osteology of the late cretaceous alvarezsauroid linhenykus ... · 2023 g street nw, washington, dc...

Osteology of the Late Cretaceous alvarezsauroidLinhenykus monodactylus from China and commentson alvarezsauroid biogeography

XING XU PAUL UPCHURCH QINGYU MA MICHAEL PITTMAN JONAH CHOINIERE CORWIN

SULLIVAN DAVID WE HONE QINGWEI TAN LIN TAN DONG XIAO and FENGLU HAN

Xu X Upchurch P Ma Q Pittman M Choiniere J Sullivan C Hone DWE Tan Q Tan L Xiao D and Han F2013 Osteology of the Late Cretaceous alvarezsauroid Linhenykus monodactylus from China and comments on alvarezminussauroid biogeography Acta Palaeontologica Polonica 58 (1) 25ndash46

The alvarezsauroid theropod Linhenykus monodactylus from the Upper Cretaceous of Inner Mongolia China is the firstknown monodactyl nonminusavian dinosaur providing important information on the complex patterns of manual evolution seenin alvarezsauroids Here we provide a detailed description of the osteology of this taxon Linhenykus shows a number of feaminustures that are transitional between parvicursorine and nonminusparvicursorine alvarezsauroids but detailed comparisons also reminusveal that some characters had a more complex distribution We also use eventminusbased treeminusfitting to perform a quantitativeanalysis of alvarezsauroid biogeography incorporating several recently discovered taxa The results suggest that there is nostatistical support for previous biogeographic hypotheses that favour pure vicariance or pure dispersal scenarios as explanaminustions for the distributions of alvarezsauroids across South America North America and Asia Instead statistically significantbiogeographic reconstructions suggest a dominant role for sympatric (or ldquowithin areardquo) events combined with a mix ofvicariance dispersal and regional extinction At present the alvarezsauroid data set is too small to completely resolve thebiogeographic history of this group future studies will need to create larger data sets that encompass additional clades

Key words Dinosauria Theropoda Parvicursorinae alvarezsauroid biogeography Treefitter dispersal vicariancesympatry Cretaceous Wulansuhai Formation Inner Mongolia China

Xing Xu [xingxuvipsinacom] Qingyu Ma [maqingyuivppgmailcom] Corwin Sullivan [csullivanivppaccn]David WE Hone [dwe_honeyahoocom] and Fenglu Han [hfl0501gmailcom] Key Laboratory of EvolutionarySystematics of Vertebrates Institute of Vertebrate Paleontology amp Paleoanthropology Chinese Academy of Sciences142 Xiwai Street Beijing 100044 ChinaPaul Upchurch [pupchurchuclacuk] and Michael Pittman [mdpittmanhotmailcom] Department of Earth Sciminusences University College London Gower Street London WC1E 6BT UKJonah Choiniere [jonahchoinieregmailcom] Department of Biological Sciences George Washington University2023 G Street NW Washington DC 20052 USAQingwei Tan [firsttansinacom] and Lin Tan Long Hao Institute of Geology and Paleontology Hohhot Inner Mongominuslia 010010 ChinaDong Xiao Department of Land and Resources Linhe Inner Mongolia 015000 China

Received 19 July 2011 accepted 1 December 2011 available online 13 December 2011

Copyright copy 2013 Xing Xu et al This is an openminusaccess article distributed under the terms of the Creative Commons Atminustribution License which permits unrestricted use distribution and reproduction in any medium provided the original auminusthor and source are credited

Introduction

With only one exception (Choiniere et al 2010) all knownalvarezsauroid theropods are restricted to Upper Cretaceousdeposits (Bonaparte 1991 Novas 1997 Naish and Dyke2004 Kessler et al 2005 Martinelli and Vera 2007 Longminusrich and Currie 2009 Salgado et al 2009) Among these LateCretaceous alvarezsauroids Asian species possess the mostspecialized bodyminusplans and are best known from the UpperCretaceous of Mongolia The Mongolian alvarezsauroid asminus

semblage includes taxa such as Mononykus olecranus fromthe Nemegt Formation (Perle et al 1993 1994) and Shuvuuiadeserti from the Djadokhta Formation (Chiappe et al 1998)whose systematic positions have been highly debated Djaminusdokhtaminusequivalent deposits in the neighboring Chinese reminusgion of Inner Mongolia (Jerzykiewicz et al 1993) containa broadly similar vertebrate fauna but few alvarezsauroidspecimens have been recovered from Inner Mongolia

In the 2008 summer field season a joint expedition teamfrom the Institute of Vertebrate Paleontology and Paleominus

httpdxdoiorg104202app20110083Acta Palaeontol Pol 58 (1) 25ndash46 2013

anthropology the Longhao Institute of Geology and Paleonminustology and the Bureau of Land and Resources of BayannurInner Mongolia prospected in the Upper Cretaceous dinominussaurminusbearing deposits of the Bayan Mandahu area in northminuscentral Inner Mongolia On August 14th two of the authors(MP and JC) discovered and collected some small theropodbones in the red beds at the Gate Locality (Jerzykiewicz et al1993) and they revisited the site and collected a few more elminusements belonging to the same specimen on August 18th Thespecimen was found on a low outwash plain east of the highcliffs at the Gate Locality and was fully exposed on a surfaceof extensively weathered red massive sandstone directly beminuslow a thick resistant sandstone layer (less than 1 m in thickminusness) with numerous calciteminusrich concretions Digging belowthe exposed bone failed to reveal additional elements

The preservation and distribution of the bones suggest thatthe specimen was originally preserved in a concretion that hadfully weathered out on the surface of the outwash plain Themajor elements of the skeleton were clustered within a 30 cmradius while some smaller pieces discovered during the initialsearch on the 14th had washed down slope to a distance of lessthan 1 m During the second site visit a wider survey revealed afew further elements up to 4 m from the skeleton A disminusarticulated and highly fragmentary turtle specimen was found~1 m from the specimen but the extensive weathering of theturtle elements suggests that they had been exposed for a muchgreater time and were not associated with the theropod Thepresence of a robust medial manual digit and a fused blockminuslike carpometacarpus identify this specimen as an alvarezminussaurid theropod but several features distinguish it from otherknown alvarezsaurids and a phylogenetic analysis places it atthe base of the derived clade Parvicursorinae (Xu et al 2011a)Furthermore only one metacarpal of this specimen possessesmanual phalanges indicating that this animal has only one phaminuslanxminusbearing digit and thus represents the first known monominusdactyl nonminusavian dinosaur This specimen was the subject of abrief description (Xu et al 2011a) however given the signifiminuscance of this taxon for understanding alvarezsauroid evolutionwe provide here a more detailed account of the osteologicalfeatures of the specimen Furthermore we provide in the presminusent paper a quantitative analysis of alvarezsauroid biogeominusgraphy incorporating information from this find

Institutional abbreviationsmdashIVPP Institute of VertebratePaleontology and Paleoanthropology Beijing China

Other abbreviationsmdashCI consistency index RI retentionindex

Systematic palaeontologyDinosauria Owen 1842Theropoda Marsh 1881Alvarezsauridae Bonaparte 1991Parvicursorinae Karhu and Rautian 1996

RemarksmdashThe Parvicursorinae were originally proposed byKarhu and Rautian (1996) but a phylogenetic taxonomic defiminusnition for the clade was first proposed by Choiniere et al(2010) which is a nodeminusbased definition (Choiniere et al2010) Linhenykus and other recently reported parvicursorinessuch as Albertonykus (Longrich and Currie 2009) and Xixiaminusnykus (Xu et al 2010) are morphologically very similar tomembers of this nodeminusbased Parvicursorinae but they are exminuscluded from the nodeminusbased clade based on the recoveredalvarezsauroid phylogeny We believe that it is not necessaryto establish a new clade and the most informative option is totreat Parvicursorinae as a stemminusbased taxon allowing for theinclusion of Linhenykus and other taxa (Xu et al 2011a b)Thus we define Parvicursorinae as the most inclusive cladeincluding Parvicursor remotus but not Patagonykus puertai

Revised Parvicursorinae diagnosismdashDorsal vertebrae opisminusthocoelous and without hypospheneminushypantrum articulationsdorsal parapophyses elevated to level of diapophyses dorsalinfradiapophyseal fossae hypertrophied posterior sacral verminustebrae with hypertrophied ventral keels anterior caudal verteminusbrae with anteriorly displaced transverse processes metacarminuspal II that is mediolaterally wider than proximodistally longand dorsoventrally shallow bears two articular facets on theproximal surface and lacks collateral ligamental fossae manminusual ungual II with ventral axial groove and without flexortubercle tibial distal end with lateral malleolus anteroposminusteriorly expanded astragalar ascending process Lminusshaped andspecialized arctometatarsalian condition with metatarsal IIIshorter than metatarsals II and IV Other possible parvicursominusrine synapomorphies include posterior dorsal vertebrae withsmall and posterodorsally oriented neural spines biconvexvertebra near posterior end of dorsal series keeled and boatminuslike sternum with median groove along midline manual phaminuslanx IIminus2 longer than IIminus1 supracetabular crest of ilium moreprominent anteriorly than posteriorly metatarsus longer thanfemur metatarsals II and IV subequal in length and pedaldigit III more slender than digits II and IV

Genus Linhenykus Xu Sullivan Pittman ChoiniereHone Upchurch Tan Xiao Tan and Han 2011aType species Linhenykus monodactylus Xu Sullivan Pittman Choiminusniere Hone Upchurch Tan Xiao Tan and Han 2011a see below

Linhenykus monodactylus Xu Sullivan PittmanChoiniere Hone Upchurch Tan Xiao Tan andHan 2011aFigs 1ndash12

Type localitymdashCoordinates in decimal degrees N4174392E10674436Bayan Mandahu ldquoGate Localityrdquo Wulansuhai Formation

Type horizonmdashCampanian Late Cretaceous (Jerzykiewicz et al 1993)

