamina. Parallel and medial to this su-ture is a preserved endocast on both sides ofPSS-MAE 101 (figs. 10, 12) and the rightside of 113 (fig. 11), representing the fillingof the orbitotemporal canal. The orbitotem-poral canal appears to have been boundedlaterally by the parietal only, with the ante-rior lamina lateral to that. Throughout mostof its course, there is no distinct medial wallfor the orbitotemporal canal, suggesting thatthe orbitotemporal vessels ran endocraniallywithin a sulcus on the medial surface of theparietal. Toward the front, however, the fron-tal provides the medial wall of a true orbi-totemporal canal, and the anterior lamina, pa-rietal, and frontal complete the anterior open-ing of this canal. Directly lateral to the an-terior opening of the orbitotemporal canal isa thickened lateral process of the anteriorlamina that ventrally supports the large post-orbital process of the parietal. This processon the anterior lamina is incomplete on theleft side of PSS-MAE 101, exposing dorsally

the deeply pitted articular surface of the pa-rietal.

Slightly anteroventral to the anterior open-ing of the orbitotemporal canal is the suturebetween the anterior lamina and frontal (fig.12). Along this arched junction, the anteriorlamina lies lateral to the frontal. Farther ven-trally, the anterior lamina meets the orbito-sphenoid at a suture that is concave anteri-orly (only the dorsal portion of which is pre-served in PSS-MAE 101). As is apparentfrom one of the specimens showing the en-docranial surface (fig. 25; PSS-MAE 123),the anterior lamina overlaps the orbitosphe-noid laterally. In the area of the sphenorbitalfissure, the anterior lamina is sharply inflect-ed medially so that a portion of it faces an-teriorly (figs. 10, 12). This surface has a deepfossa that is continuous ventrally with thesurface for muscle attachment in the alisphe-noid, the pterygoid fossa. The likely occu-pant of this area was the lateral pterygoidmuscle. The suture between the alisphenoidand anterior lamina runs obliquely across thismuscular fossa from the anteroventral marginof the sphenorbital fissure to the margin ofthe foramen masticatorium. Midway alongits length, the suture is interrupted by an an-teriorly facing foramen described above ashaving transmitted the buccal branch of themandibular nerve (figs. 10, 36A).

The ventral edge of the anterior lamina(the lateral edge of the lateral flange) formsa gentle arch from the contact with the ali-sphenoid in front to the root of the zygomaticarch behind. Anteriorly, this margin isnotched by the foramen masticatorium.Slightly dorsal and posterior to the edge ofthe foramen masticatorium is another small,anteriorly directed foramen, the supraglenoidforamen for a ramus temporalis of the ramussuperior (‘‘sgf’’ in figs. 8, 18, 36A). This fo-ramen was likely continuous with the ventralascending canal.

Posteriorly, the anterior lamina’s contactwith the squamosal can be divided into twoportions: a ventral one that is essentially hor-izontal, and a dorsal, essentially vertical one.Forming the horizontal portion is a flat flangeof the anterior lamina that buttresses the frontof the root of the zygoma and extends lat-erally toward, but falls short of, the glenoidfossa (‘‘gl’’ in figs. 6, 14, 37A). The hori-


zontal portion of the suture is gently arched,concave posteriorly, and meets the dorsalportion in a small foramen, also for a ramustemporalis (‘‘frt’’ in figs. 8, 36A). The dorsalportion runs along the course of the dorsalascending canal, which, because of breakagein PSS-MAE 101, is shown to be formed bythe squamosal and petrosal. On the left sideof this specimen (fig. 12) there is a notch atthe posterodorsal corner of the anterior lam-ina that opens to the dorsal ascending canaland probably was enclosed in a foramen bythe squamosal posteriorly (‘‘fdac’’ in figs. 8,36A). This dorsal ascending canal foramenalso transmitted a ramus temporalis. Similarforamina are found, for example, in Kamp-tobaatar (ZPAL MgM-I/33) and Lambdop-salis (Miao, 1988).

The external surface of the anterior laminahas a fairly complex topology and, in addi-tion to the lateral pterygoid muscle, providedattachment for the temporalis muscle. Thetemporalis attachment can be divided intotwo major parts: an anterodorsal one that isconvex and best expressed under the post-orbital process, and a ventral concave one.These two parts correspond to the attachmentareas identified as for the pars anterior andpars posterior of the temporalis, respectively,in Nemegtbaatar by Gambaryan and Kielan-Jaworowska (1995).

Occipital View (figs. 16, 17, 35, 37A):Well preserved in both PSS-MAE 101 and113, the mastoid exposure of the petrosalforms the lateral portion of the occiput andcomprises approximately half of the occi-put’s bony surface. The mastoid exposure issubtriangular with the base medially andwith the apex laterally. It contacts the squa-mosal laterally on the ventral portion of thenuchal crest, the parietal dorsomedially onthe dorsal portion of the nuchal crest, and thesupraoccipital and exoccipital medially. Theventral edge of the mastoid exposure isformed by the paroccipital process and thecrests arising from it, namely the caudal tym-panic process of the petrosal medially andthe crista parotica laterally.

There are three depressions on the mastoidexposure. The dorsomedial one continuesonto the exoccipital and is subvertical withits deepest point along the suture between thepetrosal and exoccipital; it lodges the atlas

when the skull is maximally extended (dor-siflexed) on the neck. The dorsolateral de-pression is shallow, roughly circular, and isseparated from the atlantal fossa by a verticalridge. Opening into the dorsolateral depres-sion is the posterior opening of the posttem-poral canal, which is wholly in the petrosal(‘‘ptc’’ in figs. 14, 16, 36A). An elongated,ventral depression runs parallel to the cristaparotica and probably represented the site ofattachment for the sternomastoid muscle(Evans and Christensen, 1979).

JUGAL

The jugal is a thin, essentially oval, lami-nar bone on the inner surface of the zygo-matic arch. In PSS-MAE 101, both zygo-matic arches are preserved, but a nearly com-plete jugal is present only on the right side(fig. 22). In PSS-MAE 113, only the left archis complete and the jugal preserved (fig. 23).

On the right side of PSS-MAE 101 (fig.22A), the jugal is not as tall as the zygoma;it is recessed from both the ventral and dorsalmargins of the zygomatic arch, but is closerto the latter. The anterior extent of the jugalis at the level of the posterior half of M1;posteriorly, it reaches almost to the same lev-el as does the ventral margin of the zygo-matic process of the maxilla. The zygomaticarch, and so the jugal, does not lie in a sag-ittal plane, but is tilted such that its dorsalmargin is slightly lateral to the ventral. Onthe left side of PSS-MAE 101 (fig. 22B), thejugal is not preserved, and a shallow, ovalfacet is exposed. The bulk of the facet is onthe maxilla, with only the posterodorsal fifthon the squamosal. In PSS-MAE 113 (fig. 23),the jugal conforms in most features to that inPSS-MAE 101, but the dorsal margin of thejugal is exposed in lateral view, forming thedorsal edge of the zygomatic arch (fig. 19).Moreover, the dorsal margin is slightly thick-er than the portion of the jugal directly me-dial to the maxilla and squamosal.

SQUAMOSAL

The squamosal is appressed to the poste-rior part of the side wall of the braincase andcan arbitrarily be divided into two parts: thezygomatic process and the dorsal flange. Thesquamosal is well preserved in both PSS-


Fig. 22. Pencil drawings of the right and left zygomatic arches (A and B) of Kryptobaatar dash-zevegi PSS-MAE 101 in dorsal view, with accompanying line drawings. Gray pattern represents matrix;parallel lines denote breakage. Abbreviations: al anterior lamina; fr frontal; ju jugal; juf jugal facet;lac lacrimal; lacf lacrimal foramen; mx maxilla; pa parietal; son supraorbital notch; sq squamosal.


Fig. 23. Pencil drawing of the left zygomatic arch of Kryptobaatar dashzevegi PSS-MAE 113 indorsal view, with accompanying line drawing. Abbreviations: fr frontal; ju jugal; lac lacrimal; mxmaxilla; pa parietal; sq squamosal; tr temporal ridge.

MAE 101 and 113, although in the latter thezygomatic process is incomplete on the rightside.

The squamosal’s contacts are as follows(figs. 8, 12, 32–34): anteriorly, with the max-illa and the jugal via the zygomatic process,and with the anterior lamina via the dorsalflange; medially and posteriorly, with the pe-trosal via the dorsal flange; and dorsally, withthe parietal via the dorsal flange. The squa-mosal has been partially lost in several spec-imens (PSS-MAE 113, 124, 125), exposingan extensive underlying facet on the petrosal.From this, it is evident that the squamosalhas no direct contribution to the side wall ofthe braincase.

The dorsal flange of the squamosal is lam-inar, tongue-shaped, and appressed to theside wall of the braincase. In ventral view(figs. 20, 21, 34, 37A), the dorsal flange con-tacts the petrosal through an L-shaped suture,with the long arm oriented sagittally and themore anteriorly located short arm transverse-ly. The short arm abuts a laterally directed,wedge-shaped projection from the lateralflange of the petrosal. In posterior view (figs.

16, 17, 35, 37A), the dorsal flange contactsthe mastoid exposure of the petrosal to formthe inferolateral extent of the nuchal crest,just lateral to the posterior opening into theposttemporal canal. In dorsal view (figs. 8,9, 31), the chief contact is with the petrosal,in front with the anterior lamina and behindwith the edge of the mastoid exposure form-ing the nuchal crest. The dorsal flange alsohas a narrow contact with the parietal dor-somedially. The suture with the anterior lam-ina can be divided into two arched portions:the anterior one is chiefly horizontal and iscontinuous with the front edge of the zygo-matic process; the posterior one is verticallydirected (fig. 8). Where these two portions ofthe squamosal–petrosal suture meet is a smallforamen, probably for a ramus temporalis(‘‘frt’’ in figs. 8, 36A). The posterior edge ofthe dorsal flange, together with the dorsaledge of the mastoid exposure of the petrosal,forms the inferolateral portion of the nuchalcrest, which is low and sharp. The ventral-most portion of the nuchal crest partially de-limits the notch that presumably lodged theexternal acoustic meatus.


The transition between the dorsal flangeand zygomatic process of the squamosal isindicated by a neck projecting anterolaterallyfrom the braincase to the glenoid fossa. Inventral view (figs. 15, 21), a shallow depres-sion, likely marking the course of the exter-nal acoustic meatus, runs along the long axisof the neck and is continued medially on thepetrosal. The anterior limit of the depressionis marked by a blunt, low transverse ridge,which extends between the medial margin ofthe glenoid and the crista parotica on the pe-trosal. The glenoid fossa is essentially flat,with the anteromedial and posterolateral cor-ners projecting slightly ventrally and the pos-teromedial corner being elevated. Its outlineis slightly teardrop-shaped, with the majoraxis running from anterolateral to postero-medial. This axis forms an abrupt angle withthe squamosal portion of the zygomatic arch,an unusual feature among mammals, whichimparts a distinctive outline to the skull ofKryptobaatar and several other MongolianLate Cretaceous multituberculates. In frontof the glenoid, the zygomatic process isflangelike and has an extensive oblique con-tact with the maxilla; it is considerably short-er and weaker than is the maxilla’s contri-bution to the zygoma. As described above,the squamosal and maxilla support the jugal,which lies medial to both bones but has thegreater part of its contact with the maxilla.In external view, lateral to the glenoid fossaand extending anterior to it on the zygomaticprocess, there is an arched ridge, concave in-feriorly, the intermediate zygomatic ridge(‘‘izr’’ in fig. 12) of Gambaryan and Kielan-Jaworowska (1995). This ridge marks thedorsal border of a shallow depression, whichthese authors suggested was for the origin ofthe posterior part of the superficial massetermuscle. This intermediate ridge is confluentwith the anterior zygomatic ridge on themaxilla, in contrast with the condition re-ported by Gambaryan and Kielan-Jaworows-ka (1995) in Nemegtbaatar, Chulsanbaatar,and Catopsbaatar where the anterior and in-termediate ridges are not confluent. A third,more posterior ridge, the posterior zygomaticridge, described for Nemegtbaatar, Chulsan-baatar, and Catopsbaatar (Gambaryan andKielan-Jaworowska, 1995), is not discerniblein Kryptobaatar.

PARIETAL

The parietals are laminar bones that formthe bulk of the roof of the cranial cavity.They are essentially lacking in PSS-MAE101 (fig. 8) and are considerably damaged inPSS-MAE 113 (fig. 9); however, enough ispreserved to provide the major morphologi-cal details.

The parietals are limited to the dorsal por-tion of the braincase (figs. 8, 9, 32) and areonly moderately convex. Anteriorly, theycontact the frontals at a broad U-shaped su-ture; lateral to the arms of the U, a narrowanterior process of the parietal extends for-ward, nearly to the lacrimal in the orbital rim,to form the posterior part of the supraorbitalnotch and supraorbital crest. Behind the an-terior process, level with the posterior borderof the frontals, is the triangular, posterolat-erally and ventrally directed postorbital pro-cess (‘‘pop’’ in figs. 12, 19). As originallypreserved, the process was long in PSS-MAE113 but was damaged during preparation. Itis completely preserved on both sides in askull referred to Kryptobaatar (PSS-MAE127). The posterior margin of the postorbitalprocess is continuous with weakly developedtemporal ridges (‘‘tr’’ in fig. 23). The tem-poral ridges are not fully preserved in PSS-MAE 113, but in PSS-MAE 127 the tem-poral ridges do not meet on the midline toform a sagittal crest. Instead, the ridges ap-proximate each other posteriorly, delimitinga broad middorsal ridge. Posteriorly, the pa-rietals contact the supraoccipital at a mostlytransverse suture, and together with the su-praoccipital form the dorsal tip of the nuchalcrest. An interparietal is not present. On itsventrolateral surface, the parietal contacts,from front to back, the frontal and the ante-rior lamina with an essentially straight su-ture, and the squamosal with a suture that isconcave medially. The contact with the fron-tal and squamosal is fairly narrow, whereasthat with the anterior lamina is extensive.

SUPRAOCCIPITAL

The supraoccipital is a laminar bone thatis essentially confined to the occiput; it iswell preserved in both PSS-MAE 101 (fig.16) and 113 (fig. 17).

The supraoccipital is a hexagonal element


(fig. 35). Its contacts on the occiput are withthe petrosals laterally and with the exocci-pitals ventrolaterally. A small portion of thesupraoccipital forms the dorsal margin of theforamen magnum and, with the parietal, thedorsal part of the nuchal crest. The nuchalcrests flare out posterolaterally along thecontact between the mastoid and squamosal;the medial continuation of these crests alongthe supraoccipital–parietal suture is lessprominent and is moderately notched in thesagittal plane. The supraoccipital is slightlyconvex dorsoventrally and concave medio-laterally. The occipital plane in PSS-MAE101 as determined by the supraoccipital in-clines anteriorly at 358 from the vertical. InPSS-MAE 113, the supraoccipital is moreconcave mediolaterally and a little taller thanin PSS-MAE 101.

EXOCCIPITAL

The exoccipital has contributions to theocciput and basicranium and can arbitrarilybe divided into three parts: the occipitalplate, the condyle, and the contribution to thejugular fossa. The sutures delimiting theexoccipital from its neighbors are distinct onthe occiput, but are not as discernible in otherareas. The exoccipital is well preserved inPSS-MAE 101 (figs. 14, 16) and 113 (figs.15, 17), but its contribution to the jugularfossa has not been cleaned of matrix on theleft side of the former.

The occipital plate of the exoccipital is asubrectangular bony lamina that forms one-third of the rim of the foramen magnum andextends anterolaterally from the edges of thataperture (fig. 35). Its contacts on the occiputare with the supraoccipital dorsomedially andwith the petrosal ventrolaterally. The suturewith the supraoccipital is straight and runsfrom dorsolaterally to ventromedially; the su-ture with the petrosal is arched and extendsventrolaterally from the point where theexoccipital, supraoccipital, and petrosal meetabove the level of the condyle to the medialslope of the paroccipital process. Along thepetrosal suture, there is a deep pit that is con-tinuous with the atlantal fossa on the petrosalthat probably lodged part of the atlas whenthe skull was maximally extended.

The condyle is a rounded structure

(‘‘ocon’’ in figs. 10, 12, 16) with the majoraxis oriented from dorsolateral to ventrome-dial. The articular surface on the condyle canbe divided into two parts: (1) a dorsal onecontributing to the lateral margin of the fo-ramen magnum and extending anterolaterallywith a similar orientation as the occipitalplate, and (2) a ventral one contributing tothe floor of the foramen magnum and lyingchiefly in a horizontal plane. The left andright condyles bound approximately half ofthe foramen magnum (figs. 16, 17). They donot contact each other in the midline, but areseparated by a distinct odontoid notch (‘‘on’’in fig. 37A). Despite the fact that the suturescannot be traced, it is likely that the medial-most portion of the condyles and the odon-toid notch are actually formed by the basi-occipital as, for example, in monotremes(Kuhn, 1971; Zeller, 1989).

The exoccipital is one of the major com-ponents of the jugular fossa, a deep excava-tion of the basicranium posterior to the prom-ontorium of the petrosal (figs. 14, 15, 20, 21,37A). The medial wall of the jugular fossa isformed by the exoccipital and basioccipital;the posterior wall largely by the exoccipitaland the paroccipital process of the petrosal;and the lateral wall entirely by the petrosal,the crista interfenestralis in the posterior partand the promontorium anteriorly. In additionto contributing to the posterior wall, the an-terior margin of the condyle projects as a thinlamina that partially floors the jugular fossa.In fact, the jugular fossa is deeply recessedinto all the surrounding bones, so that its sizeis considerably larger than what is visible inventral view. Inside the fossa, the exoccipitalis the major constituent of the roof. It con-tacts the petrosal through an oblique suturerunning anteriorly from lateral to medial.Along this suture is the small jugular fora-men, which occupies the posterolateral cor-ner of the fossa, posteromedial to the peri-lymphatic foramen (fig. 37A). A possible su-ture between the basioccipital and exoccipitalis visible in the roof of the jugular fossa onthe right side of PSS-MAE 101; it begins justanteromedial to the jugular foramen and ex-tends posteromedially but cannot be tracedonto the condyle.

On the right side of PSS-MAE 101, deepin the jugular fossa and posterolateral to the


jugular foramen, is a small, circular foramen(not visible in the figures). An opening in asimilar position may be present on the leftside of PSS-MAE 113, but it is not as clear.The foramen in PSS-MAE 101 is interpretedas for the hypoglossal nerve. The right jug-ular fossa in PSS-MAE 113 seems to be wellpreserved, but in the area where a hypoglos-sal foramen is expected, no aperture is found.It is possible, although unlikely, that the fo-ramina mentioned above are artifacts andthat the true hypoglossal foramina are con-cealed by matrix in the caudalmost portionof the jugular fossa.

According to Kielan-Jaworowska et al.(1986), the large size of the jugular fossa inMongolian Late Cretaceous multitubercula-tes suggests the presence of large ganglia onthe nerves below the jugular foramen. How-ever, large ganglia alone cannot account forthe immense jugular fossa in Kryptobaatar,especially considering the minute size of thejugular foramen. Therefore, we think that thestructural continuity between the jugular fos-sa and the middle-ear space suggests that thefossa housed a diverticulum of the cavumtympani (see also Rougier et al., 1996c).

BASIOCCIPITAL

The basioccipital is a long, narrow, lami-nar bone that forms the base of the skull an-terior to the foramen magnum and medial tothe ear regions. It can be divided into twoparts: one on the basicranial axis, and theother contributing to the jugular fossa. Thebasioccipital is best preserved in PSS-MAE101 (fig. 14), having been distorted in PSS-MAE 113 in which the left petrosal is dis-placed (fig. 15).

The basioccipital contacts the exoccipitalposterolaterally, the petrosal laterally, and thebasisphenoid anteriorly (figs. 14, 34). In ad-dition, there may even be a contact with thepterygoid along the lateral edge of the basi-occipital in front of the petrosal contact, de-pending on the composition of the crest ex-tending posteriorly from the pterygopalatineridge (see Sphenoid Complex). The suturewith the basisphenoid is straight and trans-verse, just behind the level of the back of thepterygopalatine ridges. The suture with thepromontorium of the petrosal is gently

curved, with the basioccipital becomingslightly wider posteriorly along its petrosalcontact. The maximum width of the basioc-cipital is at the level of the jugular fossa, andthen it tapers posteriorly to the odontoidnotch. As stated above, the suture betweenthe basioccipital and exoccipital in the con-dylar region is not discernible, but it waslikely oblique, running anterolaterally fromthe odontoid notch and the medial aspect ofthe condyle. A possible oblique suture sep-arates these same two bones in the jugularfossa on the right side of PSS-MAE 101.

In the basicranial axis, the anterior part ofthe basioccipital is essentially flat, whereasposteriorly there is a concavity on the mid-line. In the jugular fossa, the basioccipitalcontributes a fairly vertical wall along themedial aspect and the anteromedial part ofthe roof. The ventral edges of the basiocci-pital’s contribution to the jugular fossastrongly project laterally to form a partialfloor for that depression.

ENDOCRANIUM

Described here are structures on the en-docranial surface of the braincase preservedin three skulls of Kryptobaatar, PSS-MAE123 (fig. 25), 124, and 125 (fig. 26). Com-parative specimens also employed includetwo indeterminate skulls resembling Krypto-baatar, PSS-MAE 126 and 128. A recon-struction of the endocranial floor is shown infigure 27. As is evident from the exterior, thebraincase in Kryptobaatar is formed by theexoccipitals, basioccipital, supraoccipital, pe-trosals, parietals, frontals, and the sphenoidcomplex; as stated above, the squamosaldoes not contribute directly to the braincasewall. With the exception of the petrosals, su-tures delimiting the individual braincase el-ements are subject to some degree of uncer-tainty in the specimens showing the endocra-nium. Consequently, we base our descrip-tions on the internal morphology andattribute features to individual elements inlight of our understanding of similar struc-tures in extant mammals in which the bound-aries are known. The dorsal components ofthe braincase (i.e., the frontal and parietal)were effectively missing in the specimensdescribed herein, providing easy access to


Fig. 24. Schematic drawings of embryonicchondrocrania in left lateral view. A, generalizedsauropsid; B, generalized multituberculate, as re-constructed from here; C, generalized monotreme;D, generalized marsupial; E, generalized placen-tal. Abbreviations: ais anterior intercavernous si-nus; cev capsuloparietal emissary vein; ea eth-moidal artery; en ethmoidal nerve; ev ethmoidalvein; ips inferior petrosal sinus; nc nasal capsule;oa ophthalmic artery; oc otic capsule; pan pilaantotica; pis posterior intercavernous sinus; pmpila metoptica; pp pila preoptica; ps prootic sinus;pv pituito-orbital vein; II optic nerve; III oculo-motor nerve; IV trochlear nerve; V trigeminalnerve; VI abducens nerve, VII facial nerve.

the more complex floor and side wall of theendocranium.

Because endocrania are seldom preserved,let alone described, we add the following re-marks by way of introduction to this anatom-ical region. The orbitotemporal region of themammalian skull has a primary braincasewall formed by the chondrocranium and duramater, and external to that, a secondarybraincase wall formed by different patternsof several skeletal elements in different taxa(De Beer, 1937; Moore, 1981; Kuhn and Ze-ller, 1987). The extradural space between theprimary and secondary wall is the cavumepiptericum (Gaupp, 1902, 1905), whichhouses the trigeminal and facial ganglia andis traversed by various cranial nerves andblood vessels. The cranial nerves enter thecavum epiptericum from the brain throughspecific gaps between near vertical bars orpillars of chondrocranial cartilage (fig. 24).The pattern of these gaps (i.e., how many arepresent and their contents) differs in mono-tremes, marsupials, and placentals (see be-low; Kuhn, 1971; Kuhn and Zeller, 1987).

There is a general consensus among mor-phologists (e.g., Starck, 1967, 1978; Kuhn,1971; Moore, 1981) that the chondrocraniumin the common ancestor of mammals hadthree pillars in the orbitotemporal region be-tween the nasal and otic capsules, as occursin most extant sauropsids (fig. 24A; De Beer,1926, 1937; Bellairs and Kamal, 1981).From anterior to posterior, these are the pilapreoptica, pila metoptica, and pila antotica(fig. 24B). These three pilae provide bordersfor the apertures that transmitted cranialnerves through the primary braincase wall(Kuhn and Zeller, 1987; Zeller, 1989). An-teriorly, between the nasal capsule and thepila preoptica is the orbitonasal foramentransmitting the ethmoidal branch of the oph-thalmic nerve into the nasal cavity. Posteriorto that, between the pila preoptica and me-toptica is the optic foramen transmitting theoptic nerve into the orbit. Next, between thepilae metoptica and antotica is the metopticforamen transmitting the oculomotor nerveinto the orbit (or into the front of the cavumepiptericum). Finally, the gap between thepila antotica and the otic capsule, the prooticforamen, transmitted the trochlear, trigemi-nal, and abducens nerves into the cavum


epiptericum. Immediately posterior to that,the facial nerve entered the back of the ca-vum epiptericum through a separate opening,the primary facial foramen, between the oticcapsule and a bar of cartilage, the prefacialcommissure, connected to the front of theotic capsule.

A reconstruction of the vessels passingthrough the primary wall in basal mammalshas not yet been proposed, and we offer thefollowing model here (fig. 24A, B). Judgingfrom the anatomy of extant amniotes, bloodvessels ran through all of the apertures in theprimary wall named above except the pri-mary facial foramen. Also in the orbitonasalforamen were ethmoidal arteries and veins;in the optic foramen, the ophthalmic branchof the internal carotid artery; in the metopticforamen, the pituito-orbital vein; and in theprootic foramen, the prootic vein and inferiorpetrosal sinus. Our justification for this mod-el is as follows.

(1) Ethmoidal vessels in the orbitonasalforamen: Vessels accompanying the ethmoid-al nerve are typically present in both saurop-sids (Shindo, 1914) and mammals (Tandler,1899).

(2) Ophthalmic artery in the optic fora-men: The ophthalmic artery has been recon-structed by Miao (1988), citing De Beer(1937), with the oculomotor nerve and thepituitary vein in the metoptic foramen inLambdopsalis. However, we disagree withthis placement of the ophthalmic artery inLambdopsalis and in basal mammals. Thereare some sauropsids in which the ophthalmicartery enters the orbit via the metoptic fora-men (e.g., Crocodilus, Shiino, 1914) or evenvia a separate foramen that secondarily fuseswith the metoptic foramen (e.g., Chrysemys,Shaner, 1926; Sphenodon, Bellairs and Ka-mal, 1981). Yet, other sauropsids (e.g., La-certa, Shindo, 1914; Platydactylus, Hafferl,1921) and mammals (Tandler, 1899; Wible,1984) have the ophthalmic artery and opticnerve intimately associated.

(3) Pituito-orbital vein in the metoptic fo-ramen: A pituitary vein running from the pi-tuitary (hypophysis) to the orbit is broadlydistributed among sauropsids (Bruner, 1907),and is either in the metoptic foramen (e.g.,Lacerta, De Beer, 1937) or in a separate fo-ramen that secondarily fuses with the metop-

tic foramen (e.g., Sphenodon, Bellairs andKamal, 1981). In mammals, a comparablevein drains medially from the cavernous si-nus within the cavum epiptericum, immedi-ately anterior to the pila antotica into the hy-pophyseal fossa in the echidna (Gaupp,1908) and the platypus (personal obs.); in thelatter, the vein exits the skull via the carotidforamen. This is not the only vein drainingthe pituitary, and, following a suggestion ofone of our reviewers, Robert Presley, we re-fer to it as the pituito-orbital vein in light ofits pathway. Marsupials and placentals do nothave a strictly comparable vein, but it is pos-sible that the stem of the anterior intercav-ernous sinus of therians (Shindo, 1915) is ho-mologous with the pituito-orbital vein ofmonotremes and sauropsids.

(4) Prootic sinus and inferior petrosal si-nus in the prootic foramen: The prootic sinus(middle cerebral vein) exits the prootic fo-ramen to join the lateral head vein lateral tothe otic capsule in sauropsids, monotremes,and some marsupials (Shindo, 1915; Wible,1990; Wible and Hopson, 1995). Moreover,a canal housing this vessel is widely distrib-uted among the extinct outgroups to mam-mals (Wible and Hopson, 1995; Rougier etal., 1996a). Among extant amniotes, an in-ferior petrosal sinus is known only in mam-mals, in which it runs posteriorly from thecavernous sinus within the cavum epipteri-cum, through the prootic foramen into thecranial cavity (Shindo, 1915; Rougier et al.,1996a).

As stated above, the pattern of the aper-tures in the primary braincase wall of basalmammals is altered in different ways inmonotremes, marsupials, and placentals. Allthree retain the pila preoptica, but that is theextent of the resemblance between them.

(1) Monotremes (fig. 24C): In addition tothe pila preoptica, a complete pila antoticaforms in the chondrocranium of both theechidna (Gaupp, 1908; Kuhn, 1971) andplatypus (Zeller, 1989). However, the pila an-totica regresses during ontogeny, and in theadult it is represented by only its slender os-sified base on the basisphenoid, the middleclinoid process (‘‘mclp’’ in fig. 28; Kuhn,1971; Zeller, 1989). The pila metoptica failsto form, leaving confluent the optic and me-toptic foramina, which together are termed


the pseudoptic foramen (‘‘psf’’ in fig. 28;Gaupp, 1908). The pseudoptic foramentransmits the optic and oculomotor nerves(Kuhn, 1971; Zeller, 1989), the pituito-orbit-al vein (Gaupp, 1908; personal obs.), and inthe echidna the ophthalmic artery (Tandler,1901); in the platypus, the ophthalmic arterybranches off the maxillary artery in the orbit(Wible, 1984). The contents of the prooticforamen do not differ from that inferred forbasal mammals.

(2) Marsupials (fig. 24D): The pila preop-tica is the only pillar to form in the orbito-temporal region of the primary wall in mar-supials (Kuhn and Zeller, 1987; Maier, 1987).Consequently, the optic, metoptic, and pro-otic foramina are confluent. Transmittedthrough this large gap are the second throughsixth cranial nerves, the ophthalmic artery(Tandler, 1899), the anterior and posterior in-tercavernous sinuses, the inferior petrosal si-nus, and the prootic sinus (Shindo, 1915).The prootic sinus is present during early on-togenetic stages in all marsupials investigat-ed to date, but is retained only in adult di-delphids, caenolestids, and some dasyurids(Wible, 1990; Wible and Hopson, 1995).This single large aperture transmitting allthese structures in the marsupial chondrocra-nium has been referred to (e.g., Cords, 1915)as the sphenoparietal fenestra. Followingmost recent authors (e.g., Kuhn and Zeller,1987), we reserve the similar term of sphe-noparietal foramen for the opening in thechondrocranium of eutherians that transmitsthe third through sixth cranial nerves (see be-low).

(3) Placentals (fig. 24E): In addition to thepila preoptica, a complete pila metopticaforms in placentals and thus completes a trueoptic foramen in the orbitosphenoid of theadult (De Beer, 1937; Starck, 1967), whichin most forms transmits the ophthalmic ar-tery along with the optic nerve (Tandler,1899, 1901; Wible, 1984). The pila antotica,however, fails to develop, leaving confluentthe metoptic and prootic foramina, which to-gether are termed the sphenoparietal foramen(Voit, 1909). The sphenoparietal foramentransmits the third through sixth cranialnerves, the anterior and posterior intercav-ernous sinuses, and the inferior petrosal sinus(Shindo, 1915). Unlike monotremes and

marsupials, the prootic sinus involutes in pla-centals and is replaced by a secondary vessel,the capsuloparietal emissary vein (Gelderen,1924).

EXOCCIPITAL

In PSS-MAE 123 (fig. 25B), a small frag-ment of the left exoccipital has been pre-served in articulation with the petrosal, pos-teromedial to the subarcuate fossa. Along thesuture between the exoccipital and petrosal,a broad groove directed ventromedially is in-terpreted as having housed the sigmoid sinus.In PSS-MAE 125 (fig. 26), an even smallerportion of the right exoccipital completingthe posteromedial margin of the jugular fo-ramen has been preserved. The contributionof this bone to the jugular foramen is clearlyindicated by sutures. Along the suture be-tween the exoccipital and petrosal, directlyposterior to the jugular foramen, is an ante-riorly directed, digitiform process of theexoccipital that impinges on the lumen of theforamen. Also in this specimen, a longer seg-ment of the sulcus for the sigmoid sinus isclearly preserved (‘‘sss’’ in fig. 26). It doesnot approach the very small jugular foramen,but instead is directed toward the placewhere the missing foramen magnum wouldbe expected. An endocast of the sigmoid si-nus is preserved on both sides in the undes-cribed skull of multituberculate, n. sp. (PSS-MAE 126), and it follows the pattern de-scribed above in PSS-MAE 125.

BASIOCCIPITAL

In PSS-MAE 123, the rostral portion ofthe basioccipital is preserved; a smaller frag-ment is found in PSS-MAE 125; and the en-tire bone is likely present in PSS-MAE 124,but is not accessible for study. As preservedin PSS-MAE 123 (fig. 25), the basioccipitalis a flat, slightly concave bone that is ratherfeatureless. As gleaned from both specimens,the suture between the basioccipital and ba-sisphenoid is situated immediately behind thedorsum sellae and runs transversely. The su-ture between the basioccipital and petrosalruns subparallel to the sagittal plane, con-verging slightly anteriorly. The basioccipitalforms only a narrow portion of the skullbase. In PSS-MAE 123, the basioccipital is


Fig. 25. Stereophotograph of the floor of the endocranium of Kryptobaatar dashzevegi PSS-MAE123 in dorsal (A) and oblique dorsal (B) views, with accompanying line drawings. Gray pattern repre-sents matrix; parallel lines denote breakage. Abbreviations: al anterior lamina; bo basioccipital; ce cavumepiptericum; ds dorsum sellae; exoc exoccipital; ff facial foramen; fica foramen for internal carotidartery; fpv foramen for pituito-orbital vein; fr frontal; fV3 foramen for mandibular nerve; hf hypophy-seal fossa; iam internal acoustic meatus; jn jugular notch; jsp jugum sphenoidale; mef metoptic foramen;mx maxilla; na nasal; opf optic foramen; or orbitosphenoid; ow orbital wing (pila preoptica); pan pilaantotica; prc prootic canal; pm pila metoptica; sf subarcuate fossa; sphf sphenorbital fissure; tus tu-berculum sellae.


Fig. 25. Continued.

broken across the anteriormost portion of thejugular fossa, and the preserved fragment in-dicates that the fossa deeply excavates thebasioccipital, leaving only a thin lamella sep-arating the cranial cavity from the middle-ear cavity. Left and right jugular fossae alsoconverge medially, being separated in themidline by a bridge of the basioccipital boneless than 1 mm in thickness. Along the ba-sioccipital–petrosal suture is a shallow de-pression likely occupied by the inferior pe-trosal sinus.

PSS-MAE 128, which is not Kryptobaatarbut is referred here as an indeterminate mul-tituberculate, conforms to the pattern of theskull base described above. The only sub-stantive exception is that instead of a shallowdepression, there is a deep trough on the ba-sioccipital running to the basioccipital-petro-sal suture. This unambiguously indicates theendocranial course of the inferior petrosal si-nus.

PETROSAL

As was the case in our description of theexterior of the petrosal, when describing theinterior, the anterior lamina is consideredalong with the petrosal proper. The followingdescriptions are based on the three specimensof Kryptobaatar with exposed the endocra-nial surfaces (i.e., PSS-MAE 123, 124, and125).

The endocranial surface of the petrosal canbe subdivided into four quadrants by two in-tersecting crests. One crest is the crista pe-trosa, which runs along the prefacial com-missure (‘‘pfc’’ in fig. 26) and the anteriormargin of the subarcuate fossa (‘‘sf’’ in figs.25–27); the other crest separates the subar-cuate fossa from the internal acoustic meatus(‘‘iam’’ in figs. 25–27) and continues ante-roventrolaterally as the dorsolateral marginof the cavum epiptericum (‘‘ce’’ in figs. 25B,26, 27). The crests intersect dorsolateral tothe internal acoustic meatus in a rounded em-inence. The four quadrants so delimited areidentified by their most conspicuous features:the subarcuate fossa, the internal acousticmeatus, the cavum epiptericum, and the an-terior lamina (fig. 27).

The subarcuate fossa, which accommodat-ed the paraflocculus of the cerebellum, is alarge, deep, subspherical depression thatopens into the braincase through an ellipticalaperture, the major axis of which is dorso-ventrally oriented (figs. 25, 26). The dorsal-most margin of the subarcuate fossa is notpreserved or fully prepared in any specimen.Considering the presence of the large vas-cular groove in the exoccipital already de-scribed, it is likely that a groove for the sig-moid sinus was present on the dorsal edge ofthe subarcuate fossa. A similar structure ispresent in the skull of multituberculate, n. sp.


←

Fig. 26. Stereophotograph of the floor of the endocranium of Kryptobaatar dashzevegi PSS-MAE125 in oblique dorsal view. Gray pattern represents matrix; parallel lines denote breakage. Abbreviations:al anterior lamina; bs basisphenoid; ce cavum epiptericum; ds dorsum sellae; fr frontal; iam internalacoustic meatus; jfo jugular foramen; mef metoptic foramen; mx maxilla; or orbitosphenoid; pan pilaantotica; pet petrosal; pfc prefacial commissure; sf subarcuate fossa; sss sulcus for sigmoid sinus; tsctransverse sinus canal.

(PSS-MAE 126). Along the posteroventraledge of the subarcuate fossa, in the crest sep-arating the subarcuate fossa and the internalacoustic meatus, there is small foramen, thevestibular aqueduct, with a posterodorsallydirected groove, interpreted as having housedthe endolymphatic duct.

The quadrant housing the internal acousticmeatus is the part that contributes to the floorof the braincase, and it does so in an obliquefashion (figs. 25–27). Medially, the meatalquadrant contacts the basioccipital and exoc-cipital via a straight parasagittal suture; an-teriorly, it cannot be differentiated from thesphenoid complex. Apparently, the sphenoidcomplex and petrosal are fused endocra-nially, although a distinct morphological dis-continuity that forms the dorsum sellae prob-ably indicates the anterior extent of the pe-trosal. A cochlear aqueduct is not identifiablein any specimen, and hence the aperture onthe back of the promontorium is the perilym-phatic foramen, as occurs in monotremes(Kuhn, 1971; Zeller, 1985, 1989, 1991). Theposteromedial margin of the meatal quadrantis notched by the small jugular foramen (fig.26). The internal acoustic meatus is a sub-circular depression, although its medial wallis not very tall, and the meatus is continuousmedially with the petrosal’s contribution tothe braincase floor. At the posteromedialedge of the meatus is an anteroventrally di-rected foramen, probably for the cochlearnerve. Separating this small foramen fromthe deeper part of the meatus is a low crestthat forms the posteromedial margin of a sec-ond aperture, probably for the facial and ves-tibular nerves.