MaterialmdashThe taxon is only represented by the holotypespecimen IVPP V17608 a partial postcranial skeleton includminusing 4 partial cervical vertebrae 3 partial dorsal vertebrae 5partial sacral vertebrae 13 caudal vertebrae a fragmentary anminusterior or middle chevron part of the left scapulocoracoid a

26 ACTA PALAEONTOLOGICA POLONICA 58 (1) 2013

nearly complete sternum the distal ends of the right humerusand left ulna the left radiale the left metacarpals II and III andthe phalanges of the left digit II part of the right metacarpal IIthe incomplete right phalanx IIminus1 and complete right phalanxIIminus2 part of the left ilium the proximal end of the left pubisthe almost complete right and left femora and tibiae parts ofthe right and left astragalusminuscalcaneum complexes the nearlycomplete left metatarsus the distal end of the right metatarsalIV and all of the left pedal phalanges other than IVminus4 andIVminus5 although phalanges IIminus2 IIIminus1 and IIIminus3 are only parminustially preserved (Fig 1)

DiagnosismdashA small parvicursorine apomorphically possesminussing a transversely compressed metacarpal III without a disminustal articular surface also differs from all other parvicursominusrines in having a longitudinal ventral furrow along the entirelength of each cervical centrum diapophyseal ridges on eachcervical vertebra that extend to the posterodorsal rim ofthe centrum extremely weak ridgeminuslike epipophyses on thepostzygapophyses of the middle cervical vertebrae largepneumatic foramina in the midminusdorsal vertebrae andanteriormost caudal vertebrae whose centra are amphiminusplatyan and whose neural spines are located completely posminusterior to the neural pedicles

Description and comparisons

The specimen is small with an estimated femoral length onlyslightly exceeding 70 mm On the basis of this estimate

IVPP V17608 is only about half of the size of the holotype ofMononykus olecranus (Perle et al 1993 1994) but is considminuserably larger than that of Parvicursor remotus (Suzuki et al2002) IVPP V17608 is estimated to have a body mass of apminusproximately 450 g based on an empirical equation using femminusoral length as a predictor (Christiansen and Farintildea 2004) Allof the preserved vertebrae have completely closed neurominuscentral sutures However intercentral sutures are evident beminustween successive sacral vertebrae metacarpal III is not fusedto metacarpal II and the proximal tarsals are not completelycominusossified with the tibia Although caution is warrantedwhen relying on neurocentral fusion as a sole indicator ofontogenetic stage (Irmis 2007) the presence of other ontominusgeneticallyminusinformative fusion in the vertebral and appendiminuscular skeleton (above) suggests a relatively late ontogeneticstage for IVPP V17608 with subminusadult being more likelythan adult

Axial skeleton

The axial skeleton is represented by 4 partial cervical verteminusbrae 3 partial dorsal vertebrae 4 partial sacral vertebrae and13 caudal vertebrae Most of the preserved neural arches areincomplete and some vertebrae are only represented by brominusken centra

Cervical vertebraemdashFour cervical vertebrae are preservedbut only one is relatively complete They might represent anminusterior and middle cervical vertebrae based on the morpholminus

httpdxdoiorg104202app20110083

XU ET ALmdashALVAREZSAUROID THEROPOD FROM CHINA 27

Fig 1 Alvarezsauroid theropod Linhenykus monodactylus Xu SullivanPittman Choiniere Hone Upchurch Tan Xiao Tan and Han 2011aBayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holominustype (IVPP V17608) Skeletal silhouette showing preserved bones (missingportions shown in grey)

ogy of the ventral surfaces of the centra and the locations ofthe postzygapophyses (Fig 2)

The preserved cervical centra are strongly opisthocoeminuslous a condition also seen in Shuvuuia and Mononykus(Chiappe et al 2002) but not in Haplocheirus (Choiniere etal 2010) and Alvarezsaurus (Bonaparte 1991) Unknowncervical articular geometries for other alvarezsauroids makesit difficult to reconstruct the distribution of cervical opisthominuscoely across the alvarezsauroid tree but it appears to havebeen present by at least the Parvicursorinae node althoughcervical opisthocoely has also been proposed tentatively tobe present in Patagonykus (Novas 1996 1997 Chiappe et al2002) Future discoveries will be important in testing this hyminuspothesis The cervicals also resemble their counterparts inother alvarezsaurids in that the condyle on the anterior surminusface of each cervical centrum is distinctly smaller than the arminusticular surface The cervical centra are mediolaterally comminuspressed in such a way that their lateral surfaces are moderminusately concave and their ventral surfaces are narrow as inMononykus Parvicursor and probably Patagonykus (Novas1997) In lateral view the cervical centra are long and low asin other alvarezsauroids An oblique ridge (posterior centrominusdiapophyseal lamina) runs from the diapophysis to theposterodorsal corner of each middle cervical centrum Inventral view two sharp ridges run along the edges of thecentrum flanking a midline furrow on the narrow ventralsurface These two ridges approach each other near themidminuslength of the centrum so that the furrow widens anteriminusorly and posteriorly In Shuvuuia and Mononykus a similarfurrow is present but occupies only the anterior half of theventral surface (Chiappe et al 2002) In Linhenykus carotidprocesses appear to be present but are confluent with the anminusterior ends of the ridges Carotid processes are known in deminusrived alvarezsauroids Shuvuuia and Mononykus and someornithomimosaurs oviraptorosaurs dromaeosaurids troominus

dontids and avialans as well Subcircular pneumatic foramminusina are located immediately posterior to the parapophyses asin Alvarezsaurus and Shuvuuia Pneumatic foramina havenot been reported in Mononykus or Parvicursor The foramminusina are proportionally large (about 3 mm in diameter on a cerminusvical centrum with an estimated length of 9 mm) but this apminuspears to be partly an artifact of preservation

The neural pedicles are mediolaterally broad and dorsominusventrally low The anterior edge of each pedicle is flush withthe anterior articular surface (excluding the condyle) of thecentrum but the posterior edge is located considerably anteminusrior to the posterior articular surface The parapophyses arelow laterally projecting eminences and the diapophyses arerepresented by thin tubercles that are connected to the preminuszygapophyses by prezygodiapophyseal laminae The preminuszygapophyses are set well lateral to the centra Their articularsurfaces face nearly medially and only slightly dorsally sugminusgesting that lateral bending within the cervical column wasrather restricted The prezygapophyseal processes have astrong dorsal orientation particularly in the more posteriorcervical vertebrae in which the prezygapophyseal processesbarely protrude anteriorly beyond the intercentral articulaminustion The postzygapophyseal processes diverge stronglyaway from the midline and also have a strong dorsal orientaminustion In the anterior cervical vertebrae the postzygapophysealprocesses are located anterior to the intercentral articulationwhereas in the middle ones the postzygapophyseal processare at the level of the intercentral articulation The postminuszygapophyses each have a nearly straight medial edge and aconvex lateral edge in dorsal view as in other Asian alvarezminussaurids (Chiappe et al 2002) In Alvarezsaurus the postminuszygapophyses are paddleminusshaped due to the presence of aconvex medial edge (Bonaparte 1991) No epipophyses areapparent on the postzygapophyses of the posterior cervicalvertebrae of Linhenykus but extremely weak ridgeminuslikeepipophyses are present in the middle cervical vertebrae Theneural spines of the middle cervical vertebrae are extremelylow and very short anteroposteriorly and possess shallowinterspinous ligament fossae on their posterior margins Theneural canal is proportionally large in the cervical centra afeature of alvarezsaurids generally (Chiappe et al 2002)

Cervical ribs are not preserved but there is no evidencethat they were fused to the cervical vertebrae In Shuvuuia(IGM 100977) the ribs of the anterior cervical vertebrae arepartially fused to the diapophyses and fully fused to theparapophyses

Dorsal vertebraemdashOne of the three preserved dorsal verteminusbrae probably comes from the middle part of the dorsal regionwhile the other two are preserved in articulation with eachother and probably represent posteriormost dorsal vertebrae(Fig 3) The neurocentral sutures are all closed leaving novisible traces Each dorsal vertebra has a mediolaterally comminuspressed centrum with a sharp ventral midline ridge as inMononykus (Perle et al 1994) but in contrast to Shuvuuia(Chiappe et al 2002 Suzuki et al 2002) and Xixianykus (Xu etal 2010) in which the posteriormost dorsal centra are transminus


postzygapophysisprezygapophysis

ventral keel

parapophysis

pneumaticforamen

epipophysis

pneumatic foramendiapophyseal ridge

ventral groove5 mm

Fig 2 Cervical vertebrae of an alvarezsauroid theropod Linhenykus monominusdactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tanand Han 2011 Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Camminuspanian) holotype (IVPP V17608) Two posterior cervical vertebrae in rightlateral (A) left lateral (B) ventral (C) and dorsal (D) views

versely narrow but not ventrally ridged The articular surfacesof the dorsal centra are subminustriangular in outline unlike theovoid ones in Shuvuuia (IGM 100977) and Xixianykus (Xu etal 2010) The middle dorsal vertebra is opisthocoelous witha convex knobminuslike anterior projection that is significantlysmaller than the articular surface The condyle appears to beeven smaller relative to the anterior intercentral articular surminusface in the presacral vertebrae of other known parvicursorinessuch as Mononykus and Shuvuuia (Perle et al 1994 Chiappeet al 2002) The centrum of the middle dorsal vertebra isslightly greater in transverse width than in dorsoventral depthunlike Shuvuuia (IGM 100977) which has transversely muchwider posterior articular surface relative to the dorsoventraldepth Large openings which are probably pneumatic foramminusina are present on both sides of the centrum which terminateon the neurocentral suture and have a diameter equal to apminusproximately half the length of the centrum The left and rightforamina are continuous across the midline of the centrum butthis condition is presumably a result of postmortem damageThis is the first known example of a pneumatic dorsal vertebraamong alvarezsauroids (Chiappe et al 2002 Choiniere et al2010) The posterior dorsal vertebrae probably represent thelast and the second to last in the dorsal sequence The penultiminusmate dorsal has a biconvex centrum Biconvex dorsal verteminusbrae are also known in Xixianykus (Xu et al 2010) Monominusnykus and Shuvuuia (Chiappe et al 2002) but in Xixianykus(Xu et al 2010) and probably Mononykus and Shuvuuia(Chiappe et al 2002) it is the last dorsal vertebra that has abiconvex centrum The last dorsal has an anteriorly concavecentrum but its posterior end is broken Neither posterior dorminussal vertebra bears a pneumatic fossa or foramen on the lateralsurface of the centrum The anterior and posterior articularsurfaces of the penultimate dorsal vertebra each bear a condylethat occupies the whole area of the articular surface in contrastto the reduced anterior condyle of the middle dorsal vertebra acondition that is present in all the dorsals of Mononykus andXixianykus (Perle et al 1994 Xu et al 2010) The centrum ofthe last dorsal vertebra is considerably larger than the otherpreserved dorsal centra and is mediolaterally wider thandorsoventrally deep

The parapophyses are located on small projections at thebases of the prezygapophyseal processes and lie approximinusmately on a level with the diapophyses as in other parviminuscursorines (Chiappe et al 2002 Longrich and Currie 2009) Ahorizontal paradiapophyseal lamina (Wilson 1999) overhangsthe large pneumatic foramen on the preserved middle dorsalvertebra but is lacking in the penultimate dorsal vertebra Theneural arches of all three dorsal vertebrae are low in lateralview The prezygapophyses of the middle dorsal vertebra areconsiderably lower than the postzygapophyses which arevery close to the midline The neural spine is small and lowand posterodorsally oriented on the middle dorsal vertebra andprobably also on the posterior dorsal vertebrae though theneural spines of the posterior dorsals are not complete InMononykus and Shuvuuia (Perle et al 1994 Chiappe et al2002) the neural spines of the middle dorsals are large and

high As in other parvicursorines (Chiappe et al 2002 Longminusrich and Currie 2009) there are no hypospheneminushypantrum arminusticulations unlike the basal alvarezsauroids Haplocheirus andPatagonykus where such articulations are present (Novas1997 Choiniere 2010) The neural canal is large relative to thecentrum as in other alvarezsaurids (Chiappe et al 2002)

Sacral vertebraemdashOne isolated sacral vertebra and two setsof two fragmentary sacral vertebrae in articulation are present(Fig 4) Two of the fragmented ends fit closely together sugminusgesting that the two sets might actually represent three consecminusutive sacral vertebrae but this cannot be confirmed Thereminusfore a minimum of four and a maximum of five sacral centraare represented The isolated one is likely to be the second tolast sacral vertebra and the other three are likely to be the thirdfourth and fifth to last one For convenience we refer to thesesacral vertebrae as sacral A through sacral D with A being themost anterior in the sequence Shuvuuia and Xixianykus haveseven sacral vertebrae (Chiappe et al 2002 Xu et al 2010)but more primitive alvarezsaurids have as few as five sacrals(Novas 1996 Chiappe et al 2002) so we cannot be sure howmany were present in Linhenykus All of the preserved sacralvertebrae are opisthocoelous though sacral D is only veryweakly so The opisthocoelous condition appears to be presentin at least the first sacral vertebra of Parvicursor (Karhu andRautian 1996 Chiappe et al 2002) Most other alvarezsauridsincluding Patagonykus and Xixianykus have procoelous saminus



neuralspine

postzygapophysis

parapophysiscondyle

neural spineprezygapophysis

ventral keel

neural spine

condyle

condyleventral keel

condyle

diapophysis

condyle5 mm

pneumatic foramen

Fig 3Dorsal vertebrae of an alvarezsauroid theropod Linhenykus monominusdactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tanand Han 2011a Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Camminuspanian) holotype (IVPP V17608) A Middle dorsal vertebra in left lateral(A1) right lateral (A2) posterior (A3) ventral (A4) and dorsal (A5) viewsB Two posterior dorsal vertebrae in lateral (B1) and ventral (B2) views

cral vertebrae (Novas 1996 Chiappe et al 2002 Xu et al2010) All of the preserved sacral centra are laterally comminuspressed as in other alvarezsaurids (Chiappe et al 2002) andthe compression becomes stronger posteriorly However theanterior ends of sacrals B and C are only slightly compressedso that they form a prominent anterior margin relative to thelaterally compressed central main body This feature is presentto a degree in the first sacral vertebrae of other alvarezsauridssuch as Xixianykus (Xu et al 2010) but otherwise is not charminusacteristic of the group The mediolateral widthdorsoventralheight ratio of the anterior end of sacral B is about 19 Relaminustively wide anterior and middle sacral centra are also known inXixianykus (Xu et al 2010) The anterior articular surfaces ofthe four sacral vertebrae are semiminuscircular in outline with themediolateral width much greater than the dorsoventral heightThe posterior articular surfaces of sacrals A and B are subminustrapezoidal in outline with the mediolateral width again muchgreater than the dorsoventral height However the posteriorsurface of sacral D is triangular in outline with the mediominuslateral width much smaller than the dorsoventral height Thereappears to be a groove on the ventral surface of sacral Awhich extends posteriorly onto the ventral surface of sacral BA ventral groove is also known in Alvarezsaurus (Bonaparte1991) Xixianykus (Xu et al 2010) and an unnamed alvarezminussaurid from Mongolia (Longrich and Currie 2009) though thelength of the groove varies among these taxa The preservedanterior half of sacral C is ventrally keeled whereas the venminustral surface of sacral D is strongly keeled over all but the

anteriormost part of its length The sacral rib of sacral B arisesfrom nearly the whole length of the lateral surface of thecentrum but is slightly offset towards the anterior end of thevertebra The sacral fenestrae appear to be large and the sucminuscessive sacral ribs appear not to be fused to form a largelamina as in Xixianykus (Xu et al 2010) In contrast to thesmooth transition from the lateral surface of the centrum to theventral surface of the sacral rib a distinct notch is visible in anminusterior view between the lateral edge of the anterior end of thecentrum and the ventral surface of the sacral rib

Caudal vertebraemdashThere are 13 caudal vertebrae preservedsome of which are fragmentary (Figs 5 and 6) We have numminusbered these caudal vertebrae in an anatomically anteroposminusterior order (C1 through C13) although they may not repreminussent a continuous series Position in the numbering sequencewas determined on the basis of size in addition to the relativedevelopment and positioning of such features as neural spinestransverse processes and ventral keels C1ndash3 probably repreminussent the true first through third caudal vertebrae C4ndash6 possiminusbly come from the middle region of the anterior caudal seriesC7ndash9 from the posterior region of the anterior caudal seriesC10ndash12 from the middle caudal series and C13 from the anteminusrior region of the posterior caudal series

All of the preserved caudal vertebrae are mediolaterallycompressed and the compression is greater in the more anteminusrior caudal vertebrae C1 displays an extremely weak opiminussthocoelous condition C2 has a slightly convex anteriorintercentral articular surface and a prominent condyle on the


5 mm

Fig 4Sacral vertebrae of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tan andHan 2011 Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Presumed fourth and fifth last sacral vertebraein left lateral (A1) right lateral (A2) ventral (A3) and dorsal (A4) views B Presumed third and fourth last sacral vertebrae in right lateral (B1) left lateral(B2) and dorsal (B3) view C Presumed second last sacral vertebra in right lateral (C1) anterior (C2) posterior (C3) and ventral (C4) views

posterior intercentral articular surface and all of the othercaudal vertebrae are procoelous In other alvarezsaurids inminuscluding Alvarezsaurus and Patagonykus all of the caudalvertebrae are procoelous (Novas 1996 Chiappe et al 2002)Most of the preserved caudal vertebrae with the exception ofC1ndash3 are marked by a longitudinal furrow that runs alongthe entire ventral surface of the centrum and is bordered latminuserally by two ventral keels This feature is also known insome other alvarezsaurids such as Shuvuuia (Chiappe et al2002) All of the preserved neural arch pedicles are posiminustioned anteriorly on the centra as in other alvarezsaurids(Chiappe et al 2002) The pedicles of the anterior caudals(C1ndash9) are each restricted to the anterior half of the correminussponding centrum but the pedicles of the more posteriorcaudal vertebrae each cover most of the correspondingcentrumrsquos length The intercentral foramina are large due tothe strong dorsal orientation of the zygapophyses

C1 preserves only the centrum It is subminusequal in anterominusposterior length to the second last sacral vertebra but is muchgreater in central height The centrum is strongly compressedlaterally and the ventral margin forms a sharp keel that renminusders the anterior and posterior intercentral articular surfacestriangular in outline C2 and C3 slightly exceed C1 in anterominus

posterior length C2 and C3 each have a sharp ventral keel likeC1 but only the anterior intercentral articular surface of C2shares the triangular outline of the intercentral articular surminusfaces of C1 Caudal vertebrae with sharp ventral keels are alsopresent in Shuvuuia Parvicursor Xixianykus and Alvarezminussaurus but there are only two such caudals known in Shuminusvuuia and only one each in Parvicursor Xixianykus andAlvarezsaurus (Chiappe et al 2002 Xu et al 2010) Theposterodorsallyminusoriented neural spine of C2 is short anterominusposteriorly and low dorsoventrally The neural spine is locatedcompletely posterior to the neural pedicles lying at the level ofthe postzygapophyses The same condition is seen in someother parvicursorines such as Shuvuuia (Chiappe et al 2002)though in other taxa the neural spine is slightly more anteriorlylocated C3 exhibits transverse processes located at the basesof the prezygapophyseal processes The long axis of the crosssection of each transverse process is slightly oblique (anterominusdorsallyminusposteroventrally) As in Xixianykus (Xu et al 2010)and Shuvuuia (Chiappe et al 2002) the bases of the transverseprocesses are anteriorly located even projecting slightly beminusyond the anterior articular surface of the centrum The preminuszygapophyseal processes have a strong dorsal orientation andextend considerably beyond the intercentral articulation



neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle

Fig 5Caudal vertebrae 1ndash3 of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tanand Han 2011a Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Caudal vertebra 1 in left lateral (A1) andright lateral (A2) views B Caudal vertebrae 2 and 3 in anterior (B1) left lateral (B2) right lateral (B3) ventral (B4) and dorsal (B5) views