Lateral to the meatal quadrant is the ca-vum epiptericum (figs. 25–27), the very deepfossa that lodged the trigeminal or semilunarganglion and provided passage to variousnerves and vessels. The cavum opens ante-riorly, ventrally, and laterally via several ap-

ertures. The ventral (‘‘fv3’’ in fig. 25) andlateral openings transmitted branches of themandibular nerve, and the large, subcircularanterior opening transmitted the trochlear,abducens, ophthalmic, and maxillary nerves,given that the cavum shows the same rela-tionship to the pila antotica (see SphenoidComplex below) as in monotremes (Kuhnand Zeller, 1987). The anterior opening, thesphenorbital fissure (‘‘sphf’’ in figs. 25A,27), probably also transmitted the ramus in-fraorbitalis of the stapedial artery and theophthalmic veins (Rougier et al., 1992; Wi-ble and Hopson, 1995). The cavum epipte-ricum is longer than wide, and its anteriorportion is roofed by the laminar, laterally ex-panded pilae antotica and metoptica, whichconnect the medial and lateral crests delim-iting the trigeminal fossa. Based on PSS-MAE 123, it is apparent that a separate ca-vum supracochleare for the facial ganglion islacking in Kryptobaatar; the primary facialforamen opens directly into the back of thecavum epiptericum where the facial ganglionwould have been located. The secondary exitof the facial nerve from the cavum epipteri-cum was likely in the posterior part of thecavum’s floor, but it was inaccessible to prep-aration.

The anterior lamina quadrant is the largestand contributes to the side wall of the brain-case (figs. 25–27). It is concave in medialview and extends from the subarcuate fossaforward to contact the primary wall of thebraincase represented by the pila antotica/or-bitosphenoid, which the anterior laminaoverlaps laterally. Anterolateral to the subar-cuate fossa in the anterior lamina quadrant isa deep, broad, anteroventrally directed sulcusleading into a foramen (‘‘prc’’ in figs. 25,27). This is the sulcus and foramen for theprootic sinus, which leads into the prooticcanal. As preserved in the skull of PSS-MAE123 (fig. 25), the foramen for the prootic si-


Fig. 27. Reconstruction of the floor of the endocranium of Kryptobaatar dashzevegi. Parallel linesrepresent the cut edge of the braincase. Abbreviations: al anterior lamina; bo basioccipital; ce cavumepiptericum; ds dorsum sellae; exoc exoccipital; fica foramen for internal carotid artery; fpv foramenfor pituito-orbital vein; fr frontal; hf hypophyseal fossa; iam internal acoustic meatus; jfo jugular fo-ramen; jsp jugum sphenoidale; mef metoptic foramen; ocon occipital condyle; opf optic foramen; ororbitosphenoid; pa parietal; pan pila antotica; pet petrosal; pm pila metoptica; pop postorbital process;prc prootic canal; sf subarcuate fossa; son supraorbital notch; sphf sphenorbital fissure; sq squamosal;sup supraoccipital.

nus opens at a level ventral to the lower mar-gin of the subarcuate fossa. The anterior lam-ina in front of the prootic sulcus is verysmooth and essentially featureless.

SPHENOID COMPLEX

Given that the hypophyseal or pituitaryfossa is universally lodged in the basisphe-noid among recent mammals (Starck, 1967),we identify the corresponding part of thesphenoid complex in Kryptobaatar as the ba-sisphenoid. The hypophyseal fossa (‘‘hf’’ infigs. 25, 27) is a deep, hemispherical depres-sion on the skull base that is limited poste-riorly by a prominent dorsum sellae, also onthe basisphenoid (‘‘ds’’ in figs. 25, 26). Thedorsum sellae forms a steep angle with thefloor of the braincase behind it; the dorsumis lower in the midline, and taller and thickerlaterally where it becomes confluent with the

ossified pila antotica. Inside the hypophysealfossa, the carotid foramina open in the pos-terolateral corner of the floor (‘‘fica’’ in figs.25, 27). Laterally on the anterior wall of thehypophyseal fossa, in front of each carotidforamen, there is a small foramen in PSS-MAE 123 (‘‘fpv’’ in figs. 25, 27) and 125,the only specimens preserving the relevantarea. This foramen connects the hypophysealfossa with the orbitotemporal fossa, and inPSS-MAE 123 it is also connected to the ca-rotid foramen via a distinct sulcus in the lat-eral floor of the hypophyseal fossa. The ex-ternal aperture of this foramen is shown onthe right side of PSS-MAE 101 (in additionto 123 and 125). Based on our observationsof serially sectioned platypuses, as well asGaupp’s (1908) observations of the echidna,we interpret this foramen and sulcus as forthe pituito-orbital vein.


Fig. 28. The floor of the endocranium of the platypus Ornithorhynchus anatinus in dorsal view(modified from Zeller, 1989: fig. 6, with the author’s permission). Parallel lines represent the cut edgeof the braincase. Abbreviations: al anterior lamina; bo basioccipital; fV2 foramen for maxillary nerve;fV3 foramen for mandibular nerve; hf hypophyseal fossa; iam internal acoustic meatus; jfo 1 hyfconfluent jugular and hypoglossal foramina; jsp jugum sphenoidale; mlcp middle clinoid process (os-sified remnant of pila antotica); onsf orbitonasal foramen; pet petrosal; pp pila preoptica; prc prooticcanal; psf pseudoptic foramen (for II, III, IV, V1, VI); ptc posttemporal canal; sf subarcuate fossa; sqsquamosal; tus tuberculum sellae; vaq vestibular aqueduct.

The tall laminar walls forming the sides ofthe hypophyseal fossa are here considered asthe ossified pilae antotica (‘‘pan’’ in figs. 25–27), a component of the primary wall of thebraincase (fig. 24A–C). These walls projectanteriorly and laterally as broad, winglikestructures that connect the primary wall withthe dermal elements that form the secondarywall; sutures demarcating the lateral extentof the pila are clearly visible in PSS-MAE125 (fig. 26). The pila antotica is particularlythick posterodorsally, where in conjunctionwith the anterior lamina it roofs the cavumepiptericum. Running behind the pila anto-tica into the cavum epiptericum in Krypto-baatar were the trochlear, trigeminal, and ab-ducens nerves, based on the relationshipsthat these structures exhibit in monotremes(fig. 24C; Kuhn and Zeller, 1987; Zeller,1989). In describing the similarly situated,

ossified pila antotica in other multitubercu-lates, Kielan-Jaworowska et al. (1986), Miao(1988), and Hurum (1998a) have employedthe term taenia clino-orbitalis. This term wasoriginally coined by Gaupp (1902, 1908) forthe well-developed pila antotica that he en-countered in the chondrocranium of theechidna. Gaupp (1908: fig. 56) also appliedthe term to the remnant of the pila antoticain the adult echidna skull, but the primaryusage was for a cartilaginous structure. Con-sequently, we continue to use the more gen-eral term pila antotica rather than the morerestricted taenia clino-orbitalis.

Lateral to the hypophyseal fossa, on thelaminar extension of the pila antotica, thereis a deep trough that leads anteriorly to asizable round foramen at the level of the an-terior margin of the hypophyseal fossa(‘‘mef’’ in fig. 25–27) that opens externally


into the orbitotemporal fossa (‘‘mef’’ in fig.36A). This foramen is present in the threespecimens of Kryptobaatar with the endo-cranium exposed as well as in the indeter-minate multituberculate (PSS-MAE 128).Miao (1988) interpreted a similarly placedaperture in Lambdopsalis as the metoptic fo-ramen, the main occupant of which was theoculomotor nerve. We report here a similaropening in the right orbit of Kamptobaatar(ZPAL MgM-I/33). Even though a metopticforamen is not known for any ontogeneticstage in extant mammals, we accept Miao’sinterpretation for Lambdopsalis and extend ithere for the same structures in Kryptobaatar,Kamptobaatar, and PSS-MAE 128. As statedabove, the metoptic foramen is found in thechondrocranium of most extant sauropsidsbetween the pilae antotica and metoptica (fig.24A; De Beer, 1926, 1937; Bellairs and Ka-mal, 1981) and is generally considered tohave been present in the chondrocrania of thecommon ancestor of monotremes and the-rians (Kuhn, 1971; Kuhn and Zeller, 1987).Following this, the bone enclosing the an-teromedial aspect of the foramen in Krypto-baatar is the ossified pila metoptica (‘‘pm’’in figs. 25B, 27).

In amniotes, the two most significant fo-ramina in the interval between the passage-ways for the optic and trigeminal nerves arethe metoptic foramen and the foramen for theabducens nerve (De Beer, 1926, 1937). Al-though the abducens nerve runs through theprootic foramen in monotremes and manysauropsids (fig. 24A, C), it pierces the baseof the pila antotica at the level of the dorsumsellae in various sauropsids (De Beer, 1926,1937; Save-Soderburgh, 1947; Oelrich,1956), and it has been similarly interpretedin various extinct forms, including the prim-itive non-mammalian cynodont Thrinaxodon(Parrington, 1946), the tritylodontid Oli-gokyphus (Kuhne, 1956), and the tritheledon-tid Diarthrognathus (Crompton, 1958). Theaperture in question in Kryptobaatar, how-ever, is placed too far dorsally and forward,at the level of the front of the hypophysealfossa, to be a foramen for the abducens nerve(figs. 25–27). Therefore, we reconstruct theabducens nerve in Kryptobaatar as havingentered the cavum epiptericum with thetrochlear and trigeminal nerves, posterior to

the pila antotica, and the oculomotor as hav-ing entered the orbit via a separate metopticforamen.

The region anterior to the hypophysealfossa is damaged in all available specimens,but a full restoration (fig. 27) can be offeredbased on the fragments preserved in twospecimens of Kryptobaatar (PSS-MAE 123and 125; figs. 25, 26) as well as in the in-determinate multituberculate (PSS-MAE128). Extending forward from the hypophy-seal fossa on the midline is a rodlike portionof the sphenoid complex, the tuberculum sel-lae (‘‘tus’’ in fig. 25A). It extends anteriorlytoward paired, laterally directed projections,the orbital wings or alae orbitales (‘‘ow’’ infig. 25B). These wings are part of the orbi-tosphenoid and represent the ossified pilaepreoptica. The edge of the wings, the orbi-tosphenoidal crest, dorsally delimits large,paired foramina that are not preserved intactin any specimen, although they are nearlypreserved in PSS-MAE 123 (‘‘opf’’ in fig.25). These anterolaterally directed foramina,for the optic nerves and ophthalmic arteries,are situated between the orbital wings andthe tuberculum sellae, close to the midline.The portion of the tuberculum sellae directlymedial to the optic foramina is groovedtransversely by a shallow sulcus, the chias-matic sulcus for the optic chiasm.

In front of the optic foramina, the brain-case floor between the right and left orbitalwings is marked by a midline trough of un-certain function, which becomes narrowerand shallower rostrally. This midline portionof the sphenoid complex in this region is theyoke or jugum sphenoidale (‘‘jsp’’ in figs.25A, 27). The overall orientation of the ju-gum sphenoidale is anterodorsally. Slightlyrostral to the end of the midline trough, thecranial cavity is constricted, demarcating theposterior boundary of the olfactory bulbs.The rostralmost extent of the cranial cavityhousing the olfactory bulbs is not preparedor preserved in any available specimens. Inthe indeterminate multituberculate (PSS-MAE 128), part of the floor in front of theolfactory bulb constriction is preserved onthe left side. The surface is smooth, devoidof perforations, and corresponds to the lam-ina infracribrosa. Laterally, where the laminainfracribrosa contacts the jugum sphenoidale


behind, there is a foramen from which a siz-able sulcus runs anterodorsally. This foramenand sulcus are interpreted as for the endocra-nial portion of the ethmoidal nerve and ves-sels. None of the available specimens exhib-its any indication of an ossified cribriformplate. Our observation should be regarded asprovisional, because the preservation of theavailable specimens hindered the investiga-tion of this region of the endocranium.

In the three specimens of Kryptobaatarshowing the endocranial surfaces (PSS-MAE123, 124, and 125) and in the indeterminatemultituberculate (PSS-MAE 128), a broadcanal connects the right and left sphenorbitalrecesses, that is, the space medial to the wallsdemarcating the lateral rims of the sphenor-bital fissures. This canal is low on the brain-case floor and traverses the primary wallventral to the tuberculum sellae; a thin wallseparates this canal from the hypophysealfossa (‘‘tsc’’ in fig. 26). Among living mam-mals, the transverse canal is in a similar po-sition, crossing the midline immediately an-terior to the hypophyseal fossa, and it trans-mits a vein called the transverse canal veinin some marsupials (Archer, 1976; Marshallet al., 1990; Marshall and Muizon, 1995) andsome placentals (McDowell, 1958; MacPhee,1994). This is the only likely model for thesimilarly situated canal in Kryptobaatar (seefig. 36) and the indeterminate multitubercu-late.

The occurrence of turbinals or ridges forthe turbinals on the appropriate bones in thenasal cavity could not be studied in any ofthe specimens of Kryptobaatar considered inthis report. However, the natural nasal en-docast of an indeterminate multituberculatefrom Ukhaa Tolgod (PSS-MAE 134) showsridges in the maxillary, nasal, and frontalportions of the nasal cavity (Rougier et al.,1997b). These ridges correlate closely withsimilar structures present in living mammalssupporting the turbinals (Paulli, 1900;Moore, 1981; Hillenius, 1994). PSS-MAE134 confirms the presence of ethmoturbinalsand nasoturbinals, and a crest running par-allel to the sulcus for the nasolacrimal ductsuggests the existence of maxilloturbinals.Also, fragments of turbinals appear to be pre-served. Among multituberculates, turbinalridges and fragments have been reported pre-

viously for Lambdopsalis (Miao, 1988) andNemegtbaatar (Hurum, 1994), and fragmentsonly for Chulsanbaatar (Hurum, 1994).Among Mesozoic mammaliaforms, turbinalshave been preserved in the docodontid Hal-danodon (Lillegraven and Krusat, 1991) andthe zalambdalestid Barunlestes (Kielan-Ja-worowska and Trofimov, 1980).

MANDIBLE

As is typical for the lower jaws or man-dible in multituberculates, that of Kryptobaa-tar has a robust, deep horizontal ramus andan ascending ramus that is proportionallysmaller. Housed in the lower jaw are fiveteeth: an enlarged, procumbent incisor (i1),two premolars (p3–p4) with the mesial onegreatly reduced, and two molars (m1–2).Both lower jaws are complete and in placein PSS-MAE 101 (fig. 14). Only the left oneis preserved in PSS-MAE 113; it is well pre-served except that the ventral margin underthe incisor alveolus is missing (fig. 5). Wedescribe the lower jaw first in lateral and thenin medial views; the terminology used fol-lows that of Gambaryan and Kielan-Jawo-rowska (1995) unless noted otherwise.

In lateral view (figs. 10, 12, 29), the al-veolus of the incisor is separated from p3 bya long, dorsally concave diastema. In PSS-MAE 113, in which the ventral margin of thelower jaw is broken, the root of the incisorcan be traced posteriorly at least to the mid-dle of m1. The mental foramen (‘‘mf’’ in fig.12) is positioned in front of the lateral bulgeover the roots of the premolars; it lies closerto the superior than to the inferior margin ofthe lower jaw. In PSS-MAE 101 (fig. 10), itis even closer to the superior margin and tothe lateral bulge over the premolar roots thanin PSS-MAE 113 (fig. 29). The mental fo-ramen is quite large and faces anterolaterally.

The area on the horizontal ramus betweenthe lateral bulge over the premolar roots andthe base of the coronoid process is very shortand concave. This is the masseteric fovea(‘‘mafo’’ in fig. 12), which Gambaryan andKielan-Jaworowska (1995) interpreted as forthe pars anterior of the medial masseter mus-cle. As noted by these authors, the massetericfovea in Kryptobaatar is confluent posteri-orly with the masseteric fossa. The masse-


Fig. 29. Stereophotograph of the left lower jaw of Kryptobaatar dashzevegi PSS-MAE 113 in lateralview.

teric fossa (‘‘maf’’ in fig. 12) is shallow,broad, and subdivided by a low, obliqueridge running anteroventrally from the con-dyle, which we call the condyloid crest fol-lowing the terminology used for the dog(Evans and Christensen, 1979). The portionof the masseteric fossa anterodorsal to thecondyloid crest is larger and deeper than theposteroventral one; these two fossae proba-bly accommodated different parts of the mas-seter muscle (see Gambaryan and Kielan-Ja-worowska, 1995). The anteroventral marginof the masseteric fossa is formed by the mas-seteric crest (‘‘mc’’ in fig. 10), a well-devel-oped ridge that is concave posteriorly andconfluent posteriorly with the weaker mas-seteric line, as seen in PSS-MAE 101. The

masseteric line is nearly straight and is con-tinuous posteriorly with the nearly vertical,straight rear edge of the condylar process.

The coronoid process (‘‘cor’’ in figs. 8, 16)originates below the alveolar margin and itsanterior extent lies at the back edge of theanterior root of m1 (fig. 29). Between thelateral alveolar margin and the coronoid pro-cess there is a trough running obliquely pos-terodorsally, the temporal groove (‘‘tg’’ infig. 10), which is said to be for the pars an-terior of the temporalis muscle (Gambaryanand Kielan-Jaworowska, 1995). The coro-noid process is subtriangular and only slight-ly higher than the condyle and the molarcusps (fig. 29). The angle of the anterior bor-der of the coronoid process relative to the


Fig. 30. Stereophotograph of the left lower jaw of Kryptobaatar dashzevegi PSS-MAE 113 in medialview.

toothrow is approximately 608 (contra Kie-lan-Jaworowska and Hurum, 1997, who re-port it as less than 458). A sharp crest con-nects the tip of the coronoid process with thecondyle, delimiting a shallow mandibularnotch. The condyle (‘‘con’’ in figs. 6, 16) isrobust, lacks a distinct neck, and, as noted byKielan-Jaworowska and Hurum (1997), ispositioned at approximately the same level asthe molars. The articular surface on the con-dyle is teardrop-shaped and directed poste-rodorsally. In posterior view, the articularsurface is distributed symmetrically with re-spect to the plane of the dorsal border of theascending ramus. This is in contrast to thecondition in most mammals in which there

is an asymmetrical distribution (Cromptonand Hylander, 1986).

In medial view (fig. 30), the area of thesymphysis is only visible in the left lowerjaw of PSS-MAE 113. Most of this surfacehas been lost, but enough remains to char-acterize the symphysis as vertical and narrow(‘‘msy’’ in fig. 14). The dorsal surface of thediastema forms a roughly horizontal shelfthat narrows posteriorly; the posterior nar-rowing of this shelf is a result of the obliqueorientation of the alveoli of the premolarsand molars in the lower jaw. The most dis-tinctive feature of the medial side of themandible is the very deep, enlarged ptery-goid fossa located behind the alveolar pro-


cess. This fossa, for the medial pterygoidmuscle (Gambaryan and Kielan-Jaworows-ka, 1995), is delimited anteriorly by the pter-ygoid crest (‘‘pc’’ in fig. 16), which extendsposteroventrally as the medial margin of theprominent pterygoid shelf. The dorsal mar-gin of the pterygoid fossa is formed by alongitudinal, inconspicuous elevation, thetransversal elevation, and the posterior mar-gin is formed by a crest that is the medialequivalent of the condyloid crest describedon the lateral surface. Continuing forwardfrom the pterygoid fossa just behind the levelof m2 is the mandibular canal. Behind the

pterygoid fossa is a concave area below thecondyle, the pterygoid fovea for the lateralpterygoid muscle (Gambaryan and Kielan-Jaworowska, 1995). The fovea is both small-er and shallower than the pterygoid fossa andis delimited posteriorly by the posterior mar-gin of the lower jaw.

Another measure of the lower jaw in mul-tituberculates reported by Kielan-Jaworows-ka and Hurum (1997) is the angle betweenthe lower margin of the mandible and theocclusal surface of the molars. In Kryptobaa-tar, this angle is very low, approximately118.

VASCULAR RECONSTRUCTIONS

Comprehensive reconstructions of the cra-nial vascular system in multituberculateswere first attempted by Kielan-Jaworowskaet al. (1984, 1986). These authors surveyedthe relevant osteology in the then currentlyknown Mongolian Late Cretaceous taxa, butthey relied most heavily on cranial remains,endocasts, and sectioned skulls of two forms,Nemegtbaatar and Chulsanbaatar, alongwith isolated petrosals tentatively referred tothe Late Cretaceous North American ‘‘Ca-topsalis’’ joyneri and Mesodma thompsoni(see Wible and Hopson, 1995). Kielan-Ja-worowska et al. (1984, 1986) discovered thatmost of the presumed vascular channels inthe multituberculates also occurred in otherMesozoic mammaliamorphs, with Mamma-liamorpha being the clade comprising the lastcommon ancestor of Tritylodontidae andMammalia plus all descendants (Rowe,1988). Given that remnants of these channelstransport vessels in extant mammals, Kielan-Jaworowska et al. (1984, 1986) were able tooffer reconstructions of the major cranial ar-teries and veins, including the dural sinuses,in the multituberculates. In general, these re-constructions have been followed by subse-quent researchers and applied to other Me-sozoic mammaliamorphs, although some re-finements have been proposed by Wible(1989) and co-workers (Rougier et al., 1992;Hopson and Wible, 1993; Wible and Hopson,1995). With these refinements, vascular re-constructions have been offered for the fol-lowing multituberculates: the Paleocene tae-

niolabidid Lambdopsalis (Miao, 1988), ageneralized ‘‘taeniolabidoid’’ (Rougier et al.,1992), and ‘‘Catopsalis’’ joyneri and Mesod-ma thompsoni (Wible and Hopson, 1995).

The vascular reconstructions proposedhere for Kryptobaatar dashzevegi (figs. 36B,37B) are indebted to those already published,because the channels that previous authorsidentified in various multituberculates are,with minor amendments, repeated in Krypto-baatar. The basis for the reconstructions andthe terminology employed is our own pub-lished and unpublished studies of the cranialvessels occurring in extant mammals (Wible,1984, 1986, 1987, 1989, 1990; Rougier et al.,1992; Wible and Hopson, 1995; and refer-ences cited therein).

VEINS

In extant mammals, the veins of the headcan be divided into the intracranial dural si-nus system and the extracranial veins drain-ing the dural sinuses and superficial struc-tures. Although the components of the duralsinus system are conservative among extantmammals, the extracranial veins draining itvary considerably (Gelderen, 1924; Wible,1990; Wible and Hopson, 1995). In mono-tremes, the principal drainage is through theprootic canal and foramen magnum (Hochs-tetter, 1896), whereas it is through the post-glenoid foramen and foramen magnum inmarsupials (Gelderen, 1924; Archer, 1976)and through the postglenoid and internal jug-


Fig. 31. Reconstruction of the skull of Kryptobaatar dashzevegi in anterior view. Abbreviations: frfrontal; lac lacrimal; man mandible; mx maxilla; na nasal; pmx premaxilla; sq squamosal.

ular foramina and foramen magnum in mostplacentals (Gelderen, 1924; Butler, 1967).

DURAL SINUS SYSTEM

The dural sinus system reconstructed forKryptobaatar resembles in most regards thatproposed for Nemegtbaatar, Chulsanbaatar(Kielan-Jaworowska, 1986; Kielan-Jawo-rowska et al., 1986), and Lambdopsalis(Miao, 1988). Included are superior (dorsal)sagittal, transverse, sigmoid, and prootic si-nuses; the first three are nearly ubiquitous intheir incidence among extant mammals (Gel-deren, 1924), whereas the prootic sinus isknown only for adult monotremes and someadult marsupials (Wible, 1990; Wible andHopson, 1995). The evidence for these duralsinuses in Kryptobaatar is in the form ofgrooves on the endocranial surfaces andmolds of vascular channels on the endocaststhat resemble those occurring in extant mam-mals. The superior sagittal and transverse si-nuses are visible in dorsal view in the ex-posed endocast of PSS-MAE 113; the supe-rior sagittal sinus lies between the right andleft cerebral hemispheres, and the pairedtransverse sinuses are between the cerebralhemispheres and the vermis (central lobe) ofthe cerebellum. Evidence for the sigmoid andprootic sinuses, the end distributaries of thetransverse sinus, is found on the specimens

showing the endocranial surfaces. The pro-otic sinus is indicated by a groove and fo-ramen into the prootic canal in the anteriorlamina quadrant (figs. 25, 27), resemblingthose structures in monotremes (Wible andHopson, 1995). The sigmoid sinus is indi-cated by grooves on the petrosal and exoc-cipital (fig. 26), and it is apparent that thesinus left the cranial cavity via the foramenmagnum based on the extremely reduced sizeof the jugular foramen and the sulcus pre-served in PSS-MAE 125 (fig. 26). This pat-tern is also clearly shown on the endocast ofan undescribed skull of a new species ofMongolian Late Cretaceous multituberculate(PSS-MAE 126). Kryptobaatar shows no in-dication of a tentorial sinus, described asdraining into the prootic sinus from in frontin Chulsanbaatar and Nemegtbaatar (Kie-lan-Jaworowska et al., 1986).

Other major components of the dural sinussystem that are ubiquitous among extantmammals are the cavernous sinus, the majorvenous pool in the cavum epiptericum, andits two principal conduits, the inferior (ven-tral) petrosal sinus (petrobasilar sinus) andthe ophthalmic veins (Shindo, 1915; Gelde-ren, 1924). The cavernous sinus and oph-thalmic veins are not indicated in Krypto-baatar, but were likely present, with the oph-thalmic veins entering the cavernous sinus


Fig. 32. Reconstruction of the skull of Kryptobaatar dashzevegi in dorsal view. Abbreviations: alanterior lamina; exoc exoccipital; fr frontal; ju jugal; lac lacrimal; mx maxilla; na nasal; pa parietal;pmx premaxilla; sq squamosal; sup supraoccipital.

via the sphenorbital fissure. The course of theinferior petrosal sinus is indicated by a shal-low sulcus on the endocranial surface of thebasioccipital–petrosal suture in PSS-MAE123 (fig. 25) and by a trough in the sameplace in the indeterminate multituberculatePSS-MAE 128. Similarly situated sulci forthe inferior petrosal sinus are found in manyextant mammals, including both monotremes(Hochstetter, 1896; Kuhn, 1971; Zeller,1989). Given the small size of the jugularforamen in Kryptobaatar, we think that theinferior petrosal sinus left the cranial cavityvia the foramen magnum as vertebral veins

(‘‘vv’’ in fig. 37B), as occurs in both mono-tremes (Hochstetter, 1896). A transversegroove on the back of the dorsum sellae inPSS-MAE 123 (fig. 25) and 124 likely heldthe posterior intercavernous sinus, as in thedog (Evans and Christensen, 1979) and var-ious marsupials (Archer, 1976); this grooveopens into the area of the sulcus for the in-ferior petrosal sinus along the basioccipital–petrosal suture.

LATERAL HEAD VEIN

In addition to the venous drainage fromthe sigmoid and inferior petrosal sinuses ex-


Fig. 33. Reconstruction of the skull of Kryptobaatar dashzevegi in lateral view. Abbreviations: alanterior lamina; exoc exoccipital; fr frontal; lac lacrimal; man mandible; mx maxilla; na nasal; ororbitosphenoid; pa parietal; pet petrosal; pmx premaxilla; sq squamosal; sth stylohyal.

iting the skull via the foramen magnum, theother major egress for the dural sinuses inKryptobaatar was via the prootic canal. Asin monotremes and some marsupials (Wible,1990; Wible and Hopson, 1995), the prooticsinus in Kryptobaatar (‘‘ps’’ in fig. 37B)drained into the lateral head vein (‘‘lhv’’ infig. 37B) in the middle-ear cavity through theprootic canal, which in extant taxa forms inthe posterior aspect of the prootic foramen inthe primary braincase wall (fig. 24A–C).Judging from the size of the endocranial ap-erture of the prootic canal (figs. 25, 27), thelateral head vein was well developed inKryptobaatar; the tympanic aperture of theprootic canal is partially hidden within a re-cess. The intratympanic course of the lateralhead vein (fig. 37B) was posteromedially,dorsal to the facial nerve. The vein then con-tinued ventrally into the neck as the internaljugular vein. The lateral head vein probablydid not leave the tympanic cavity with thefacial nerve through the stylomastoid notch,contra Kielan-Jaworowska et al. (1986) andMiao (1988), but had a separate, more medialexit (Rougier et al., 1992; Wible and Hopson,1995). Whether a post-trigeminal vein waspresent in Kryptobaatar is uncertain, al-though we have reconstructed one here(‘‘ptv’’ in figs. 36B, 37B). Along with the

prootic sinus, the post-trigeminal vein is theother major tributary of the lateral head veinin the echidna, draining posteriorly from thecavernous sinus (Wible and Hopson, 1995);we have observed a small vein in a similarposition accompanying the ramus inferior ofthe stapedial artery in the juvenile platypusdescribed by Zeller (1989). We think that theramus inferior was present in Kryptobaatar(see below), and a post-trigeminal vein hasbeen reconstructed accompanying that arteryin other multituberculates by Wible and Hop-son (1995). However, given the absence ofthis vein in therians, without specific osteo-logical evidence for it, the presence of thepost-trigeminal vein in Kryptobaatar andother multituberculates is equivocal.

TRANSVERSE CANAL VEIN

A canal in front of the hypophyseal fossaconnecting the right and left orbitotemporalregions is occupied among living mammalsby a large vein known as the transverse canalvein (McDowell, 1958; Archer, 1976). Giventhat a similar canal is present in specimensof Kryptobaatar (figs. 26, 36A) and in PSS-MAE 128, we restore that vessel here (‘‘tcv’’in fig. 36B). Judging from the size of the


Fig. 34. Reconstruction of the skull of Kryptobaatar dashzevegi in ventral view. Abbreviations: alanterior lamina; ali alisphenoid; bo basioccipital; bs basisphenoid; exoc exoccipital; fr frontal; mxmaxilla; pal palatine; pet petrosal; pmx premaxilla; pt pterygoid; sq squamosal; vo vomer.

transverse canal, the vein in Kryptobaatarwas a substantial one.

PITUITO-ORBITAL VEIN

A vein to the pituitary gland from the or-bital sinus, passing through the primarybraincase wall, is known for various saurop-sids (Bruner, 1907; Bellairs and Kamal,1981). Among mammals, monotremes havea similar vein (Gaupp, 1908), which in theplatypus exits the hypophyseal fossa via thecarotid foramen (personal obs.). Kryptobaa-tar has a foramen in the anterolateral wall ofthe hypophyseal fossa connecting to the or-

bitotemporal fossa, which in one specimen(PSS-MAE 123) leads into a sulcus runningposteriorly to the carotid foramen (fig. 25).This arrangement resembles that which wehave observed in the platypus. Consequently,we restore the pituito-orbital vein in this fo-ramen in Kryptobaatar.

VEINS OF ASCENDING, ORBITOTEMPORAL, AND

POSTTEMPORAL CANALS

Veins likely also accompanied at leastsome of the major arteries passing throughthe lateral braincase wall in Kryptobaatar(fig. 36B). Following a model proposed by


Fig. 35. Reconstruction of the skull of Kryptobaatar dashzevegi in posterior view. Abbreviations:bo basioccipital; exoc exoccipital; fr frontal; lac lacrimal; man mandible; pa parietal; pet petrosal; sqsquamosal; sup supraoccipital.

Watson (1911, 1920), various authors (e.g.,Kermack et al., 1981) have reconstructedveins as the sole occupants of the ascending,orbitotemporal, and posttemporal canals inMesozoic mammaliaforms. However, Kie-lan-Jaworowska et al. (1984, 1986) arguedeffectively, on the basis of the anatomy ofextant mammals, that arteries and not veinswere the primary occupants of these canalsin multituberculates and other Mesozoic syn-apsids (see also Wible, 1989). Following themodel of Kielan-Jaworowska et al. (1984,1986), we reconstruct in Kryptobaatar venaecommitantes with the ramus superior, ramussupraorbitalis, and arteria diploetica magnain the ascending, orbitotemporal, and post-temporal canals, respectively (fig. 36B).These veins drained from the orbit to themiddle ear and occiput, and they likely weresubstantial, as evidenced by the preservedendocast of the orbitotemporal canal showingtwo subequal vessels in the undescribed skullof a new species of Mongolian Late Creta-ceous multituberculate (PSS-MAE 126).

SUBARCUATE FENESTRATION

In Nemegtbaatar and Chulsanbaatar, Kie-lan-Jaworowska et al. (1986) identified an-

other vascular canal between the subarcuatefossa and the posttemporal canal, which inthe embryos of some extant bats and ele-phant shrews transmits a venous connectionbetween the dural sinuses and the mastoidemissary system. These authors also reporteda minute fenestration of the subarcuate fossain ‘‘Catopsalis’’ joyneri. Luo (1989) found alarge fenestration in Ptilodus montanus andMesodma, and he (1996) proposed this con-dition as a synapomorphy of ptilodontids andeucosmodontids. In Kryptobaatar, the subar-cuate fossa is very deep, hindering identifi-cation of a subarcuate fenestration; neverthe-less, as far as we can tell, such an openingis lacking.

ARTERIES

In extant mammals, three major paired ar-teries supply blood to the head: the internalcarotid, the external carotid, and the verte-bral. In most forms, the internal carotid andvertebral supply the brain, whereas the ex-ternal carotid along with the stapedial artery,the principal extracranial branch of the inter-nal carotid, supply the meninges and super-ficial structures (Tandler, 1899, 1901). How-ever, in the echidna, marsupials, and certain


placentals (e.g., xenarthrans, carnivorans, un-gulates), the stapedial artery is lacking andmost or all of its end branches are annexedto the external carotid (Bugge, 1978, 1979;Wible, 1984, 1987).

INTERNAL CAROTID ARTERY

In extant mammals, the internal carotid ar-tery has an extracranial course ventral to thebasicranium en route to the carotid foramen,which opens endocranially posterolateral tothe hypophysis (pituitary). Both the positionof the extracranial course and the foramenvary. The artery’s course is either on thepromontorium (transpromontorial), throughthe substance of the tympanic wall (perbul-lar), or medial to the tympanic wall (extra-bullar); additionally, the extracranial courseof the artery is not always indicated by agroove or canal (Wible, 1986). The carotidforamen is either within the basisphenoid orbetween the basisphenoid and petrosal (DeBeer, 1937), and its endocranial aperture iseither within the hypophyseal fossa or pos-terolateral to it.

In Kryptobaatar (‘‘ica’’ in fig. 37B), theoccupant of the well-developed groove onthe anterior half of the promontorium, the ca-nal between the petrosal, alisphenoid, andpterygoid, and the endocranial foramen with-in the hypophyseal fossa was the internal ca-rotid artery, vein, and nerve. Among extantmammals, similar transpromontorial groovesare found only in certain placentals (e.g., li-potyphlans, aardvarks) and, with but one ex-ception, are occupied by the internal carotidartery along with its companion vein andsympathetic nerve (Wible, 1986). The excep-tion occurs in some strepsirhine primates, inwhich the artery is lost and the nerve is theprincipal occupant of the transpromontorialgroove (Conroy and Wible, 1978). However,given that this is an unusual occurrenceamong extant mammals, we deem it morelikely that both the internal carotid artery andnerve were present in Kryptobaatar. Addi-tionally, the connections and sizes of the var-ious sulci on the promontorium in Krypto-baatar strongly support a vascular interpre-tation. The internal carotid artery was likelynot the sole supplier of blood to the brain inKryptobaatar, given that the vertebral artery

is present in extant monotremes, marsupials,and the vast majority of placentals (Hochs-tetter, 1896; Tandler, 1899, 1901; Gillilan,1972), with cetaceans being an exception(McFarland et al., 1979).

STAPEDIAL ARTERY

During development in all but one of theextant mammals investigated, a well-devel-oped stapedial artery arises from the extra-cranial portion of the internal carotid andsends branches with the trigeminal nerve(Tandler, 1902; Wible, 1984, 1987); the oneexception is the echidna, in which the sta-pedial artery apparently does not form andthe end branches accompanying the trigem-inal nerve are supplied through the externalcarotid (Hochstetter, 1896). In placentals,there are two major branches of the embry-onic stapedial system: the ramus superior,supplying a ramus supraorbitalis which ac-companies the ophthalmic nerve, and the ra-mus inferior, supplying a ramus infraorbitalisand mandibularis in company with the max-illary and mandibular nerves, respectively(Tandler, 1902; Wible, 1984, 1987). The ra-mus superior fails to form in marsupials andis a late embryological addition in the platy-pus that does not extend rostrally as far asthe ophthalmic nerve (Wible, 1984, 1987).The embryonic stapedial system is retainedin few adult mammals, and in most, at leastone of the end branches of the stapedial ar-tery is annexed to a branch of the externalcarotid (Tandler, 1899, 1901; Bugge, 1974;Wible, 1984, 1987). In fact, the main stemof the stapedial (proximal stapedial) is re-tained only in the adult platypus and certainadult placentals (e.g., lipotyphlans, scanden-tians) (Tandler, 1899; Bugge, 1974; Wible,1987). The two principal sites of anastomosisbetween the external carotid and stapedialsystems are the maxillary artery, which con-nects to the ramus inferior of the stapedialbelow the mandibular nerve exit, and the ar-teria diploetica magna, which connects to theramus superior of the stapedial through theposttemporal canal (Wible, 1987).