C4ndash13 are similar to each other in general morphologythough there are minor structural variations among the indiminusvidual vertebrae As in parvicursorines but unlike in Alvarezminussaurus and Haplocheirus (Bonaparte 1991 Choiniere 2010Choiniere et al 2010) the vertebrae gradually becomeshorter anteroposteriorly and lower dorsoventrally along thecaudal series although C10 is longer than the preceding verminustebra The anterior intercentral articular surfaces of C3ndash13are subminustrapezoidal in outline The neural arch pedicles exminustend further posteriorly on the centra of the posterior caudalvertebrae than on the centra of the anterior ones Transverseprocesses are present in all preserved caudal vertebraethough they become greatly reduced in size posteriorly Theposition of the transverse processes becomes more posteriorand slightly more ventral in the more posterior caudal verteminusbrae The transverse processes are located close to the anteminusrior intercentral articular surface in C4 but are close to theposterior articular surface in C13 These processes areposterolaterally oriented throughout the caudal series as inShuvuuia and Alvarezsaurus (Chiappe et al 2002) In Parviminuscursor the transverse processes have been interpreted asanterolaterally oriented (Karhu and Rautian 1996) The longaxis of the cross section of each transverse process remainsclearly oblique (oriented anterodorsallyminusposteroventrally) as

far back as C9 but becomes nearly horizontal in C10ndash13The prezygapophyseal processes retain a strong dorsal orienminustation throughout the series and the articular facets of boththe preminus and postzygapophyses are steeply inclined throughminusout the caudal sequence The prezygapophyseal processesbecome shorter in the posterior caudals but extend beyondthe intercentral articulation even in C12 the last caudal verminustebra in which prezygapophyses are preserved The postminuszygapophyses assume a more posterior location in the posteminusrior caudal vertebrae projecting just beyond the intercentralarticular surface in C13 The neural spines of C4ndash13 are posminusteriorly positioned occurring at the level of the postzygapominusphyses as in C2 and C3 Passing posteriorly along the seriesthe neural spines become lower dorsoventrally but longeranteroposteriorly In C13 the neural spine is still visible in theform of a sharp low ridge and extends over about twominusthirdsof the total length of the centrum The posterior margin of theneural spine of C13 is much thicker transversely than the anminusterior margin

A fragmentary chevron is preserved between C7 and C8A proximal bridge connects the left and right halves of thechevron dorsal to the hemal canal and the anterior margin ofthe proximal end is slightly bifurcated The distal portion ofthe chevron is broken away


5 mm

Fig 6Caudal vertebrae 4ndash13 of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tanand Han 2011a Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Vertebrae 4ndash6 in left lateral (A1) right latminuseral (A2) ventral (A3) and dorsal (A4) views B Vertebra 7 in ventrolateral (B1) dorsolateral (B2) ventral (B3) and posterior (B4) views C Vertebrae 8 and9 in left lateral (C1) right lateral (C2) ventral (C3) and dorsal (C4) view D Vertebrae 10ndash12 in right lateral (D1) left lateral (D2) and ventral (D3) viewsE Vertebra 13 in right lateral (E1) left lateral (E2) ventral (E3) anterior (E4) and posterior (E5) views

Pectoral girdle and sternum

Scapula and coracoidmdashUnlike those of other alvarezminussauroids the scapula and coracoid of V17608 are fused tominusgether although the evidence for this comes exclusivelyfrom a small preserved portion of the left scapulocoracoid(Fig 7A) The preserved scapulocoracoid is reminiscent ofthe scapula and coracoid of basal alvarezsauroids (Choiniere2010 Choiniere et al 2010) rather than those of otherparvicursorines (Chiappe et al 2002) in having a weakly deminusveloped glenoid lip that protrudes only slightly beyond theventral edge of the scapular blade In most other parviminuscursorines (Chiappe et al 2002) the glenoid lip is muchmore prominent The small preserved portion of the acrominusmion process suggests an abrupt dorsal expansion of thescapular blade The glenoid fossa faces posteroventrally as inother alvarezsauroids (Chiappe et al 2002 Choiniere et al2010) The part of the scapular blade adjacent to the glenoidis convex laterally and flat medially with a thick somewhatrounded ventral margin and a transversely narrow dorsalmargin that bears a sharp longitudinal ridge The broken surminusface of the scapular blade reveals a hollow interior thoughthis is possibly due to erosion The coracoid is representedonly by a small piece of bone fused to the scapula in theglenoid region and little can be said about its morphology

SternummdashThe sternum is a single fused element lacking anytrace of a midline suture (Fig 7B1 B2) It appears to be almostcomplete though its left margin is slightly broken In generalform it is a small stout bone tapering to a broad posteriorpoint The sternum is only slightly anteroposteriorly longerthan it is transversely wide whereas in other Asian parviminuscursorines this element is considerably longer than wide (Perleet al 1994 Chiappe et al 2002) The dorsal surface of the sterminusnum is transversely concave a feature also seen in other Asianalvarezsaurids (Perle et al 1994 Chiappe et al 2002) Theventral surface of the sternum is pitted and bears a moderatelydeveloped carina (sternal keel) that bifurcates anteriorly intotwo lateral crests enclosing a median groove This region isslightly eroded in IVPP V17608 so that the degree of develminus

opment of the crests and groove cannot be fully determinedThe carina is much weaker than in other parvicursorines (Perleet al 1994 Chiappe et al 2002) representing little more than asharp line of inflection between the right and left halves of thesternum The maximum depth along the midline of the sterminusnum is less than 30 of the maximum width of the sternumwhereas the two measurements are subequal in other parviminuscursorines The medial groove is however proportionallylonger than in other parvicursorines (Perle et al 1994Chiappe et al 2002) The anterior margin of the sternum isconvex without a midline indentation as seen in Mononykus(Perle et al 1994) As in Shuvuuia and Mononykus (Perle et al1994 Chiappe et al 2002) the sternum lacks facets for rib arminusticulation If ossified sternal ribs were present they evidentlyleft no impressions on the bony sternum

Forelimb

HumerusmdashOnly the distal end of a right humerus is preserved(Fig 8A1ndashA3) However it is clear that the distal end curvesanteriorly relative to the shaft in lateral view a feature alsoseen in Mononykus and a variety of other theropods includingmore primitive tetanurans such as Baryonyx The shaft is trunminuscated at a point 7 mm from the distal end revealing a subminusovalcross section A large condyle occupies the whole medial sideof the distal end so that there is no distinct entepicondyle ofthe kind present in Patagonykus (Novas 1997) The distalcondyle projects mainly anteriorly and is subtriangular in anminusterior view as in Mononykus (Perle et al 1994) although inMononykus the condyle extends further proximally than inLinhenykus The articular surface of the distal condyle appearsto continue a long way onto the distal surface of the humerusProximal to the distal condyle the shaft of the humerus is relaminustively flat The condyle is also anteriorly prominent close tothe medial side of the humerus and is subminustriangular in outlinein distal view On the distal surface of the humerus thecondylersquos lateral border is defined by an oblique crest that prominusjects further distally than the main part of the condyle In distalview a small posterior notch separates the crest from theanteroposteriorly expanded main body of the condyle as inMononykus (Perle et al 1994) The posterior surface of thedistal end is only slightly concave a condition intermediatebetween those seen in other Asian alvarezsaurids and in Pataminusgonykus (Chiappe et al 2002)

UlnamdashA long bone fragment is tentatively identified as thedistal end of the left ulna (Fig 8B1ndashB3) The shaft brokenthrough approximately 10 mm from the distal end is subminuscirminuscular in cross section In anterior or posterior view the shaftcurves medially The shaft expands anteroposteriorly towardthe distal end close to which is a shallow fossa on the medialsurface As in other alvarezsaurids (Chiappe et al 2002) thedistal end bears a large carpal trochlea on the posteromedialcorner of which is a distinct triangular articular facet that exminustends further proximally on the medial surface than does themain part of the trochlea In contrast to other alvarezsaurids(Chiappe et al 2002) there appears to be an indentation in



sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

Fig 7 Left scapulocoracoid and sternum of an alvarezsauroid theropodLinhenykus monodactylus Xu Sullivan Pittman Choiniere Hone UpchurchTan Xiao Tan and Han 2011a Bayan Mandahu (ldquoGate Localityrdquo) LateCretaceous (Campanian) holotype (IVPP V17608) A Left scapulocoracoidin lateral view B Sternum in ventral (B1) and anterior (B2) views

the middle of the distal trochlear surface between moreprominent anterior and posterior areas

ManusmdashThe preserved manual elements (Fig 9) include theleft radiale left metacarpals II and III left digit II incompleteright metacarpal II incomplete right phalanx IIminus1 and comminusplete right phalanx IIminus2 Note here we refer to the digits of thetetanuran hand as IIminusIV (Kundraacutet et al 2002 Xu et al 2009)rather than as IndashIII as preferred by many previous workers(Gauthier 1986 Chatterjee 1998 Wagner and Gauthier 1999Vargas and Fallon 2005 Vargas et al 2008 Tamura et al2011) All of the elements belonging to the left manus and carminuspus appear to have weathered out of a single nodule in whichthey were preserved close to a state of natural articulation

The left radiale is a small ovoid bone with a maximum diminusameter of about 2 mm No unfused proximal carpal elementshave previously been reported in any alvarezsaurid (Chiappeet al 2002) What appears to be the proximoventral surface ofthe radiale bears a distinct fossa A separate ldquosemilunaterdquo carminuspal is absent but is probably fused to metacarpal II as sugminusgested by previous studies (Chiappe et al 2002 Choiniere2010 Choiniere et al 2010) This inference is supported bythe close topological resemblance between the middle part ofthe proximal articular surface of the main carpometacarpal elminusement (formed by metacarpal II and the ldquosemilunaterdquo carpal inLinhenykus) which is transversely trochleated and stronglyconvex in dorsal or ventral view and the corresponding surminusface of the ldquosemilunaterdquo carpal in many less specializedtetanurans (Chure 2001) However there is no visible suturebetween the ldquosemilunaterdquo carpal and metacarpal II suggestingthat fusion occurred very early in ontogeny

The main carpometacarpal element is proximodistallyshort mediolaterally broad dorsoventrally shallow and subminuspentagonal in dorsal or ventral outline It is proportionallymuch narrower transversely than its counterpart in Mononykus(lengthwidth ratio about 080 in Linhenykus compared toabout 060 in Mononykus) Three articular facets are presenton the proximal surface of the main carpometacarpal elementThe middle and lateral facets appear to correspond to the ldquomeminusdial facetrdquo and ldquocentral facetrdquo of Chiappe et al (2002) a termiminus

nology we adopt here (the lateral facet defined by Chiappe etal [2002] is formed exclusively by metacarpal IV which isprobably absent in Linhenykus) However a distinctive accesminussory facet is present medial to the medial facet in IVPPV17608 and an equivalent surface is also present in Monominusnykus but was not separately considered by Chiappe et al(2002) We refer to this facet as the ldquoaccessory medial facetrdquoThe accessory medial facet comprises a relatively deep fossaand faces more medially than proximally It is separated fromthe medial facet by a dorsoventral constriction of the proximalsurface of the main carpometacarpal element The medialfacet is a transversely grooved elongate trochlea in proximalview and represents the area that corresponds to the proximalsurface of the ancestral semilunate carpal This trochlear strucminusture is obliquely oriented tilting in a ventromedialminusdorsominuslateral direction so that the groove extends slightly onto thedorsal surface of the main carpometacarpal element In dorsalview the medial facet is proximally convex As in Monominusnykus the proximal articular surface is dorsoventrally conminusstricted between the medial and central facets (Perle et al1994) The central facet is a large distinct fossa Interestinglydespite the almost identical morphology of the proximal articminusular surface of the main carpometacarpal in Linhenykus andMononykus this concave central facet has a straight dorsalborder and a convex ventral border in Linhenykus while theopposite is true in Mononykus (Perle et al 1994)

The distal end of metacarpal II bears two prominent conminusdyles separated by a broad wellminusdeveloped groove Collatminuseral ligamental fossae are absent as in other alvarezsauridsincluding Patagonykus (Novas 1997) The medial condyle ofthe distal end is more dorsally prominent than the lateralcondyle However the lateral condyle extends significantlymore distally than the lateral one as in Haplocheirus Pataminusgonykus and many other nonminusavian theropods (Choiniere2010 Choiniere et al 2010) whereas in Mononykus the disminustal condyles are approximately equal in size (Chiappe et al2002) The medial condyle is obliquely oriented and its dorminussal surface is transversely narrow The lateral condyle haslimited dorsal exposure and is transversely broad This conminusdyle is located at the lateral edge of the bone so that there isno accessory process situated proximolateral to it as inMononykus (Perle et al 1994) The entire lateral surface ofmetacarpal II forms a sharply defined flat articular facet formetacarpal III as in Mononykus and Shuvuuia (Perle et al1994 Chiappe et al 2002)

Metacarpal III is completely preserved but is not incorpominusrated into the main carpometacarpal block as in Mononykus(Perle et al 1994) Metacarpal III is much shorter and moreslender than metacarpal II Its proximal articular surface issubminustriangular in outline predominantly convex and tilted toface somewhat laterally The ventromedial corner of theproximal surface bears a small depression that probably conminustributes to the lateral portion of the central facet describedabove Most of the shaft of metacarpal III is subminustriangular incross section The dorsal margin is sharply ridged along itsentire length The distal end of metacarpal III is strongly


Fig 8 Right humerus and left ulna of an alvarezsauroid theropod Linheminusnykus monodactylus Xu Sullivan Pittman Choiniere Hone UpchurchTan Xiao Tan and Han 2011 Bayan Mandahu (ldquoGate Localityrdquo) LateCretaceous (Campanian) holotype (IVPP V17608) A Right humerus inanterior (A1) posterior (A2) and medial (A3) views B left ulna in anterior(B1) lateral (B2) and posterior (B3) views Scale bars 5 mm

compressed transversely so that the distal end is a plateminuslikestructure with a considerable ventral expansion The distalend is sharply truncated in lateral view The dramatic reducminustion of the distal articular surface to a thin vertical edge sugminusgests that no phalanges are associated with this digit In conminustrast Mononykus and Shuvuuia have wellminusdefined condylardistal articulations on the third and fourth metacarpals (Perleet al 1994 Chiappe et al 2002) which are fused to metacarminuspal II and distal carpals to form the carpometacarpus InShuvuuia a complete set of manual digits III and IV areknown (Suzuki et al 2002) These data suggest that Monominusnykus has phalanges on digits III and IV that were not preminusserved in the holotype

Manual phalanx IIminus1 is a large robust element approximinusmately equal in thickness to the humerus It has a maximumlengthwidth ratio of about 24 compared to a value of 15for the equivalent bone in Mononykus (Perle et al 1994Chiappe et al 2002) Phalanx IIminus1 is expanded transverselyat the proximal end particularly in the lateral direction Themaximum widthdepth ratio is about 11 and a moderatelydeveloped dorsal process projects from the dorsolateral corminusner of the proximal articulation There is also a correspondminusing dorsally directed medial process separated from the

lateral one by an intervening depression The ventral surminusface of the phalanx bears ridges that extend along the meminusdial and lateral borders The lateral ridge is much sharperthan the medial one and the two are separated by a broadaxial furrow a feature unique to alvarezsauroids (Novas1996 Choiniere 2010 Choiniere et al 2010) The dorsalsurface is narrowest at its distal end in contrast to the transminusversely widened distal articular surface seen in Mononykus(Perle et al 1994) However in Linhenykus the ventral partof the distal articulation is considerably wider than the dorminussal part The distal articulation is strongly ginglymoid withtwo distinct hemicondyles Both hemicondyles are narrowtransversely and project further ventrally than dorsally butthe medial hemicondyle is narrower than the lateral oneCollateral ligamental fossae are present on both sides Bothare large and shallow and the lateral fossa is relatively dorminussal in position The distal articular surface extends well ontothe dorsal surface of the phalanx and a distinct extensorfossa is present immediately proximal to the distal articulaminustion A poorly developed flexor fossa is seen on the ventralsurface of the element

Manual ungual phalanx IIminus2 is the longest manual elementbut is nevertheless proportionally smaller than in other parviminus



radialemedial facet

radiale

radiale

radiale

radiale

accessory medial facet

extensor fossa

distal end of metacarpal III metacarpal IIIphalanx II-1

centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)

Fig 9 Manus of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tan and Han 2011aBayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Right manus in dorsal (A1) and dorsolateral (A2) viewsB Right main carpometacarpal element in dorsal (B1) ventral (B2) proximal (B3) medial (B4) lateral (B5) and distal (B6) views C Left maincarpometacarpal element in dorsal (C1) proximal (C2) and distal (C3) views D Metacarpal III in lateral (D1) and medial (D2) views E Right manual phaminuslanx IIminus1 in ventral view F Left manual phalanx IIminus2 in lateral (F1) proximal (F2) and ventral (F3) views

cursorines (Chiappe et al 2002 Suzuki et al 2002) It is lessthan 12 times as long as phalanx IIminus1 but in Mononykus theungual is nearly twice the length of the proximal phalanx(Perle et al 1994) In Linhenykus phalanx IIminus2 is robust andhas a stronger curvature than in Mononykus (Perle et al 1994)assuming the proximal articular surface is oriented verticallyin both taxa The transversely broad proximal articular surfaceis divided into two nearly equal facets by a prominent verticalmedian ridge A flexor tubercle is absent as in all alvarezminussaurids in which phalanx IIminus2 is known (Novas 1996 Longminusrich and Currie 2008) Instead the ventral surface is flattenedand a distinct but very shallow fossa is present on the ventralsurface at a level just distal to the proximal ends of the groovesfor the keratinous sheath From this fossa a shallow short andnarrow groove extends along the midline of the ventral surminusface The lateral and medial sheath grooves are proximallydeep and curve ventrally at their proximal ends each forminga distinct notch in the margin of the proximal part of the venminustral surface These notches are homologous to the closed fominusramina seen on the ventral surface of phalanx IIminus2 in Monominusnykus and Shuvuuia (Longrich and Currie 2009)

Pelvic girdle

IliummdashThe ilium is represented only by the middle part ofthe left element including part of the preacetabular processand almost all of the pubic peduncle (Fig 10A1ndashA3) Thepreacetabular process is nearly vertical in orientation butpart of the iliac blade located dorsal to the pubic peduncle isinclined so that the lateral surface of the blade also facessomewhat dorsally as in other parvicursorines (Xu et al2010) In lateral view the supraacetabular crest extends anteminusriorly to the level of the pubic peduncle and overhangs theacetabular fossa The crest is more prominent anteriorly thanposteriorly as in Mononykus (Perle et al 1994 Longrich andCurrie 2009) In ventral view the supraacetabular crestrecedes into the anteroventrallyminusoriented pubic pedunclewhich is more ventral and less anterior than is the case inMononykus (Perle et al 1994) The peduncle is greatly elonminusgated anteroposteriorly and narrow mediolaterally appearminusing almost ridgeminuslike in ventral view A cuppedicus fossa isabsent as in other alvarezsaurids (Xu et al 2010)