Kryptobaatar has grooves, canals, and fo-ramina that indicate the presence of a well-developed stapedial system, including theproximal stapedial artery, ramus inferior, ra-


Fig. 36. Reconstruction of the skull of Kryptobaatar dashzevegi in left lateral view (A) with themajor cranial arteries (white outline) and veins (black) added (B). Zygomatic arch is cut to show vesselsin the floor of the orbit. Vessels in shadow pass through intramural canals, with the exception of thepart of the ramus superior immediately rostral to the dorsalmost ramus temporalis branch which isendocranial. Abbreviations: adm arteria diploetica magna (and accompanying vein); ea ethmoidal artery(and accompanying vein); ef ethmoidal foramen; fbu foramen buccinatorium; fdac foramen of dorsalascending canal; fdv foramen for frontal diploic vein; fma foramen masticatorium; frt foramen forramus temporalis; ioa infraorbital artery (and accompanying vein); iof infraorbital foramen; lhv lateralhead vein; mef metoptic foramen (which transmitted pituito-orbital vein); oa ophthalmic artery; ocaoccipital artery (and accompanying vein); ompf orbital opening of minor palatine foramen; opf opticforamen; otc orbitotemporal canal; psa proximal stapedial artery; ptc posttemporal canal; ptv post-trigeminal vein; ri ramus inferior; rio ramus infraorbitalis (and accompanying vein); rm ramus man-dibularis (and accompanying vein); rs ramus superior (and accompanying vein); rso ramus supraorbitalis(and accompanying vein); rt ramus temporalis (and accompanying vein); sgf supraglenoid foramen; sonsupraorbital notch; spa sphenopalatine artery (and accompanying vein); spf sphenopalatine foramen;sphf sphenorbital fissure; tc transverse canal; tcv transverse canal vein.


mus superior, and arteria diploetica magna(figs. 36, 37). The proximal stapedial arteryoccupied the groove on the promontoriumthat runs posterolaterally from the groove forthe internal carotid artery toward the fenestravestibuli (‘‘psa’’ in fig. 37B). In extant mam-mals, grooves on the promontorium directedat the fenestra vestibuli invariably containthe proximal stapedial artery, but suchgrooves are only found in some placentalsretaining that artery in the adult (e.g., li-potyphlans; Wible, 1987). From the rim ofthe fenestra vestibuli, the proximal stapedialartery in Kryptobaatar ran through what ap-pears to be a bicrurate stapes (Rougier et al.,1996c); this is the artery’s course in the pla-centals that retain it, but in the platypus theartery runs posterodorsal to the columnar sta-pes (Wible, 1987).

Two canals lateral to the fenestra vestibuliin Kryptobaatar indicate that beyond the sta-pes the proximal stapedial divided into theramus superior and the ramus inferior (‘‘rs’’and ‘‘ri’’ in figs. 36B, 37B), as occurs in theplatypus and certain placentals (Wible,1987). The canal for the ramus inferior inKryptobaatar is directed anteriorly and liesdorsal to the epitympanic recess on the in-folded lateral flange. Posteriorly, this canal isconfluent with the canal for the facial nerve,and anteriorly it presumably opens into thecavum epiptericum. In extant mammals, sim-ilarly situated canals are found in the echidnaand certain placentals (e.g., scandentians,macroscelidids). In the placentals, the mainoccupant is arterial, with the ramus inferiorin front and sometimes the proximal stape-dial behind; the course of the facial nerve isvery close to the arterial canal, and, in fact,in macroscelidids, the facial nerve and prox-imal stapedial share a canal (MacPhee,1981). In the echidna, in which the ramusinferior is lacking, the main occupant is ve-nous, the post-trigeminal in front and the lat-eral head vein behind, with the posterior partof the canal also transmitting the facial nerve(Wible and Hopson, 1995). The only form inwhich the ramus inferior and post-trigeminalvein co-occur is the platypus, and thesestructures travel together in an open sulcus(Wible and Hopson, 1995; personal obs.). Wethink that in Kryptobaatar the ramus inferiorwas the major vascular occupant of the canal

in question, because the large size of thegroove for the proximal stapedial on thepromontorium and the small size of the fo-ramen for the ramus superior suggest that theother major end branch of the proximal sta-pedial, the ramus inferior, was present. Theevidence for the occurrence of the post-tri-geminal vein in Kryptobaatar has been dis-cussed above.

The course of the ramus inferior rostral tothe petrosal bone in Kryptobaatar was pre-sumably through the cavum epiptericum andthen the sphenorbital fissure (fig. 36B), be-cause other suitable foramina of exit in themesocranium are lacking. An endocranialcourse for the ramus inferior (or its rostralcontinuation, the ramus infraorbitalis) is un-usual among extant mammals, but is knownfor some chiropterans (Kallen, 1977; Wibleand Davis, 2000) and dipodoid rodents (Bug-ge, 1971). Within the orbit, the ramus infe-rior, now the ramus infraorbitalis (‘‘rio’’ infig. 36B), ran forward in the sphenopalatinegroove along the suture between the maxillaand frontal. En route, it likely supplied theminor palatine artery into a foramen in theorbital process of the maxilla, the major pal-atine and sphenopalatine artery in the sphe-nopalatine foramen (‘‘spa’’ in fig. 36B), andits terminal branch, the infraorbital arteryinto the infraorbital canal (‘‘ioa’’ in fig. 36B).It also likely provided a ramus orbitalis con-necting with the rostral continuation of theramus superior (see below).

Isolated petrosals of Late Cretaceous mul-tituberculates, in particular ‘‘Catopsalis’’joyneri, have been critical for reconstructingthe ramus superior in Kryptobaatar, as muchof this vessel is hidden within the lateral wallof the braincase. In ‘‘C.’’ joyneri, en routefrom the middle ear to the orbit, the ramussuperior passed through a complex canal,which for descriptive purposes is treated asthree successive canals: the ventral ascend-ing, dorsal ascending, and orbitotemporal(Rougier et al., 1992; Wible and Hopson,1995). The ventral ascending canal begins inthe middle ear, runs back posteriorly in a hor-izontal plane, and then turns dorsally; open-ing into it from the temporal fossa is the su-praglenoid foramen, which transmitted a ra-mus temporalis to the temporalis muscle. Thedorsal ascending canal is essentially vertical,


←

mpt medial pterygopalatine trough; oca occipitalartery (and accompanying vein); ocon occipitalcondyle; on odontoid notch; pacc posterior aper-ture of carotid canal; pef perilymphatic foramen;ppr paroccipital process; pr promontorium of pe-trosal; ps prootic sinus; psa proximal stapedial ar-tery; ptc posttemporal canal; ptca pterygoid ca-nal; ptr pterygopalatine ridge; ptv post-trigeminalvein; ri ramus inferior; rs ramus superior; rvnfrecess for vascular and nervous foramina (prooticcanal, ventral ascending canal, canal for ramus in-ferior, and facial canal); th tympanohyal; ttf ten-sor tympani fossa; vo vomer; vv vertebral vein.

Fig. 37. Reconstruction of the right basicra-nium of Kryptobaatar dashzevegi in ventral view(A) with the major cranial arteries (white outline)and veins (black) added (B). Abbreviations: admarteria diploetica magna (and accompanyingvein); ali alisphenoid; aptca artery of pterygoidcanal; cp crista parotica; ctpp caudal tympanicprocess of petrosal; er epitympanic recess; fmaforamen masticatorium; foi foramen ovale infer-ium; fv fenestra vestibuli; gica groove for internalcarotid artery; gl glenoid fossa; gpsa groove forproximal stapedial artery; ica internal carotid ar-tery; jf jugular fossa; jfo jugular foramen; lhv lat-eral head vein; lpt lateral pterygopalatine trough;

curving forward into the anteriorly directedorbitotemporal canal. Opening from behindand marking the limits of the ventral and dor-sal ascending canals is the posttemporal ca-nal, which transmitted the arteria diploeticamagna from the occiput. The ventral ascend-ing canal is largely within the anterior lam-ina, the dorsal ascending canal is between thepetrosal and presumably the squamosal, andthe posttemporal canal is entirely in the pe-trosal. The reconstruction of arteries intothese various canals is based largely on thepattern in monotremes; between the platypusand echidna all the canals present in ‘‘C.’’joyneri can be found (Rougier et al., 1992;Wible and Hopson, 1995).

The ramus superior in Kryptobaatar exhib-ited essentially the same pattern as in ‘‘Ca-topsalis’’ joyneri, as reconstructed from thevarious surface foramina into this system ofcanals (figs. 36A, 37A). The ventral ascend-ing canal in Kryptobaatar is indicated by itstympanic aperture and by two foramina in theanterior lamina on the lateral braincase wall,which transmitted rami temporales from thecanal (‘‘rt’’ in fig. 36B). The position of thesethree foramina suggests that the ventral as-cending canal runs posteriorly in a largelyhorizontal plane. Breakage in one specimen,PSS-MAE 101, provides access to the nearvertical dorsal ascending canal between thepetrosal and squamosal and shows a notchthat was probably enclosed to complete a fo-ramen, which transmitted a third ramus tem-poralis from the canal (fig. 36). The anteriorlydirected orbitotemporal canal is indicated bya horizontal endocast representing its fillingand by its anterior aperture in the orbit. The


endocast reveals that the anterior part of theorbitotemporal canal was enclosed, but thatthe posterior part had no medial wall andtherefore is an endocranial orbitotemporalgroove. Finally, the posttemporal canal is in-dicated by its posterior aperture on the occi-put, entirely within the petrosal (figs. 16, 17).Judging from the size of this aperture, the ar-teria diploetica magna was of substantial size(‘‘adm’’ in figs. 36B, 37B).

The rostral continuation of the ramus su-perior within the orbit is the ramus supraor-bitalis (‘‘rso’’ in fig. 36B). In the extant mam-mals retaining this vessel (i.e., monotremesand some placentals), it provides up to foursorts of branches: the lacrimal, frontal, andethmoidal arteries running with the ophthal-mic nerve branches of the same name, andthe ramus orbitalis anastomosing with the ra-mus infraorbitalis (Tandler, 1899, 1902; Bug-ge, 1974; Wible, 1987). Given that all fourbranches are present in both the platypus andechidna and in diverse placentals (e.g., ar-madillos, lemurs), it is probable that Krypto-baatar exhibited the same condition.

ARTERY OF THE PTERYGOID CANAL

An artery accompanies the nerve of thepterygoid canal in many extant mammals,but in most it is derived from the externalcarotid system and runs posteriorly into thecanal from the orbit (Wible, 1984). In only a

few instances (e.g., some lipotyphlans, Mc-Dowell, 1958; MacPhee, 1981; Novacek,1986) is the artery derived from the internalcarotid, running anteriorly into the canalfrom behind. Given the large size of the sul-cus running forward from the carotid canalvisible in PSS-MAE 113 (fig. 21), it seemslikely that Kryptobaatar exhibited the patternas some lipotyphlans (‘‘aptca’’ in fig. 37B).However, in light of the rare occurrence ofthis origin for the artery of the pterygoid ca-nal among extant mammals, we are less con-fident in this particular aspect of the vascularreconstruction. Moreover, note that the majoroccupant of the pterygoid canal in extantmammals is the nerve (Wible, 1984).

OPHTHALMIC ARTERY

An ophthalmic artery derived from the in-ternal carotid (circulus arteriosus) and ac-companying the optic nerve is widely dis-tributed among extant mammals, includingthe echidna, all marsupials investigated, andmost placentals (Tandler, 1899, 1902; Bugge,1974; Archer, 1976; Wible, 1984). In light ofthis distribution, an ophthalmic artery prob-ably ran with the optic nerve in the opticforamen in Kryptobaatar as well (‘‘oa’’ infig. 36B). It may have anastomosed withbranches of the ramus supraorbitalis withinthe orbit via a ramus orbitalis (not shown infig. 36B).

COMPARISONS

The following section is devoted to high-lighting particular characters or charactersystems of the multituberculate skull towhich our descriptions of Kryptobaatar arerelevant or have prompted reevaluation ofcurrent evidence. An ongoing project beyondthe scope of this report is our reevaluation ofthe phylogenetic interrelationships withinMultituberculata. In conducting this reeval-uation, we have observed many of the rele-vant multituberculate specimens, and someof our observations are included here. Toplace these taxa in a phylogenetic context,we have reproduced a recent hypothesis ofmultituberculate interrelationships (fig. 38)and of the higher-level relationships of mul-tituberculates (fig. 39).

As a cautionary note, the taxonomy of theLate Jurassic Paulchoffatiidae has been re-vised recently by Hahn (1993), who has beenpublishing on this group for more than aquarter century. Hahn (1993) has createdseveral new genera for previously describedspecimens and reassigned other specimens topreviously named genera. We follow Hahn’s(1993) revision in our taxonomic usages.

SNOUT AND PALATE

PRENASAL PROCESS OF THE PREMAXILLA

The tip of the rostrum is rarely preservedin multituberculates. PSS-MAE 101 (fig. 6)is an exception and clearly shows that a pre-


Fig. 38. Cladogram of phylogenetic relationships within Multituberculata (a), based on the analysisin Rougier et al. (1997), with an enlargement of the tree section containing Kryptobaatar and its nearestrelatives, all of which are Mongolian Late Cretaceous taxa with the exception of Pentacosmodon fromthe North American Paleocene (b). A similar Mongolian Late Cretaceous grouping was called Dja-dochtatheria by Kielan-Jaworowska and Hurum (1997). In their phylogeny, Kryptobaatar is in a clade,Djadochtatheriidae, with Djadochtatherium, Catopsbaatar, and Tombaatar.

nasal process of the premaxilla (internarialbar) is lacking in Kryptobaatar. A prenasalprocess dividing the external nares is foundin the outgroups to Mammalia, such as Hal-danodon (Lillegraven and Krusat, 1991),Morganucodon (Miao, 1988; Wible, 1991)and Sinoconodon (Wible et al., 1990; Cromp-

ton and Luo, 1993), but is lacking in extantmammals and the prototribosphenidan Vin-celestes (Rougier, 1993).

To date, an internarial bar has been recon-structed for only one multituberculate, thePaleocene taeniolabidid Lambdopsalis. Miao(1988) based this on the presence of a wedge


of bone between the anterior ends of the na-sals, which he interpreted to be the brokendorsal tip of the prenasal process. Miao(1988) speculated that the prenasal processin Lambdopsalis either was an evolutionaryreversal or that the process was more wide-spread in multituberculates than indicated inpreviously described specimens. In supportof the latter, Miao (1988: 9) suggested thatthe left nasal in stereophotographs of Chul-sanbaatar (ZPAL MgM-I/89) published inKielan-Jaworowska et al. (1986: fig. 16) ap-peared ‘‘indicative of a sutural facet at theanterior end,’’ presumably for the prenasalprocess. However, we interpret the left nasalin this specimen as broken, in light of otherspecimens of Chulsanbaatar (e.g., ZPALMgM-I/139) that show a more complete na-sal with no sutural facets at the anterior end(see Kielan-Jaworowska, 1974: pl. XI).Therefore, we find no evidence of an inter-narial bar in Chulsanbaatar or other multi-tuberculates described to date. The presenceof an internarial bar in Lambdopsalis is con-troversial (see below), and the absence of thisstructure primitively in multituberculates iscorroborated by our observations of snoutsof Late Jurassic paulchoffatiids, includingtwo species of Pseudobolodon (V.J. 111-155,V.J. 450-155) and one of Kuehneodon (V.J.112-155).

The element in question in Lambdopsalisis illustrated in two rostra (Miao, 1988; figs.1, 2), but it exhibits differences in size andposition between the two, as well as betweenthe right and left sides of the same individ-ual. Given these differences, the purportedprenasal process of the premaxilla may be abroken part of the nasal. Despite our doubtabout the element in Lambdopsalis, it islargely irrelevant as a systematic feature, be-cause it is not shared with any other multi-tuberculate and was absent primitively in thegroup.

Although the internarial bar is lacking inadult extant mammals, remnants of it occurduring development in monotremes and mar-supials. In the platypus, it forms prenatallyas a dorsal outgrowth from the rostral part ofthe premaxilla (os carunculae of Wilson,1900), which supports the egg-tooth; it per-sists in young individuals after the egg-toothis shed and is subsequently resorbed (Green,

1930; Hill and De Beer, 1949; Zeller, 1989).This element does not come into contact withthe nasals. In the echidna, the os carunculaeforms independent of, but in contact with, theprenasal process of the premaxilla (Gaupp,1908; Kuhn, 1971). Among marsupials, anindependent ossification in the cleft betweenthe anterior cupulae of the nasal capsule, alsocalled the os carunculae, has been reportedin embryos of the didelphid Caluromys phi-lander (Denison and Terry, 1921). In addi-tion, condensed connective tissue in the sameregion has been reported and homologizedwith the prenasal process of the premaxillain Trichosurus vulpecula (Broom, 1909), Di-delphis marsupialis (Toeplitz, 1920), Didel-phis aurita, and Perameles nasuta (Hill andDe Beer, 1949). Remnants of an egg-toothpapilla beneath the premaxilla have also beendiscovered in Trichosurus and Phascolarctoscinereus (De Beer, 1937; Hill and De Beer,1949). Finally, in placentals, remnants of theos carunculae, egg-tooth, and internarial barare unknown (De Beer, 1937). In light of thedistribution of the characters associated withthe os carunculae, egg-tooth, and internarialbar, as well as the phylogenetic position ofmultituberculates (fig. 39; Rowe, 1988, 1993;Rougier et al., 1996a, 1996c; Hu et al., 1997;Ji et al., 1999), it is likely that an os carun-culae and egg-tooth, or remnants thereof,were at least temporarily present in multitu-berculates at some point in ontogeny. Theimplications of these features for assessingwhether multituberculates were oviparous orviviparous are inconclusive, because differ-ent conditions for the os carunculae, egg-tooth, and internarial bar are present in theextant phylogenetic bracket for multituber-culates (i.e., monotremes and therians).Based on the reconstructed small size of thepelvic outlet in Kryptobaatar (ZPAL MgM-I/41) relative to that in monotremes, Kielan-Jaworowska (1979) proposed that multitu-berculates were viviparous with extremelysmall neonates. Lillegraven et al. (1987) ar-gued that the same condition was found inthe common ancestor of marsupials and pla-centals.

SEPTOMAXILLA

In multituberculates, the premaxilla has alarge dorsal process that forms the lateral


Fig. 39. Cladogram of phylogenetic relationships within Mammaliaformes based on analyses byRougier et al. (1996a) and Hu et al. (1997).

margin of the external nares (fig. 33), as inmost therians. In contrast, in monotremesand most groups of Mesozoic mammals, thisposition is occupied by a separate bone, theseptomaxilla. Based on recent phylogenetichypotheses of multituberculate relationships(e.g., Rougier et al., 1996a, 1996c; Hu et al.,

1997; Ji et al., 1999), the presence of a sep-tomaxilla in the group would be expected. Infact, such a bone has been recently describedfor the Late Jurassic paulchoffatiid Pseudo-bolodon krebsi (Hahn and Hahn, 1994). Itspurported septomaxilla is cuneiform and liesbetween the dorsal process of the premaxilla,


the nasal, and the maxilla; that is, it is ex-cluded entirely from the external nares. Thiswould be a unique position for the septo-maxilla among synapsids, which otherwisehave the septomaxilla in the margin of theexternal nares (Wible et al., 1990). Our ob-servations of the specimen of Pseudobolodonkrebsi (V.J. 451-155) described by Hahn andHahn (1994) suggest that the bone in ques-tion is likely part of the premaxilla and thatwhat is illustrated by these authors (1994: ta-bles 1, 2) as a suture between the purportedseptomaxilla and premaxilla is merely acrack within the premaxilla. Therefore, pend-ing new information, we consider the septo-maxilla to be absent in multituberculates. Itis also absent in marsupials and placentals(Wible et al., 1990), although the homologiesof the purported septomaxilla of some xe-narthrans remain controversial (Zeller et al.,1993).

NASAL OVERHANG AND ANTERIOR NASAL

NOTCH

In primitive mammaliaforms (e.g., Hal-danodon, Lillegraven and Krusat, 1991; Si-noconodon, Crompton and Luo, 1993), theanterior end of the nasal extends rostrally be-yond the septomaxilla and the posterodorsalprocess of the premaxilla, dorsal to the ex-ternal narial aperture. We call this conditionthe nasal overhang of the nares. Additionally,at least in Haldanodon, there is a deep an-terior nasal notch in the rostral margin that,according to Lillegraven and Krusat (1991),may have transmitted ethmoidal vessels.

Among multituberculates, a nasal over-hang of the nares is illustrated for the LateJurassic paulchoffatiid Pseudobolodon krebsi(Hahn and Hahn, 1994: fig. 2b), and we ob-served it in Kuehneodon simpsoni (V.J. 112-155) and Henkelodon nais (V.J. 401-155);these three taxa also have an anterior nasalnotch. In contrast, a nasal overhang is absentor greatly reduced in all younger taxa pre-serving the anterior tip of the nasal: Krypto-baatar (fig. 33), Kamptobaatar (Kielan-Ja-worowska, 1971: pl. III), Nemegtbaatar(Kielan-Jaworowska, 1974: pl. IX), Chulsan-baatar (ibid.: pl. XII), Catopsbaatar (ibid.:pl. XIX), and Lambdopsalis (Miao, 1988: fig.4). However, the anterior nasal notch is re-

tained, at least in Kryptobaatar (‘‘ano’’ infigs. 8, 12), Kamptobaatar (Kielan-Jawo-rowska, 1969: pl. XVI), and Catopsbaatar(Kielan-Jaworowska, 1974: pl. XVIII).

NASAL FORAMINA

Once considered to be highly unusualamong mammals (Kielan-Jaworowska, 1971;Miao, 1988), nasal foramina are now knownto be distributed widely among Mammalia-formes, including Sinoconodon (Cromptonand Luo, 1993), Haldanodon (Lillegravenand Krusat, 1991), Vincelestes (Rougier,1993), and Ornithorhynchus (Zeller, 1989).Nasal foramina have also been describedwidely among Late Cretaceous and Cenozoicmultituberculates. However, until recentlythey had not been reported for any Late Ju-rassic paulchoffatiid. In 1994, Hahn andHahn reported a single pair of small nasalforamina in Pseudobolodon krebsi (V.J. 451-155), and we have seen the same conditionin another species of the same genus, P.oreas (V.J. 461-155). However, we reporthere that nasal foramina are not ubiquitousamong paulchoffatiids, as they are whollylacking in Kuehneodon dryas (V.J. 454-155).Another feature of the nasal foramina thatvaries considerably among multituberculatesis the number of pairs present (Kielan-Ja-worowska and Hurum, 1997), from a singlepair in Pseudobolodon to seven or eight pairsin Lambdopsalis (Miao, 1988). Kielan-Ja-worowska and Hurum (1997) reported twopairs of nasal foramina in Kryptobaatar andin Chulsanbaatar, Catopsbaatar, Tombaa-tar, and the ptilodontoid Ptilodus. However,we found only one pair in Catopsbaatar(ZPAL MgM-I/78), and in Ptilodus, wefound three pairs in P. gracilis (USNM6076) and two pairs, asymmetrically placedon the right and left sides, in P. montanus(AMNH 35490). Finally, in KryptobaatarPSS-MAE 101, we found two foramina onthe right and three on the left (‘‘naf’’ in figs.6, 8).

Foramina in the nasals were first describedin multituberculates by Simpson (1925) inDjadochtatherium matthewi. However, it wasnot until 1937 that Simpson applied the termnasal foramina to these apertures in his de-scription of Ptilodus. Miao (1988) cited


Simpson (1937) as employing two differentterms for these foramina: either vascular fo-ramina or nasal foramina. From our readingof Simpson (1937), we think that the term heintended for these apertures was nasal foram-ina, and that vascular foramina was meant asa descriptor of their function and not an al-ternative term. Rather than vascular, a ner-vous function was suggested for the nasal fo-ramina by Kielan-Jaworowska (1971).

INFRAORBITAL FORAMINA

Some Jurassic multituberculates have twoinfraorbital foramina on the facial process ofthe maxilla, which vary in size from sube-qual to the anterior one being considerablythe larger (Hahn, 1985). The latter conditionis also found in Early Cretaceous Arginbaa-tar from Khoobur in Mongolia (Kielan-Ja-worowska et al., 1987). The smaller posteriorforamen is absent in Kryptobaatar (fig. 33)and other Mongolian Late Cretaceous mul-tituberculates, and in Cimolodonta (sensuSimmons, 1993; i.e., ptilodontoids and tae-niolabidoids), except for Meniscoessus ro-bustus (Archibald, 1982; Simmons, 1993).

Multiple facial exits for the infraorbital ca-nal system are present in basal mammalia-morphs (Kermack et al., 1981; Lillegravenand Krusat, 1991; Crompton and Luo, 1993),monotremes (Kuhn, 1971; Zeller, 1989), andin members of the therian lineage, includingpaurodontids (personal obs. of an undescrib-ed new genus from the Morrison Formation)and Vincelestes (Rougier, 1993). A foramenfor the lacrimal branch of the infraorbitalsystem is present in most forms with multipleexits; it opens between the maxilla, the lac-rimal, and sometimes the jugal. Some Juras-sic multituberculates and monotremes havemultiple exits, but lack the foramen for thelacrimal branch.

Miao (1988) questioned the utility of char-acters of the infraorbital foramina in phylo-genetic analyses, because there are instancesof individual variability. For example, on oneside of a maxilla of Lambdopsalis of the 13maxillae reported by Miao (1988), two ratherthan a single infraorbital foramina were pre-sent, though the smaller posterior foramenwas distinguished by only a tiny bridge.However, as pointed out by Rougier et al.

(1997a), the discrepancies noted by Miao(1988) have no bearing when considered ina wider phylogenetic framework, because thevariations occur in only a few very disparategroups. We add here that after review of themajor collections of Mongolian Late Creta-ceous multituberculates (MAE and ZPAL),we have found no instances of more than asingle infraorbital foramen.

Another feature of the infraorbital systempreserved in Kryptobaatar is several smallforamina linking the infraorbital canal witheither the maxillary sinus or nasal cavity.Rougier et al. (1997a) reported similar ap-ertures in another form from Ukhaa Tolgod,Tombaatar, and speculated that they maylead into alveolar canals and have carriednerves and vessels to the upper teeth. Con-firmation of the presence of alveolar canalsfor Kryptobaatar is provided by high-reso-lution CT scans of PSS-MAE 101.

PALATINE FORAMINA

There is considerable confusion about theforamina in and around the palatine’s contri-bution to the palate because different termshave been used by various authors. As ourpoint of reference, we have used the dog(Evans and Christensen, 1979), which hasthree sets of palatine foramina: a large an-terior one transmitting the major palatinenerve and vessels, a small one behind ittransmitting the accessory palatine nerve andartery, and at the back of the palate a notchor large foramen for the minor palatine nerveand vessels (see figs. 15, 34). Additionally,we have named these foramina based on theiroccupants (contra Evans and Christensen,1979): that is, major, accessory, and minorpalatine foramina, respectively. Similarlyplaced foramina are widely distributedamong mammaliaforms, but all three do notnecessarily occur together. For example,among primitive mammaliaforms, Sinoco-nodon (Crompton and Luo, 1993) has allthree, but Morganucodon (Kermack et al.,1981) and Haldanodon (Lillegraven andKrusat, 1991) have only the major and minorpalatine foramina.

Among multituberculates, paulchoffatiidsappear to have only the major and minor fo-ramina (Hahn, 1987). Most of the Mongolian


Late Cretaceous taxa with well-preservedpalates (Kryptobaatar; Kamptobaatar, Kie-lan-Jaworowska, 1971; Chulsanbaatar, Ne-megtbaatar, Kielan-Jaworowska et al., 1986)have all three (fig. 15), although the numberof accessory foramina varies. Another vari-ant is presented by Ptilodus montanus, whichseems to have only the major and accessorypalatine foramina (Simpson, 1937; AMNH34590). This is the same condition reportedfor Lambdopsalis, although all three fora-mina are visible in stereophotographs and re-constructions in Miao (1988: figs. 3, 14, 18).

According to Kielan-Jaworowska et al.(1986), the opening we call the minor pala-tine foramen does not penetrate bone fully inNemegtbaatar as shown from their study ofa sectioned specimen. To distinguish thisopening from a true foramen, Kielan-Jawo-rowska et al. (1986) employed the term pa-latonasal notch. Additionally, they general-ized from their observations of Nemegtbaa-tar, suggesting that the purported foramenpresent in other multituberculates is alsomerely a palatonasal notch. We cannot sayunequivocally that the minor palatine fora-men in Kryptobaatar penetrates bone, be-cause full preparation of this passageway isessentially impossible. However, Kryptobaa-tar has a foramen in the orbital process ofthe maxilla (fig. 12) that closely resemblesthe one leading into the minor palatine canalin various therians (Novacek, 1986). Conse-quently, we are fairly confident that the mi-nor palatine foramen on the palate transmit-ted the same named nerves and vessels inKryptobaatar. We also think that a true mi-nor palatine foramen is more widespreadamong multituberculates, but justificationawaits better preserved orbital processes ofthe maxilla.

FACIAL PROCESS OF LACRIMAL

In their phylogenetic analysis of selectedmultituberculate taxa, Kielan-Jaworowskaand Hurum (1997) reported a large, roughlyrectangular facial process of the lacrimal sep-arating the frontal and the maxilla as char-acteristic of Djadochtatheria, the monophy-letic clade that includes 10 Mongolian LateCretaceous genera. This is in contrast to thesmall, arcuate facial process of some other

multituberculates. We offer the followingamendments to Kielan-Jaworowska and Hu-rum’s scoring.

Among Djadochtatheria, whereas the typ-ical djadochtatherian facial process is said tobe present in Sloanbaatar, Catopsbaatar,and Bulganbaatar (Kielan-Jaworwoska andHurum, 1997), we were not able to unequiv-ocally identify a lacrimal in the specimens ofthese taxa housed in Warsaw. According toKielan-Jaworowska (personal commun.), thetypical djadochtatherian facial process is pre-sent on the Catopsbaatar specimen housedin Moscow (PIN 4537-5).

Among Taeniolabididae, Kielan-Jawo-rowska and Hurum (1997) reported a small,arcuate facial process in Lambdopsalis andTaeniolabis. However, the lacrimal is whollylacking in Lambdopsalis (Miao, 1988), andif one was present in Taeniolabis, it had nofacial exposure (Broom, 1914).

Among Ptilodontoidea, for the observationof a small facial process of the lacrimal inPtilodus, Kielan-Jaworowska and Hurum(1997) cited Simpson (1937). Although Gid-ley (1909) originally identified a lacrimal inP. gracilis, Simpson (1937: 740), in reana-lyzing the same specimen and other speci-mens of P. montanus, stated that ‘‘the ante-rior orbital rim is not perfectly preserved inany case and certainty is impossible, butthere is no suggestion of facial exposure ofthe lacrimal and the maxilla probably formsthis rim.’’ In our observations of the speci-mens of P. montanus and P. gracilis de-scribed by Simpson (1937), we were unableto either deny or confirm the presence of alacrimal. This is also the case in the skull ofthe ptilodontoid Ectypodus tardus (YPM-PU14724), which was originally described bySloan (1979) as having no lacrimal.

Among Paulchoffatiidae, a small, arcuatefacial process contacting the posteriorly ex-panded nasals has been reported for Pseu-dobolodon krebsi (Hahn, 1969; Hahn andHahn, 1994), but we have also seen it in P.oreas (V.J. 460-155), Kuehneodon dryas(V.J. 454-155), and Meketichoffatia krausei(V.J. 446-155).

ORBITOTEMPORAL REGIONORBITAL MOSAIC

The sutural pattern of the bony elementscontributing to the orbital mosaic has been


reported for fewer than a handful of multi-tuberculates. The two with the most evidentsutures are Kryptobaatar (this report; figs.18, 19, 33) and Lambdopsalis (Miao, 1988).In both taxa, the palatine is wholly excludedfrom the orbit, and the major contributing el-ements are the frontal and the maxilla, withthe latter forming the anterior part of the or-bital roof and extending posteriorly throughthe orbit to floor the sphenorbital recess.However, the proportions of several elementsare strikingly different between the two taxa.In Lambdopsalis, there is no lacrimal; the or-bitosphenoid has a restricted orbital expo-sure; and the alisphenoid is expanded dor-sally, contacting the frontal and parietal andexcluding the anterior lamina from the orbit-al mosaic. In contrast, in Kryptobaatar (figs.18, 19, 33), there is a lacrimal with a smallorbital process, the orbitosphenoid has amore extensive orbital exposure; the anteriorlamina contacts the frontal, and the alisphe-noid is confined to the sphenorbital recess.

How widespread these patterns are amongmultituberculates is not entirely clear be-cause of preservational problems. A lacrimalwas primitively present in multituberculates,given its distribution reported above in paul-choffatiids and Mongolian Late Cretaceoustaxa, but the extent of the orbital contributionis uncertain. A maxilla resembling that inKryptobaatar and Lambdopsalis is also re-ported in Chulsanbaatar and Nemegtbaatar(Hurum, 1994, 1998a), but the maxilla doesnot form a bony roof over the anterior orbitalspace in the paulchoffatiids Pseudobolodon(V.J. 447-155, 451-155, 460-155) and Me-ketichoffatia (V.J. 446-155) or in the ptilo-dontoids Ectypodus (Sloan, 1979) and Pti-lodus (Simpson, 1937). The orbital exposureof the orbitosphenoid in Pseudobolodon(originally described as Kuehneodon inHahn, 1977), Kamptobaatar (Kielan-Jawo-rowska, 1971), Chulsanbaatar, and Nemegt-baatar (Hurum, 1994, 1998a) appears to bemore substantial than that in Kryptobaatar.Finally, as in Kryptobaatar, the alisphenoidis confined to the sphenorbital recess inKamptobaatar and Nemegtbaatar (see be-low), and the anterior lamina sends a processforward that contacts the frontal high in theorbit in Nemegtbaatar and Chulsanbaatar(Hurum, 1998a).

The primitive condition of the palatine formultituberculates is still controversial, butMiao’s (1988, 1993) suggestion that the ab-sence of an orbital process may be primitiveis still viable. In addition to Kryptobaatar,an orbital process of the palatine is lackingin Catopsbaatar (ZPAL MgM-I/78) andKamptobaatar (ZPAL MgM-I/33) (contraKielan-Jaworowska, 1971, and Hurum,1994), but it is said to be present in a sec-tioned skull of Nemegtbaatar (Hurum, 1994,1998a) and is illustrated in Ectypodus (Sloan,1979: fig. 1). We have not had the opportu-nity to study the specimen of Nemegtbaatardescribed by Hurum (1994, 1998a); however,we were not able to identify sutures distin-guishing any orbital elements in Ectypodustardus (YPM-PU 14724). If Hurum’s (1994,1998a) interpretations of the orbital suturesin Nemegtbaatar are correct, then the Mon-golian Late Cretaceous taxa would be poly-morphic regarding the presence of the pala-tine in the orbit. Because an orbital processhas been reported in Nemegtbaatar and Ecty-podus, we carefully scrutinized the conditionin Kryptobaatar. On the left side of PSS-MAE 101 (fig. 12) and possibly also 113,there is a line along the ventral edge of thepart of the orbit bulging over the frontal lobethat we interpret as a fracture within the or-bitosphenoid. The bone anteroventral to thisline is the only orbital element that conceiv-ably could be palatine, but the high-resolu-tion CT sections of PSS-MAE 101 do notsupport this interpretation. Interestingly, thisis not the position the palatine is reported toexhibit in Nemegtbaatar, which is in the an-teroventral corner of the orbit (Hurum,1998a). In Kryptobaatar, the correspondingpart of the orbit is most assuredly maxilla.

One feature of the orbit that clearly varieswithin multituberculates is the orbital pocketof Gambaryan and Kielan-Jaworowska(1995), the depression in the orbital processof the frontal for the pars anterior of the me-dial masseter muscle. Among MongolianLate Cretaceous taxa, an orbital pocket is re-ported for Kamptobaatar (Kielan-Jaworows-ka, 1971), Catopsbaatar (Gambaryan andKielan-Jaworowska, 1995), Tombaatar(Rougier et al., 1997a), Nemegtbaatar, Chul-sanbaatar, Sloanbaatar (Kielan-Jaworowskaet al., 1986), and Kryptobaatar (figs. 10–14).


Among ptilodontoids, an orbital pocket is re-ported for Ectypodus by Sloan (1979), but ithas not been noted in Ptilodus (Simpson,1937; Wall and Krause, 1992). Among tae-niolabidids, Sloan (1981) noted that the largeorbital pocket in Taeniolabis taoensis hadbeen misidentified as the orbit proper byBroom (1914) and Granger and Simpson(1929). In Lambdopsalis, an orbital pocket isnot apparent in the reconstructions in Miao(1988), but a reduced one was said to be pre-sent in a personal communication by that au-thor to Gambaryan and Kielan-Jaworowska(1995: 65) dated December 1994. Finally, anorbital pocket is absent in the paulchoffatiidsMeketichoffatia (V.J. 443-155) and Pseudo-bolodon (V.J. 447-155, 451-155, 460-155).

ORBITAL FORAMINA

Given that few orbits are known in detailfor multituberculates, it is not surprising thatinformation concerning orbital foramina isalso limited. The reported distributions ofvarious orbital foramina in multituberculatesare treated below.

Lacrimal Foramen: A lacrimal foramenhas been noted thus far for only the paul-choffatiids Meketichoffatia (originally de-scribed as Paulchoffatia by Hahn, 1969) andPseudobolodon (Hahn and Hahn, 1994),Lambdopsalis (1988), Nemegtbaatar (Hu-rum, 1994), and Kryptobaatar (‘‘lacf’’ in fig.16). In the mammaliaforms Morganucodon(Kermack et al., 1981), Haldanodon (Lille-graven and Krusat, 1991), and Vincelestes(Rougier, 1993), there are two lacrimal fo-ramina in the orbital process of the lacrimal,although Sinoconodon has only one (Cromp-ton and Luo, 1993). The paulchoffatiids havea single opening in the ventral part of theorbital process (Hahn and Hahn, 1994),whereas in Nemegtbaatar (Hurum, 1994) andKryptobaatar the foramen is in the ventralpart of the lacrimal’s contribution to the or-bital rim. Lambdopsalis, in which the lacri-mal is absent, has a single lacrimal foramenin the orbital process of the maxilla (Miao,1988). The lacrimal is also lacking in mono-tremes, and their lacrimal foramen is be-tween the maxilla and frontal (Kuhn, 1971;Zeller, 1989). As reported above, the pres-ence of the lacrimal is uncertain in Ptilodus;

however, we found a lacrimal foramen in theorbital process of the isolated maxillae of P.montanus (USNM 9735).