PubismdashOnly the proximal part of the left pubis is present(Fig 10B) As in other alvarezsaurids (Hutchinson andChiappe 1998) this part of the bone is strongly laterally comminuspressed In contrast to other parvicursorines (Hutchinson andChiappe 1998) the proximal articular surface is subtrianminusgular in outline rather than kidneyminusshaped but this morpholminusogy results at least partly from damage to the anteromedialpart of the proximal surface The articular surface for theilium appears to be concave and is much longer than wide Itseems more proximally (as opposed to proximoposteriorly)oriented than is the case in Mononykus and Patagonykus(Perle et al 1994 Novas 1997) Its lateral margin is not conminuscave in contrast to other parvicursorines (Hutchinson andChiappe 1998) The acetabular fossa of the pubis is slightlyinclined ventrolaterally Its posterolateral corner is built out

into a distinct laterally expanded process Anteriorly thepubis differs from those of other parvicursorines in lacking apreacetabular tubercle (Hutchinson and Chiappe 1998) Thepubic ischial peduncle is proximodistally long but mediominuslaterally very thin The dorsal part of the articular surface forthe ischium is transversely wide and markedly concavewhereas the ventral part is strongly compressed and slightlyconvex The lateral surface of the proximal part of the pubisis flattened and overhung by the laterally expanded processat the posterolateral corner of the acetabular fossa Althoughthere are some morphological differences between the pubisof Linhenykus and that of other parvicursorines the close reminussemblance of the element in Linhenykus and other parviminuscursorines suggests that Linhenykus also had strong opisthominuspuby as in other parvicursorines (Chiappe et al 2002)

Hindlimb

FemurmdashBoth femora are preserved but neither is complete(Fig 11A1ndashA6 B1ndashB3) The femur is estimated to be slightlymore than 70 mm long A wellminusdeveloped trochanteric crestappears to be present but the part of the crest formed by thelesser trochanter is missing Consequently it is uncertainwhether the greater and lesser trochanters are completelyfused to each other without any intervening notch as in otherparvicursorines (Chiappe et al 2002) The femoral head ismedially directed and is perpendicular to both the femoralshaft and the trochanteric crest The femoral head differsfrom that of Mononykus in being proportionally more stoutand in being separated from the shaft by a slight neck As inMononykus (Chiappe et al 2002) however the femoral headand trochanteric crest give the proximal end of the femur anLminusshaped outline The proximolateral margin of the femur issmooth and slightly concave without any trace of a posteriortrochanter The femoral shaft is bowed anteriorly Threeridges originate from the proximal end extending along theposterolateral posteromedial and anteromedial edges of the


supracetabular crest

acetabular fossa

supracetabular crestpubic peduncle

5 mm

Fig 10 Pelvic elements of an alvarezsauroid theropod Linhenykus monominusdactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan XiaoTan and Han 2011a Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous(Campanian) holotype (IVPP V17608) A Left ilium in lateral (A1) dorsal(A2) and ventral (A3) views B Left pubis in lateral view



femoral head

pneumaticforamen

lateral condylemedial condyle

fibularcrest

ascendingprocess

ascendingprocess

10 mm

Fig 11 Femora and tibiotarsi of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tan andHan 2011a Bayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Right femur in proximal (A1) anterior (A2) posteminusrior (A3) lateral (A4) medial (A5) and distal (A6) views B Left femur in anterior (B1) posterior (B2) and lateral (B3) views C Left tibiotarsus in anterior (C1)posterior (C2) medial (C3) and lateral (C4) views D Right tibiotarsus in anterior (D1) posterior (D2) and medial (D3) views E Left astragalus in anterior view

femoral shaft The posterolateral ridge is sharp and extendsfor more than half of the femoral length whereas the posterominusmedial ridge is the weakest of the three The anteromedialridge is the most prominent but also the shortest and its distalend curves posteriorly onto the medial surface of the femurLateral to the distal part of the posteromedial ridge lies a disminustinct 3 mm groove containing an elongated foramen that isprobably homologous to a similarly positioned foramen inMononykus (Perle et al 1994) No fourth trochanter is presminusent in Linhenykus or Parvicursor whereas an incipientfourth trochanter is present in a juvenile specimen of Shuminusvuuia and a weak fourth trochanter is present in all otheralvarezsaurids in which the proximal femoral shaft is preminusserved (Chiappe et al 2002 Xu et al 2010) Posterior andparallel to the anteromedial ridge is a long wide and shallowgroove The distal part of this groove widens slightly to forma small fossa The mediolateral width of the femur increasesconsiderably toward the distal end of the bone The anteriorsurface near the distal end is transversely convex over its latminuseral half and slightly concave over its medial half The posteminusrior surface of the distal end of the femur forms a large widepopliteal fossa bordered on either side by medial and lateralridges that extend to the distal condyles The medial ridge issharper than the lateral Unlike in Mononykus but resemminusbling the condition seen in Parvicursor and Patagonykus(Chiappe et al 2002) the popliteal fossa is widely open disminustally In Xixianykus the popliteal fossa is partially open disminustally (Xu et al 2010) The medial condyle is transversely narminusrow in distal view and subminustriangular in posterior view Thelateral condyle has three projections an ectocondylar tuberprojecting posteriorly a weak coneminusshaped distal projection(present in all alvarezsauroids) and a weak ridgeminuslike lateralprojection oriented almost anteroposteriorly possibly anapomorphic feature for Linhenykus

TibiamdashBoth tibiae are almost completely preserved (Fig11C D) The tibia is about 14 times as long as the femur Theproximal end appears to be subminustrapezoidal in proximal viewand is estimated to be only slightly longer anteroposteriorlythan transversely The proximal articular surface is inclinedanterolaterally The cnemial crest extends distally along thetibial shaft for about 14 mm The fibular condyle is anterominusposteriorly long and appears to bear an accessory crest anteriminusorly unlike the condition in other parvicursorines (Chiappe etal 2002 Xu et al 2010) As in Mononykus and Parvicursor(Chiappe et al 2002) the medial side of the posterior condyleforms a proximally directed eminence The fibular and posteminusrior condyles are separated from each other by a relativelydeep posterior notch A fossa is present on the tibial shaft justdistal to the notch between the two condyles and is partlyoverhung by them The 18 mm long fibular crest is narrowlyseparated from the fibular condyle by a 3 mm gap The crest issharp and recedes at both ends A shallow groove with a disminustinctly deepened distal end lies posterior and parallel to the fibminusular crest on the lateral surface of the tibial shaft The tibialshaft is subminuscircular in cross section in the midshaft region A20 mm long weak ridge arises near the midpoint of the tibiarsquos

length and runs close to the medial edge of the anterior surminusface The distal part of the tibial shaft has a flat posterior surminusface whereas the anterior surface has a slight transverse conminuscavity to accommodate the ascending process of the astraminusgalus A short distomedially directed crest is present along thedistalmost part of the anteromedial margin of the tibia and along sharp ridge separates the anterior and lateral surfaces ofthe distal end The lateral portion of the distal end is relativelyrobust being subequal in anteroposterior thickness to the meminusdial portion The distal end of the right tibia has a wellminuspreminusserved lateral surface and there is no indication that the fibulareached this part of the tibiotarsus

Astragalus and calcaneummdashThe astragalusminuscalcaneum comminusplex is represented only by pieces of bone attached to the distalends of both tibiae (Fig 11C1 D1 E) The astragalusminuscalcaminusneum complex is tightly articulated with the tibia and the lefttibiotarsus even shows cominusossification among the three eleminusments The main body of the astragalus is poorly preserved inboth the left and right hindlimbs As in Mononykus (Chiappeet al 2002) the laminar ascending process is approximatelyLminusshaped in anterior view due to a strong indentation on themedial side The ascending process arises only from the lateralhalf of the body of the astragalus so that the medial edge of theascending process is close to the midline of the tibial shaft Ashallow anterior fossa is present at the base of the ascendingprocess

PesmdashThe preserved pedal elements include nearly completeleft metatarsus distal end of right metatarsal IV and otherprobable right metatarsal fragments left pedal phalanges Iminus1and Iminus2 left pedal phalanx IIminus1 partial IIminus2 and complete IIminus3partial left pedal phalanges IIIminus1 and IIIminus3 complete IIIminus2 andcomplete left pedal ungual IIIminus4 left pedal phalanges IVminus1IVminus2 IVminus3 and partial IVminus4 in articulation (Fig 12)

The metatarsus is 73 mm long and its transverse width isless than 5 mm in the midshaft region As in other parviminuscursorines (Chiappe et al 2002 Longrich and Currie 2009Xu et al 2010) the metatarsus is longer than the femur andexhibits a specialized arctometatarsalian condition in whichmetatarsal III terminates well distal to the proximal end of themetatarsus as in other parvicursorines (Longrich and Currie2009 Turner et al 2009 Xu et al 2010) In less specializedarctometatarsal taxa metatarsal III normally extends right tothe proximal end of the metatarsus in an attenuated form(Hutchinson and Padian 1997)

Metatarsal II is straight for most of its length although theproximal and distal ends both have a strong medial curvaminusture The proximal end of metatarsal II appears to have asubminustriangular outline though its lateral portion is missing inthe only preserved example of ths element Relative to themidshaft the proximal end is strongly expanded both dorminussally and laterally Just 7 mm away from the proximal end isa short groove along the dorsomedial margin The shaft isdeeper dorsoventrally than wide transversely for its entirelength although the discrepancy is reduced in the distal thirdof the shaft A weak crest is present along the ventromedial


margin over about the middle third of the shaft A slight colminuslateral ligamental pit appears to be present on the medial sideof the distal end of the metatarsal

Metatarsal III is visible over about the distal twominusfifths ofthe metatarsus in dorsal view and over less than the distaloneminusfifth in ventral view Proximally this metatarsal iswedgeminusshaped fitting tightly between metatarsals II and IVFrom the sharply pointed proximal end of metatarsal III thedorsal surface of the shaft rapidly widens and indeed becomesbroader than the other metatarsals in dorsal view but subseminusquently undergoes a slight constriction and remains narrow forapproximately the distal threeminusfifths of its length The dorsalsurface proximal to the constriction is slightly concave transminusversely a feature also seen in other Asian parvicursorines(Longrich and Currie 2009) Ventrally the shaft forms a weaklongitudinal ridge extending toward the distal end As preminusserved the distal portion of metatarsal III curves considerablyventrally as does that of metatarsal IV The distal end is conminussiderably wider transversely than those of metatarsals II andIV The articular surface is not ginglymoid though obliquehemicondylar ridges are weakly developed on the ventral surminusface The distal articular surface extends some distance proximinusmally onto the dorsal surface of the shaft Large and deep colminus

lateral ligmental fossae are present on the medial and lateralsides of the distal end of the metatarsal

Metatarsal IV is only slightly longer than metatarsal II andis closely appressed to the latter for most of its length althoughin the distalmost part of the metatarsus the two are separatedby metatarsal III Metatarsal IV is much deeper dorsoventrallythan wide transversely for most of its length and its distal endappears to curve laterally Metatarsal IV is transversely widerthan metatarsal II over the proximal half of the metatarsus butnarrower over the distal half A weak crest runs along theproximal half of the ventrolateral margin The distal end ofmetatarsal IV is slightly expanded dorsally and strongly exminuspanded ventrally The distal articulation is nonminusginglymoidThe dorsal exposure of the articular surface is strongly skewedtowards the lateral side of the metatarsal On the ventral surminusface of the distal end two prominent ridgeminuslike hemicondylesare developed the lateral hemicondyle being more ventrallyprominent than the medial There is no collateral ligamentalpit on the lateral side of the distal end of the metatarsal

The pedal phalanges are relatively long and slender comminuspared to those of Mononykus (Perle et al 1994) All of thenonminusungual phalanges are subminusrectangular in cross sectionand have strongly ginglymoid distal ends with hemicondyles



median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)

Fig 12 Pes of an alvarezsauroid theropod Linhenykus monodactylus Xu Sullivan Pittman Choiniere Hone Upchurch Tan Xiao Tan and Han 2011aBayan Mandahu (ldquoGate Localityrdquo) Late Cretaceous (Campanian) holotype (IVPP V17608) A Left metatarsus in dorsal (A1) ventral (A2) medial (A3)and lateral (A4) views B Left pedal phalanges Iminus1 and Iminus2 in right (B1) and left (B2) lateral views C right pedal phalanges IIminus1 IIminus2 and IIminus3 in lateral (C1)dorsal (C2) ventral (C3) views D left pedal phalanges IVminus3 IVminus4 and IVminus5 in dorsal (D1) ventral (D2) and lateral (D3) views E right pedal phalanxIVminus4 in proximal (E1) dorsal (E2) lateral (E3) medial (E4) and distal (E5) views F left pedal phalanges IIIminus3 and IIIminus4 in lateral view

that are of considerable extent both dorsoventrally and proximinusmodistally Collateral ligamental fossae are well developedon most of the nonminusungual phalanges although the pits on thepenultimate phalanges are shallow The proximal phalangeshave distinct extensor fossae immediately proximal to theirdistal articular surfaces and less distinct fossae are also presminusent on the intermediate phalanges

Pedal phalanx Iminus1 is small Its proximal articular surface isnearly flat and inclined to face slightly dorsally The dorsaland ventral margins do not form prominent lips suggestingthat the distal end of metatarsal I is nonminusginglymoid The distalhemicondyles of phalanx Iminus1 are more prominent dorsally thanventrally and the medial hemicondyle is larger than the latminuseral A shallow collateral ligamental fossa is located ventrallyon the medial surface of the distal articulation whereas the latminuseral surface is flat Phalanx IIminus1 is much deeper dorsoventrallyat the proximal end than at the distal end Distally the phalanxbears not only a distinct extensor fossa but also a prominentflexor fossa bounded laterally and medially by significantproximal extensions of the hemicondyles The ventral lip ofthe proximal end of phalanx IIminus2 is asymmetrical with its tipmedially positioned Phalanx IIIminus1 has a strongly expandedproximal end but the shaft narrows significantly as it passesdistally Phalanx IVminus1 is a relatively robust element Its proximinusmal end is deeply notched on the ventral side with a long androbust medial process and a short transversely thin lateral prominuscess as in other parvicursorines (Longrich and Currie 2009)The lateral collateral ligamental fossa is shallow and locatedcentrally on the lateral surface of the distal end of the phalanxwhereas the medial fossa appears deep

The pedal ungual phalanges are moderately curved andlaterally compressed each having a subminuselliptical cross secminustion The preserved examples all lack flexor tubercles andtheir arcuate sheath grooves are relatively weak Phalanx Iminus2is asymmetrical having a relatively flat lateral margin but amore convex medial margin

Stratigraphic and geographic rangemdashType locality only

Morphological evolutionA numerical phylogenetic analysis produced two most parsiminusmonious trees each with a tree length of 103 steps a CI of079 and a RI of 086 (Fig 13A B) and recovers Linhenykusas a basal parvicursorine alvarezsauroid (Xu et al 2011a)The optimized synapomorphies for the Alvarezsauroideaand more exclusive clades are listed in Supplementary Onminusline Material (SOM 1) available at httpapppanplSOMapp58minusXu_etal_SOMpdf

The synapomorphies shared with other parvicursorines inminusclude dorsal vertebrae opisthocoelous and without hypominusspheneminushypantrum articulations dorsal parapophyses eleminusvated to level of diapophyses dorsal infradiapophyseal fossaehypertrophied posterior sacral vertebrae with hypertrophiedventral keels anterior caudal vertebrae with anteriorly disminus

placed transverse processes metacarpal II that is mediominuslaterally wider than proximodistally long and dorsoventrallyshallow bears two articular facets on the proximal surfaceand lacks collateral ligamental fossae manual ungual II withventral axial groove and without flexor tubercle tibial distalend with lateral malleolus anteroposteriorly expanded astraminusgalar ascending process Lminusshaped and specialized arctometaminustarsalian condition with metatarsal III shorter than metatarsalsII and IV Other possible parvicursorine synapomorphies seenin Linhenykus include posterior dorsal vertebrae with smalland posterodorsally oriented neural spines biconvex vertebranear posterior end of dorsal series keeled and boatlike sterminusnum with median groove along midline manual phalanx IIminus2longer than IIminus1 supracetabular crest of ilium more prominentanteriorly than posteriorly metatarsus longer than femurmetatarsals II and IV subequal in length and pedal digit IIImore slender than digits II and IV

These synapomorphies are mostly related to the appendiminuscular skeleton which is unsurprising given that Shuvuuia isthe only alvarezsaurid in which the skull is wellminusknown(Chiappe et al 2002) Indeed parvicursorines apparentlydiffer from other nonminusavian theropods in having highlyspecialized forelimbs and hindlimbs (Chiappe et al 2002)While their forelimb morphology has received much attenminustion (Zhou 1995 Senter 2005 Longrich and Currie 2009)little attention has so far been paid to their hindlimbs alminusthough hindlimb morphology has been extensively studiedin other theropod groups (Gatesy 1991 Hutchinson 2001Carrano and Hutchinson 2002)

The highly modified hindlimbs of parvicursorines havebeen interpreted as an adaptation for rapid running (Xu et al2010) One commonly used indicator of cursoriality is thelength of the crural segment of the hindlimb relative to that ofthe femoral segment though this is probably more effective inpromoting energy efficiency during running than in permittinga high maximum speed Previous analyses have shown thatderived theropods possess relatively greater tibiafemur lengthratios than primitive ones and this evolutionary trend continminusues into the Aves (Gatesy 1991 Jones et al 2000 Christiansenand Bonde 2002) However many theropod groups display anopposite trend during their evolutionary history basal memminusbers of these theropod groups (for example the Ceratosauriathe Tyrannosauroidea the Therizinosauroidea the Oviraptominusrosauria the Troodontidae and the Dromaeosauridae) haverelatively longer distal segments of the hindlimb than derivedmembers of each group Alvarezsauroidea and Aves are exminusceptional among the major theropod groups in elongating thecrural segment of the hindlimb in their evolutionary historywhich is probably related to a sizeminusdecrease trend within thesetwo groups Linhenykus has a tibiafemur length ratio of about140 similar to other parvicursorines but significantly greaterthan in basal alvarezsauroids and most other nonminusavian therominuspods (Gatesy 1991 Jones et al 2000 Christiansen and Bonde2002) In Aves the evolutionary pattern was more complexsignificant distal limb elongation took place but was precededby significant distal limb shortening close to the base of the


Aves (Hu et al 2010) We interpret this temporary shorteningtrend as resulting from either a substantial improvement inflight capability (and thus decreased reliance on the hind limbfor primary locomotion) or a shift toward an arboreal life style(a very long crus and foot might be awkward in climbing andreducing the overall mass of the hind limb would be helpful toa flier) and the distal limb elongation in more derived birds asresulting from a shift toward avianminusstyle locomotion in whichmovements of the lower hindlimb segments predominated inrunning and particularly in walking (Jones et al 2000) In deminusrived alvarezsauroids pronounced elongation of the tibiominustarsus relative to the femur is more likely to indicate increasedcursorial capability Given that parvicursorines have highertibiafemur ratios than most other nonminusavian theropods theyare probably among the most capable cursors within this evominuslutionary grade (Xu et al 2010) Further indicators of cursorialspecialization include the extremely elongate metatarsus (lonminusger than the femur) specialized arctometatarsalian pes andproportionally short and robust pedal phalanges of alvarezminussaurids Some of these features are also known in ornithominusmimosaurs (Makovicky et al 2004) another nonminusavian therominuspod group regarded as highly cursorial In less specializedcursorial nonminusavian theropods such as troodontids the metaminustarsus is shorter and the pedal phalanges are more slender