Sphenopalatine Foramen: A sphenopala-tine foramen in the anteroventral portion ofthe orbital wall is widely distributed amongmultituberculates. What may differ amongtaxa are the enclosing bony elements. Ac-cording to Kielan-Jaworowska et al. (1986),the foramen is at the junction of the maxilla,frontal, orbitosphenoid, and palatine in tae-niolabidoids, based presumably on Kampto-baatar (Kielan-Jaworowska, 1971) and Ne-megtbaatar (see also Hurum, 1994), whereasit lies in or near the frontomaxillary suturein Lambdopsalis (Miao, 1988) and Krypto-baatar (‘‘spf’’ in figs. 10, 12, 18, 19), thetwo forms whose orbital mosaics are knownbest. A true characterization of the bordersof the sphenopalatine foramen among mul-tituberculates awaits better preserved orbits.

Foramen for Frontal Diploic Vein: An ap-erture in the frontal near the dorsal orbitalrim resembling that described here inKryptobaatar for the frontal diploic vein(‘‘fdv’’ in figs. 18, 36A) is also found in Ca-topsbaatar (ZPAL MgM-I/78), Chulsanbaa-tar (ZPAL MgM-I/168), Ptilodus (AMNH35490, USNM 6076), and the paulchoffatiidsKuehneodon (V.J. 454-155), Meketichoffatia(V.J. 446-155), and Pseudobolodon (V.J.447-155, 451-155, 458-155, 460-155). Miao(1988) did not report one in Lambdopsalis,and we did not see any in the specimens ofKamptobaatar and Nemegtbaatar availablefor study. Among primitive mammaliaforms,there are two foramina in the frontal in Mor-ganucodon (Kermack et al., 1981) and onein Haldanodon (Lillegraven and Krusat,1991) that may have transmitted the frontaldiploic vein.

Anterior Opening of Orbitotemporal Ca-nal: Kamptobaatar (ZPAL MgM-I/33) andNemegtbaatar (Kielan-Jaworowska et al.,1986) have a foramen below the postorbitalprocess at the anterior end of the orbitotem-poral canal, which transmitted the orbitotem-poral vessels, resembling that reported herein Kryptobaatar. According to Kielan-Ja-worowska et al. (1986), this aperture (theirpostorbital foramen) is in the parietal in Ne-megtbaatar; it is between the parietal, fron-tal, and anterior lamina in Kryptobaatar


(‘‘otc’’ in fig. 12). Lambdopsalis (Miao,1988) also has an opening at the anterior endof the orbitotemporal canal, but it lies wellposterior to the reduced postorbital process,between the parietal, frontal, and alisphe-noid. Hahn and Hahn (1994) reported a post-orbital foramen in Pseudobolodon krebsi, butthe relevant specimen (V.J. 451-155) showsno communication between this aperture andan orbitotemporal canal. If this is a vascularforamen, it served a nutritive function only.The only basal mammaliamorph in which ananterior opening of the orbitotemporal canalhas been reported is Adelobasileus, in whichthe aperture is positioned in the parietal wellposterior to the orbit (Lucas and Luo, 1993).In addition, although the foramen has notbeen described, orbitotemporal vessels enter-ing the orbit have been restored for Morga-nucodon (Kermack et al., 1981; Rougier etal., 1992).

Ethmoidal Foramen: Simmons (1993) em-ployed the presence/absence of an ethmoidalforamen in her phylogenetic analysis of Mul-tituberculata, noting its absence in only twotaxa, the ptilodontoids Ptilodus and Ectypo-dus. This result is surprising, given that anethmoidal foramen to our knowledge is uni-versally present among living mammals, rep-resenting the only pathway for branches ofthe ophthalmic nerve to reach the nasal cav-ity from the orbit. However, our inspectionof the relevant specimens reveals an eth-moidal foramen to be present in both P. mon-tanus (AMNH 35490; see also Simpson,1937) and E. tardus (YPM-PU 14724).Therefore, an ethmoidal foramen is nowknown for all multituberculates preservingthe relevant portion of the orbit, includingKryptobaatar (‘‘ef’’ in figs. 12, 18, 19, 36A).

Optic Foramen: Lambdopsalis (Miao,1988), Kamptobaatar (ZPAL MgM-I/33),and Kryptobaatar (‘‘opf’’ in figs. 10, 12,36A) have a foramen in the orbitosphenoidthat, given its position, likely transmitted theoptic nerve. An optic foramen has been re-ported for two other multituberculates, butthese do not resemble the openings in thethree taxa named above. The supposed opticforamen in Meketichoffatia (originally de-scribed as Pseudobolodon by Hahn, 1981) istoo small and too far forward, whereas thatin Nemegtbaatar (Kielan-Jaworowska et al.,

1986) connects to the upper part of the ca-vum epiptericum rather than to the orbit. Be-yond excluding these openings as optic fo-ramina, we are uncertain of their function.Among cynodonts, an optic foramen occursin Probainognathus (MCZ 4274), Sinoco-nodon (Crompton and Luo, 1993), Adeloba-sileus (Lucas and Luo, 1993), and eutherians(Novacek et al., 1997), and was likely morewidespread among multituberculates andmammaliaforms than is currently known.

Metoptic Foramen: Among mammalia-forms, a metoptic foramen has thus far beenidentified in only five taxa, all of them mul-tituberculates. In Kryptobaatar, the metopticforamen is anteromedial to the sphenorbitalfissure (‘‘mef’’ in fig. 36A), and on the en-docranial surface is anterolateral to the hy-pophyseal fossa, with a deep sulcus on theossified pila antotica running into the fora-men from behind (‘‘mef’’ in figs. 25–27). InLambdopsalis (Miao, 1988), the metoptic fo-ramen is also anterolateral to the hypophy-seal fossa, but is posterior to the sphenorbitalfissure. The metoptic foramen in Kampto-baatar (ZPAL MgM-I/33) is only knownfrom the orbital surface; as in Kryptobaatar,it is anteromedial to the sphenorbital fissure.Finally, the metoptic foramen in Nemegtbaa-tar and Chulsanbaatar has only been de-scribed from the endocranial surface, and itlies posterolateral to the hypophyseal fossa(Hurum, 1998a). Miao (1988) also suggestedthat tiny pores in the anterior floor of thehypophyseal fossa reported in Meketichoffa-tia (originally described as Pseudobolodonby Hahn, 1981) may represent metoptic fo-ramina. In our study of this specimen (V.J.443-155), we think these pores are artifac-tual. A metoptic foramen is a primitive am-niote feature that was likely more widespreadamong multituberculates and other mamma-liaforms than is currently known, althoughthe ossification of the pila metoptica or lackthereof may be a variable feature.

Sphenorbital Fissure: The opening of thecavum epiptericum into the orbit, the sphe-norbital fissure, reported for Lambdopsalis(Miao, 1988), Ptilodus (Simpson, 1937),Ectypodus (Sloan, 1979), and other Mongo-lian Late Cretaceous multituberculates (Kie-lan-Jaworowska et al., 1986), resembles thatdescribed here for Kryptobaatar (‘‘sphf’’ in


fig. 36A). It is not directly visible in lateralview, but is concealed behind the alisphenoidand/or anterior lamina (figs. 10, 12, 36A).One potential difference is the occupants ofthe sphenorbital fissure. As mentioned above,Lambdopsalis (Miao, 1988) and Kryptobaa-tar have separate optic and metoptic fora-mina; if these apertures are truly lacking inother multituberculates, the optic and oculo-motor nerves and the ophthalmic arteryprobably left the cranial cavity through thecavum epiptericum via the sphenorbital fis-sure.

Miao (1988) noted that the multitubercu-late sphenorbital fissure is unusual in that itis concealed in lateral view. This differs fromthe condition in basal mammaliaforms, suchas Sinoconodon (Crompton and Luo, 1993)and Haldanodon (Lillegraven and Krusat,1991), and in modern mammals in which theforamen is generally visible from the side.Miao (1988: 54) speculated that ‘‘this is dueto the retention of a medially bony walledcavum epiptericum in multituberculates,among early mammals, which prevents theorbitosphenoid from articulating with the ali-sphenoid in most parts.’’ As an alternativeexplanation, this concealment may resultfrom the apparent foreshortening of the tem-poral region in the multituberculates forwhich the position of the sphenorbital fissureis known.

POSTORBITAL PROCESS

The postorbital process on the frontal boneis a widespread feature of the mammalianskull (Novacek, 1986). Presumably, as in thedog (Evans and Christensen, 1979), it servesas an attachment for the orbital ligament, de-limiting the back of the orbit, and marks theanterior limit of the temporalis muscle at-tachment.

In multituberculates, the postorbital pro-cess has been said by Miao (1988, 1993) tobe reduced or absent. Gambaryan and Kie-lan-Jaworowska (1995) pointed out, contraMiao (1988, 1993), that a parietal postorbitalprocess is characteristic of taeniolabidoids. Inour preliminary comparisons among multi-tuberculates, we have identified three condi-tions for the postorbital process: on the fron-tal and inconspicuous, on the parietal and

short, and on the parietal and long. An in-conspicuous postorbital process on the fron-tal is found in paulchoffatiids (Meketichof-fatia krausei, V.J. 446-155; Pseudobolodonoreas, V.J. 460-155; Pseudobolodon krebsi,V.J. 451-155), Ptilodus (Simpson, 1937), andEctypodus tardus (YPM-PU 14724). In con-trast, in those Mongolian Late Cretaceoustaxa preserving the postorbital process, it ison the parietal. It is short in Chulsanbaatar,Kamptobaatar, and Nemegtbaatar (Kielan-Jaworowska and Hurum, 1997) and is longin Catopsbaatar (Kielan-Jaworowska andSloan, 1979) and Kryptobaatar. Kielan-Ja-worowska and Hurum (1997) characterizedthe postorbital process of Kryptobaatar asshort, but PSS-MAE 113 had long ones priorto preparation damage (see reconstruction infig. 33). The only non-Mongolian multitu-berculate with a postorbital process on theparietal, and a short one at that, is Taenio-labis taoensis from the Paleocene of NorthAmerica (Broom, 1914). We are uncertain ofthe condition in Lambdopsalis, which hasbeen said to have a short but distinct post-orbital process on the frontal (Miao, 1988)and on the parietal (Meng, personal commun.cited in Gambaryan and Kielan-Jaworowska,1995).

Miao (1988) suggested that a true postor-bital process is lacking in Mongolian LateCretaceous multituberculates and that theprocess on the parietal in these forms doesnot delimit the back of the orbit. In support,he noted that the postorbital constriction ison the posterior side of the parietal process,whereas it is on the anterior side of a truepostorbital process on the frontal. Based onKryptobaatar, however, we think that the el-ement on the parietal represents a true post-orbital process that has shifted onto a differ-ent skull roof bone. In the dog (Evans andChristensen, 1979) and the platypus (Zeller,1989; personal obs.), anterior to the postor-bital process, are the foramen for the frontaldiploic vein and the passageway by whichthe supraorbital nerve and vessels leave theorbit. The same structures are found imme-diately in front of the process in question onthe parietal in Kryptobaatar (fig. 18). Addi-tionally, Miao (1988) has misidentified theposition of the postorbital constriction in theMongolian Late Cretaceous taxa; it does lie


posterior to the process on the parietal, in thevery foreshortened temporal fossa (fig. 32).Incidentally, the postorbital process is not in-variably on the frontal in extant mammals;in hyracoids, for example, both the frontaland the parietal contribute to the formationof this process (Fischer, 1986: fig. 27;MacPhee, 1994: fig. 6).

NASOPARIETAL CONTACT

One of the more unusual features of theskull roof above the orbit in the taeniolabi-dids Taeniolabis (Broom, 1914; Simpson,1937) and Lambdopsalis (Miao, 1988) is thelong anterior process of the parietal thatoverlies the frontal and extends forward tocontact the nasal. The absence/presence ofthis feature has been used in recent phylo-genetic analyses within Multituberculata bySimmons (1993), Rougier et al. (1997a), andKielan-Jaworowska and Hurum (1997). Werestudied the incidence of nasoparietal con-tact in paulchoffatiids, following a sugges-tion from J. A. Hopson, and offer the follow-ing amendments to the scoring of this fea-ture. Whereas nasoparietal contact has beenwidely held to be a unique feature of tae-niolabidids, we have also observed it in sev-eral paulchoffatiids: Kuehneodon dryas (V.J.454-155), Pseudobolodon krebsi (V.J. 447-155, 451-155), and Pseudobolodon, n. sp.(V.J. 400-155, 450-155). The condition inMeketichoffatia krausei (V.J. 445-155, 446-155) is uncertain.

JUGAL

Until recently, the jugal was widely con-sidered (e.g., Clemens and Kielan-Jaworows-ka, 1979; Hahn, 1983) to be wholly lackingin multituberculates. J. A. Hopson (personalcommun. cited in Kielan-Jaworowska et al.,1986) noted the presence of a jugal or a facetfor it on the medial side of the zygomaticarch in several previously described speci-mens. From Hopson’s observations, Hahn(1987) reported the presence of a jugal faceton the medial side of the zygomatic processof the maxilla in several paulchoffatiids,Kuehneodon dryas, Pseudobolodon oreas,and ?Pseudobolodon, sp. indet., and we haveobserved this facet in Meketichoffatia krau-sei (V.J. 446-155). Hopson et al. (1989) re-

ported displaced, broken jugals in Ptilodusmontanus, which contacted the medial sur-face of both the maxilla and the squamosalin the zygomatic arch and extended forwardinto the anteroventral part of the orbit. Theyalso reported an incomplete jugal in Ne-megtbaatar, which contacted both the max-illa and the squamosal, as well as small frag-ments of a jugal in Chulsanbaatar. The ju-gals described here in Kryptobaatar (PSS-MAE 101 and 113) are the first completeones reported for multituberculates (figs. 22,23). They conform with the jugals in Ptilo-dus montanus, except that they do not extendas far forward into the anteroventral part ofthe orbit.

Outside of multituberculates, monotremesare the only other mammaliaform group inwhich replacement of the jugal in the zygo-matic arch by an elongate maxillary processextending nearly to the squamosal glenoid islikely primitive (Hopson et al., 1989). How-ever, rather than internal to the maxilla andsquamosal as in multituberculates, the re-duced jugal is dorsal to them in the platypus(Zeller, 1989). The jugal is wholly absent inthe echidna (Kuhn, 1971).

BASICRANIUM AND LATERALBRAINCASE WALL

PTERYGOPALATINE RIDGES AND TROUGHS

In the mesocranium, Kryptobaatar haspaired medial and lateral pterygopalatinetroughs (‘‘mpt’’ and ‘‘lpt’’ in fig 37A) sepa-rated by the paired pterygopalatine ridges(‘‘ptr’’ in fig. 37A) and on the midline by thevomer (figs. 14, 15, 20, 21, 34, 37A). A sim-ilar arrangement is known for other Mon-golian Late Cretaceous multituberculates(e.g., Kamptobaatar, Kielan-Jaworowska,1971; Nemegtbaatar, Kielan-Jaworowska etal., 1986) and for Late Jurassic paulchoffa-tiids (e.g., Pseudobolodon oreas, Hahn,1981), except in the latter the lateral troughis significantly narrower and shallower thanthe medial. We have seen a similar arrange-ment in Ptilodus (USNM 6076). Medial andlateral pterygopalatine troughs are also foundin primitive mammaliaforms (e.g., Morgan-ucodon, Kermack et al., 1981; Sinoconodon,Crompton and Luo, 1993). In contrast, inLambdopsalis, pterygopalatine ridges are


lacking, and a single trough bordered by avery low, midline crest enters the choana(Miao, 1988).

Regarding the occupants of the medial andlateral troughs, the former held the air pas-sageway, whereas there are two hypothesesfor the occupants of the lateral in multitu-berculates. According to Barghusen (1986),the auditory (Eustachian) tube connecting thenasopharynx and the cavum tympani occu-pied the lateral trough. Other authors havereconstructed muscle there: tensor veli pala-tini by Kielan-Jaworowska (1971) and Hahn(1981), and medial pterygoid by Gambaryanand Kielan-Jaworowska (1995). We thinkthat the morphology in Kryptobaatar elimi-nates the auditory tube as a possible occu-pant; the pterygopalatine ridge extends farrostrally into the choana, an arrangement thatrequired the auditory tube to empty into thenasal cavity and not into the nasopharynx asit does in extant mammals (Starck, 1995).The more likely course for the auditory tubewas one posterior to the pterygopalatineridge to join the nasopharynx in the medialtrough. On the other hand, the morphologyof the lateral trough in Kryptobaatar, in par-ticular the fossa at its anterior end, appearsappropriate for muscle attachment. However,we are uncertain what the muscular occupantcould have been. The tensor veli palatiniarises more posteriorly in living therians (seeEvans and Christensen, 1979), and the me-dial pterygoid would seem to have been at amechanical disadvantage within the lateraltrough, with the medially inflected, lateralchoanal wall constraining the direction of themuscle’s pull. As a third hypothesis, we pro-pose that a parallel channel to the main airpassageway may have occupied the lateraltrough. Either such a channel and/or muscleseem to be the only viable options. Perhapsevaluating these options in light of the mor-phology of other mammaliaforms and mam-maliaform outgroups, which is beyond thescope of this report, may resolve this prob-lem.

The pterygopalatine ridges in Kryptobaa-tar and Kamptobaatar (ZPAL MgM-I/33)end as a rounded, posteroventrally directedprocess that is reminiscent of the pterygoidhamulus of living therians. Kielan-Jawo-rowska (1971) suggested that it served a sim-

ilar function, directing the pull of the tendonof the tensor veli palatini. However, shedoubted its homologies with the therian ham-ulus because the two are formed from dif-ferent parts of the pterygoid. We agree withKielan-Jaworowska (1971), noting that thetherian hamulus is formed from the postero-lateral part of the pterygoid, whereas the pro-cess in Kryptobaatar and Kamptobaatar isoff the posteromedial part of the bone.

ECTOPTERYGOID

A separate ectopterygoid bone in the pos-terolateral corner of the palate has been de-scribed for one specimen of Morganucodonoehleri (Kermack et al., 1981) and in the me-socranium of various non-mammalian cyno-donts (e.g., Thrinaxodon, Fourie, 1974). Anectopterygoid has also been said (e.g., Par-rington and Westoll, 1940; Presley and Steel,1978) to be present in monotremes, but thehomologies of this element have been ques-tioned (Wible, 1991). At one time, an ecto-pterygoid was also thought to be present inthe lateral wall of the choana in certain Mon-golian Late Cretaceous multituberculates(Kielan-Jaworowska, 1971; Clemens andKielan-Jaworowska, 1979), although the su-ture delimiting it from the alisphenoid behindwas said to be indistinct (Kielan-Jaworowskaet al., 1986). A separate ectopterygoid hasalso been reconstructed in the early Eoceneptilodontoid Ectypodus (Sloan, 1979). Miao(1988, 1993) suggested that the purportedmultituberculate ectopterygoid may merelybe part of the alisphenoid, which is how itwas restored in Chulsanbaatar by Rougier etal. (1992: fig. 8C) and in subsequent recon-structions by Kielan-Jaworowska and co-workers (e.g., Gambaryan and Kielan-Jawo-rowska, 1995; Kielan-Jaworowska and Hu-rum, 1997; Hurum, 1998a). This view is sup-ported by the morphology of the alisphenoidin Kryptobaatar and our observations of themultituberculates in the ZPAL collections.

Miao (1988) also suggested that althoughthe ectopterygoid was not present as a sep-arate element in multituberculates, it wasfused with the pterygoid to form the equiv-alent of the therian hamulus. The basis forthis hypothesis is that the ventrolateral por-tion of the therian pterygoid, which forms


from cartilage separate from the intramem-branous dorsomedial portion of the ptery-goid, is homologous with the ectopterygoidbone of other amniotes (Presley and Steel,1978). This is not the only view on the ho-mologies of these elements; Kuhn (1971) andZeller (1989) treated the pterygoid cartilageas a derivative of the palatoquadrate. Thiscontroversy is beyond the scope of our re-port, but we have stated above that a hamulusas occurs in therians is not present in Kryp-tobaatar or other multituberculates.

ALISPHENOID

The alisphenoid in multituberculates hasbeen said (e.g., Kermack and Kielan-Jawo-rowska, 1971; Kielan-Jaworowska et al.,1986) to be a small element with little con-tribution to the side wall of the braincase.This is in contrast to the usual condition inmammaliaforms in which the alisphenoid isexpanded dorsally and contacts the frontal(Novacek, 1986; Wible and Hopson, 1993;Hopson and Rougier, 1993). Miao (1988,1993) has questioned the purported smallsize of the alisphenoid in multituberculates,noting that there are few specimens that un-ambiguously detail the extent of the alisphe-noid in the braincase wall; one is Lambdop-salis, which has an alisphenoid extendingdorsally to meet the parietal in the skull roof,according to Miao (but see below). In addi-tion, Miao (1993: 69) included under his cat-egory of multituberculate apomorphies an‘‘elastic ossification pattern in the orbitotem-poral region as reflected by the variability inextent of the alisphenoid and the so-called‘anterior lamina of the petrosal’.’’

A review of recent reconstructions of themultituberculate braincase would seem tosupport Miao’s (1993) observation of ali-sphenoid variability. At one extreme isLambdopsalis, in which the alisphenoid issaid to extend to the skull roof (Miao, 1988);at the other extreme is Kryptobaatar, inwhich the alisphenoid has very little contri-bution to the braincase (fig. 10). Intermediatebetween these two extremes, according toHurum (1998a), is the alisphenoid in Ne-megtbaatar, with Chulsanbaatar showing anelement intermediate between Nemegtbaatarand Kryptobaatar. In Kryptobaatar, the su-

tures delimiting the alisphenoid from the an-terior lamina are very distinct and showwithout doubt that the alisphenoid is a smallelement. Moreover, the suture between thealisphenoid and anterior lamina in Krypto-baatar closely resembles that tentativelyidentified for Kamptobaatar by Kielan-Ja-worowska (1971), with which we concurbased on ZPAL MgM/I-33, and for Nemegt-baatar by Kielan-Jaworowska et al. (1986).Hurum (1998a) has provided a new recon-struction for Nemegtbaatar based on a sec-tioned specimen in which the contribution ofthe alisphenoid to the braincase is more sub-stantial than that reconstructed by Kielan-Ja-worowska et al. (1986). Because we have notseen the sectioned skull of Nemegtbaatar (orChulsanbaatar), we are not able to evaluateHurum’s reconstruction. Regarding Lamb-dopsalis, we have not had access to the entirecollection of specimens studied by Miao(1988), but we identified a possible suture inthe anterior portion of the epitympanic recessof IVPP V7151.80 (see Miao and Lillegrav-en, 1986; Rougier et al., 1996c). This pos-sible suture is in the same place that the ali-sphenoid contacts the petrosal in Kryptobaa-tar, suggesting that Lambdopsalis also mighthave a small alisphenoid. This single obser-vation does not refute Miao’s reconstructionfor Lambdopsalis, but it does raise doubtsabout it, in particular in light of the frag-mentary nature of the material. In summary,we are skeptical about the extreme variabilityreported for the multituberculate alisphenoid,and we think that Miao’s (1993) inclusion ofan elastic ossification pattern in the orbito-temporal region as a multituberculate apo-morphy seems unwarranted.

Kielan-Jaworowska et al. (1986) claimedthat the alisphenoid in multituberculates isnot pierced by any foramina. However, Miao(1988) described in Lambdopsalis a foramenovale inferium and foramen masticatoriumwholly within the alisphenoid and an ali-sphenoid canal between the alisphenoid andanterior lamina. Additionally, we report herefor Kryptobaatar and various other Mongo-lian Late Cretaceous forms, a buccinator fo-ramen between the alisphenoid and anteriorlamina (see below; ‘‘fbu’’ in figs. 10, 36A).


FORAMINA FOR BRANCHES OF THE

MANDIBULAR NERVE

Most Mesozoic mammaliaforms, includ-ing multituberculates, have two foramina inthe anterior lamina and/or petrosal that areinterpreted for branches of the trigeminalnerve (Wible and Hopson, 1993; Luo, 1994).Recently, we (Rougier et al., 1996a) haveraised doubts about which branches of thetrigeminal nerve occupied these various ap-ertures. To date, two models of nerve resto-ration have been proposed. First, Simpson(1937) restored branches of the mandibularnerve in the two foramina in Ptilodus mon-tanus, citing Hill’s (1935) observation of twoforamina for the mandibular nerve in the ali-sphenoid of some extant rodents. Second,Patterson and Olson (1961) restored the max-illary and mandibular nerves respectively inthe two foramina in the anterior lamina ofSinoconodon. This model has been followedby Kermack (1963, 1967) in reconstructingnerves in Morganucodon and by most sub-sequent authors considering Mesozoic mam-maliaforms other than multituberculates. Incontrast, authors researching multitubercu-lates (e.g., Kielan-Jaworowska, 1971; Sloan,1979) invariably have followed the modelproposed by Simpson (1937). In describingKryptobaatar, we too have followed Simp-son’s model. However, we have accepted itbecause the pattern of the foramina in Kryp-tobaatar fits the arrangement of the mandib-ular nerve foramina in some living rodents(Hill, 1935; Wahlert, 1974). Specifically, thetwo foramina in Kryptobaatar are roughly atthe same level, with the medial one ante-roventrally directed and the lateral one ven-trolaterally directed (‘‘foi’’ and ‘‘fma’’ in fig.37A); this does not correspond to the posi-tions of the maxillary and mandibular nerveexits in any extant mammals. As detailed be-low, there is some variation in the positionof the two foramina in question in other mul-tituberculates; however, the variation is notextreme enough in any taxon to warrant analternative nerve restoration. The implica-tions of this for non-multituberculate mam-maliaforms are beyond the scope of this re-port, and what is required is careful compar-ison of the relevant extinct taxa with themorphology of extant mammals and the es-

tablishment of morphological criteria to rec-ognize the different branches of the trigem-inal system based on osteology.

All multituberculates known to date withone exception have at least two apertures inthe anterior lamina and/or petrosal that havebeen interpreted as for branches of the man-dibular division of the trigeminal nerve: theforamen ovale inferium and the foramenmasticatorium. The exception is Lambdop-salis in which the two foramina are said tobe in the alisphenoid (Miao, 1988). In ourpreliminary comparisons, we have noted thatthe positions of the foramina in the petrosalvary with regard to each other and to theepitympanic recess. In Late Jurassic paul-choffatiids (e.g., Kuehneodon dryas, V.J.454-155; Pseudobolodon, n. sp., V.J. 450-155), the foramen ovale inferium is at a levelposterior to the foramen masticatorium, andboth foramina are lateral to the epitympanicrecess. In ptilodontoids (e.g., Ptilodus mon-tanus, AMNH 35490; Mesodma thompsoni,Wible and Hopson, 1995), the two foraminaare at the same level and the foramen ovaleinferium traverses the epitympanic recess.Among Mongolian Late Cretaceous multitu-berculates, the foramen ovale inferium tra-verses the epitympanic recess in Sloanbaatar(ZPAL MgM-I/20), whereas both foraminatraverse it in Kryptobaatar (fig. 37A), Kamp-tobaatar (ZPAL MgM-I/33), Nemegtbaatar(ZPAL MgM-I/81), Chulsanbaatar (ZPALMgM-I/168), and Catopsbaatar (ZPALMgM-I/78). The foramina are roughly at thesame level in Kryptobaatar, Kamptobaatar,and Nemegtbaatar, but the foramen ovale in-ferium is more posterior in Sloanbaatar,Chulsanbaatar, and Catopsbaatar. Finally,in Lambdopsalis, the two foramina are at thesame level, lateral to the epitympanic recess,but supposedly enclosed in the alisphenoid(Miao, 1988). Another variant is found inKamptobaatar (ZPAL MgM-I/33) in whichthere are more than two foramina in the epi-tympanic recess (five on the right side andfour on the left) that likely held branches ofthe mandibular nerve (Kielan-Jaworowska,1971).

We have identified in Kryptobaatar anoth-er foramen in the pterygoid fossa of the ali-sphenoid between that bone and the anteriorlamina as for the buccal branch of the man-


dibular nerve (‘‘fbu’’ in figs. 10, 36A). Thisforamen is not unique to Kryptobaatar, butalso occurs in Catopsbaatar (ZPAL MgM-I/78), Chulsanbaatar (ZPAL MgM-I/168),Kamptobaatar (ZPAL MgM-I/33), Nemegt-baatar (ZPAL MgM-I/81), and Sloanbaatar(ZPAL MgM-I/20). Kielan-Jaworowska et al.(1986) described for Chulsanbaatar (ZPALMgM-I/168; see also Hurum, 1998a: ZPALMgM-I/84, the sectioned skull) and Nemegt-baatar (ZPAL MgM-I/82 and 76, the sec-tioned skull) another more dorsally situatedforamen wholly within the anterior lamina,as possibly for the deep temporal branch ofthe mandibular nerve. We have not yet stud-ied the sectioned skulls, and we were notable to confirm the presence of this dorsalforamen of the anterior lamina in the othercited specimens.

VENOUS SYSTEM

Our reconstruction of the venous systemhas considered only a few vessels. We treatthese below, with the exception of the frontaldiploic vein, which has already been dis-cussed above (see Orbital Foramina).

Superior Sagittal and Transverse Sinuses:A superior sagittal sinus connecting withpaired transverse sinuses is nearly ubiquitousamong extant mammals (Gelderen, 1924),and the multituberculates for which the rel-evant anatomy is known are no exception.Endocasts of Kryptobaatar, Ptilodus (Simp-son, 1937; Krause and Kielan-Jaworowska,1993), Lambdopsalis (Miao, 1988), Nemegt-baatar, Chulsanbaatar (Kielan-Jaworowskaet al., 1986), and an undescribed new speciesof Mongolian Late Cretaceous multituber-culate (PSS-MAE 126) show evidence thatthis dural sinus pattern was present.

Sigmoid Sinus: In Kryptobaatar, the sig-moid sinus does not exit the skull via thesmall jugular foramen, but likely left throughthe foramen magnum in the vertebral veins(‘‘vv’’ in fig. 37B). The same arrangementhas been reported for Chulsanbaatar and Ne-megtbaatar by Kielan-Jaworowska et al.(1986) based on clear evidence from endo-cranial casts (contra Miao, 1988). A foramenmagnum exit for the sigmoid sinus is not theuniversal pattern in multituberculates.Grooves indicate that the sigmoid sinus ex-

ited through the jugular foramen in Lamb-dopsalis (Miao, 1988) and through both thejugular foramen and foramen magnum inpaulchoffatiids (Lillegraven and Hahn,1993).

Miao (1988) questioned the taxonomicsignificance of differences in the sigmoid si-nus’s egress in multituberculates, becausethis vessel exhibits considerable diversityamong extant and extinct mammals. How-ever, at least some of this diversity falls alongtaxonomic lines; for example, the principalexit of the sigmoid sinus in all extant mar-supials is via the foramen magnum (Archer,1976; Wible, 1990). Consequently, this char-acter should not be eliminated a priori fromphylogenetic analyses. In fact, Rougier et al.(1996a, 1996c) have used the egress of thesigmoid sinus in their analyses of mammal-iaform interrelationships. They observed thatin addition to paulchoffatiids, the sulcus forthe sigmoid sinus extends to the jugular fo-ramen in Morganucodon and Haldanodon.In contrast, in the other taxa for which thesigmoid sulcus is known, it does not reachthe jugular foramen.

Inferior Petrosal Sinus: Among extantmammals, the inferior petrosal sinus followsa course along the basioccipital-petrosal su-ture that may be largely either endocranial,intramural, or extracranial (Wible, 1983). Wereconstruct the inferior petrosal sinus in anendocranial course in Kryptobaatar in partbecause there is a shallow sulcus in the ap-propriate location in PSS-MAE 123 (fig. 25)and in part because there is no bony pas-sageway appropriate for an extracranial orintramural route. Moreover, given the smallsize of the jugular foramen (figs. 26, 37A),it seems likely that this vessel exited the cra-nial cavity via the foramen magnum in thevertebral veins (‘‘vv’’ in fig. 37B). We be-lieve that Nemegtbaatar and Chulsanbaatarlikely exhibited the same pattern; there is nolikely route other than an endocranial one,and the jugular foramen in these forms isalso small (Kielan-Jaworowska et al., 1986).The inferior petrosal sinus was not recon-structed for Lambdopsalis by Miao (1988),but we think it likely occupied a deep sulcuson the endocranial surface between the pe-trosal and the basioccipital (Miao, 1988: fig.23). In contrast to the Late Cretaceous taxa,


the jugular foramen in Lambdopsalis is large(Miao, 1988) and may have transmitted theinferior petrosal sinus, as occurs in most ex-tant therians (Wible, 1983). As noted byRougier et al. (1996a), canals in the petrosalof some paulchoffatiids indicate that the in-ferior petrosal sinus had an intrapetrosalcourse to the jugular foramen as occurs in allmammaliaforms with the exception of extantmammals. We have also observed such ca-nals in Ptilodus montanus (AMNH 35490).

The small size of the jugular foramen inKryptobaatar, Nemegtbaatar, and Chulsan-baatar suggests an unusual pattern for thedrainage of the dural sinuses. As statedabove, it is likely that neither the inferior pe-trosal sinus nor the sigmoid sinus exitedthrough the jugular foramen. Consequently,the lateral head vein was the major constit-uent of the internal jugular vein. The onlyextant mammal exhibiting this same patternis the echidna (Hochstetter, 1896). The platy-pus has a minor variant in that a part of theinferior petrosal sinus drains into the internaljugular vein through the enormous metoticfissure, which is formed by the confluent jug-ular and hypoglossal foramina (fig. 28), andpart is into the vertebral veins via the fora-men magnum (Rougier et al., 1992). In con-trast, in marsupials and placentals, the majorcontributor to the internal jugular vein (theinferior petrosal and/or sigmoid sinuses) ex-its the cranial cavity via the jugular foramen.

Lateral Head Vein and Prootic Sinus: Awell-developed prootic canal that transmittedthe prootic sinus to the lateral head vein isknown for all multituberculates preservingthe relevant portion of the petrosal bone.Several characters of this vessel have beenemployed in phylogenetic analyses amongmammaliaforms: most recently, the positionof the cranial and tympanic apertures of theprootic canal (Rougier et al., 1996a, 1996c).In multituberculates, the cranial aperture issaid by Rougier et al. (1996a, 1996c) to lieat the anterodorsal margin of the subarcuatefossa, whereas the tympanic aperture is con-fluent with the pterygoparoccipital foramen(ventral ascending canal). The prootic canalin Kryptobaatar conforms with these obser-vations (figs. 27, 37A). However, we addhere that in some paulchoffatiids (V.J. 73-155, 82-155) the cranial aperture of the pro-

otic canal lies between the cavum epipteri-cum and the subarcuate fossa (see also Lil-legraven and Hahn, 1993: figs. 1, 6). Also,the tympanic aperture is not confluent withthe pterygoparoccipital foramen in Chulsan-baatar (ZPAL MgM-I/168) and in an isolat-ed petrosal tentatively assigned to Menis-coessus (Luo, 1989; Wible and Hopson,1995). Additionally, the condition in paul-choffatiids and Nemegtbaatar is uncertain.

Post-trigeminal Vein: In multituberculates,the vascular canal in the petrosal above theepitympanic recess and lateral flange hasbeen called variously the canal for the ?max-illary artery (Kielan-Jaworowska et al.,1986), the post-trigeminal canal (Rougier etal., 1996a, 1996c), and the canal for the ra-mus inferior (Miao, 1988; Wible and Hop-son, 1995). In ‘‘Catopsalis’’ joyneri and Me-sodma thompsoni, the canal is wide enoughthat it likely had two occupants, the ramusinferior and the post-trigeminal vein (Wibleand Hopson, 1995). We are confident that aramus inferior occupied the canal in questionin Kryptobaatar, given the size of the groovefor the stapedial artery and the ventral as-cending canal; however, we cannot excludethe presence of a post-trigeminal vein as well(‘‘ptv’’ in figs. 36B, 37B). A canal for theramus inferior is also present in Lambdop-salis, but the incidence of this structureamong other multituberculates is uncertaindue to preservational problems. The onlyother mammaliaform with a post-trigeminalvein enclosed in a canal is the echidna (Wi-ble and Hopson, 1995).

Pituito-orbital Vein: In Kryptobaatar, wehave identified a separate foramen in the an-terolateral wall of the hypophyseal fossa forthe pituito-orbital vein (‘‘fpv’’ in figs. 25, 27,36A). Miao (1988), in his description ofLambdopsalis, is the only other author to dis-cuss this vein (his pituitary vein) in Meso-zoic mammals. He concluded that this veinshared the metoptic foramen with the ocu-lomotor nerve, based on the co-occurrence ofthese structures in the chondrocranium ofsome living sauropsids (e.g., Lacerta, DeBeer, 1937). This restoration was the onlylogical one for Miao, because a separate fo-ramen for the vein, as occurs in the chondro-cranium of some other sauropsids (e.g.,Sphenodon, Bellairs and Kamal, 1981), was


not identified in Lambdopsalis (although hedid report a nutrient foramen in the midlineof the anterior wall of the hypophyseal fos-sa). In the case of Kryptobaatar, there aretwo distinct foramina, and in our restorationwe follow the model provided by those sau-ropsids where the nerve and vein occupyseparate foramina. A likely candidate for aseparate foramen for the pituito-orbital veindoes not appear to be present in Meketichof-fatia (V.J. 446-155), although it should benoted that this region is not in pristine con-dition.

Transverse Canal Vein: Here is reportedfor the first time in multituberculates thepresence of a transverse canal vein. This veinin Kryptobaatar (‘‘tcv’’ in fig. 36B) is an un-expected occurrence among Mammaliafor-mes, because a similar canal is only knownfor some metatherians (Marshall et al., 1990;Marshall and Muizon, 1995) and placentals(McDowell, 1958; MacPhee, 1994). The ab-sence of this canal in other multituberculatesmay be an artifact of preservation or relatedto the relative inaccessibility of the medialwall of the sphenorbital recess for prepara-tion. In fact, two of the sections of Nemegt-baatar published by Hurum (1998a: figs. 2,4) show gaps across the midline dorsal to thehorizontal portion of the basisphenoid in theskull base that are reminiscent of the canalin Kryptobaatar. The meaning of these gapsawaits further consideration. The phyloge-netic implications of the transverse canalvein are unclear, and we are uncertain aboutthe homology of the transverse canal of mul-tituberculates with that of later therians.

ARTERIAL SYSTEM

The arterial reconstructions presented herefor Kryptobaatar do not differ in principlefrom that offered previously for Lambdop-salis (Miao, 1988), ‘‘Catopsalis’’ joyneri(Rougier et al., 1992; Wible and Hopson,1995), and a generalized taeniolabidoid(Rougier et al., 1992). The most striking dif-ferences among these reconstructions con-cern the positions of particular osseous pas-sageways (e.g., supraglenoid foramen, ven-tral ascending canal). The pattern recon-structed here for Kryptobaatar also generallyagrees with that interpreted for such diverse

mammaliaforms as Morganucodon and Vin-celestes (Rougier et al., 1992). In fact, onespecific feature of the arterial pattern, a hor-izontal ventral ascending canal, supports thegrouping of multituberculates with tricono-dontids and prototribosphenidans (Rougier etal., 1996a).

Internal Carotid Artery: Kryptobaatar hasprovided information about certain arteriesnot included in other vascular reconstructionsof multituberculates. Probably the most note-worthy of these is the basicranial portion ofthe internal carotid artery, which has provento be of systematic value among some groupsof extant mammals (Wible, 1986). Thegroove on the promontorium in Kryptobaa-tar clearly indicates that the internal carotidfollowed a transpromontorial course en routeto the cranial cavity (‘‘ica’’ in fig. 37B).Among multituberculates, a similar coursehas been described previously from sulcionly in Ectypodus (Sloan, 1979), and outsideof multituberculates, only in Vincelestes(Rougier et al., 1992) and certain eutherians(MacIntyre, 1972; Wible, 1986). However,the basicranial course of the internal carotidin Kryptobaatar is atypical in two regards:first, it does not run the length of the prom-ontorium but is confined to the rostral half(also in Ectypodus), and second, its rostral-most part is enclosed in a long canal betweenthe petrosal, alisphenoid, and pterygoid thatleads to the carotid foramen in the basisphe-noid. Both of these derived conditions occuronly rarely among certain placentals(MacPhee, 1981; Wible, 1984; MacPhee andCartmill, 1986). Among multituberculates, atranspromontorial course confined to the ros-tral half of the promontorium like that inKryptobaatar and Ectypodus is also indicatedby promontorial grooves in Meketichoffatia(V.J. 446-155), Ptilodus (AMNH 35490),Chulsanbaatar (ZPAL MgM-I/168), Kamp-tobaatar (ZPAL MgM-I/33), Nemegtbaatar(ZPAL MgM-I/82), and Sloanbaatar (ZPALMgM-I/20). A carotid groove is lacking inPseudobolodon oreas (V.J. 460-155), Kueh-neodon dryas (V.J. 454-155), Lambdopsalis(Miao, 1988), and the isolated petrosals re-ferred to ‘‘Catopsalis’’ joyneri and Mesodmathompsoni (Wible and Hopson, 1995). Giventhat carotid foramina are putatively lackingin Lambdopsalis (Miao, 1988), its internal


carotid artery apparently did not reach thecranial cavity, and the vertebral arteries musthave been the principal supplier of blood tothe brain. Noted, however, that given the vas-cular pattern reconstructed for Lambdopsalisby Miao (1988), including the presence of awell-developed stapedial artery, the absenceof a carotid foramen is surprising.

In Kryptobaatar, the carotid foramen isentirely within what we think is the basi-sphenoid (‘‘pacc’’ in fig. 37A); its ventral ap-erture is not on the basicranial surface but isrecessed somewhat dorsally at the end of thecarotid canal (fig. 14), and its dorsal apertureis in the posterolateral part of the hypophy-seal fossa (‘‘fica’’ in figs. 25, 27). Amongmultituberculates, the location of the dorsalaperture of the carotid foramen has been re-ported only in Meketichoffatia krausei(Hahn, 1981), Nemegtbaatar (Kielan-Jawo-rowska et al., 1986; Hurum, 1998a), andChulsanbaatar (Hurum, 1998a). As in Kryp-tobaatar, the carotid foramen is in the pos-terolateral part of the hypophyseal fossa inM. krausei and Nemegtbaatar. However, inChulsanbaatar, it is illustrated (Hurum,1998a: fig. 12) in the anterolateral part of thehypophyseal fossa, although no basis for thisposition is provided. Among other mamma-liaforms, the dorsal aperture of the carotidforamen is in the anterolateral part of the hy-pophyseal fossa in Triconodon (Kermack,1963) and Morganucodon (Kermack et al.,1981) and in the posterolateral part in mono-tremes (Zeller, 1989). In contrast, the dorsalaperture appears to be wholly lateral to thehypophyseal fossa in marsupials (see Archer,1976) and placentals (see Ellenberger andBaum, 1908; Cooper and Schiller, 1975;Evans and Christensen, 1979; Novacek,1986). Among multituberculates, the ventralaperture of the carotid foramen is on the ba-sicranial surface within the basisphenoid inMeketichoffatia krausei (Hahn, 1981) andPtilodus (Simpson, 1937), and between thebasisphenoid and the petrosal in Ectypodus(Sloan, 1979), Kamptobaatar (ZPAL MgM-I/33), Sloanbaatar (ZPAL MgM-I/20), Ne-megtbaatar, and Chulsanbaatar (Kielan-Ja-worowska et al., 1986). Noted, however, thatthe position of the carotid foramen in theMongolian taxa implies a long intraosseouscourse for the artery, which approximates the

morphology of the carotid canal in Krypto-baatar. Among other mammaliaforms, theventral aperture is generally within the ba-sisphenoid (Archer, 1976; Rougier et al.,1992; Luo, 1994) except in some placentals(De Beer, 1937; Wible, 1984).

Stapedial Artery: A stapedial groove run-ning posterolaterally on the promontorium tothe oval window like that in Kryptobaatar(‘‘gpsa’’ in fig. 37A) has been reported pre-viously for multituberculates in Ectypodus(Sloan, 1979) and in various isolated petro-sals, including those referred to ‘‘Catopsal-is’’ joyneri (Kielan-Jaworowska et al., 1986;Wible and Hopson, 1995) and Meniscoessus(Luo, 1989), as well as unreferred specimensfrom the Bug Creek Anthills (Fox and Meng,1997). We have also observed such a stape-dial groove in Meketichoffatia (V.J. 446-155), Ptilodus (AMNH 35490), Chulsanbaa-tar (ZPAL MgM-I/168), Kamptobaatar(ZPAL MgM-I/33), Nemegtbaatar (ZPALMgM-I/82), and Mesodma thompsoni(FMNH 53904). In light of the position ofthe groove in these taxa, it seems likely thattheir stapedial artery ran through a bicruratestapes, as reconstructed for Kryptobaatar. Agroove for the stapedial artery is also presentin Lambdopsalis, but it differs in that it runswholly posterior to the oval window; there-fore, the artery did not perforate the stapesbut ran behind it (Miao, 1988), a fact sub-sequently confirmed by Meng’s (1992) de-scription of a columnar stapes in Lambdop-salis. A stapedial groove is not ubiquitousamong multituberculates, being absent forexample in paulchoffatiids (Lillegraven andHahn, 1993) other than Meketichoffatia.Among other mammaliaforms, a stapedialgroove is known only for Vincelestes (Roug-ier et al., 1992) and certain eutherians (Wi-ble, 1987). However, among these taxa, theartery’s course to the oval window is usuallya lateral or anterolateral one; a posterolateralcourse is known only for the tarsier (Mac-Phee and Cartmill, 1986).

All multituberculates for which the rele-vant anatomy is preserved have a lateralflange that is bent medially to contact thepromontorium (Rougier et al., 1996a), exceptfor Lambdopsalis in which, according toMiao (1988), the alisphenoid accomplishesthis. It is the infolded lateral flange that


forms the floor of the canal for the ramusinferior (post-trigeminal canal). Because ofpreservational problems, the canal for the ra-mus inferior has been described in only a fewmultituberculates; in addition to Kryptobaa-tar, these include Nemegtbaatar (Luo, 1989),Lambdopsalis (Miao, 1988), and various iso-lated petrosals, including those referred to‘‘Catopsalis’’ joyneri (Kielan-Jaworowska etal., 1986), Mesodma thompsoni (Wible andHopson, 1995), and Meniscoessus (Luo,1989). To date, we have found only one mul-tituberculate in which the canal for the ramusinferior is without a doubt absent (Ptilodusmontanus, AMNH 35490), although the ab-sence of the canal does not necessarily implythe absence of the artery, which could haverun ventral to the tympanic roof. Among oth-er mammaliaforms, similar canals in the tym-panic roof are known only for the echidnaand certain placentals (e.g., scandentians,macroscelidids), where they are formed byappositional bone growth from the petrosal(Kuhn, 1971; MacPhee, 1981). In the echid-na, the major occupant is venous (Wible andHopson, 1995), whereas in placentals, it isthe ramus inferior of the stapedial artery(MacPhee, 1981).

Miao’s (1988) interpretation of the osseousstructures associated with the course of theramus inferior in Lambdopsalis differs sub-stantially from that presented here for othermultituberculates. He proposed that the ra-mus inferior en route to the cavum epipteri-cum passed through an alisphenoid canal, in-dicated by a groove on the medial side of thealisphenoid and an opening anterior to theforamen ovale inferium. Additionally, the ra-mus inferior was exposed laterally throughan elongated fenestra slightly larger than theforamen ovale inferium. Behind this fenestra,the ramus inferior was enclosed in a secondcanal that Miao terms ‘‘canal of maxillaryartery,’’ following Kielan-Jawarowska et al.(1986). Although we have not seen all therelevant specimens of Lambdopsalis, wespeculate that the fenestra is an artifact andthat the ramus inferior was fully enclosed inan osseous canal as in other multitubercula-tes, except Ptilodus. The relationships of theosseous canal in Lambdopsalis (i.e., to thefacial nerve, prootic canal, and stapedial sys-tem) are the same as the canal for the ramus

inferior in other multituberculates, and we in-terpret them as homologous. The homoge-neity of this region across Multituberculata,invariably formed by the petrosal, casts ad-ditional doubts on Miao’s (1988) interpreta-tion of the alisphenoid in Lambdopsalis (seeabove). What Miao (1988) terms the alisphe-noid canal, we think merely reflects thecourse of the ramus inferior through the ca-vum epiptericum en route to the sphenorbitalfissure.

A pattern for the canals associated with theramus superior and arteria diploetica magnaresembling that described here for ‘‘Catop-salis’’ joyneri and Kryptobaatar (fig. 36A)has been described for other multitubercu-lates, including Chulsanbaatar, Nemegtbaa-tar (Kielan-Jaworowska et al., 1986; Rougieret al., 1992), Lambdopsalis (Miao, 1988),Ptilodus (Wible and Hopson, 1995), and iso-lated petrosals referred to Meniscoessus(Luo, 1989) and Mesodma thompsoni (Wibleand Hopson, 1995). Differences among taxaconcern the positions of foramina, the com-munications between canals, and the enclos-ing bony elements. For example, the tym-panic aperture of the ventral ascending canal(pterygoparoccipital foramen of non-mam-malian cynodonts) for the ramus superior isconfluent with that for the prootic canal inmost taxa including Kryptobaatar (fig. 37),but is posterior to the prootic canal in Chul-sanbaatar (ZPAL MgM-I/168) and Menis-coessus (Luo, 1989). The ventral ascendingcanal has a communication with the prooticcanal intramurally in Nemegtbaatar (Kielan-Jaworowska et al., 1986) and Lambdopsalis(Miao, 1988), but remains separate in ‘‘Ca-topsalis’’ joyneri and Mesodma thompsoni(Wible and Hopson, 1995); the condition inKryptobaatar is unknown. The supraglenoidforamen for a ramus temporalis opens in theanterior lamina at a level anterior to the fe-nestra vestibuli in most taxa, including Kryp-tobaatar and the paulchoffatiids Meketichof-fatia (V.J. 446-155), Kuehneodon (V.J. 454-155), and Pseudobolodon (V.J. 450-155), butis level with the fenestra vestibuli in Chul-sanbaatar (ZPAL MgM-I/168), Sloanbaatar(ZPAL MgM-I/20), Lambdopsalis (Miao,1988), and Mesodma thompsoni (Wible andHopson, 1995). As in most mammaliaforms(Luo, 1994), the posterior opening into the


posttemporal canal for the arteria diploeticamagna is between the squamosal and petrosalin paulchoffatiids (Lillegraven and Hahn,1993), Nemegtbaatar (Hurum, 1998a), andKamptobaatar (ZPAL MgM-I/33), but iswithin the petrosal in other multituberculatesincluding Kryptobaatar (figs. 35, 36A). Also,the posterior opening is located low on theocciput in most taxa, including Kryptobaatarand Pseudobolodon (V.J. 460-155), but isdorsally situated in Ptilodus (AMNH 35490)and Lambdopsalis (Miao, 1988); the positionof the posterior opening into the posttempo-ral canal dictates the length of ventral as-cending canal. A foramen of the dorsal as-cending canal for a ramus temporalis isfound, in addition to Kryptobaatar, in Kamp-tobaatar (ZPAL MgM-I/33), Chulsanbaatar(ZPAL MgM-I/168), Ptilodus (AMNH35490), and Lambdopsalis (Miao, 1988), butis lacking in specimens of Nemegtbaatar. Fi-nally, as discussed above, an orbitotemporalchannel that transmitted the ramus superiorto the orbit has yet to be identified in paul-choffatiids, but it was likely present giventhe widespread presence of this structure.

The general pattern for the ramus superiorand arteria diploetica magna that occurs inmultituberculates is repeated in most regardsin other mammaliamorphs (Rougier et al.,1992, 1996a; Wible and Hopson, 1995). Themost striking differences among taxa concernthe course of the ramus superior and its ros-tral continuation, the ramus supraorbitalis,which may have been wholly or partially ex-tracranial, intramural, or endocranial. Thetaxon that most closely resembles the mul-tituberculate pattern is the prototribospheni-dan Vincelestes (Rougier et al., 1992, 1996a).Like multituberculates, Vincelestes has an in-tramural ventral ascending canal within theanterior lamina that initially runs posteriorlyin a horizontal plane; it also has multiple fo-ramina for rami temporales, including a fo-ramen of the dorsal ascending canal on thesuture between the anterior lamina, squa-mosal, and parietal (Rougier et al., 1992).The only other taxa with a horizontal coursefor the ventral ascending canal are tricono-dontids (Rougier et al., 1996a, 1996c), buttheir canal differs in that it is intramural be-tween the petrosal and squamosal (Rougieret al., 1996a).

PTERYGOID CANAL

Arising from the carotid groove on the an-terolateral pole of the promontorium in Kryp-tobaatar is a canal of subequal diameter thatis directed anterolaterally toward the orbit;we have interpreted this as the pterygoid ca-nal (‘‘ptca’’ in fig. 37A) based on the resem-blance to that, for example, in lipotyphlans(McDowell, 1958; MacPhee, 1981; Novacek,1986). We think that some other MongolianLate Cretaceous taxa have a similar arrange-ment, with the only difference being that theposterior aperture of the pterygoid canal isfarther away from the posterior aperture ofthe carotid canal, and these two openings areconnected by a groove for the contents of thepterygoid canal. The morphology in questionwas described for Chulsanbaatar and Ne-megtbaatar by Kielan-Jaworowska et al.(1986). In these forms, the carotid grooveleads to the posterior aperture of the carotidcanal situated more posteriorly than in Kryp-tobaatar. Originating at the posterior aper-ture of the carotid canal is a narrower groovethat curves anteromedially across the lateraland anterior surface of the promontorium’santerior pole and ends at the pterygoid canalbetween the petrosal and basisphenoid. Kie-lan-Jaworowska et al. (1986) interpretedwhat we think to be the posterior aperture ofthe carotid canal as the ‘hiatus Fallopii’(quotes in original) and the pterygoid canalas the carotid foramen. That the ‘hiatus Fal-lopii’ of Kielan-Jaworowska et al. (1986) isnot homologous with that of other mammalswas recognized by these authors and hasbeen discussed by others (e.g., Miao, 1988;Luo, 1989; Wible and Hopson, 1995), butthere is no consensus on alternative occu-pants for this aperture in Chulsanbaatar andNemegtbaatar. Given the remarkable posi-tional similarity with the posterior apertureof the carotid canal in Kryptobaatar, wethink that it served the same function inChulsanbaatar and Nemegtbaatar. We haveseen the same arrangement, that is, a ptery-goid canal arising from the carotid groove,in Sloanbaatar (ZPAL MgM-I/20) andKamptobaatar (ZPAL MgM-I/33). Given thesize of the aperture into the pterygoid canalin these forms, its occupants likely includedan artery and nerve (i.e., the deep petrosal


nerve). As with Kryptobaatar, the greater pe-trosal nerve must have entered the canal di-rectly from the cavum epiptericum and/or ca-vum supracochleare.

PETROSAL

During the last few years, we have pub-lished several phylogenetic analyses of Me-sozoic mammaliamorphs relying principallyon features of the petrosal bone (Wible andHopson, 1993; Wible et al., 1995; Rougier etal., 1996a, 1996c). Among multituberculates,the petrosal has been described in some de-tail in various paulchoffatiids (Hahn, 1988;Lillegraven and Hahn, 1993), Ptilodus(Simpson, 1937; Luo, 1989; Wible and Hop-son, 1995), Chulsanbaatar, Nemegtbaatar(Kielan-Jaworowska et al., 1986; Hurum,1998a, 1998b), Lambdopsalis (Miao, 1988;Meng and Wyss, 1995), and from an arrayof isolated elements from the Late Creta-ceous of North America (Kielan-Jaworowskaet al., 1986; Luo, 1989, 1996; Luo and Ket-ten, 1991; Fox and Meng, 1997), includingthose referred to ‘‘Catopsalis’’ joyneri andMesodma thompsoni (Wible and Hopson,1993, 1995). A striking result of these studieshas been how uniform the petrosal morphol-ogy is across Multituberculata, a conclusionreinforced by our descriptions of Kryptobaa-tar. However, the only derived petrosal fea-ture shared by all multituberculates known todate is an infolded bony flange that contactsthe promontorium and housed the epitym-panic recess (Rougier et al., 1996a, 1996c).This bony flange is continuous with the an-terior lamina and crista parotica in all mul-tituberculates with the sole exception ofLambdopsalis, following Miao’s (1988) in-terpretation (but see our comments in theAlisphenoid section above). Some of the pe-trosal features discussed below may be di-agnostic of Multituberculata but are depen-dent on the phylogenetic positions of the ex-ceptions. Other petrosal features have beendiscussed already under the Venous Systemand Arterial System sections.

Despite the general uniformity/similarityin petrosal morphology within Multituber-culata, our comparisons of Kryptobaatarwith other multituberculates have also high-lighted differences (see also Kielan-Jawo-

rowska et al., 1986; Luo, 1989, 1996; Lille-graven and Hahn, 1993; Wible and Hopson,1995; Kielan-Jaworowska and Hurum,1997). Some of these differences are dis-cussed here.

Jugular Fossa: Kielan-Jaworowska andHurum (1997) used the size and depth of thejugular fossa in their phylogenetic analysis ofselected multituberculate taxa. They reportedit to be ‘‘small and shallow’’ in the hypo-thetical ancestor, based on Late Jurassic Pla-giaulacoidea and Ptilodus, and ‘‘large anddeep’’ in Kryptobaatar (‘‘jf’’ in fig. 37A),Nemegtbaatar, Chulsanbaatar, Catopsbaa-tar, Kamptobaatar, and Lambdopsalis. Totheir list we add Sloanbaatar (ZPAL MgM-I/20). The only plagiaulacoid for which wecould confirm a small, shallow jugular fossawas Pseudobolodon (V.J. 450-155, 460-155).In Ptilodus gracilis (USNM 6076) and Ec-typodus (YPM-PU 14724), the jugular fossais shallow but not small. Finally, we couldnot confirm the size and depth of the jugularfossa in Catopsbaatar.

Channel for Perilymphatic Duct: Prior toour description of Kryptobaatar, two char-acter states of the channel for the perilym-phatic duct in multituberculates had beenidentified by us (Rougier et al., 1996a,1996c): (1) an open sulcus, and (2) a sulcuspartially enclosed by bony lappets. The for-mer state occurs in Ptilodus and Mesodmathompsoni, the latter in ‘‘Catopsalis’’ joy-neri. We have subsequently seen bony lap-pets on the perilymphatic channel in Meke-tichoffatia krausei (V.J. 446-155) and Pseu-dobolodon oreas (V.J. 460-155). Kryptobaa-tar exhibits a third state: no indication of theperilymphatic channel whatsoever.

Recessus Scalae Tympani: Ornithorhyn-chus possesses a distinct depression imme-diately external to the perilymphatic fora-men, the recessus scalae tympani (perilym-phatic recess of Fox and Meng, 1997), wherethe perilymphatic duct contacts the cavumtympani (Zeller, 1991, 1993). We have ob-served a similar depression in the tricono-dontid Priacodon (Rougier et al., 1996a). Wealso report a conspicuous recessus scalaetympani in the following multituberculates(see also Fox and Meng, 1997): Ptilodusmontanus (AMNH 35490), Ectypodus(YPM-PU 14724), Mesodma thompsoni


(FMNH 53904), ‘‘Catopsalis’’ joyneri(AMNH 119445), and Kamptobaatar (ZPALMgM-I/33). In contrast, the depression is in-conspicuous in Kryptobaatar, Pseudobolo-don (V.J. 460-155), Meketichoffatia krausei(V.J. 446-155), Nemegtbaatar (ZPAL MgM-I/81), Sloanbaatar (ZPAL MgM-I/20), andLambdopsalis (Miao, 1988).

Post-promontorial Tympanic Recess: Ouranalyses (e.g., Wible and Hopson, 1993;Rougier et al., 1996a, 1996c) have indicatedthat the tall, horizontal crista interfenestralismarks the posterior boundary of the cavumtympani in multituberculates and that a post-promontorial tympanic recess behind theperilymphatic foramen was lacking. How-ever, the crista interfenestralis is not very tallin any multituberculates, and at least in a fewforms (e.g., Pseudobolodon, V.J. 450-155,460-155) it seems likely that the cavum tym-pani extended into a post-promontorial re-cess. Moreover, if the depression identifiedabove as the recessus scalae tympani had thesame function as in Ornithorhynchus (i.e.,housing a diverticulum of the perilymphaticduct that contacted the cavum tympani, Zell-er, 1991), then the crista interfenestralis didnot prevent the cavum tympani from reach-ing the perilymphatic duct in those multitu-berculates with a conspicuous recessus scalaetympani. A post-promontorial recess hasheretofore been known only for prototribo-sphenidans (Wible, 1990) and isolated petro-sals from the Early Cretaceous Khoobur lo-cality (Wible et al., 1995). In light of themorphology in multituberculates, this char-acters needs to be reevaluated before use infuture phylogenetic studies.

Caudal Tympanic Process: Until recently,our analyses (e.g., Wible et al., 1995; Roug-ier et al., 1996a) had shown that a crest me-dial to the paroccipital process forming theposterior wall of the tympanic cavity was asynapomorphy of prototribosphenidans plusisolated petrosals from the Early CretaceousKhoobur locality. In addition, a caudal tym-panic process has been reported in the sym-metrodont Zhangheotherium (Hu et al.,1997). A caudal tympanic process had beenunknown for multituberculates, but Rougieret al. (1996a, 1996c) reported one in Kryp-tobaatar (‘‘ctpp’’ in fig. 37A; PSS-MAE113) and Kamptobaatar (ZPAL MgM-I/33).

To the list of multituberculates with a caudaltympanic process, we now add Meketichof-fatia krausei (V.J. 446-155), Chulsanbaatar(ZPAL MgM-I/145), Nemegtbaatar (contraRougier et al., 1996a; ZPAL MgM-I/82), andPtilodus gracilis (USNM 6076). On the listof those without this process are Pseudobol-odon (V.J. 450-155, 460-155) and Ectypodus(YPM-PU 14724).

Paroccipital Process: The form of the par-occipital process varies among multituber-culates. We have described it above as tri-angular in Kryptobaatar (‘‘ppr’’ in fig. 37A);this shape also occurs in Meketichoffatiakrausei (V.J. 446-155), Chulsanbaatar, Ne-megtbaatar, Sloanbaatar, and Kamptobaatar(see Kielan-Jaworowska and Hurum, 1997:fig. 11). The paroccipital process is a finger-like projection in Ectypodus (YPM-PU14724), and is expanded posterolaterally inPseudobolodon (V.J. 450-155) and Ptilodusmontanus (AMNH 35490). Two of theseconditions appear to characterize the paroc-cipital process of Lambdopsalis, where it isboth triangular and posterolaterally expanded(Miao, 1988).

Medial Wall of Epitympanic Recess:Rougier et al. (1996c) reported that the me-dial wall of the epitympanic recess in Kryp-tobaatar (fig. 21; PSS-MAE 113) is formedby a sharp, ventrally projecting crest com-posed of the crista parotica posteriorly (‘‘cp’’in fig. 37A) and the lateral flange anteriorly.They also noted that the vertical nature ofthis wall would have severely constrained thepossible positions of the stapes and incus.Hurum et al. (1996) observed a similar ar-rangement to the medial wall of the epitym-panic recess in Kamptobaatar and Chulsan-baatar and suggested ‘‘that there must havebeen a significant vertical component in theplane of the manubrium and therefore of thetympanic membrane’’ (p. 269). We add herethat a tall medial wall of the epitympanic re-cess also occurs in Meketichoffatia (V.J. 110-155), Ptilodus (AMNH 35490), Mesodmathompsoni (FMNH 53904), Nemegtbaatar(ZPAL MgM-I/82), Catopsbaatar (ZPALMgM-I/78), and Lambdopsalis (Miao, 1988).In fact, the only multituberculate preservingthe relevant anatomy that has a low medialwall is Ectypodus (YPM-PU 14724). Wehave not seen similarly developed medial


walls of the epitympanic recess in other Me-sozoic mammaliaforms.

Lateral Wall of Epitympanic Recess: Wefound two different conditions for the lateralwall of the epitympanic recess in multituber-culates: (1) either it is a low crest and theepitympanic recess is not sharply demarcat-ed, so that the lateral braincase wall and epi-tympanic recess are confluent, or (2) promi-nent ridges separate the epitympanic recessfrom the lateral braincase wall. The formercondition is found in Meketichoffatia krausei(V.J. 446-155), Kuehneodon dryas (V.J. 454-155), Ectypodus (YPM-PU 14724), Mesod-ma thompsoni (FMNH 53904), and Kamp-tobaatar (ZPAL MgM-I/33); the latter con-dition is found in Kryptobaatar (figs. 20, 21),Pseudobolodon (V.J. 450-155), Ptilodusmontanus (AMNH 35490), Chulsanbaatar(ZPAL MgM-I/168), Nemegtbaatar (ZPALMgM-I/76), Sloanbaatar (ZPAL MgM-I/20),Catopsbaatar (ZPAL MgM-I/78), ‘‘Catop-salis’’ joyneri (AMNH 119445), and Lamb-dopsalis (Miao, 1988).

Tensor Tympani Fossa: Our analyses (e.g.,Wible and Hopson, 1993; Rougier et al.,1996a, 1996c) have indicated that the tensortympani muscle occupied a deep recess inmultituberculates (‘‘ttf’’ in figs. 14, 37A).This remains true for most taxa, but a few(e.g., Pseudobolodon, V.J. 450-155, 460-155) have a shallow tensor tympani fossa.

Facial Ganglion Floor: In Rougier et al.(1996a, 1996c), we characterized the facialganglion floor in multituberculates as a nar-row petrosal bridge beneath the primary fa-cial foramen. There are only a few forms inwhich this morphology can be verified, mostnotably various isolated petrosals from theNorth American Late Cretaceous studied byLuo (1989) and Wible and Hopson (1993,1995). However, Kryptobaatar exhibits a dif-ferent morphology; the apparent confluenceof the facial canal with the canal for the ra-mus inferior produces a broad petrosal bridgebeneath the facial ganglion. It is uncertainhow widespread this morphology is withinMultituberculata.

Cavum Supracochleare: Our analyses(e.g., Wible and Hopson, 1993; Rougier etal., 1996a, 1996c) have characterized multi-tuberculates as having a cavum supracoch-leare, which housed the facial (geniculate)

ganglion, continuous with the cavum epi-ptericum. As with the facial ganglion floor,there are few taxa in which this morphologycan be verified, most notably paulchoffatiids(Lillegraven and Hahn, 1993), ‘‘Catopsalis’’joyneri (AMNH 119445), Mesodma thomp-soni (FMNH 53904), and Kryptobaatar(PSS-MAE 123). Luo (1996: 49A) reportedthat a ‘‘geniculate ganglion completely en-closed in the cavum supracochleare’’ is asynapomorphy of Cimolomyidae and Tae-niolabididae. Given that this statement wastaken from a published abstract, we cannotevaluate the supporting evidence. However,Miao (1988, 1993) clearly has described thecavum supracochleare continuous with thecavum epiptericum in the taeniolabididLambdopsalis. The only cimolomyid repre-sented by a petrosal is Meniscoessus (Luo,1989), and the pattern of its cavum supra-cochleare has not yet been described. In ourcomparisons, we have discovered one mul-tituberculate in which the cavum supracoch-leare is partially walled off from the cavumepiptericum and enclosed within the petrosal(i.e., Ptilodus montanus, AMNH 35490). Apartially enclosed cavum supracochleare hasbeen reported previously for Vincelestes, iso-lated petrosals from the Early Cretaceous ofMongolia, and a few marsupials, and a com-pletely enclosed one was reported only in tri-conodontids, Tachyglossus, and most theri-ans (Wible et al., 1995; Rougier et al.,1996a).

Hypertrophied Vestibule: Miao (1988) re-ported a greatly expanded vestibular appa-ratus in the Paleocene taxon Lambdopsalisand suggested it represented an adaptation tothe perception of low-frequency vibrations,as has been posited for certain extant verte-brates with enlarged vestibules. Luo and Ket-ten (1991) examined isolated petrosals fromthe North American Late Cretaceous Menis-coessus and ‘‘Catopsalis’’ joyneri with CTand reported a more than 10-fold expansionin vestibular size over the condition in opos-sums and humans. They proposed that thisextraordinary inflation was a diagnostic fea-ture of known Cretaceous and Tertiary mul-tituberculates and a possible synapomorphyof Multituberculata. The latter is unlikely inlight of Lillegraven and Hahn’s (1993; seealso Hurum et al., 1996; Fox and Meng,


1997) observation that the vestibule is notenlarged in the Late Jurassic paulchoffatiids.Moreover, we have seen no evidence of ves-tibular inflation from the outer contour of thepetrosal in Kryptobaatar. Hurum et al.(1996) and Hurum (1998b) reported fromanalyses of serial sections that the vestibulein Nemegtbaatar and Chulsanbaatar is notmuch expanded, and Hurum (1998b: 90)stated that the dimensions of the inner ear‘‘are within the limits expected for an extant,small mammal.’’ Therefore, the hypertro-phied vestibule seen in Lambdopsalis (Miao,1988), Meniscoessus, and ‘‘Catopsalis’’ joy-neri (Luo and Ketten, 1991) most likely is aunique adaptation of these taxa for low-fre-quency hearing, as would be anticipated, forexample, in burrowing forms.

HYPOGLOSSAL FORAMEN

Most Mesozoic mammaliaforms and nearoutgroups have a single hypoglossal fora-men, although two are present in Adeloba-sileus (Luo, 1994). Hahn (1969) reported twosmall openings behind the jugular foramenin Meketichoffatia krausei, both of which hetermed condyloid foramina. He (see alsoHahn, 1988) reconstructed the spinal acces-sory nerve in the more anterior opening andthe hypoglossal nerve in the posterior one.We are aware of no instance among extantmammals where the spinal accessory nervehas such a course, and in forms with twoforamina (e.g., Didelphis, Wible, 1990) bothtransmit components of the hypoglossalnerve. Among other multituberculates, a sin-gle hypoglossal foramen has been describedfor Ptilodus (Simpson, 1937) and Lambdop-salis (Miao, 1988) and was illustrated for Ec-typodus (Sloan, 1979: fig. 1). Among Mon-golian Late Cretaceous taxa, the hypoglossalforamen is reported to be either absent, con-fluent with the jugular foramen (as in mono-tremes; ‘‘jfo 1 hyf’’ in fig. 28), or indiscern-ible (Kielan-Jaworowska et al., 1986). Kryp-tobaatar (PSS-MAE 101) is the exception,with a small aperture in the appropriate placethat we have tentatively identified as a hy-poglossal foramen.

ENDOCRANIUM

CAVUM EPIPTERICUM AND PRIMARY

BRAINCASE WALL

Among multituberculates, the structure ofthe cavum epiptericum has been consideredin detail in Kryptobaatar, Meketichoffatiakrausei (Hahn, 1981), Lambdopsalis (Miao,1988), Nemegtbaatar and Chulsanbaatar(Hurum, 1998a). In Kryptobaatar, the cavumepiptericum is an oval endocranial space out-side the primary wall of the braincase, pos-terolateral to the hypophyseal fossa (‘‘ce’’ infigs. 25B, 26, 27). Contributing to the wallsof the cavum epiptericum in Kryptobaatarare: posteriorly, the petrosal; laterally, the an-terior lamina and alisphenoid; medially, theorbitosphenoid, presphenoid, and basisphe-noid, as well as the ossified pilae antotica andmetoptica; dorsally, the orbitosphenoid andossified pilae antotica and metoptica; andventrally, the petrosal, alisphenoid, and per-haps maxilla. As reconstructed here, the ca-vum epiptericum in Kryptobaatar containedthe trigeminal and facial ganglia and the cav-ernous sinus; traversing the cavum were thedivisions of the trigeminal nerve, the troch-lear and abducens nerves, the greater petrosalnerve, the ramus infraorbitalis of the stape-dial artery, and the ophthalmic veins.

The cavum epiptericum in Lambdopsalis,Nemegtbaatar, and Chulsanbaatar resemblesthat in Kryptobaatar in most regards. Differ-ences exhibited by Lambdopsalis include thepresence of a conspicuous recess for the fa-cial ganglion in the posterior part of the ca-vum epiptericum; the large alisphenoid in thelateral wall of the cavum epiptericum, en-larged at the expense of the anterior lamina;and the exclusion from the cavum epipteri-cum of the ramus infraorbitalis at least in partby its course through an alisphenoid canal(but see above). Also, given the more pos-terior position of the metoptic foramen, theoculomotor nerve in Lambdopsalis wouldhave run through the front part of the cavumepiptericum.

Hahn (1981) illustrated and described thecavum epiptericum in a specimen of Pseu-dobolodon oreas, which he later identified(Hahn, 1993) as Meketichoffatia krausei. Itscavum epiptericum was shown to be a long,narrow, roofless space lateral to the hypo-


physeal fossa and tuberculum sellae that end-ed anteriorly at the sphenorbital canal. As de-scribed by Hahn (1981), the absence of aroof distinguishes the cavum of Meketichof-fatia from that of Kryptobaatar, Lambdop-salis, Nemegtbaatar, and Chulsanbaatar inwhich an extensive roof is formed by the os-sified primary wall of the braincase. We havehad the opportunity to study the basisphenoidof Meketichoffatia (V.J. 443-155) in light ofthat in Kryptobaatar. We think that a roofformed by the primary wall was present, al-though not as thick and extensive as in Kryp-tobaatar, Lambdopsalis, Nemegtbaatar, andChulsanbaatar; however, the roof has beendamaged in Meketichoffatia through com-pression. In fact, there is evidence of breaksin what we interpret to be the medial edgeof the roof at the level of the anterior extentof the hypophyseal fossa. Additionally, whatHahn (1981) described as the sphenorbitalcanal is really just the anterior part of theroofed cavum epiptericum. Also distinguish-ing the cavum of Meketichoffatia is a smalllongitudinal canal running through the floor,within which Hahn (1981) placed a bloodvessel and which Miao (1988) subsequentlyidentified as an alisphenoid canal resemblingthat purported for Lambdopsalis. We are un-certain of the nature of this canal in Meke-tichoffatia. However, because of the canal’slength, extreme medial position, and smalldiameter, it is unlikely an alisphenoid canal.Finally, a metoptic foramen is not indicatedin Meketichoffatia, but may have been dam-aged along with the bulk of the roof of thecavum epiptericum.

The structure of the cavum epiptericum inother mammaliaforms and near outgroupsdiffers in various ways from that in multi-tuberculates. First, there is only a partial floorbelow the cavum epiptericum in Adelobasi-leus, Sinoconodon, Dinnetherium, Megazos-trodon, and Morganucodon (Luo, 1994).Second, the cavum supracochleare is sepa-rated wholly or partially from the cavumepiptericum in triconodontids, isolated petro-sals from the Early Cretaceous Khoobur lo-cality, Vincelestes, Tachyglossus, and theri-ans (Rougier et al., 1996a). Third, a coursefor the oculomotor nerve separate from theother contents of the cavum epiptericum isnot known for any other mammaliaforms, al-

though as discussed below this may be apreservational artifact and is likely primitive-ly present in Mammalia. Finally, perhaps themost striking difference is the presence of amassive roof over the cavum epiptericumformed by the primary braincase in multitu-berculates, and its absence in all other mam-maliaforms known thus far, as noted by Kie-lan-Jaworowska et al. (1986). According toLucas and Luo (1993), only a very smallremnant of the primary wall, the slender os-sified base of the pila antotica, occurs in Ade-lobasileus, Sinoconodon, Megazostrodon,Morganucodon, and monotremes. In the firstthree taxa, the pila antotica has only beendescribed from the ventral view, where it isvisible through the open floor of the cavumepiptericum, and the pila may have beenmore substantial than thus far reported. How-ever, the massive ossified pila antotica ofKryptobaatar (‘‘pan’’ in figs. 25–27), Lamb-dopsalis, Nemegtbaatar, and Chulsanbaatarexceeds any other ossified pila reported thusfar among cynodonts, suggesting that this isa secondary specialization of these multitu-berculates. This thickening and extension ofthe pila antotica might be a derived featurefor Multituberculata, of which the conditionin Meketichoffatia might represent an earlystage.

Considering that the common ancestor ofall mammals is thought (e.g., Starck, 1967,1978; Kuhn, 1971; Zeller, 1989) to have hada fully developed primary wall (the pilaepreoptica, metoptica, and antotica; fig. 24A,B), the presence of these structures is ex-pected at least in all pre-mammalian synap-sids if these elements are homologous acrossAmniota. However, the term pila refers to achondrocranial structure that need not be os-sified in the adult, as occurs in most adultsauropsids (De Beer, 1937; Bellairs and Ka-mal, 1981). In Mammaliaformes, the orbitalregion both externally and internally is rarelypreserved or accessible for preparation, ham-pering the recognition of the presence of theprimary wall and its foramina. Based on thepresence of an optic foramen in the orbito-sphenoid of Probainognathus, Sinoconodon,and Adelobasileus, we expect the optic fo-ramina to have been present in the orbito-sphenoid of the mammaliaforms occupying aphylogenetic position between these Triassic/


Early Jurassic forms and Mammalia. This inturn implies an independent loss of an ossi-fied pila metoptica in monotremes, marsupi-als, and Vincelestes, or the absence of thepila metoptica in the ancestry of multituber-culates and therians, with the reacquisition ofa neomorphic pila metoptica in eutherians assuggested by Maier (1987) and in multitu-berculates. Accepting the phylogenetic posi-tion for multituberculates (fig. 39) proposedby Rowe (1988, 1993), Sereno and McKenna(1995), Rougier et al. (1996a, 1996c), Hu etal. (1997), and Ji et al. (1999), the latter sce-nario is favored. The loss of the pila antoticain marsupials and eutherians can be consid-ered a synapomorphy (Rowe, 1988). Theproblem of the presence/absence of a metop-tic foramen is more complex because even ifperfectly preserved in most cases, it will beconfined deep within the cavum epiptericumand hidden in lateral view by the secondarybraincase wall. As noted by Miao (1988),given the presence of a metoptic foramen inmultituberculates, it is likely that this fora-men was also present in pre-multituberculatemammaliaforms.

HYPOPHYSEAL FOSSA, TUBERCULUM SELLAE,AND JUGUM SPHENOIDALE

The entire structure of the hypophysealfossa is known for only five multitubercu-lates: Kryptobaatar (‘‘hf’’ in figs. 25, 27),Meketichoffatia krausei (originally describedas Pseudobolodon oreas by Hahn, 1981),Lambdopsalis (Miao, 1988), Nemegtbaatar,and Chulsanbaatar (Hurum, 1998a). In alltaxa, the hypophyseal fossa is a very deepexcavation in the dorsal surface of the brain-case floor, bordered laterally by tall wallsformed by the primary wall of the braincase.The dorsum sellae, which forms the posteriorwall, is also very prominent in Kryptobaatar(‘‘ds’’ in figs. 25, 26), Meketichoffatia, Ne-megtbaatar, and Chulsanbaatar, but is lowin Lambdopsalis. Well-developed carotid fo-ramina enter the posterolateral aspect of thehypophyseal fossa in Kryptobaatar (‘‘fica’’in figs. 25, 27), Nemegtbaatar, and, as inter-preted here, Meketichoffatia; Hahn (1981) in-correctly identified the endocranial entranceof the carotid as in the dorsum sellae in Mek-etichoffatia. Carotid foramina are more an-

teriorly placed in the hypophyseal fossa inChulsanbaatar (Hurum, 1998a) and are de-scribed by Miao (1988) as wholly lacking inLambdopsalis.

Among Mesozoic mammaliaforms, thestructure of the hypophyseal fossa is knownonly in Morganucodon and Triconodon. Inthe former, as reconstructed from several in-complete basisphenoids by Kermack et al.(1981), the hypophyseal fossa is bordered bytall anterior and lateral walls, with the dor-sum sellae being very low; the carotid fora-mina enter the anteriormost aspect of the fos-sa’s floor. A single basisphenoid is known forTriconodon (Kermack, 1963; Kermack et al.,1981); it shows a fairly shallow hypophysealfossa with anteriorly placed carotid foraminain the floor, but this specimen has been com-pressed, hindering evaluation. Among extantmammals, the hypophyseal fossa is generallyshallow, with the dorsum sellae being themost prominent wall (Kuhn, 1971; Archer,1976; Novacek, 1986; Zeller, 1989); the ca-rotid foramina enter the posterior aspect ofthe hypophyseal fossa in monotremes buttend to lie wholly lateral to the fossa in ther-ians.

In addition to Kryptobaatar, the endocra-nial surface anterior to the hypophyseal fos-sa, that is, the tuberculum sellae and the ju-gum sphenoidale (‘‘tus’’ and ‘‘jsp’’ in fig.25A), has been described in detail only inMeketichoffatia by Hahn (1981). The majordifference between Kryptobaatar and Meke-tichoffatia is the proportional shortening ofthe mesocranium in the former. In Meketi-choffatia, the tuberculum sellae, hypophysealfossa, and cavum epiptericum are long,whereas in Kryptobaatar these structures areshorter and broader (fig. 27). Damage to thearea labeled as the fossa hypochiasmatica inMeketichoffatia (Hahn, 1981: fig. 4) obscuresthe morphology of the rostral portion of thetuberculum sellae; crushing has collapsed theorbital wings over this region. This is thearea where the optic foramina and chiasmaticsulcus would be expected. Consequently,crushing may account for the apparent ab-sence of structures in Meketichoffatia that wehave found in Kryptobaatar.

The jugum sphenoidale in Meketichoffatiahas a broad, shallow sulcus on the midline,whereas the jugum is narrow in Kryptobaa-


tar and even narrower in PSS-MAE 128. Theorbital wings in Meketichoffatia are broadand continue anterodorsally into the laminainfracribrosa, much as in Kryptobaatar.Hahn (1981) describes the cranial aperture ofthe ethmoidal foramen in Meketichoffatia,which seems to occupy a position similar tothat in Kryptobaatar.

CRIBRIFORM PLATE

An ossified cribriform plate is ubiquitousin living mammals except for the platypus,which is known to have a cartilaginous onethat resorbs during ontogeny (Zeller, 1988).The cribriform plate is seldom preserved infossils, because of its delicate nature and itsinaccessibility to preparation. Among multi-tuberculates, a purported cribriform plate has

been reported in a serially sectioned Ne-megtbaatar (Kielan-Jaworowska et al., 1986;Hurum, 1994) based on tiny bone fragmentsin the appropriate region. Kryptobaatar andLambdopsalis (Miao, 1988), which showpristine preservation of delicate endocranialstructures, have failed to provide any evi-dence of an ossified cribriform plate. How-ever, given the extant phylogenetic bracketfor multituberculates, the presence of a car-tilaginous or ossified cribriform plate is ex-pected. In fact, the presence of at least a car-tilaginous cribriform plate has been suggest-ed in Diarthrognathus and Thrinaxodonbased on differential matrix preservation inthe nasal and cranial cavities (Crompton,1958). Consequently, the presence of a crib-riform plate cartilaginous and/or osseous islikely an ancient trait in cynodont history.

CONCLUSIONS

The cranial anatomy of Mongolian LateCretaceous multituberculates, in particularNemegtbaatar and Chulsanbaatar, has pre-viously been evaluated in some detail in aseries of papers by Kielan-Jaworowska (e.g.,1974) and various co-workers (e.g., Kielan-Jaworowska et al., 1986; Gambaryan andKielan-Jaworowska, 1995; Hurum et al.,1996). In the tradition of these studies, ourdescriptions of Kryptobaatar dashzevegi pre-sent the first bone-by-bone treatment of cra-nial anatomy in the most abundant Mongo-lian Late Cretaceous multituberculate, whichis derived from specimens revealing most ofthe osseous sutural relationships as well asmost of the grooves, canals, and foraminarelevant to the cranial nervous, arterial, andvenous systems. The only significant com-ponents of the skull of Kryptobaatar lackingfrom our description are the nasal fossa andparanasal sinuses. Yet, without recourse toserial sectioning, which allowed the descrip-tion of this region in Nemegtbaatar andChulsanbaatar (Hurum, 1994), or high-res-olution CT scanning, which has revealed de-tails of internal anatomy in the skull of, forexample, the primitive non-mammalian cy-nodont Thrinaxodon (Rowe et al., 1995),none of the available specimens of Krypto-baatar exhibited sufficient exposure of the

nasal fossa to enable description. The speci-mens of K. dashzevegi studied by us provid-ed little new information on brain structureto address the questions raised by Kielan-Ja-worowska (1997) concerning neglected char-acters of the multituberculate brain.

The exquisite state of preservation ofspecimens of Kryptobaatar dashzevegi hasoffered us a rare glimpse of details of cranialanatomy that are seldom preserved in an un-distorted manner in Mesozoic mammali-aforms, with Morganucodon (Kermack et al.,1981) and Vincelestes (Rougier et al., 1992;Rougier, 1993; Hopson and Rougier, 1993)being among the more noteworthy excep-tions. The anatomical details uncovered byour study impact knowledge of the skull ofmultituberculates, and of Mesozoic mam-maliaforms in general, in various ways.

(1) We have been able to describe osseouselements poorly known or not known at allfor other multituberculates or other Mesozoicmammaliaforms. Elsewhere (Rougier et al.,1996c), we have provided descriptions of thestylohyal and partial stapes, malleus, and ec-totympanic, the first for some of these ele-ments among Multituberculata and for thestylohyal among Mesozoic Mammaliafor-mes.

Although partial jugals have been reported


for Ptilodus, Nemegtbaatar, and Chulsan-baatar (Hopson et al., 1989), those in Kryp-tobaatar are the first complete jugals knownfor multituberculates. In all instances, the ju-gal is a reduced element lying internal to themaxilla and the squamosal in the zygomaticarch. Among other mammaliaforms, mono-tremes resemble multituberculates in that thejugal is replaced in the zygomatic arch by themaxilla and the squamosal (Hopson et al.,1989). However, the pattern of replacementdiffers somewhat; in the platypus, the re-duced jugal is situated dorsal to the maxillaand the squamosal (Zeller, 1989), and in theechidna, the jugal is wholly absent (Kuhn,1971).

The orbital mosaic and foramina and theendocranium were largely unknown in mul-tituberculates prior to our study, with the ma-jor exceptions being Lambdopsalis (Miao,1988), Nemegtbaatar, and Chulsanbaatar(Hurum, 1994, 1998a), and were also poorlyknown in other Mesozoic mammaliaforms.We have confirmed Miao’s (1988) observa-tion in Lambdopsalis of an optic foramen, ametoptic foramen, and an extensive, thick os-sified primary wall of the braincase includingthe pilae metoptica and antotica; the lattertwo also occur in Nemegtbaatar and Chul-sanbaatar (Hurum, 1998a). The presence ofthese structures in these taxa, as well as em-bryological evidence, suggests that the opticforamen, the metoptic foramen, and an ex-tensive ossified primary wall may be morewidespread among Multituberculata andMammaliaformes. The robust nature of thepila antotica in Kryptobaatar, Lambdopsalis,Nemegtbaatar, Chulsanbaatar, and perhapsMeketichoffatia may be a derived feature ofMultituberculata, because the ossified rem-nants of the pila in the outgroups are not assubstantive. We have also described for thefirst time in multituberculates a foramen forthe pituito-orbital vein, unknown in othermammaliaforms, and a transverse canal, oc-curring only in some metatherians and pla-centals (McDowell, 1958; Archer, 1976).

(2) We have been able to confirm the ab-sence of several elements from the multitu-berculate skull whose incidence has beencontroversial. Kryptobaatar clearly lacks anectopterygoid, a prenasal process of the pre-maxilla, and a septomaxilla, all of which

have been posited for multituberculates (see,respectively, Kielan-Jaworowska, 1971;Miao, 1988; Hahn and Hahn, 1994). Theseelements now appear to be ubiquitously ab-sent in multituberculates or, at the minimumin the case of the prenasal process, not pre-sent primitively in the group. Multitubercu-lates share the absence of an ectopterygoidwith other mammaliaforms except apparentlyMorganucodon (Kermack et al., 1981), theabsence of a prenasal process with othermammals (Wible, 1991), and the absence ofa septomaxilla with marsupials and placen-tals (Wible et al., 1990), with the possibleexception of some xenarthrans (Zeller et al.,1993).

Kryptobaatar also clearly lacks an orbitalexposure of the palatine, which may repre-sent a synapomorphy of multituberculates assuggested by Miao (1988, 1993). Hurum(1994, 1998a) argued against this interpre-tation, in light of his description of the pal-atine in the orbit of Nemegtbaatar. However,it appears that Hurum’s interpretation of Ne-megtbaatar, if correct, is unusual amongmultituberculates.

(3) We have been able to confirm the pres-ence of several elements in the multituber-culate skull whose incidence has been con-troversial. Miao (1988, 1993) questionedwhether any multituberculate really has analisphenoid with a reduced contribution tothe lateral wall of the braincase, as initiallydescribed by Kermack and Kielan-Jawo-rowska (1971) and Kielan-Jaworowska(1971). Kryptobaatar clearly has such a re-duced alisphenoid, and its presence confirmsthe tentative observations of such an elementin other Mongolian Late Cretaceous taxa(Kielan-Jaworowska, 1971; Kielan-Jawo-rowska et al., 1986). Kryptobaatar also clear-ly shows that the anterior lamina is expandedforward dorsal to the alisphenoid, again con-firming previous observations in other Mon-golian Late Cretaceous taxa (Kielan-Jawo-rowska, 1971; Kielan-Jaworowska et al.,1986; Hurum, 1998a). Both of these ele-ments, a reduced alisphenoid and expandedanterior lamina, are shared by MongolianLate Cretaceous multituberculates and mono-tremes (Wible and Hopson, 1993). Moreover,these characters have been used to support amonotreme-multituberculate clade (Wible


and Hopson, 1993; Meng and Wyss, 1995),which more recent analyses, some of whichinclude this feature, do not support (fig. 39;Rougier, 1996a, 1996c; Hu et al., 1997; Ji etal., 1999).

(4) We have been able to resolve compet-ing anatomical hypotheses for several ele-ments of the multituberculate and mamma-liaform skull. The troughs lateral to the pter-ygopalatine ridges that occur in Kryptobaa-tar and most Mesozoic mammaliaforms havebeen said to have accommodated the audi-tory tube (Barghusen, 1986) or to have pro-vided muscle attachment (Kielan-Jaworow-ska, 1971; Hahn, 1981; Gambaryan and Kie-lan-Jaworowska, 1995). The morphology inKryptobaatar eliminates the auditory tube asa possible occupant, because the troughs ex-tend too far anteriorly into the nasal cavityrather than ending in the nasopharynx, thesite of origin for the auditory tube in extantmammals (Starck, 1995). We think that thelateral troughs provided muscle attachmentand/or an accessory route for air in Krypto-baatar and other Mesozoic mammaliaforms.

Miao (1988) has argued that the elementidentified (e.g., Kielan-Jaworowska et al.,1986) as a postorbital process on the parietalin Mongolian Late Cretaceous multitubercu-lates is not a true postorbital process, which,he suggested, is reduced or absent in Multi-tuberculata. However, the process on the pa-rietal in question in Kryptobaatar has thesame positional relationship to the frontal di-ploic vein and the supraorbital nerves andvessels, as does the true postorbital processon the frontal in extant mammals. Conse-quently, the postorbital process, which is in-conspicuous and on the frontal in other mul-tituberculates, has shifted onto the parietaland become enlarged in Mongolian Late Cre-taceous taxa.

Kielan-Jaworowska et al. (1986) have ar-gued, on the basis of serial sections of Ne-megtbaatar, that the aperture at the back ofthe palate identified (e.g., Kielan-Jaworows-ka, 1971) as a minor palatine (palatonasal)foramen in multituberculates is merely anotch that does not fully penetrate bone.Along with a minor palatine foramen, wehave found a second aperture in the orbitalprocess of the maxilla in Kryptobaatar re-sembling that leading into the minor palatine

canal in therians (Novacek, 1986). Therefore,we deem it likely that the palatal aperture inquestion in Kryptobaatar is a true foramenthat transmitted the minor palatine nerve andvessels, as occurs in most mammaliaformsincluding other multituberculates (exceptpossibly Nemegtbaatar).

Most Mesozoic mammaliaforms, includ-ing multituberculates, have multiple aper-tures in and around the anterior lamina thathave been interpreted as having transmittedbranches of the trigeminal nerve (Luo,1994). Two models for the nerves in theseapertures have been proposed: branches ofthe mandibular nerve transmitted in multi-tuberculates (Simpson, 1937; Kielan-Jawo-rowska, 1971), and the maxillary and man-dibular nerves transmitted as in other mam-maliaforms (Patterson and Olson, 1961; Ker-mack et al., 1981). The morphology inKryptobaatar and the vast majority of mul-tituberculates support the former restorationfor multituberculates, because the foraminain question are at the same anteroposteriorlevel, which is not the pattern that the max-illary and mandibular nerve exits exhibit inany extant mammal. The implications of thisfor the reconstruction of nerves in other Me-sozoic mammaliaforms are yet to be ex-plored, but at a minimum legitimate concernshave been raised (see also Rougier et al.,1996a).

Kielan-Jaworowska et al. (1986) suggestedthat the enormous jugular fossa in MongolianLate Cretaceous multituberculates housedlarge ganglia on the cranial nerves below thejugular foramen. The exceedingly large di-mensions of the jugular fossa and its con-nection with the middle-ear cavity in Kryp-tobaatar suggest that this recess housed a di-verticulum of the cavum tympani (Rougier etal., 1996c). We think that this interpretationcan be extended to other multituberculateswith large jugular fossae.

(5) We have been able to reconstruct themajor components of both the cranial arterialand venous patterns in Kryptobaatar. Thecranial vascular system of Kryptobaatar ingeneral resembled that restored previouslyfor multituberculates (e.g., Kielan-Jaworow-ska et al., 1986; Miao, 1988; Wible and Hop-son, 1995) or for other Mesozoic mammali-aforms (e.g., Rougier et al., 1992). However,


the specimens of Kryptobaatar have provid-ed insight into the course of several vesselspreviously unreconstructed for multituber-culates. Kryptobaatar clearly had a trans-promontorial internal carotid artery, andbased upon this observation, we have dis-covered transpromontorial carotid grooves ina variety of other multituberculates. Agroove for a transpromontorial internal ca-rotid is a feature that multituberculates sharewith Vincelestes (Rougier et al., 1992) andcertain eutherians (Wible, 1986), but it is ab-sent in basal mammaliaforms, symmetro-donts, triconodontids, monotremes, and mar-supials. Consequently, a transpromontorialsulcus has been acquired convergently inmultituberculates and prototribosphenidans.

The endocranial aperture of the carotid fo-ramen is in the posterolateral aspect of thedeep hypophyseal fossa in Kryptobaatar,Pseudobolodon (Hahn, 1981), and Nemegt-baatar (Kielan-Jaworowska et al., 1986; Hu-rum, 1998a), whereas it is more anteriorlyplaced in Chulsanbaatar (Hurum, 1998a).Our comparisons with other Mesozoic mam-maliaforms have led us to an additional ob-servation of potential phylogenetic impact:the endocranial aperture of the carotid fora-men is in the hypophyseal fossa in most taxawith the exception of marsupials and placen-tals, in which it is wholly lateral to the fossa.

The positions of the groove for the stape-dial artery and the foramina for its endbranches, along with the partial bicrurate sta-pes described for Kryptobaatar elsewhere(Rougier et al., 1996c), support a course forthe artery through the stapes. We have founda similar arrangement of sulci and foraminain a wide variety of other multituberculates,with the exception being Lambdopsalis, inwhich the artery ran posterodorsal to the ovalwindow (Miao, 1988), as in the platypus(Wible, 1987). Consequently, the vast major-ity of multituberculates exhibited the samerelationship between the stapedial artery andstapes as in all extant eutherians in which theartery is present in the adult (Wible, 1987).The same pathway has been inferred for Vin-celestes (Rougier et al., 1992). Multituber-culates differ, however, from Vincelestes andmost eutherians in the course of the stapedialartery beneath the promontorium of the pe-trosal. The multituberculate vessel ran pos-

terolaterally to reach the fenestra vestibuli,whereas it had a lateral or anterolateralcourse in the remaining taxa.

The cranial venous system of Kryptobaa-tar and other multituberculates resembledthat in monotremes, with the primary exitsfrom the cranial cavity being the prootic ca-nal and the foramen magnum. However,Kryptobaatar is unique among mammali-aforms known to date in having a separateforamen for the pituito-orbital vein and isunique among non-therians in having a trans-verse canal vein.

Lambdopsalis, whose cranial anatomy hasbeen described in detail (Miao, 1988), hasserved as a model for Multituberculata inseveral phylogenetic analyses (e.g., Wible,1991; Meng, 1992; Meng and Wyss, 1995),despite this taxon’s extreme specializationsfor burrowing and numerous apomorphiccranial features compared with MongolianLate Cretaceous taxa (Miao, 1988; Luo,1996). A goal of our report on Kryptobaatarhas been to provide a detailed description ofthe skull and lower jaws of a less specializedform that can serve as an appropriate modelfor future comparative morphological andphylogenetic studies on Multituberculata andon Mammaliaformes in general. In most as-pects of its cranial anatomy, Kryptobaatar,and not Lambdopsalis, resembles the vastmajority of other multituberculates. Cranialfeatures that Kryptobaatar shares with mostmultituberculates, but which differ in Lamb-dopsalis, include: the absence of the prenasalprocess of the premaxilla; an anterior nasalnotch; fewer nasal foramina; the presence ofa lacrimal; a more extensive orbital exposureof the orbitosphenoid; a foramen for the fron-tal diploic vein; the anterior opening of theorbitotemporal canal below the postorbitalprocess; the presence of a jugal; the presenceof pterygopalatine ridges; a reduced alisphe-noid; mandibular nerve foramina in the an-terior lamina; a transpromontorial internalcarotid artery; carotid foramina in the pos-terolateral hypophyseal fossa; a stapedial ar-tery that ran through a bicrurate stapes; a su-praglenoid foramen opening at a level ante-rior to the oval window; the posterior open-ing of the posttemporal canal low on theocciput; an epitympanic recess on the lateralflange; a crista parotica continuous with the


lateral flange; a caudal tympanic process ofthe petrosal; a triangular paroccipital process;a nonhypertrophied vestibule; and a tall dor-sum sellae. In contrast, there are only a fewfeatures in which Lambdopsalis, and notKryptobaatar, exhibits the likely primitivemultituberculate condition. These include: aninconspicuous postorbital process on thefrontal; the absence of a buccinator foramenbetween the anterior lamina and alisphenoid;the absence of a transverse canal foramen;and a large jugular foramen that transmittedthe sigmoid sinus.

In our descriptions of Kryptobaatar andcomparisons with other multituberculates, anumber of features of the cranial anatomyhave been identified that are shared at vari-ous levels within Multituberculata. Evaluat-ing the phylogenetic significance of thesefeatures both within Multituberculata andwithin Mammaliaformes is beyond the scopeof the current report and awaits an in-depthanalysis of multituberculate interrelation-ships.

In lieu of a such an analysis, we compareour descriptions of Kryptobaatar to the re-cent diagnoses of Djadochtatheria, Djadoch-tatheriidae, and Kryptobaatar published byKielan-Jaworowska and Hurum (1997),based on their cladistic analysis of 43 char-acters, (18 dental and 25 cranial). Djadoch-tatheria includes 10 of the 11 MongolianLate Cretaceous multituberculate genera(with the exception being Buginbaatar) andtentatively 2 North American genera (Para-cimexomys and Pentacosmodon). Djadoch-tatheriidae includes Kryptobaatar, Djadoch-tatherium, Catopsbaatar, and Tombaatar.Rougier et al. (1997a) have also placed Kryp-tobaatar within a monophyletic MongolianLate Cretaceous clade, which also includesPentacosmodon (fig. 38). However, the char-acters supporting the interrelationships ofthese taxa are largely from the dentition and,therefore, less relevant to our considerationof the cranial anatomy of Kryptobaatar.

For Djadochtatheria, Kielan-Jaworowskaand Hurum (1997: 206) cited the followingcranial synapomorphies: ‘‘large frontals,pointed anteriorly in the middle, deeply in-serted between the nasals; U-shaped fronto-parietal suture; a sharp edge between the lat-eral and palatal walls of premaxilla (rounded

in other multituberculates). A large roughlyrectangular facial surface of the lacrimal, ex-posed on the cranial roof, separating thefrontal from the maxilla is also characteristicfor Djadochtatheria, but this may be a ple-siomorphy.’’ Of these proposed synapomor-phies, only two are found exclusively in dja-dochtatherians: the frontoparietal suture is V-shaped in non-djadochtatherians, includingthe paulchoffatiid Pseudobolodon (V.J.447.155, 451-1255, 460-155), Ptilodus(Simpson, 1937), and taeniolabidids (Grang-er and Simpson, 1929; Miao, 1988); and thefacial exposure of the lacrimal is arcuate innon-djadochtatherians, including the paul-choffatiids Kuehneodon (V.J. 454-155), Mek-etichoffatia (V.J. 446-155), and Pseudobolo-don (Hahn and Hahn, 1994; V.J. 460-155).However, given that a V-shaped process ofthe frontals also contacts the nasals in thepaulchoffatiids Kuehneodon (V.J. 454-155),Meketichoffatia (V.J. 446-155), and Pseudo-bolodon (V.J. 447.155, 451-1255, 460-155),this purported djadochtatherian feature maybe plesiomorphous for Multituberculata. Inaddition to djadochtatherians, the sharp edgebetween the lateral and palatal walls of thepremaxilla occurs in Lambdopsalis (Miao,1988).

For Djadochtatheriidae, Kielan-Jaworow-ska and Hurum (1997: 208) cited the follow-ing: ‘‘. . . differs from all other multituber-culates (and all other mammals) in having asubtrapezoidal snout in dorsal view, with theanterior and lateral margins confluent withthe zygomatic arches rather than incurved infront of the arches . . . [differs] from othermembers of Djadochtatheria in having thesnout extending for 50% or more of the skulllength, anterior part of promontorium (sensuHurum et al., 1996) irregular, with incurva-tures on both sides rather than oval (a char-acter shared with Lambdopsalis).’’ As notedby Kielan-Jaworowska and Hurum (1997),the features of the snout distinguish Krypto-baatar, Djadochtatherium, and Catopsbaatarfrom other multituberculates, but the prom-ontorial feature is known only for Krypto-baatar and Catopsbaatar. None of these dja-dochtatheriid characters has yet been de-scribed for Tombaatar.

For Kryptobaatar, Kielan-Jaworowska andHurum (1997: 208) provided the following


revised diagnosis: ‘‘The smallest member ofthe Djadochtatheriidae nov. (skull length 25–32 mm), most similar to Djadochtatherium,with which it shares an arcuate (rather thantrapezoidal as in Catopsbaatar) p4. The p4in Kryptobaatar is relatively longer than inDjadochtatherium and has eight serrations(number unknown in Djadochtatherium).Differs from Catopsbaatar in having smallerfacial surface of the lacrimal; differs fromDjadochtatherium and Catopsbaatar in hav-ing a shorter postorbital process. Shares withMLCM [Mongolian Late Cretaceous multi-tuberculates] except Tombaatar the alveolusfor I3 formed by premaxilla, rather than byboth premaxilla and maxilla (autapomorphyof Tombaatar). Differs from Catopsbaatarand Tombaatar in having four upper pre-molars. Differs from Catopsbaatar in havingan inner row of cusps in M1 extending forless than a half, or a half of the tooth length,rather than a long row; shares short inner rowof cusps in M1 with Tombaatar and theNorth American Paracimexomys (Lillegrav-en 1969; Archibald 1982). Differs fromChulsanbaatar and Sloanbaatar in havingcusps on the inner ridge in M1, rather thana smooth ridge. Differs from Tombaatar inhaving M1 cusp formula 4–5:4:3–5 ratherthan 5:5:2. Differs from Djadochtatheriumand Catopsbaatar in having a shorter snout(but longer than in non-djadochtatheriid dja-dochtatherians), a relatively less robust lowerincisor, and less prominent masseteric andparietal crests.’’

Our additions and amendments to their re-vised diagnosis follows: Differs from otherdjadochtatherians in having either confluentcarotid and pterygoid canals or a very shortseparation between the two, a separate hy-poglossal foramen, a foramen for the pituito-orbital vein, and a transverse canal. Differsfrom Nemegtbaatar in that the anterior open-ing of the posttemporal canal is between theparietal, frontal, and anterior lamina. Differsfrom Nemegtbaatar, Chulsanbaatar, andKamptobaatar in having a smaller orbital ex-posure of the orbitosphenoid. Differs fromNemegtbaatar, Chulsanbaatar, and Catops-baatar in that the anterior and intermediatezygomatic ridges are confluent and the pos-terior ridge is absent. Differs from Nemegt-baatar and Kamptobaatar in that the poste-

rior opening into the posttemporal canal iswithin the petrosal. Differs from Chulsan-baatar and Sloanbaatar in that the supragle-noid foramen is at a level anterior to the fe-nestra vestibuli. Resembles Sloanbaatar inhaving I2s directed ventromedially and incontact with each other, leaving a triangularopen space between them and the edge of thepremaxilla. Resembles Djadochtatheriumand Catopsbaatar in having a long postor-bital process on the parietal (contra Kielan-Jaworowska and Hurum, 1997). ResemblesNemegtbaatar and Sloanbaatar in the ab-sence of a conspicuous recessus scalae tym-pani. Resembles Nemegtbaatar and Chulsan-baatar in having a sigmoid sinus that exitsvia the foramen magnum and not the jugularforamen. Resembles Kamptobaatar andChulsanbaatar in having a foramen of thedorsal ascending canal. Resembles Kampto-baatar, Nemegtbaatar, Chulsanbaatar, andCatopsbaatar in having both the foramenovale inferium and foramen masticatoriumtraverse the epitympanic recess.

ACKNOWLEDGMENTS

For the great privilege of studying the ex-quisite MAE skulls of Kryptobaatar, we aregrateful to Dr. Demberelyin Dashzeveg ofthe Geological Institute of the MongolianAcademy of Sciences, Ulaan Baatar, Mon-golia, and Dr. Michael J. Novacek of theAmerican Museum of Natural History. Forthe high-resolution CT scans of Kryptobaa-tar, we thank Dr. Timothy Rowe of the Uni-versity of Texas, Austin. For access to com-parative materials, we thank G. Hahn, Institutfur Geologie und Palaontologie, Phillipps-Universitat, Marburg; J. A. Hopson, Univer-sity of Chicago, Chicago; Z. Kielan-Jawo-rowska, Institute of Paleobiology, PolishAcademy of Sciences, Warsaw; B. Krebs andT. Martin, Institut fur Palaontologie, FreieUniversitat, Berlin; R.D.E. MacPhee andN.B. Simmons, Department of Mammalogy,AMNH; W. Maier, Lehrstuhl fur SpezielleZoologie, Karl-Universitat, Tubingen; M.J.Novacek and M.C. McKenna, Department ofVertebrate Paleontology, AMNH; and U.Zeller, Museum fur Naturkunde, Berlin. Foramassing the vast collection of Kryptobaatarspecimens described above, we thank all the


members of the MAE field crews. The de-scription of Kryptobaatar was made possiblethrough the skilled preparation completed byAmy Davidson at the AMNH. Also at theAMNH, Edward Heck produced figures 1–23 and 29–37, Lorraine Meeker producedfigures 25–26, and G.W.R. produced the re-mainder. Discussions over the years with E.F.Allin, G. Hahn, J.A. Hopson, J. Hurum, Z.Kielan-Jaworowska, Z. Luo, J. Meng, D.Miao, M.J. Novacek, and R. Presley have en-hanced our comprehension of the multitu-berculate skull. For comments on an earlier

version of this manuscript, we are extremelygrateful to E.F. Allin, J.A. Hopson, Z. Kie-lan-Jaworowska, J. Meng, and R. Presley.This research was supported by NSF GrantsBSR 91-19212, DEB 93-0070, DEB 94-0799, DEB 95-27811, DEB 96-25431, DEB99-96051, and DEB 99-96172. GRW’s re-search has also been supported by a RalphE. Powe Junior Faculty Award from OakRidge Associated Universities. Finally, weacknowledge the extreme patience and un-derstanding of our spouses, Stephanie Davisand Adela Reale.

REFERENCES

Archer, M.1976. The basicranial region of marsupicar-

nivores (Marsupialia), interrelation-ships of carnivorous marsupials, andaffinities of the insectivorous perame-lids. Zool. J. Linn. Soc. 59: 217–322.

Archibald, J. D.1982. A study of Mammalia and geology

across the Cretaceous–Tertiary bound-ary in Garfield County, Montana. Univ.California Publ. Geol. Sci. 122: 1–286.

Averianov, A. O.1997. New Late Cretaceous mammals of

southern Kazakhstan. Acta Palaeontol.Pol. 42: 243–256.

Barghusen, H. R.1986. On the evolutionary origin of the ther-

ian tensor veli palatini and tensor tym-pani muscles. In N. Hotton III, P. D.MacLean, J. J. Roth, and E. C. Roth(eds.), The ecology and biology ofmammal-like reptiles: 253–262. Wash-ington, D.C.: Smithsonian InstitutionPress.

Bellairs, A. d’A., and A. M. Kamal1981. The chondrocranium and the develop-

ment of the skull in recent reptiles. InC. Gans and T. S. Parsons (eds.), Bi-ology of the Reptilia: 1–263. London:Academic Press.

Bonaparte, J. F.1986. Sobre Mesungulatum houssayi y nue-

vos mamıferos Cretacicos de Patagon-ia. Actas IV Congr. Argent. Paleontol.2: 48–61.

Broom, R.1909. Observations on the development of

the marsupial skull. Proc. Linn. Soc.N.S.W. 34: 195–214.

1914. On the structure and affinities of theMultituberculata. Bull. Am. Mus. Nat.Hist. 33: 115–134.

Bruner, H. L.1907. On the cephalic veins and sinuses of

reptiles, with description of a mecha-nism for raising the venous blood-pres-sure in the head. Am. J. Anat. 7: 1–117.

Bryant, H. N., and A. P. Russell1992. The role of phylogenetic analysis in the

inference of unpreserved attributes ofextinct taxa. Philos. Trans. R. Soc. Lon-don B337: 405–418.

Bugge, J.1971. The cephalic arterial system in mole-

rats (Spalacidae), bamboo rats (Rhizo-myidae), jumping mice and jerboas(Dipodoidea) and dormice (Gliroidea)with special reference to the systematicclassification of rodents. Acta Anat. 79:165–180.

1974. The cephalic arterial system in insecti-vores, primates, rodents and lago-morphs, with special reference to thesystematic classification. Ibid. 87(suppl. 62): 1–159.

1978. The cephalic arterial system in carni-vores, with special reference to the sys-tematic classification. Ibid. 101: 45–61.

1979. Cephalic arterial pattern in New Worldedentates and Old World pangolinswith special reference to their phylo-genetic relationships and taxonomy.Ibid. 105: 37–46.

Butler, H.1967. The development of the mammalian

dural venous sinuses with especial ref-erence to the post-glenoid vein. J. Anat.102: 33–56.


Clemens, W. A., and Z. Kielan-Jaworowska1979. Multituberculata. In J. A. Lillegraven,

Z. Kielan-Jaworowska, and W. A.Clemens (eds.), Mesozoic mammals,the first two-thirds of mammalian his-tory: 99–149. Berkeley: Univ. Califor-nia Press.

Clemens, W. A., and J. A. Lillegraven1986. New Late Cretaceous, North American

advanced therian mammals that fit nei-ther the marsupial nor eutherian molds.In K. M. Flanagan and J. A. Lillegrav-en (eds.), Vertebrates, phylogeny, andphilosophy. Contrib. Geol. Univ. Wy-oming Spec. Pap. 3: 55–85.

Conroy, G. C, and J. R. Wible1978. Middle ear morphology of Lemur var-

iegatus: implications for primate pale-ontology. Folia Primatol. 29: 81–85.

Cooper, G, and A. L. Schiller1975. Anatomy of the guinea pig. Cambridge,

MA: Harvard Univ. Press.Cope, E. D.

1880. On the foramina perforating the poste-rior part of the squamosal bone of theMammalia. Proc. Am. Philos. Soc. 18:452–461.

Cords, E.1915. Uber das Primordialcranium von Per-

ameles spec.? unter Berucksichtigungder Deckknochen. Anat. Hefte 52: 1–84.

Crompton, A. W.1958. The cranial morphology of a new genus

and species of ictidosaurian. Proc.Zool. Soc. London 130: 183–216.

Crompton, A. W., and W. L. Hylander1986. Changes in mandibular function fol-

lowing the acquisition of a dentary-squamosal jaw articulation. In N. Hot-ton III, P. D. MacLean, J. J. Roth, andE. C. Roth (eds.), The ecology and bi-ology of mammal-like reptiles: 263–282. Washington, D.C.: SmithsonianInstitution Press.

Crompton, A. W., and Z. Luo1993. Relationships of the Liassic mammals

Sinoconodon, Morganucodon oehleri,and Dinnetherium. In F. S. Szalay, M.J. Novacek, and M. C. McKenna (eds.),Mammal phylogeny: Mesozoic differ-entiation, multituberculates, mono-tremes, early therians, and marsupials:30–44. New York: Springer.

Dashzeveg, D., M. J. Novacek, M. A. Norell, J.M. Clark, L. M. Chiappe, A. Davidson,M. C. McKenna, L. Dingus, C. Swish-er, and A. Perle

1995. Extraordinary preservation in a new

vertebrate assemblage from the LateCretaceous of Mongolia. Nature 374:446–449.

De Beer, G. R.1926. Studies on the vertebrate head. II. The

orbitotemporal region of the skull. Q. J.Microsc. Sci. 70: 263–370.

1937. The development of the vertebrateskull. Oxford: Clarendon Press.

Denison, W., and R. J. Terry1921. The chondrocranium of Caluromys.

Washington Univ. Stud. 8: 161–181.Eisthen, H. L.

1992. Phylogeny of the vomeronasal systemand of receptor cell types in the olfac-tory and vomeronasal epithelia of ver-tebrates. Microsc. Res. Tech. 23: 1–21.

Ellenberger, W., and H. Baum1908. Handbuch der vergleichenden Anato-

mie der Haustiere. Berlin: Verlag VonAugust Hirschwald.

Evans, H. E., and G. C. Christensen1979. Anatomy of the dog. Philadelphia: W.

B. Saunders.Fischer, M. S.

1986. Die Stellung der Schliefer (Hyracoidea)im phylogenestischen System der Eu-theria. Cour. Forschungsinst. Senckenb.84: 1–132.

Fourie, S.1974. The cranial morphology of Thrinaxo-

don liorhinus Seeley. Ann. S. Afr. Mus.65: 337–400.

Fox, R. C., and J. Meng1997. An X-radiographic and SEM study of

the osseous inner ear of multitubercu-lates and monotremes (Mammalia): im-plications for mammalian phylogenyand evolution of hearing. Zool. J. Linn.Soc. 121: 249–291.

Freeman, E. F.1976. A mammalian fossil from the Forest

Marble (Middle Jurassic) of Dorset.Proc. Geol. Assoc. 87: 231–235.

1979. A Middle Jurassic mammal bed fromOxfordshire. Palaeontology 22: 135–166.

Gambaryan, P. P., and Z. Kielan-Jaworowska1995. Masticatory musculature of Asian tae-

niolabidoid multituberculate mammals.Acta Palaeontol. Pol. 40: 45–108.

Gaupp, E.1902. Uber die Ala temporalis des Saugetier-

schadels und die Regio orbitalis einigeranderer Wirbeltierschadels. Anat. Hefte19: 155–230.

1905. Neue Deutungen auf dem Gebiete der


Lehre vom Saugetierschadel. Anat.Anz. 27: 273–310.

1908. Zur Entwicklungsgeschichte und ver-gleichenden Morphologie des Schadelsvon Echidna aculeata var. typica. Se-mon Zool. Forschungsreisen in Aus-tralien 6: 539–788.

Gauthier, J. A., A. G. Kluge, and T. Rowe1988. Amniote phylogeny and the importance

of fossils. Cladistics 4: 105–208.Gelderen, C. van

1924. Die Morphologie der Sinus durae ma-tris. Zweiter Teil. Die vergleichendeOntogenie der neurokraniellen Venender Vogel und Saugetiere. Z. Anat.-Entwickelungsgesch. 74: 432–508.

Gidley, J. W.1909. Notes on the fossil mammalian genus

Ptilodus, with descriptions of a newspecies. Proc. U.S. Natl. Mus. 36: 611–626.

Gillilan, L. A.1972. Blood supply to primitive mammalian

brains. J. Comp. Neurol. 145: 209–222.Godefroit, P.

1997. Reptilian, therapsid and mammalianteeth from the Upper Triassic of Var-angeville (northeastern France). Bull.Inst. R. Sci. Nat. Belg. Sci. Terre 67:83–102.

Granger, W., and G. G. Simpson1929. A revision of the Tertiary Multituber-

culata. Bull. Am. Mus. Nat. Hist. 56:601–676.

Green, H. L.1930. A description of the egg-tooth of Or-

nithorhynchus, together with somenotes on the development of the pala-tine processes of the premaxillae. J.Anat. 64: 512–522.

Greenwald, N. S.1987. Patterns of tooth eruption and replace-

ment in multituberculate mammals. J.Vertebr. Paleontol. 8: 265–277.

Gregory, W. K.1910. The orders of mammals. Bull. Am.

Mus. Nat. Hist. 27: 1–524.Griffiths, M.

1978. The biology of monotremes. NewYork: Academic Press.

Hafferl, A.1921. Zur Entwickelungsgeschichte der

Kopfgefasse des Gecko. (Platydactylusannularis.). Anat. Hefte 59: 1–42.

1933. Das Arteriensystem. In L. Bolk, E.Goppert, E. Kallius, and W. Lubosch(eds.), Handbuch der vergleichendenAnatomie der Wirbeltiere, vol. 6: 563–

684. Reimpression 1967. Amsterdam:Ascher.

Hahn, G.1969. Beitrage zur Fauna der Grube Guimar-

ota Nr. 3. Die Multituberculata. Pa-laeontogr. Abt. A 133: 1–100.

1977. Neue Schadel-Reste von Multitubercu-laten (Mamm.) aus dem Malm Portu-gals. Geol. Palaeontol. 11: 161–186.

1978. Milch-Bezahnungen von Paulchoffati-idae (Multituberculata; Ober-Jura).Neues Jahrb. Geol. Palaontol. Mh.1978: 25–34.

1981. Zum Bau der Schadel-Basis bei denPaulchoffatiidae (Multituberculata;Ober-Jura). Senckenb. Lethaea 61:227–245.

1983. Uberblick uber die Erforschungs-Ges-chichte der Multituberculaten (Mam-malia). Schriftenr. Geol. Wiss. 19/20:217–246.

1985. Zum Bau des Infraorbital-Foramens beiden Paulchoffatiidae (Multituberculata,Ober Jura). Berl. Geowiss. Abh. 60: 5–27.

1987. Neue Beobachtungen zum Schadel-und Gebiss-Bau der Paulchoffatiidae(Multituberculata, Ober-Jura). Palaeo-vertebr. Montpellier 17: 155–196.

1988. Die Ohr-region der Paulchoffatiidae(Multituberculata, Ober-Jura). Ibid. 18:155–185.

1993. The systematic arrangement of thePaulchoffatiidae (Multituberculata) re-visited. Geol. Palaeontol. 27: 201–214.

Hahn, G., and R. Hahn1994. Nachweis des Septomaxillare bei Pseu-

dobolodon krebsi n. sp. (Multitubercu-lata) aus dem Malm Portugals. Berl.Geowiss. Abh. E13: 9–29.

1999. Neue Beobachtungen an Plagiaulacoi-dea (Multituberculata) des Ober-Juras.1. Zum Zahn-Wechsel bei Kielanodon.Ibid. E28: 1–7.

Hahn, G., J. C. Lepage, and G. Wouters1987. Ein Multituberculaten-Zahn aus der

Ober-Trias von Gaume (S-Belgien).Bull. Soc. Belg. Geol. 96: 39–47.

Hill, J. E.1935. The cranial foramina in rodents. J.

Mammal. 16: 121–129.Hill, J. P., and G. R. De Beer

1949. Development of the Monotremata. PartVII. The development and structure ofthe egg-tooth and caruncle in themonotremes and on the occurrence ofvestiges of the egg-tooth and caruncle


in marsupials. Trans. Zool. Soc. Lon-don 26: 503–544.

Hillenius, W. J.1994. Turbinates in therapsids: evidence for

Late Permian origins of mammalian en-dothermy. Evolution 48: 207–229.

Hochstetter, F.1896. Beitrage zur Anatomie und Entwicke-

lungsgeschichte des Blutgefasssystemsder Monotremen. Semon Zool. For-schungsreisen in Australien 5: 189–243.

Hopson, J. A., Z. Kielan-Jaworowska, and E. F.Allin

1989. The cryptic jugal of multituberculates.J. Vertebr. Paleontol. 9: 201–209.

Hopson, J. A., and G. W. Rougier1993. Braincase structure in the oldest known

skull of a therian mammal: implicationsfor mammalian systematics and cranialevolution. In P. Dodson and P. Ginger-ich (eds.), Functional morphology andevolution. Am. J. Sci. 293-A: 268–299.

Hopson, J. A., and J. R. Wible1993. Cranial vascular patterns in Multituber-

culata: the postglenoid foramen thatwasn’t there. J. Vertebr. Paleontol. 13(suppl. 3): 42A.

Hu, Y., Y. Wang, Z. Luo, and C. Li1997. A new symmetrodont mammal from

China and its implications for mam-malian evolution. Nature 390: 137–142.

Hurum, J. H.1994. Snout and orbit of Cretaceous Asian

multituberculates studied by serial sec-tions. Acta Palaeontol. Pol. 39: 181–221.

1998a. The braincase of two Late CretaceousAsian multituberculates studied by se-rial sections. Ibid. 43: 21–52.

1998b. The inner ear in two Late Cretaceousmultituberculate mammals, and its im-plications for multituberculate hearing.J. Mamm. Evol. 5: 65–93.

Hurum, J. H., R. Presley, and Z. Kielan-Jawo-rowska

1995. Multituberculate ear ossicles. In A. Sunand Y. Wang (eds.), Sixth symposiumon Mesozoic terrestrial ecosystems andbiota, short papers: 243–246. Beijing:China Ocean Press.

1996. The middle ear in multituberculatemammals. Acta Palaeontol. Pol. 41:253–275.

Jenkins, F. A., Jr.1990. Monotremes and the biology of Meso-

zoic mammals. Neth. J. Zool. 40: 5–31.

Jenkins, F. A., Jr., and C. R. Schaff1988. The Early Cretaceous mammal Gobi-

conodon (Mammalia, Triconodonta)from the Cloverly Formation in Mon-tana. J. Vertebr. Paleontol. 8: 1–24.

Jerzykiewicz, T., and D. A. Russell1991. Late Mesozoic stratigraphy and verte-

brates of the Gobi Basin. CretaceousRes. 12: 345–377.

Jerzykiewicz, T., P. J. Currie, D. A. Eberth, P. A.Johnston, E. H. Koster, and J.-J. Zheng

1993. Djadokhta Formation correlative stratain Chinese Inner Mongolia: an over-view of the stratigraphy, sedimentarygeology, and paleontology and compar-isons with the type locality in the pre-Altai Gobi. Can. J. Earth Sci. 30:2180–2195.

Ji, Q., Z. Luo, and S. Ji1999. A Chinese triconodont mammal and

mosaic evolution of the mammalianskeleton. Nature 398: 326–330.

Kallen, F. C.1977. The cardiovascular system of bats:

structure and function. In W. A. Wim-satt (ed.), Biology of bats, vol. 3: 289–483. New York: Academic Press.

Kermack, K. A.1963. The cranial structure of the tricono-

donts. Philos. Trans. R. Soc. LondonB246: 83–102.

1967. The interrelations of early mammals.Zool. J. Linn. Soc. 47: 241–249.

Kermack, K. A., and Z. Kielan-Jaworowska1971. Therian and non-therian mammals. In

D. M. Kermack and K. A. Kermack(eds.), Early mammals. Zool. J. Linn.Soc. 50 (suppl. 1): 103–115.

Kermack, K. A., F. Mussett, and H. W. Rigney1981. The skull of Morganucodon. Zool. J.

Linn. Soc. 71: 1–158.Kielan-Jaworowska, Z.

1969. Discovery of a multituberculate mar-supial bone. Nature 222: 1091–1092.

1970. New Upper Cretaceous multitubercu-late genera from Bayn Dzak, Gobi De-sert. Palaeontol. Pol. 21: 35–49.

1971. Skull structure and affinities of theMultituberculata. Ibid. 25: 4–41.

1974. Multituberculate succession in the LateCretaceous of the Gobi Desert (Mon-golia). Ibid. 30: 25–44.

1979. Pelvic structure and nature of reproduc-tion in Multituberculata. Nature 277:402–403.

1980. Absence of ptilodontoidean multituber-culates from Asia and its palaeogeo-


graphic implications. Lethaia 13: 169–173.

1983. Multituberculate endocranial casts. Pa-laeovertebr. Montpellier 13: 1–12.

1986. Brain evolution in Mesozoic mammals.In K. M. Flanagan and J. A. Lillegrav-en (eds.), Vertebrates, phylogeny, andphilosophy. Contrib. Geol. Univ. Wy-oming Spec. Pap. 3: 21–34.

1992. Interrelationships of Mesozoic mam-mals. Hist. Biol. 6: 185–202.

1994. A new generic name for the multitu-berculate mammal ‘Djadochtatherium’catopsaloides. Acta Palaeontol. Pol. 39:134–136.

1997. Characters of multituberculates neg-lected in phylogenetic analyses of earlymammals. Lethaia 29: 249–266.

1998. Humeral torsion in multituberculatemammals. Acta Palaeontol. Pol. 43:131–134.

Kielan-Jaworowska, Z., and J. F. Bonaparte1996. Partial dentary of a multituberculate

mammal from the Late Cretaceous ofArgentina and its taxonomic implica-tions. Mus. Argent. Cienc. Nat. ‘‘Ber-nardino Rivadavia’’ Inst. Nac. Invest.Cienc. Nat., n. ser. 145: 1–9.

Kielan-Jaworowska, Z., and D. Dashzeveg1978. New Late Cretaceous mammal locality

in Mongolia and a description of a newmultituberculate. Acta Palaeontol. Pol.23: 115–130.

Kielan-Jaworowska, Z., and P. P. Gambaryan1994. Postcranial anatomy and habits of

Asian multituberculate mammals. Fos-sils Strata 36: 92 pp.

Kielan-Jaworowska, Z., and J. H. Hurum1997. Djadochtatheria - a new suborder of

multituberculate mammals. Acta Pa-laeontol. Pol. 42: 201–242.

Kielan-Jaworowska, Z., and T. Qi1990. Fossorial adaptations of a taeniolabi-

doid multituberculate mammal fromthe Eocene of China. Vertebr. PalAsiat.28: 81–94.

Kielan-Jaworowska, Z., and R. E. Sloan1979. Catopsalis (Multituberculata) from

Asia and North America and the prob-lem of taeniolabidoid dispersal in theLate Cretaceous. Acta Palaeontol. Pol.24: 187–197.

Kielan-Jaworowska, Z., and A. V. Sochava1969. The first multituberculate from the Up-

permost Cretaceous of the Gobi Desert(Mongolia). Ibid. 14: 356–367.

Kielan-Jaworowska, Z., and B. A. Trofimov1980. Cranial morphology of the Cretaceous

eutherian mammal Barunlestes. Ibid.25: 167–185.

Kielan-Jaworowska, Z., C. Poplin, R. Presley, andA. de Ricqles

1984. Preliminary note on multituberculatecranial anatomy studied by serial sec-tions. In W.-E. Reif and F. Westphal(eds.), Third symposium on Mesozoicterrestrial ecosystems: 123–128. Tu-bingen: Attempto Verlag.

Kielan-Jaworowska, Z., R. Presley, and C. Poplin1986. The cranial vascular system in taenio-

labidoid multituberculate mammals.Philos. Trans. R. Soc. London B313:525–602.

Kielan-Jaworowska, Z., D. Dashzeveg, and B. A.Trofimov

1987. Early Cretaceous multituberculatesfrom Mongolia and a comparison withLate Jurassic forms. Acta Palaeontol.Pol. 32: 3–47.

Kielan-Jaworowska, Z., M. J. Novacek, B. A. Tro-fimov, and D. Dashzeveg

2000. Mammals from the Mesozoic of Mon-golia. In M. J. Benton, E. N. Kuro-chkin, M. A. Shishkin, and D. M. Un-win (eds.), The age of dinosaurs inRussia and Mongolia. London: Cam-bridge Univ. Press.

Krause, D. W., and F. E. Grine1996. The first multituberculates from Mad-

agascar: implications for Cretaceousbiogeography. J. Vertebr. Paleontol. 16(suppl. 3): 46A.

Krause, D. W., and Z. Kielan-Jaworowska1993. The endocranial cast and encephaliza-

tion quotient of Ptilodus (Multituber-culata, Mammalia). Palaeovertebr.Montpellier 22: 99–112.

Krause, D. W., Z. Kielan-Jaworowska, and J. F.Bonaparte

1989. Ferugliotherium Bonaparte, the firstknown multituberculate from SouthAmerica. J. Vertebr. Paleontol. 12:351–376.

Krause, D. W., G. V. Prasad, W. von Koenigswald,A. Sahni, and F. E. Grine

1997. Cosmopolitanism among GondwananLate Cretaceous mammals. Nature 390:504–507.

Kristhalka, L., R. J. Emry, J. E. Storer, and J. F.Sutton

1982. Oligocene multituberculates (Mamma-lia: Allotheria): youngest known re-cord. J. Paleontol. 56: 791–794.

Kuhn, H.-J.1971. Die Entwicklung und Morphologie des

Schadels von Tachyglossus aculeatus.


Abh. Senckenb. Naturforsch. Ges. 528:1–192.

Kuhn, H.-J., and U. Zeller1987. The cavum epiptericum in monotremes

and therian mammals. In H.-J. Kuhnand U. Zeller (eds.), Morphogenesis ofthe mammalian skull. Mamm. Depicta13: 50–70.

Kuhne, W. G.1956. The Liassic therapsid Oligokyphus.

London: Brit. Mus.Lillegraven, J. A., and G. Hahn

1993. Evolutionary analysis of the middleand inner ear of Late Jurassic multitu-berculates. J. Mamm. Evol. 1: 47–74.

Lillegraven, J. A., and G. Krusat1991. Cranio-mandibular anatomy of Haldan-

odon exspectatus (Docodonta; Mam-malia) from the Late Jurassic of Por-tugal and its implications to the evolu-tion of mammalian characters. Contrib.Geol. Univ. Wyoming 28: 39–138.

Lillegraven, J. A., S. D. Thompson, B. K. McNab,and J. L. Patton

1987. The origin of eutherian mammals. Biol.J. Linn. Soc. 32: 281–336.

Lucas, S. G., and Z. Luo1993. Adelobasileus from the Upper Triassic

of west Texas: the oldest mammal. J.Vertebr. Paleontol. 13: 309–334.

Luckett, W. P.1993. An ontogenetic assessment of dental

homologies in therian mammals. In F.S. Szalay, M. J. Novacek, and M. C.McKenna (eds.), Mammal phylogeny:Mesozoic differentiation, multituber-culates, monotremes, early therians,and marsupials: 182–204. New York:Springer.

Luo, Z.1989. Structure of the petrosals of Multitu-

berculata (Mammalia) and morphologyof the molars of early arctocyonids(Condylarthra, Mammalia). Ph.D. diss.,Univ. California, Berkeley, 426 pp.

1994. Sister-group relationships of mammalsand transformations of diagnosticmammalian characters. In N. C. Fraserand H.-D. Sues (eds.), In the shadow ofthe dinosaurs — early Mesozoic tetra-pods: 98–128. Cambridge: CambridgeUniv. Press.

1996. Phylogenetic implications of the petro-sal features on the relationships of themultituberculate families of the LateCretaceous. J. Vertebr. Paleontol. 16(suppl. 3): 49A.

Luo, Z., and D. R. Ketten1991. CT scanning and computerized recon-

structions of the inner ear of multitu-berculate mammals. J. Vertebr. Paleon-tol. 11: 220–228.

MacIntyre, G. T.1972. The trisulcate petrosal pattern of mam-

mals. In T. Dobzhansky, M. K. Hecht,and W. C. Steere (eds.), Evolutionarybiology, 6: 275–303. New York: Ap-pleton-Century-Crofts.

MacPhee, R.D.E.1981. Auditory region of primates and

eutherian insectivores. Contrib. Prima-tol. 18: 282 pp.

1994. Morphology, adaptations, and relation-ships of Plesiorycteropus, and a diag-nosis of a new order of eutherian mam-mals. Bull. Am. Mus. Nat. Hist. 220:214 pp.

MacPhee, R.D.E., and M. Cartmill1986. Basicranial structures and primate sys-

tematics. In D. R. Swindler and J. Er-win (eds.), Comparative primate biol-ogy, vol. 1. Systematics, evolution, andanatomy: 219–275. New York: Alan R.Liss.

Maier, W.1987. The ontogenetic development of the or-

bitotemporal region in the skull ofMonodelphis domestica (Didelphidae,Marsupialia), and the problem of themammalian alisphenoid. In H.-J. Kuhnand U. Zeller (eds.), Morphogenesis ofthe mammalian skull. Mamm. Depicta13: 71–90.

Marshall, L. G., J. A. Case, and M. O. Woodburne1990. Phylogenetic relationships of the fami-

lies of marsupials. Curr. Mamm. 2:433–502.

Marshall, L. G., and C. de Muizon1995. Part II: The skull. In C. de Muizon

(ed.), Pucadelphys andinus (Marsupia-lia, Mammalia) from the early Paleo-cene of Bolivia. Mem. Mus. Natl. Hist.Nat. 165: 21–90.

Martinson, G. G.1982. Pozdnemelovye molliuski Mongolii

(The Upper Cretaceous mollusks ofMongolia). Joint Soviet-Mongolian Pa-leontol. Exped. Trans. 17: 5–76. [inRussian].

McDowell, S. B., Jr.1958. The Greater Antillean insectivores.

Bull. Am. Mus. Nat. Hist. 115: 113–214.


McFarland, W. L., M. S. Jacobs, and P. J. Mor-gane

1979. Blood supply to the brain of the dol-phin, Tursiops truncatus, with compar-ative observations on special aspects ofthe cerebrovascular supply of other ver-tebrates. Neurosci. Biobehav. Rev. 3(suppl. 1): 1–93.

McKenna, M. C., and S. K. Bell1997. Classification of mammals above the

species level. New York: ColumbiaUniv. Press.

Meng, J.1992. The stapes of Lambdopsalis bulla

(Multituberculata) and transformationalanalyses on some stapedial features inMammaliaformes. J. Vertebr. Paleontol.12: 459–471.

Meng, J., and A. R. Wyss1995. Monotreme affinities and low frequen-

cy hearing suggested by multitubercu-late ear. Nature 377: 141–144.

1996. Multituberculate phylogeny. Ibid. 379:407.

Miao, D.1986. Dental anatomy and ontogeny of Lamb-

dopsalis bulla (Mammalia, Multituber-culata). Contrib. Geol. Univ. Wyoming24: 65–76.

1988. Skull morphology of Lambdopsalisbulla (Mammalia, Multituberculata)and its implications to mammalian evo-lution. Contrib. Geol. Univ. WyomingSpec. Pap. 4: 1–104.

1993. Cranial morphology and multitubercu-late relationships. In F. S. Szalay, M. J.Novacek, and M. C. McKenna (eds.),Mammal phylogeny: Mesozoic differ-entiation, multituberculates, mono-tremes, early therians, and marsupials:63–74. New York: Springer.

Miao, D., and J. A. Lillegraven1986. Discovery of three ear ossicles in a

multituberculate mammal. Natl. Geogr.Res. 2: 500–507.

Monastersky, R.1996. The lost tribe of mammals. Sci. News

150: 378–379.Moore, W. J.

1981. The mammalian skull. Cambridge:Cambridge Univ. Press.

Nomina Anatomica, 5th ed.1983. Baltimore: Williams & Wilkins.

Nomina Anatomica Veterinaria, 2nd ed.1973. Vienna: Adolf Holzhausen’s Succes-

sors.Novacek, M. J.

1977. Aspects of the problem of variation, or-

igin and evolution of the eutherian au-ditory bulla. Mammal Rev. 7: 131–149.

1986. The skull of leptictid insectivorans andthe classification of eutherian mam-mals. Bull. Am. Mus. Nat. Hist. 183:1–112.

1993. Patterns of skull diversity in the mam-malian skull. In J. Hanken and B. K.Hall (eds.), The skull, vol. 2, Patternsof structural and systematic diversity:438–545. Chicago: Univ. ChicagoPress.

1997. Mammalian evolution: an early recordbristling with evidence. Curr. Bio. 7:R489–R491.

Novacek, M. J., M. Norell, M. C. McKenna, andJ. Clark

1994. Fossils of the Flaming Cliffs. Sci. Am.271: 60–69.

Novacek, M. J., M. A. Norell, L. Dingus, and D.Dashzeveg

1996. Dinosaurs, mammals, birds, and lizardsfrom the Late Cretaceous Ukhaa Tol-god fauna, Mongolia. J. Vertebr. Pa-leontol. 19 (suppl. 3): 56A.

Novacek, M. J., G. W. Rougier, J. R. Wible, M.C. McKenna, D. Dashzeveg, and I. Ho-rovitz

1997. Epipubic bones in eutherian mammalsfrom the Late Cretaceous of Mongolia.Nature 389: 483–486.

Oelrich, T. M.1956. The anatomy of the head of Ctenosaura

pectinata (Iguanidae). Misc. Publ. Mus.Zool. Univ. Michigan 94: 122 pp.

Parrington, F. R.1946. On the cranial anatomy of cynodonts.

Proc. Zool. Soc. London 116: 181–197.Parrington, F. R., and T. S. Westoll

1940. The evolution of the mammalian pal-ate. Proc. R. Soc. London B230: 305–355.

Pascual, R., F. J. Goin, D. W. Krause, E. Ortiz-Jaureguizar, and A. A. Carlini

1999. The first gnathic remains of Sudameri-ca: implications for gondwanathere re-lationships. J. Vertebr. Paleontol. 19:373–382.

Patterson, B., and E. C. Olson1961. A triconodontid mammal from the Tri-

assic of Yunnan. In G. Vandebroek(ed.), International colloquium on theevolution of lower and specializedmammals: 129–191. Brussels: K.Vlaamse Acad. Wettensh., Lett. SchoneKunsten Belgie.

Paulli, S.1900. Uber die Pneumaticitat des Schadels


bei den Saugetieren. I. Uber den Baudes Siebbeins. Uber die Morphologiedes Siebbeins und die der Pneumaticitatbei den Monotremen und den Marsu-pialiern. Gegenbaurs Morphol. Jahrb.28: 147–178.

Presley, R.1981. Alisphenoid equivalents in placentals,

marsupials, monotremes and fossils.Nature 294: 668–670.

Presley, R., and F.L.D. Steel1978. The pterygoid and ectopterygoid in

mammals. Anat. Embryol. 154: 95–110.

Prothero, D. R., and C. Swisher III1992. Magnetostratigraphy and geochronolo-

gy of the terrestrial Eocene–Oligocenetransition in North America. In D. R.Prothero and W. A. Beggren (eds.), Eo-cene–Oligocene climatic and bioticevolution: 46–73. Princeton: PrincetonUniv. Press.

Rose, K. D., and H.-D. Sues1996. Vertebrate paleontology. Geotimes

(Feb.): 31–34.Rougier, G. W.

1993. Vincelestes neuquenianus Bonaparte(Mammalia, Theria) un primitivo mam-ıfero del Cretacico Inferior de la Cuen-ca Neuquina. Ph.D. diss., Univ. BuenosAires, 720 pp.

Rougier, G. W., J. R. Wible, and J. A. Hopson1992. Reconstruction of the cranial vessels in

the Early Cretaceous mammal Vince-lestes neuquenianus: implications forthe evolution of the mammalian cranialvascular system. J. Vertebr. Paleontol.12: 188–216.

1996a. Basicranial anatomy of Priacodon frui-taensis (Triconodontidae, Mammalia)from the Late Jurassic of Colorado, anda reappraisal of mammaliaform inter-relationships. Am. Mus. Novitates3183: 38 pp.

Rougier, G. W., J. R. Wible, and M. J. Novacek1996b. Multituberculate phylogeny. Nature

379: 406.1996c. Middle-ear ossicles of Kryptobaatar

dashzevegi (Mammalia, Multitubercu-lata): implications for mammaliamorphrelationships and the evolution of theauditory apparatus. Am. Mus. Novita-tes 3187: 43 pp.

Rougier, G. W., M. J. Novacek, and D. Dashzeveg1997a. A new multituberculate from the Late

Cretaceous locality Ukhaa Tolgod,Mongolia. Considerations on multitu-

berculate interrelationships. Am. Mus.Novitates 3191: 26 pp.

Rougier, G. W., J. R. Wible, and M. J. Novacek1997b. Nasal and endocranial morphology of

Late Cretaceous multituberculates fromUkhaa Tolgod, Mongolia. J. Vertebr.Paleontol. 13 (suppl. 3): 72A.

Rowe, T.1988. Definition, diagnosis and origin of

Mammalia. J. Vertebr. Paleontol. 8:241–264.

1993. Phylogenetic systematics and the earlyhistory of mammals. In F. S. Szalay, M.J. Novacek, and M. C. McKenna (eds.),Mammal phylogeny: Mesozoic differ-entiation, multituberculates, mono-tremes, early therians, and marsupials:129–145. New York: Springer.

Rowe, T., W. D. Carlson, and W. W. Bottorff1995. Thrinaxodon: Digital atlas of the skull

(CD-ROM). Austin: Univ. Texas Press.Save-Soderbergh, G.

1947. Notes on the brain-case in Sphenodonand certain Lacertilia. Zool. Bidr. Upps.25: 489–516.

Segall, W.1970. Morphological parallelisms of the bulla

and auditory ossicles in some insecti-vores and marsupials. Fieldiana Zool.51: 169–205.

Sereno, P. C., and M. C. McKenna1995. Cretaceous multituberculate skeleton

and the early evolution of the mam-malian shoulder girdle. Nature 377:144–147.

1996. Multituberculate phylogeny. Ibid. 379:406–407.

Shaner, R. F.1926. The development of the skull of the tur-

tle, with remarks on fossil reptileskulls. Anat. Rec. 32: 343–367.

Shiino, K.1914. Studien zur Kenntnis des Wirbeltier-

kopfes. I. Das Chondrocranium vonCrocodilus mit Berucksichtigung derGehirnnerven und der Kopfgefasse.Anat. Hefte 50: 253–382.

Shindo, T.1914. Zur vergleichenden Anatomie der ar-

teriellen Kopfgefasse der Reptilien.Anat. Hefte 51: 267–356.

1915. Uber die Bedeutung des Sinus caver-nosus der Sauger mit vergleichendenanatomischer Berucksichtigung andererKopfvenen. Ibid. 52: 319–495.

Sigogneau-Russell, D.1991. First evidence of Multituberculata

(Mammalia) in the Mesozoic of Africa.


Neues Jahrb. Geol. Palaontol. Mh.1991: 119–125.

Sigogneau-Russell, D., R. Frank, and J. Hemmerle1986. A new family of mammals from the

lower part of the French Rhaetic. In K.Padian (ed.), The beginning of the ageof dinosaurs, faunal change across theTriassic–Jurassic boundary: 99–108.Cambridge: Cambridge Univ. Press.

Simmons, N. B.1993. Phylogeny of Multituberculata. In F. S.

Szalay, M. J. Novacek, and M. C. Mc-Kenna (eds.), Mammal phylogeny: Me-sozoic differentiation, multitubercula-tes, monotremes, early therians, andmarsupials: 146–164. New York:Springer.

Simpson, G. G.1925. A Mesozoic mammal skull from Mon-

golia. Am. Mus. Novitates 201: 12 pp.1937. Skull structure of the Multituberculata.

Bull. Am. Mus. Nat. Hist. 73: 727–763.Sloan, R. E.

1979. Multituberculata. In R. W. Fairbridgeand D. Jablonski (eds.), The encyclo-pedia of paleontology: 492–498.Stroudsberg, PA: Dowden, Hutchinson& Ross.

1981. Systematics of Paleocene multituber-culates from the San Juan Basin, NewMexico. In S. G. Lucas, J. K. Rigby,Jr., and B. S. Kues (eds.), Advances inSan Juan Basin paleontology: 127–160.Albuquerque, Univ. New MexicoPress.

Starck, D.1967. Le crane des mammiferes. In P.-P. Gras-

se (ed.), Traite de Zoologie, vol. 16, pt.1: 405–549, 1095–1102. Paris: Masson.

1978. Das evolutive Plateau Saugetier. Son-derb. Naturwiss. Ver. Hamb. 3: 7–33.

1995. Lehrbuch der speziellen Zoologie.Band II: Wirbeltiere. Teil 5/1: Sauge-tiere. Jena: Gustav Fischer.

Tandler, J.1899. Zur vergleichenden Anatomie der

Kopfarterien bei den Mammalia.Denkschr. K. Akad. Wiss. Wien Math.-Natwiss. Kl. 67: 677–784.

1901. Zur vergleichenden Anatomie derKopfarterien bei den Mammalia. Anat.Hefte 18: 327–368.

1902. Zur Entwicklungsgeschichte der Kopf-arterien bei den Mammalia. Gegen-baurs Morphol. Jahrb. 30: 275–373.

Thewissen, J. G. M.1989. Mammalian frontal diploic vein and the

human foramen caecum. Anat. Rec.223: 242–244.

Toeplitz, C.1920. Bau und Entwicklung des Knorpel-

schadels von Didelphis marsupialis.Zoologica (Stuttg.) 27: 1–84.

Trofimov, B. A.1975. New data on Buginbaatar Kielan-Ja-

worowska et Sochava, 1969 (Mamma-lia, Multituberculata) from Mongolia.Joint Soviet-Mongolian Paleontol. Ex-ped. Trans. 2: 7–13. [In Russian].

Voit, M.1909. Das Primordialcranium des Kaninchens

unter Berucksichtigung der Deck-knochen. Anat. Hefte 38: 425–616.

Wahlert, J. H.1974. The cranial foramina of protrogomor-

phous rodents; an anatomical and phy-logenetic study. Bull. Mus. Comp.Zool. 146: 363–410.

Wall, C. E., and D. W. Krause1992. A biomechanical analysis of the mas-

ticatory apparatus of Ptilodus (Multi-tuberculata). J. Vertebr. Paleontol. 12:172–187.

Watson, D.M.S.1911. The skull of Diademodon, with notes

on those of some other cynodonts. Ann.Mag. Nat. Hist. 8: 293–330.

1920. On the Cynodontia. Ibid. 9: 506–524.Wible, J. R.

1983. The internal carotid artery in early eu-therians. Acta Palaeontol. Pol. 28: 281–293.

1984. The ontogeny and phylogeny of themammalian cranial arterial pattern.Ph.D. diss., Duke Univ., Durham, NC,705 pp.

1986. Transformations in the extracranialcourse of the internal carotid artery inmammalian phylogeny. J. Vertebr. Pa-leontol. 6: 313–325.

1987. The eutherian stapedial artery: charac-ter analysis and implications for super-ordinal relationships. Zool. J. Linn.Soc. 91: 107–135.

1989. Vessels on the side wall of the brain-case in cynodonts and primitive mam-mals. In H. Splechtna and H. Hilgers(eds.), Trends in vertebrate morpholo-gy. Fortschr. Zool. 35: 406–408.

1990. Late Cretaceous marsupial petrosalbones from North America and a cla-distic analysis of the petrosal in therianmammals. J. Vertebr. Paleontol. 10:183–205.


1991. Origin of Mammalia: the craniodentalevidence reexamined. Ibid. 11: 1–28.

Wible, J. R., and D. L. Davis2000. Ontogeny of the chiropteran basicra-

nium, with reference to the Indian falsevampire bat Megaderma lyra. In R. A.Adams and S. C. Pedersen (eds.), On-togeny, functional ecology and evolu-tion of bats: 214–246. New York: Cam-bridge Univ. Press.

Wible, J. R., and J. A. Hopson1993. Basicranial evidence for early mammal

phylogeny. In F. S. Szalay, M. J. No-vacek, and M. C. McKenna (eds.),Mammal phylogeny: Mesozoic differ-entiation, multituberculates, mono-tremes, early therians, and marsupials:45–62. New York: Springer.

1995. Homologies of the prootic canal inmammals and non-mammalian cyno-donts. J. Vertebr. Paleontol. 15: 331–356.

Wible, J. R., D. Miao, and J. A. Hopson1990. The septomaxilla of fossil and Recent

synapsids and the problem of the sep-tomaxilla of monotremes and armadil-los. Zool. J. Linn. Soc. 98: 203–228.

Wible, J. R., G. W. Rougier, M. J. Novacek, M.C., McKenna, and D. Dashzeveg

1995. A mammalian petrosal from the EarlyCretaceous of Mongolia: implicationsfor the evolution of the ear region andmammaliamorph interrelationships.Am. Mus. Novitates 3149: 19 pp.

Williams, P. L., R. Warwick, M. Dyson, and L. H.Bannister (eds.)

1989. Gray’s anatomy, 37th ed. Edinburgh:Churchill-Livingstone.

Wilson, J. T.1900. On the skeleton of the snout and os ca-

runculae of the mammary foetus of themonotremes. Proc. Linn. Soc. N.S.W.25: 58–59.

Witmer, L. M.1995. The extant phylogenetic bracket and

the importance of reconstructing softtissues in fossils. In J. Thomason (ed.),Functional morphology in vertebratepaleontology: 19–33. New York: Cam-bridge Univ. Press.

Zeller, U.1985. Die Ontogenese und Morphologie der

Fenestra rotunda und des Aquaeductuscochleae von Tupaia und anderenSaugern. Gegenbaurs Morphol. Jahrb.131: 179–204.

1988. The lamina cribrosa of Ornithorhyn-chus (Monotremata, Mammalia). Anat.Embryol. 178: 513–519.

1989. Die Entwicklung und Morphologie desSchadels von Ornithorhynchus anatin-us (Mammalia: Prototheria: Mono-tremata). Abh. Senckenb. Naturforsch.Ges. 545: 1–188.

1991. Foramen perilymphaticum und Reces-sus scalae tympani von Ornithorhyn-chus anatinus (Monotremata) und an-deren Saugern. Verh. Anat. Ges. 84:441–443.

1993. Ontogenetic evidence for cranial ho-mologies in monotremes and therians.In F. S. Szalay, M. J. Novacek, and M.C. McKenna (eds.), Mammal phyloge-ny: Mesozoic differentiation, multitu-berculates, monotremes, early therians,and marsupials: 95–128. New York:Springer.

Zeller, U., J. R. Wible, and M. Elsner1993. New ontogenetic evidence on the sep-

tomaxilla of Tamandua and Choloepus(Mammalia, Xenarthra), with a reeval-uation of the problem of the mamma-lian septomaxilla. J. Mamm. Evol. 1:31–46.

GLOSSARY

Given that usages of anatomical terms varyamong researchers, we include a glossary speci-fying our usages for the cranial foramina and ca-nals. The English form is given, followed by theLatin equivalent in parentheses where appropriatebased on the Nomina Anatomica (1983) and/orNomina Anatomica Veterinaria (1973). For manyterms, we have used the morphology of the dogCanis familiaris as the basis for our discussion,relying on the detailed treatment of dog anatomyby Evans and Christensen (1979). Considering the

numerous plesiomorphies retained by monotremes(Zeller, 1989, 1993; Jenkins, 1990) and multitu-berculates (Kielan-Jaworowska, 1992; Lillegravenand Hahn, 1993; Kielan-Jaworowska and Gam-baryan, 1994), the former might represent a suit-able model for evaluating multituberculate mor-phology; however, the details of monotreme cra-nial anatomy, although well known (e.g., Kuhn,1971; Zeller, 1989), are not described and illus-trated to the same degree as for the dog.

ACCESSORY PALATINE FORAMEN (FORAMEN PALA-


TINUM ACCESSORIUS). In the dog (Evans andChristensen, 1979), the accessory palatine nerveand artery (branches of the major palatine nerveand artery, respectively) travel in an accessorypalatine canal to supply the caudal portion of themucosa on the hard palate; oddly enough, theirforamen of exit on the palate is called the minorpalatine foramen. In Kryptobaatar, there are fouror five small foramina in the lateral margin of thepalatine (minor palatine foramina of Kielan-Ja-worowska, 1970), behind the major palatine fo-ramen, that we interpret as for the accessory pal-atine nerves and arteries. Unlike the terminologyin the dog, we call the apertures that transmittedthese nerves and arteries the accessory palatineforamina and reserve the term minor palatine fo-ramen for the more posterior aperture that trans-mitted the minor palatine nerve and artery.

ASCENDING CANAL (CANALIS ASCENDENS). Firstused by Kielan-Jaworowska et al. (1986) for theintramural canal in multituberculates within thesuture between the anterior lamina and the squa-mosal. These authors suggested that this canalwas more widespread among Mesozoic and extanttaxa, and a more detailed evaluation of its ho-mologies was completed by Rougier et al. (1992).The major occupants were the ramus superior ofthe stapedial artery and accompanying veins (Wi-ble, 1989; Rougier et al., 1992; Wible and Hop-son, 1995). The canal was further subdivided intodorsal and ventral ascending canals by Rougier etal. (1992), based on the positional relationship tothe posttemporal canal.

BUCCINATOR FORAMEN (FORAMEN BUCCINATOR-IUM). Aperture in the alisphenoid, on the sidewall of the braincase, transmitting the buccalbranch of the mandibular nerve in some rodents(Hill, 1935; Wahlert, 1974). Despite the implica-tion of the name, the buccal branch of the man-dibular nerve is not the motor nerve to the buc-cinator muscle, but is sensory to the skin and mu-cosa associated with that muscle (Evans andChristensen, 1979; Williams et al., 1989). Used inKryptobaatar for the anteriorly directed aperturein the pterygoid fossa of the alisphenoid betweenthat bone and the anterior lamina.

CANAL FOR RAMUS INFERIOR (CANALIS RAMUS IN-FERIUS). Following Wible and Hopson (1995),the horizontal canal dorsal to the epitympanic re-cess and lateral flange that connects the middleear and cavum epiptericum in multituberculates.Transmitted the ramus inferior of the stapedial ar-tery and perhaps the post-trigeminal vein. InKryptobaatar, the canal for the ramus inferior hasa common tympanic aperture with the secondaryfacial foramen. Equivalent to the canal for the?maxillary artery of Kielan-Jaworowska et al.

(1986) and the post-trigeminal canal of Rougieret al. (1996a, 1996c).

CAROTID CANAL (CANALIS CAROTICUM). In hu-mans (Williams et al., 1989), the internal carotidartery and nerves run from the skull base to thecranial cavity through a perbullar carotid canal(Wible, 1986) in the petrosal (petrous temporal).In Kryptobaatar, the internal carotid artery andnerves first ran through the middle ear in a grooveon the promontorium of the petrosal and then en-tered a carotid canal via a foramen between thepetrosal and alisphenoid at the anterolateral poleof the promontorium. From this aperture, the ca-rotid canal runs anteromedially and dorsallyacross the anterior pole of the promontorium be-tween the petrosal, alisphenoid, and probably thepterygoid. The canal ends at the carotid foramenin the basisphenoid, which opens endocranially inthe posterolateral part of the hypophyseal fossa.Branching off the ventromedial part of the carotidcanal is the pterygoid canal.

DORSAL ASCENDING CANAL (CANALIS ASCENDENS

DORSALIS). Following Rougier et al. (1992), thesegment of the ascending canal dorsal to its com-munication with the posttemporal canal, trans-porting the ramus superior of the stapedial arteryto the orbitotemporal channel. In Kryptobaatar,the dorsal ascending canal is an intramural pas-sageway between the squamosal and petrosal.However, the equivalent channel can be endocra-nial, as in lipotyphlans, or an extracranial sulcus,as in the non-mammalian cynodont Probaino-gnathus (Rougier et al., 1992).

ETHMOIDAL FORAMEN (FORAMEN ETHMOIDA-LE). In the dog (Evans and Christensen, 1979),two ethmoidal foramina connect the orbit and cra-nial cavity; the smaller, ventral one is in the suturebetween the frontal and orbitosphenoid, and theother is entirely in the frontal. The ventral fora-men transmits the ethmoidal nerve, a branch ofthe ophthalmic, and the dorsal transmits the ex-ternal ethmoidal artery, a branch of the maxillary,and an accompanying vein. Humans (Williams etal., 1989) have anterior and posterior ethmoidalforamina, which transmit the anterior and poste-rior ethmoidal nerves and vessels, respectively,with the former nerve being equivalent to the eth-moidal nerve of the dog. The foramina in humansare either in the frontoethmoidal suture or in thefrontal. In Kryptobaatar, there is a single eth-moidal foramen between the frontal and orbito-sphenoid, which accommodated the ethmoidalnerve and vessels.

FENESTRA VESTIBULI (FENESTRA VESTIBU-LI). Term in the Nomina Anatomica (1983) forthe oval window, the opening in the petrosal (pe-trous temporal) that accommodates the footplateof the stapes.


FORAMEN FOR FRONTAL DIPLOIC VEIN (FORAMEN

VENA DIPLOETICA FRONTALIS). In the dog (Evansand Christensen, 1979), a small unnamed foramenin the orbital surface of the postorbital (zygomat-ic) process of the frontal transmits the frontal di-ploic vein, an emissary vein from the diploe ofthe frontal to the ophthalmic vein. In the few eu-therians for which this vein has been described, itconnects either the superior (dorsal) sagittal sinusor the frontal paranasal air sinus and the frontaldiploe with the ophthalmic vein (Thewissen,1989). Zeller (1989: fig. 55) reported a frontal di-ploic vein in Ornithorhynchus, and his illustrationof the adult skull shows an unnamed foramen inthe frontal in the same position as that transmit-ting the frontal diploic vein in the dog. A similarlyplaced foramen in the frontal in Tachyglossus ac-commodates diploic vessels according to Kuhn(1971) and, as interpreted here, in Kryptobaatar.To facilitate description, we designate the aperturein monotremes and Kryptobaatar the foramen forthe frontal diploic vein.

FORAMEN FOR RAMUS TEMPORALIS (FORAMEN RA-MUS TEMPORALIS). Openings in the squamosaland parietal of various extant therians that trans-mit temporal rami of the ramus superior of thestapedial artery and accompanying veins into thetemporal fossa (Wible, 1987). Similar foraminaare found in or along the anterior lamina of Kryp-tobaatar and are presumed to have had the samefunction. Included are the supraglenoid foramenin the anterior lamina in front of the root of thezygoma, the foramen of the dorsal ascending ca-nal between the anterior lamina and the squamo-sal at their superior limits, and an unnamed fo-ramen between the anterior lamina and squamosalabove the root of the zygoma.

FORAMEN FOR PITUITO-ORBITAL VEIN (FORAMEN

VENA PITUITO-ORBITALIS). In Kryptobaatar, theforamen in the anterolateral wall of the hypophy-seal fossa connecting to the orbitotemporal fossa,which transmitted a vein between the pituitary(hypophysis) and the orbit.

FORAMEN OF DORSAL ASCENDING CANAL (FORA-MEN CANALIS ASCENDENS DORSALIS). Miao (1988)coined the term ascending canal foramen for theopening into the dorsal ascending canal betweenthe anterior lamina, squamosal, and parietal inLambdopsalis. This aperture was interpreted byRougier et al. (1992) as having housed a temporalbranch of the ramus superior of the stapedial ar-tery. Miao’s (1988) term is modified here to fo-ramen of the dorsal ascending canal to distinguishit from other foramina on the ventral ascendingcanal. In Kryptobaatar, the foramen of the dorsalascending canal is between the anterior laminaand squamosal at their superior limits.

FORAMEN OVALE INFERIUM (FORAMEN OVALE IN-

FERIUM). In Ptilodus, Simpson (1937) reportedtwo openings in the area where the foramen ovalewas expected. He interpreted both openings as forthe mandibular nerve and named them the fora-men masticator and foramen ovale inferius. As amodel he offered the condition in some extant ro-dents where as many as three separate aperturestransmitting branches of the mandibular nervewere reported by Hill (1935). Following Kielan-Jaworowska et al. (1986), we employ the termforamen ovale inferium for the more medial of theapertures that transmitted the mandibular nervethrough the anterior lamina in multituberculates.In Kryptobaatar, the foramen ovale inferium liesin the anterior part of the epitympanic recess.

HYPOGLOSSAL CANAL (CANALIS HYPOGLOSSI). Inthe dog (Evans and Christensen, 1979), the open-ing in the exoccipital that transmits the hypoglos-sal nerve and the vein of the hypoglossal canal,which connects the condyloid and vertebral veins.In Kryptobaatar, only one hypoglossal foramenseems to be present; it is deep within the jugularfossa and completely encircled by the exoccipital.

INCISIVE FORAMEN (FORAMEN INCISIVUM). In thedog (Evans and Christensen, 1979), a large aper-ture (palatine fissure) in the anterior palate be-tween the premaxilla and maxilla transmits theseptal branch of the caudal nasal nerve, the rostralseptal branch of the major palatine artery, and thenasopalatine (incisive) duct, which connects thenasal and oral cavities with the vomeronasal (Ja-cobson’s) organ. We call the equivalent aperturein Kryptobaatar the incisive foramen followingthe Nomina Anatomica (1983). It likely transmit-ted the same structures as the palatine fissure inthe dog; there is no bony evidence other than theincisive foramen for the nasopalatine duct andvomeronasal organ, but their presence is inferredfor Kryptobaatar from the widespread distributionof these structures among extant amniotes (Gau-thier et al., 1988; Eisthen, 1992).

INFRAORBITAL FORAMEN (FORAMEN INFRAORBITA-LE). In the dog (Evans and Christensen, 1979),the opening in the maxilla dorsal to the fourthpremolar at the anterior end of the infraorbital ca-nal that connects the orbit and snout and accom-modates the infraorbital nerve, artery, and vein.In the floor of the infraorbital canal are severalsmall alveolar canals, and in the medial wall isthe incisivomaxillary canal; alveolar branches ofthe infraorbital nerve and vessels are transmittedto the premolars by the former and to the canineand incisors by the latter. In Kryptobaatar, theinfraorbital foramen is dorsoventrally compressed,visible in ventral view, and above the embrasuresof the first and second premolars or slightly pos-terior to them. In the right side of PSS-MAE 113,two openings leading medially into the maxilla


(either to the maxillary sinus or nasal cavity) arevisible. As noted by Rougier et al. (1997a), thesemay be apertures to the alveolar canals.

INTERNAL ACOUSTIC MEATUS (MEATUS ACUSTICUS

INTERNUS). In the dog (Evans and Christensen,1979), the recess on the endocranial surface of thepetrosal accommodating the facial and vestibulo-cochlear nerves as well as the labyrinthine arteryoff the basilar. A transverse crest divides the me-atus into a dorsal part for the facial nerve and partof the vestibular nerve, and a ventral part for therest of the vestibular nerve and for the cochlearnerve. In Kryptobaatar, the meatus is divided bya low transverse crest with a single aperture oneither side.

JUGULAR FORAMEN (FORAMEN JUGULARE). Inthe dog (Evans and Christensen, 1979), the open-ing between the petrosal, basioccipital, and ex-occipital that connects the cranial cavity and skullbase and transmits the glossopharyngeal, vagus,and accessory nerves along with the sigmoid si-nus. In Kryptobaatar, the jugular foramen lies be-tween the petrosal and exoccipital, and in light ofits small size likely transmitted nerves only; thesigmoid sinus and perhaps the inferior petrosal si-nus exited the cranial cavity via the foramen mag-num as, for example, in Tachyglossus (Wible andHopson, 1995).

LACRIMAL FORAMEN (FORAMEN LACRIMALE). Inthe dog (Evans and Christensen, 1979), the open-ing into the nasolacrimal (lacrimal) canal lieswithin a depression in the orbital surface of thelacrimal bone called the fossa for the lacrimal sac.Accompanying the nasolacrimal duct into the ca-nal is a delicate branch of the malar artery, offthe infraorbital. ‘‘In about 50 per cent of dogs, thenasolacrimal duct has two openings (Evans andChristensen, 1979: 1105).’’ In Kryptobaatar, thesmall opening into the nasolacrimal canal, whichwe call the lacrimal foramen following, for ex-ample, Novacek (1986), is in the lacrimal high onthe orbital rim, near the suture with the facial pro-cess of the maxilla.

MANDIBULAR CANAL (CANALIS MANDIBULAR-IS). In the dog (Evans and Christensen, 1979),the canal opening posteriorly in the ramus of themandible and anteriorly via the mental foramenthat accommodates the inferior alveolar nerve offthe mandibular nerve and accompanying vesselsfrom the maxillary artery and vein. In Kryptobaa-tar, the canal for the inferior alveolar nerve andvessels opens posteriorly just behind the level ofthe second molar.

MASTICATORY FORAMEN (FORAMEN MASTICATOR-IUM). Aperture in the alisphenoid, on the sidewall of the braincase, transmitting the masticatorybranch of the mandibular nerve in some rodents(Hill, 1935; Wahlert, 1974). Following Simpson

(1937), the term is used for the more lateral ofthe two foramina in the anterior lamina that trans-mitted branches of the mandibular nerve in mul-tituberculates (see foramen ovale inferium). Onone side of one specimen of Kryptobaatar (PSS-MAE 113), the masticatory foramen is bifurcatedby the lateral flange, which produces a ventralopening into the epitympanic recess and a lateralopening into the temporal fossa.

MAJOR PALATINE FORAMEN (FORAMEN PALATINUM

MAJOR). In the dog (Evans and Christensen,1979), the major palatine nerve and artery(branches of the maxillary nerve and artery, re-spectively) leave the orbit in the palatine canal inthe maxilla that opens at the major palatine fora-men on the palate between the maxilla and pala-tine. In Kryptobaatar, the major palatine nerveand artery left the orbit via the sphenopalatine fo-ramen and appeared on the hard palate via themajor palatine foramen in the anteromedial cornerof the suture between the maxilla and palatine atthe level of the P4/M1 embrasure; the presence ofthe dorsal aperture of the palatine canal cannot bedetermined.

MENTAL FORAMEN (FORAMEN MENTALE). In thedog (Evans and Christensen, 1979), there are usu-ally three mental foramina, each of which trans-mits nervous and vascular branches of the inferioralveolar nerves and vessels from the mandible tothe chin. In Kryptobaatar, the single mental fo-ramen is anterolaterally directed and in the dia-stema between the incisor and third premolar,closer to the superior than the inferior margin ofthe bone.

METOPTIC FORAMEN (FORAMEN METOPTI-CUM). The aperture occurring in most extantsauropsids between the pilae metoptica and an-totica that transmits the oculomotor nerve and insome instances also the trochlear nerve, ophthal-mic artery, and/or pituito-orbital vein (De Beer,1926, 1937; Bellairs and Kamal, 1981). A metop-tic foramen is not present in the chondrocraniumor bony skull of any extant mammal, because thepilae metoptica and antotica do not co-occur;however, one is generally held to have been pre-sent in the chondrocranium of the common an-cestor of Mammalia (Kuhn, 1971; Kuhn and Zell-er, 1987; Zeller, 1989). In Lambdopsalis, Miao(1988) described an aperture at the anterolateralcorner of the hypophyseal fossa as the metopticforamen and placed within it the oculomotornerve, ophthalmic artery, and pituitary vein (pi-tuito-orbital vein), citing De Beer (1937). Follow-ing Miao, and in light of the morphology in Tach-yglossus (see Gaupp, 1908) and Ornithorhynchus(personal obs.), we restore the oculomotor nervein the similarly placed foramen in Kryptobaatar,but not the ophthalmic artery and pituito-orbital


vein. Although the ophthalmic artery accompa-nies the oculomotor nerve in some sauropsids(Shiino, 1914; Bellairs and Kamal, 1981), in oth-ers and in mammals, the ophthalmic artery runswith the optic nerve (Tandler, 1899; Hafferl, 1933;Wible, 1984). The pituito-orbital vein in Krypto-baatar occupied its own foramen.

MINOR PALATINE FORAMEN (FORAMEN PALATINUM

MINUS). In the dog (Evans and Christensen,1979), the minor palatine nerve and artery(branches of the maxillary nerve and artery, re-spectively) leave the orbit and reach the soft pal-ate via an unnamed notch, rarely closed to forma foramen, in the back edge of the maxilla andpalatine. In Kryptobaatar, the minor palatinenerve and artery left the orbit via a foramen inthe posterior limit of the orbital process of themaxilla and traversed the minor palatine canal,which opens on the back edge of the palate be-tween the palatine, maxilla, and alisphenoid. Wedesignate the opening on the palate the minor pal-atine foramen.

NASAL FORAMEN (FORAMEN NASALE). First usedby Simpson (1937) for the two pairs of foraminavisible on the dorsum of the nasals in Ptilodus,suggested to have been vascular in function. Kie-lan-Jaworowska (1971) proposed the lateral eth-moidal nerve as an alternative occupant. Both ner-vous and vascular functions seem likely. Similarforamina in the nasal of the iguanid Ctenosaurapectinata accommodate the dorsal cutaneousbranch of the lateral ethmoidal nerve, as pointedout by Kielan-Jaworowska (1971), and venoustributaries to the orbital sinus (Oelrich, 1956). Inthe dog (Evans and Christensen, 1979), the exter-nal nasal nerve, off the ethmoidal nerve, andbranches of the external ethmoidal artery run ven-tral to the nasal bone and emerge at that bone’santerior limit to supply the muzzle; these seemlikely occupants for the nasal foramina in multi-tuberculates. In the specimens of Kryptobaatardescribed here, the number of nasal foramina isuncertain because of damage; there are two fo-ramina arranged asymmetrically between the twosides in PSS-MAE 101.

OPTIC FORAMEN (FORAMEN OPTICUM). In thedog (Evans and Christensen, 1979), the openingin the orbitosphenoid that connects the cranialcavity and orbit and transmits the optic nerve andthe ophthalmic branch of the internal carotid ar-tery. A similar opening occurs in the orbitosphe-noid of Kryptobaatar. The embryological precur-sor of the optic foramen is in the chondrocraniumbetween the pilae preoptica and metoptica (Starck,1967; Moore, 1981).

ORBITOTEMPORAL CANAL (CANALIS ORBITOTEM-PORALIS). Following Rougier et al. (1992), thechannel in the side wall of the braincase connect-

ing the dorsal ascending canal with the orbit andtransmitting the ramus superior of the stapedialartery, its continuation the ramus supraorbitalis,and accompanying veins. In Kryptobaatar, thischannel is largely endocranial, bounded laterallyby the parietal, except anteriorly where there is ashort intramural course between the parietal andfrontal; the anterior opening of the orbitotemporalcanal is beneath the postorbital process, betweenthe parietal, frontal, and anterior lamina. The or-bitotemporal canal, also called the sinus canal(Watson, 1911) and internal parietal groove (Kie-lan-Jaworowska et al., 1986), may be wholly ex-tracranial as, for example, in the non-mammaliancynodont Probainognathus (Rougier et al., 1992).

PERILYMPHATIC FORAMEN (FORAMEN PERILYMPHA-TICUM). Following Kuhn (1971) and Zeller(1989, 1991), the aperture in monotremes in theback of the promontorium of the petrosal thattransmits the perilymphatic duct into the inner earand provides support for the secondary tympanicmembrane. In therians, the perilymphatic duct be-comes enclosed in a separate canal, the cochlearaqueduct, by the processus recessus; the enclosureof the perilymphatic duct leaves a separate aper-ture for the secondary tympanic membrane, thefenestra cochleae (Zeller, 1985). In Kryptobaatar,a cochlear aqueduct is lacking and, therefore, theaperture on the back of the promontorium is aperilymphatic foramen.

POSTTEMPORAL CANAL (CANALIS POSTTEMPORAL-IS). Following Rougier et al. (1992), the vascularcanal between the petrosal and squamosal and/orparietal connecting the ascending canal with theocciput and transmitting the arteria and vena di-ploetica magna. In Kryptobaatar, the posterioropening is at the lateral margin of the occiput,entirely within the mastoid exposure of the petro-sal.

PRIMARY FACIAL FORAMEN (FORAMEN FACIALIS

PRIMARIUM). Following Wible (1990) and Wibleand Hopson (1993), the opening for the facialnerve in the endocranial surface of the petrosalthat connects the internal acoustic meatus with thecavum supracochleare or cavum epiptericum. InKryptobaatar, the primary facial foramen sharesa common aperture in the internal acoustic meatuswith that for the vestibular nerve, and it opensinto the back of the cavum epiptericum.

PROOTIC CANAL (CANALIS PROOTICUS). Firstused by Gaupp (1908) for the canal in the petrosalof the echidna Tachyglossus aculeatus, transmit-ting the prootic sinus from the cranial cavity tothe middle ear where it joins the post-trigeminalvein to form the lateral head vein. In Kryptobaa-tar, the endocranial aperture of the prootic canalis ventral to the lower margin of the subarcuatefossa; the ventral aperture is in a recess medial to


the crista parotica along with the foramen for theramus superior.

PROOTIC FORAMEN (FORAMEN PROOTICUM). Theopening in the orbitotemporal region of the chon-drocranium between the pila antotica and the oticcapsule. In extant mammals, a complete pila an-totica and, therefore, the prootic foramen arefound only in the chondrocrania of monotremes(Kuhn, 1971; Zeller, 1989). Transmitted are thefourth through sixth cranial nerves, the prooticvein (Wible and Hopson, 1995) and inferior pe-trosal sinus (Shindo, 1915) between the cavumepiptericum and the cranial cavity. This foramenis altered in adult monotremes by the resorptionof all but the base of the pila antotica (Kuhn andZeller, 1987) and the enclosure of the prootic si-nus into a separate prootic canal (Wible and Hop-son, 1995). The comparable region of the endo-cranium in Kryptobaatar is posterior and lateralto the ossified pila antotica, medial to the anteriorlamina, and anterior to the petrosal. Transmittedthrough this gap were the fourth through sixth cra-nial nerves and the inferior petrosal sinus. Theprootic sinus was enclosed in a separate canal inthe anterior lamina and petrosal.

PTERYGOID CANAL (CANALIS PTERYGOIDEUS). Inthe dog (Evans and Christensen, 1979), the pter-ygoid canal is in the suture between the pterygoidand basisphenoid and transmits the nerve and ar-tery of the pterygoid canal from the basicraniumto the posteroinferior floor of the orbit. The nerveenters the basicranial end of the canal and isformed by sympathetic fibers from the internal ca-rotid (deep petrosal) nerve and parasympatheticfibers from the greater (major) petrosal nerve; theartery is a branch of the maxillary that enters theorbital end of the canal. In Kryptobaatar, a canalresembling the pterygoid canal of the dog origi-nates from the carotid groove or canal and is en-closed between the pterygoid and alisphenoid; thelocation of the orbital aperture could not be found.Given the size of this pterygoid canal and its in-timate relationship to the carotid canal, the arteryof the pterygoid canal was present, probably as abranch of the internal carotid as occurs in somelipotyphlans (McDowell, 1958; MacPhee, 1981).Whereas both the deep petrosal nerve and the ar-tery of the pterygoid canal entered the pterygoidcanal from the carotid canal in Kryptobaatar, thegreater petrosal nerve must have entered from thecavum epiptericum; there are no tympanic aper-tures for the greater petrosal nerve that position itto enter the carotid canal. Consequently, the pter-ygoid canal of Kryptobaatar is not strictly ho-mologous with that of the dog.

SECONDARY FACIAL FORAMEN (FORAMEN FACIALIS

SECUNDARIUM). Following Wible (1990) and Wi-ble and Hopson (1993), the opening in the tym-

panic surface of the petrosal by which the hyo-mandibular branch of the facial nerve enters themiddle ear from the cavum supracochleare or ca-vum epiptericum. In Kryptobaatar, the secondaryfacial foramen appears to connect with the cavumepiptericum and has a common tympanic aperturewith the canal for the ramus inferior of the sta-pedial artery.

SPHENORBITAL FISSURE (FISSURA SPHENORBITAL-IS). The term sphenorbital fissure (orbital fissureor anterior lacerate foramen) has been used for thegap walled medially by the orbitosphenoid andlaterally by the alisphenoid or anterior lamina thattransmits the contents of the cavum epiptericumanteriorly into the orbit (Simpson, 1937; McDow-ell, 1958; Kielan-Jaworowska et al., 1986). Thenervous and vascular contents of the gap so de-fined vary dramatically among extant mammalsand may include some combination of the follow-ing: the optic, oculomotor, trochlear, ophthalmic,maxillary, and abducens nerves; the ramus in-fraorbitalis, arteria anastomotica, and ophthalmicartery; and the ophthalmic veins. In Kryptobaatar,the sphenorbital fissure lies between the orbito-sphenoid ventromedially, the frontal dorsomedi-ally, the anterior lamina laterally, and probablythe maxilla ventrally; it transmitted the trochlear,ophthalmic, and maxillary nerves, the ophthalmicveins, and probably the ramus infraorbitalis.

SPHENOPALATINE FORAMEN (FORAMEN SPHENOPA-LATINUM). In the dog (Evans and Christensen,1979), the opening in the palatine that connectsthe orbit and the nasal fossa and transmits the cau-dal nasal nerve and the sphenopalatine artery andvein, branches of the maxillary nerve, artery, andvein, respectively. The sphenopalatine foramenlies immediately dorsal to the caudal palatine fo-ramen, by which the major palatine nerve and ar-tery enter the palatine canal. In Kryptobaatar, thesphenopalatine and caudal palatine foramina thatoccur in the dog are fused into a single openingeither within the maxilla or between the maxillaand frontal. We call this common opening thesphenopalatine foramen, following, for example,Kielan-Jaworowska et al. (1986).

STYLOMASTOID FORAMEN (FORAMEN STYLOMAS-TOIDEUM). In the dog (Evans and Christensen,1979), the opening between the petrosal, the os-seous bulla, and the tympanohyal (tympanohyoid)cartilage by which the facial nerve (hyomandib-ular branch) leaves the middle ear and the stylo-mastoid artery off the posterior (caudal) auricularenters. In Kryptobaatar, only a stylomastoid notchformed by the petrosal, including the ossifiedhooklike tympanohyal, is evident. In addition tothe facial nerve, the notch probably accommodat-ed a stylomastoid artery, because that vessel iswidely distributed in extant mammals (Wible,


1987). However, whether that artery was derivedfrom the stapedial system as a ramus posterior orfrom the external carotid system is uncertain.

SUPRAGLENOID FORAMEN (FORAMEN SUPRAGLE-NOIDALE). First used by Cope (1880) for the pre-sumed venous aperture perforating the base of thezygomatic process of the squamosal from abovethat occurs in a variety of extant therians. Simp-son (1937) applied the name to the aperture inPtilodus in front of the base of the zygomatic pro-cess of the squamosal that faces anterolaterallyinto the temporal fossa; however, he noted thisopening does not correspond exactly with the su-praglenoid foramen of therians and questioned thehomology of the two. In Kryptobaatar, the supra-glenoid foramen is in the anterior lamina slightlyposterior and dorsal to the foramen masticatorium.Following Rougier et al. (1992) and Wible andHopson (1995), we reconstruct as the principalvessel in it an arterial branch to the temporalismuscle off the ramus superior of the stapedial,known as a ramus temporalis.

TRANSVERSE CANAL (CANALIS TRANSVER-SUS). Term used by Gregory (1910) for the ve-nous canal in the basisphenoid crossing the mid-line in some extant therians. In Kryptobaatar, thetransverse canal is a large channel in the sphenoidcomplex immediately in front of the hypophysealfossa, connecting the left and right sphenorbitalrecesses.

VENTRAL ASCENDING CANAL (CANALIS ASCENDENS

VENTRALIS). Following Rougier et al. (1992), theventral part of the ascending canal between itstympanic opening (equivalent to the pterygopar-occipital foramen, see Rougier et al., 1992; Wibleand Hopson, 1995) and the posttemporal canal,transporting the ramus superior of the stapedialartery. In Kryptobaatar, the ventral ascending ca-nal is horizontal, posterolaterally directed, andpresumably entirely within the anterior lamina; itstympanic opening shares a common recess withthe prootic canal. The equivalent channel can bemore vertical and either endocranial, as in lipo-typhlans, or extracranial, as in Morganucodon(Rougier et al., 1992).


INDEX OF ANATOMICAL TERMS

aqueductscochlear, 39, 55, 118vestibular, 55

arteriesaccessory palatine, 77, 114–115diploetica magna, 67, 71, 91–92, 118ethmoidal, 50, 72, 115, 118external carotid, 67–68frontal, 23, 72infraorbital, 70, 116internal carotid, 31, 36–37, 68, 89–90, 102, 115lacrimal, 72major palatine, 70, 117, 119maxillary, 68minor palatine, 70, 115, 118ophthalmic, 50–51, 58, 72, 119of pterygoid canal, 38, 72, 92, 119ramus inferior, 41, 65, 68–70, 88, 91, 115ramus infraorbitalis, 55, 68–70, 96, 119ramus mandibularis, 68ramus orbitalis, 72ramus superior, 41, 67–72, 91–92, 115ramus supraorbitalis, 23, 67–72, 92, 118ramus temporalis, 42–43, 45, 71, 91–92, 116,

120sphenopalatine, 119stapedial, 37, 38–39, 42, 67, 70, 90, 102vertebral, 67, 68, 90

auditory tube, 84, 101

bonesalisphenoid, 33–36, 85, 100anterior lamina, 36, 42–43, 55–56, 79, 100basioccipital, 48, 51–53basisphenoid, 31, 56, 98ectopterygoid, 84–85, 100ectotympanic, 99exoccipital, 47–48, 51frontal, 21–23, 79–80incus, 40, 94interparietal, 46jugal, 43, 83, 99–100lacrimal, 20–21, 78–79malleus, 40, 99mandible, 59–62maxilla, 23–28, 79–80nasal, 19–20, 76–77, 83orbitosphenoid, 31–33, 58, 79, 81palatine, 28–29, 79, 100parietal, 46, 82–83, 101petrosal, 36–43, 53–56, 93–96

premaxilla, 16–19, 72–74presphenoid, 31pterygoid, 30–31, 84septomaxilla, 16, 74–76, 100squamosal, 43–46stapes, 40, 70, 90, 94, 99, 102supraoccipital, 46–47vomer, 31, 83

canalsaccessory palatine, 114alisphenoid, 85, 91, 97alveolar, 25, 77, 116ascending, 67, 115carotid, 31, 37–38, 68, 89–90, 92, 115dorsal ascending, 43, 70–71, 115facial, 70hypoglossal, 116infraorbital, 23, 25, 70, 116mandibular, 62, 117of ?maxillary artery, 88, 115minor palatine, 27, 78, 101, 118orbitotemporal, 23, 42, 67, 70–71, 101, 118palatine, 26, 117, 119posttemporal, 43, 67, 70–72, 91–92, 118post-trigeminal, 88, 90, 115prootic, 41, 55, 65, 88, 91, 102, 118pterygoid, 38, 72, 92–93, 119for ramus inferior, 70, 88, 90, 115semicircular, 36, 39sinus, 118sphenoparietal, 97transverse, 59, 65–66, 89, 100, 120ventral ascending, 41, 70–71, 91, 115, 120

cavaepiptericum, 49, 53, 55, 70, 95, 96–98supracochleare, 55, 95

cerebellumparaflocculus, 53vermis, 63

cerebrum, 63choana, 28, 29, 30, 31, 34cochlea, 36condyles

of mandible, 61occipital, 47

crestscondyloid, 60masseteric, 60nuchal, 45, 46, 47orbitosphenoidal, 58


pterygoid, 62sagittal, 46supraorbital, 22, 46

cribriform plate, 59, 99cristae

interfenestralis, 39, 94parotica, 40, 93, 94petrosa, 53

dorsum sellae, 56, 64, 98ducts

endolymphatic, 55nasolacrimal, 59, 117perilymphatic, 39, 55, 93, 118

egg-tooth, 74

fenestraecochleae, 118sphenoparietal, 51vestibuli, 39, 70, 115

foraminafor abducens nerve, 58accessory palatine, 28, 77–78, 114–115buccinator, 36, 85, 115carotid, 37, 56, 68, 89–90, 102, 115for cochlear nerve, 55, 98of dorsal ascending canal, 43, 92, 116ethmoidal, 23, 33, 58–59, 81, 115for frontal diploic vein, 23, 80, 82, 115–116hypoglossal, 47–48, 96, 116incisive, 19, 116infraorbital, 23, 77, 116jugular, 33, 47, 51, 55, 63, 87–88, 117lacrimal, 21, 80, 117magnum, 47, 48, 63, 102major palatine, 27, 77–78, 117masticatorium, 36, 41, 86, 117mental, 59, 117metoptic, 33, 49, 50–51, 57–58, 81, 88, 96–97,

100, 117minor palatine, 27, 34, 70, 77–78, 101, 118nasal, 19–20, 76–77, 118optic, 33, 49, 50–51, 58, 72, 81, 97, 100, 118orbitonasal, 49, 50ovale, 36ovale accessorius, 36ovale inferium, 41, 86, 116perilymphatic, 39, 118for pituito-orbital vein, 56, 88, 100, 102, 116primary facial, 50, 55, 95, 118prootic, 49, 50–51, 65, 118–119pseudoptic, 50–51

pterygoparoccipital, 41, 91, 120for ramus temporalis, 43, 45, 46, 71, 92, 116secondary facial, 41, 118sphenopalatine, 25–26, 70, 80, 119sphenoparietal, 51stylomastoid, 119supraglenoid, 42, 70, 91, 119–120for vestibular nerve, 55

fossaeatlantal, 43, 47glenoid, 46hypochiasmatica, 98hypophyseal, 38, 56–57, 98, 102incudis, 40jugular, 39, 40, 47–48, 53, 93, 101masseteric, 59–60orbitonasal, 22pterygoid, of alisphenoid, 35–36, 86–87pterygoid, of mandible, 61–62stapedius, 40subarcuate, 53tensor tympani, 39, 95trigeminal, 55

foveaemasseteric, 59pterygoid, 62

gangliafacial, 49, 55, 95, 96trigeminal, 49, 55, 96

groovesethmoidal, 33internal carotid, 36–38, 68, 89, 92, 102, 115internal parietal, 118sphenopalatine, 26–27, 70stapedial, 37, 38, 70, 90, 102temporal, 60

hamulus, of pterygoid, 31, 84hiatus Fallopii, 92

jugum sphenoidale, 58, 98–99

laminaeinfracribrosa, 58, 99obturans, 36

lateral flange, 40, 70, 90, 94

meatusesexternal acoustic, 40, 45internal acoustic, 53, 55, 116–117

musclesincisivus superioris, 18


lateral pterygoid, 36, 42, 62medial masseter, pars anterior, 22, 59, 79medial pterygoid, 39, 62, 84sternomastoid, 43superficial masseter, pars anterior, 23, 24superficial masseter, pars posterior, 46temporalis, 43, 60, 70, 82tensor tympani, 39, 95tensor veli palatini, 84

nasal overhang, 20, 76nerves

abducens (VI), 49, 51, 58, 96accessory palatine, 77, 114buccal, of V, 36, 86–87, 115caudal nasal, 119cochlear, 55, 117deep petrosal, 38, 92–93, 119deep temporal, 86ethmoidal, 23, 33, 49, 59, 115, 118facial (VII), 40, 41, 49, 55, 117, 118, 119frontal, 23glossopharyngeal (IX), 117greater petrosal, 38, 93, 96, 119inferior alveolar, 117infraorbital, 25, 116internal carotid, 38, 68, 115major palatine, 26–27, 77, 117, 119mandibular (V3), 36, 41, 68, 86, 116, 117maxillary (V2), 55, 68minor palatine, 27, 77, 118oculomotor (III), 49, 51, 55, 58, 88, 97, 117ophthalmic (V1), 55, 68optic (II), 49, 51, 58, 72, 118of pterygoid canal, 38, 72, 119spinal accessory (XI), 48, 117trigeminal (V), 49, 51, 96trochlear (IV), 49, 51, 55, 96vagus (X), 117vestibular, 55, 117

notchesanterior nasal, 20, 76odontoid, 47, 48palatonasal, 78stylomastoid, 40, 119supraorbital, 22, 23, 46

occipital plane, 47olfactory bulbs, 58orbital pocket, 22–23, 79–80orbital wing, 58os carunculae, 74

pilaeantotica, 49, 50–51, 57–58, 97–98, 100metoptica, 49, 50–51, 58, 97–98, 100preoptica, 49, 50–51, 58, 97

postpalatine torus, 29prefacial commissure, 50, 53processes

caudal tympanic, of petrosal, 40, 94coronoid, 60–61facial, of lacrimal, 20, 78orbital, of palatine, 28, 79, 100paroccipital, of petrosal, 40, 94posterodorsal, of premaxilla, 17postorbital, of frontal, 82–83, 101postorbital, of parietal, 46, 82–83, 101prenasal, of premaxilla, 17, 72–74, 100rostral tympanic, of petrosal, 39

promontorium, of petrosal, 36

recessesepitympanic, 35, 40–41, 86, 94–95post-promontorial tympanic, 94scala tympani, 93–94sphenorbital, 25, 35

ridgesanterior zygomatic, 24–25intermediate zygomatic, 46orbital, 22posterior zygomatic, 46pterygopalatine, 30–31, 83–84, 101temporal, 46

sinusesanterior intercavernous, 50, 51cavernous, 63inferior petrosal, 50–51, 53, 63–64, 87–88, 117maxillary, 25, 77, 116posterior intercavernous, 51, 64prootic, 41, 50–51, 55, 63, 65, 88, 118sigmoid, 51, 53, 63, 87, 117superior sagittal, 63, 87tentorial, 63transverse, 63, 86

sphenoobturator membrane, 36sphenorbital fissure, 25, 33, 55, 70, 81–82, 119stylohyal, 36, 99subarcuate fenestration, 67sulci

chiasmatic, 58, 98facial, 41for inferior petrosal sinus, 53, 64, 87


for nasolacrimal duct, 59for pituito-orbital vein, 56, 66for prootic sinus, 55for sigmoid sinus, 51, 63

symphysis, mandibular, 61

taenia clino-orbitalis, 57thickening, of premaxilla, 19transversal elevation, 62troughs

lateral, of petrosal, 40–41pterygopalatine, 30–31, 34, 83–84, 101

tuberculum sellae, 58, 98turbinals, 59tympanohyal, 40, 119

veinscapsuloparietal emissary, 51frontal diploic, 23, 115–116internal jugular, 65, 88lateral head, 50, 64–65, 88, 118middle cerebral, 50ophthalmic, 55, 63, 96pituitary, 50, 117pituito-orbital, 50, 56, 66, 88–89, 116, 117–118post-trigeminal, 41, 65, 70, 88, 115, 118prootic, 50transverse canal, 59, 65–66, 89, 120vertebral, 64, 87

vestibule, 36, 95–96vomeronasal organ, 116

CRANIAL ANATOMY OF KRYPTOBAATAR DASHZEVEGI (MAMMALIA

Documents

Transcript of CRANIAL ANATOMY OF KRYPTOBAATAR DASHZEVEGI (MAMMALIA