Historical biogeographyof the AlvarezsauroideaThe historical biogeography of alvarezsauroids has receivedconsiderable attention during the past two decades Initiallysome authors (eg Bonaparte 1991) suggested that Alvarezminussaurus calvoi represented a lineage of Cretaceous theropodsthat were endemic to Gondwana perhaps converging to somedegree on the Laurasian ornithomimosaurs (Barsbold andOsmoacutelska 1990 Novas 1997) However this idea was rapidlyoverturned as a result of the discovery of first East Asian andthen North American alvarezsauroids (Novas 1997 Hutchinminusson and Chiappe 1998) This wider geographic distributionhas been interpreted in two ways preminusAptian vicariance vsLate Cretaceous dispersals The vicariance model was prominusposed by Novas (1996) who suggested that alvarezsauridsprobably originated during or before the Early Cretaceous andtherefore spread across Pangaea before geographic barriersseparated Laurasia and Gondwana from the Aptian onwardsIn contrast the dispersal hypothesis envisages alvarezsauridsoriginating in South America initially followed by at least onedispersal event into North America followed in turn by atleast one dispersal event from the latter into East Asia (Martiminusnelli and Vera 2007 Longrich and Currie 2009) This hypothminusesis is based on the observations that (i) the most basalalvarezsaurids come from South America whereas the mostderived ones occur in East Asia (Martinelli and Vera 2007Longrich and Currie 2009) and (ii) some palaeogeographicand biogeographic evidence support faunal exchanges beminus

tween North and South America (Bonaparte and KielanminusJawominusrowska 1987 Lucas and Hunt 1989 Gayet et al 1992 Sulminuslivan and Lucas 2000) and between North America and EastAsia (Russell 1993 Le Loeuff 1997 Hutchinson and Chiappe1998 Sereno 1999) across landbridges that formed during theLate Cretaceous

There is however a third interpretation of the observeddistributions of alvarezsaurids that is the null biogeographichypothesis The latter states that there is no detectable biominusgeographic pattern because the data are inadequate or beminuscause genuine distribution signals have been obscured bymissing data The null biogeographic hypothesis should berejected (via analytical and statistical approaches) before weaccept the possibility of a genuine biogeographic pattern(991 Upchurch et al 2002 Sanmartin and Ronquist2004 Upchurch 2008) but no previous study has attemptedthis for the alvarezsaurids Below we use the new alvarezminussauroid phylogeny presented in Xu et al (2011a) to test comminuspeting biogeographic hypotheses utilizing a cladistic biogeominusgraphic method and statistical evaluation based on randomminusization tests

MethodsmdashRonquist (1998) and Sanmartin and Ronquist(2004) proposed the cladistic biogeographic method knownas Treefitter In this approach each type of biogeographicevent is assigned a cost The default costs in Treefitter arevicariance = 0 sympatry = 0 extinction = 1 and dispersal =2 However the investigator can alter these costs For examminusple setting vicariance = minus1 and the three other event types =0 produces a ldquomaximum codivergencerdquo analysis analogousto that implemented in the programs Component (Page 19931995) as has been applied to dinosaurs (Upchurch et al2002 Butler et al 2006) and crocodyliforms (Turner 2004)Treefitter uses phylogenetic topologies and information onthe geographic ranges of the terminal taxa to generate themost parsimonious biogeographic reconstruction (ie thebiogeographic history with the lowest overall event cost thatexplains the distributions of the taxa)

It is conceivable that taxon distributions are random withrespect to phylogeny (the null hypothesis) therefore thispossibility must be tested before the biogeographic reconminusstruction can be legitimately interpreted as a set of genuineevolutionary events To test the null hypothesis the phylominusgenetic topology or the positions of terminal taxa with reminusspect to that topology are randomized and the event cost calminusculated for these random data This process is repeated thouminussands of times so that a population of randomized results isgenerated If the event cost required by the original (unminusrandomized) data is less than the event costs of 95 of therandomized data sets then the null hypothesis can be reminusjected with a pminusvalue of lt 005 Treefitter can also reconstructthe frequencies of each of the four types of biogeographicevent and randomization tests can again be used to determinusmine whether these frequencies depart significantly fromvalues expected from random data sets (Ronquist 1998Sanmartin and Ronquist 2004)



Analyses and resultsmdashThe phylogenetic analysis of Xu etal (2011a) produced a fully resolved cladogram apart fromthe trichotomy between Achillesaurus Alvarezsaurus andthe remaining clade of more derived alvarezsaurids For thepurposes of biogeographic analysis this gives two topolominusgies that should be explored (i) Achillesaurus and Alvarezminussaurus are sister taxa (Fig 13A) and (ii) Alvarezsaurus andthe remaining derived alvarezsaurid clade are sister taxa (ieAchillesaurus is the most basal taxon) (Fig 13B) The thirdpossibility in which Achillesaurus and derived alvarezsauminusrids are sisterminustaxa to the exclusion of Alvarezsaurus isbiogeographically equivalent to tree 2 in Fig 13 becauseboth Achillesaurus and Alvarezsaurus occur in South Amerminusica The taxa in Trees 1 and 2 have been assigned to threegeographic areas (A = Asia N = North America S = SouthAmerica) to create the Treefitter data sets shown in SOM 1These data sets have been analyzed using the exhaustivesearch option under the default and maximum codivergencecost regimes in Treefitter 12b (Ronquist 1998) For eachanalysis randomization tests were carried out using 10000replicates in which the tree topology was perturbed eachtime In addition the frequencies of the four types of biominusgeographic event have been calculated and their statisticalsignificance evaluated using 10000 random replicates

The results of these analyses are summarized in Tables 1and 2 and indicate the followingndash Each analysis produced two optimal area cladograms

(OACs) and across all analyses all three possible relationminusships between Asia North America and South Americaare represented in these OACs (Table 1) Restricting conminussideration purely to OACs that are statistically significanthowever allows rejection of the topology (Asia (NorthAmerica South America))

ndash The choice of cost regime strongly affects the statisticalsupport for the reconstructions Under the default costdata set 2 passes the randomization test and data set 1 fails

narrowly (Table 1) However under the maximum codiminusvergence cost regime both data sets produce statisticallyinsignificant reconstructions (Table 1)

ndash The alternative positions of the South American alvarezminussaurids near the base of the phylogeny (see Trees 1 and 2in Fig 13) have some effect on biogeographic reconstrucminustions In particular data set 2 produces more statisticallysignificant event frequency results than does data set 1(Table 2)

ndash The biogeographic reconstructions produced under the deminusfault cost regime have low pminusvalues which suggest that thenull hypothesis can be rejected However examination ofthe event frequencies in Table 2 demonstrates that disminuspersal does not occur more or less often than expected andin some instances vicariance and extinction occur less ofminusten than expected in random data Thus it seems thatthese biogeographic reconstructions are largely statistiminuscally significant because of their higher than expected freminusquencies of sympatric events

DiscussionmdashThe statistically significant optimal area cladominusgrams produced by the Treefitter analyses are compatiblewith both the preminusAptian vicariance and Late Cretaceous disminuspersal hypotheses outlined above For example the topology(South America (Asia North America)) could have arisen asa result of the formation of geographic barriers first betweenLaurasia and Gondwana and then between Asia and NorthAmerica The same area cladogram topology would also ocminuscur if alvarezsaurids originated in South America and thendispersed to Asia or North America or Asia North America(although it should be noted that none of the optimal reconminusstructions explain the geographic distributions via dispersalalone or via dispersal sympatry and extinction withoutvicariance) The optimal area cladogram topology is alsocompatible with reconstructions that mix vicariance and disminuspersal events (eg see the event frequencies for data set 2 unminusder default costs Table 2) Thus it is clear that the OACs by


Table 1 Summary of the results of the Treefitter analyses of data sets 1 and 2 (see SOM 2) Abbreviations DC default event costs MC maximumcodivergence event costs OAC optimal area cladogram rec cost total cost of the most parsimonious biogeographic reconstruction

Data set Cost regime OAC Rec cost pminusvalue

1 DC (South America (Asia North America)) (North America (Asia South America)) 4 00054ndash0091

1 MC (South America (Asia North America)) (Asia (North America South America)) minus2 0997ndash0999


2 MC (South America (Asia North America)) (Asia (North America South America)) minus4 0999

Table 2 Summary of event frequencies reconstructed using Treefitter Abbreviations DC default costs MC maximum codivergence costs Onlystatistically significant pminusvalues are shown ldquo+rdquo and ldquominusrdquo indicate that the frequency of the event in the real data is greater than or less than expectedfrom random data respectively

Data set Cost regime Vicariance frequency Sympatry frequency Extinction frequency Dispersal frequency

1 DC 1ndash2 8ndash9 (p = 00062ndash00077) + 0ndash4 0ndash2

1 MC 2 6ndash9 0ndash3 0ndash3

2 DC 2ndash4 16ndash17 (p = 00016ndash00025) + 06 (p = 00258) minus 1ndash4

2 MC 4 (p = 00068ndash0132 minus) minus 11ndash18 (p = 00122ndash075) + 0ndash10 (p = 00466ndash035) minus 0ndash7

themselves do not allow a convincing rejection of any ofthe biogeographic histories proposed previously Moreoverthese results identify equally valid but hitherto unconsideredhistories involving a more complex mix of vicariance reminusgional extinction and dispersal

The event frequencies are somewhat more informativethan the optimal area cladograms Table 2 indicates that thefrequencies of vicariance extinction and dispersal are nohigher than would be expected from random data irrespecminustive of which cost regime is applied or which data set is anaminuslyzed The only statistically significant results relate tohigher than expected levels of sympatry in both data setsand lower than expected rates of vicariance and extinctionin data set 2 Thus there is no statistical support for eitherthe preminusAptian vicariance or Late Cretaceous dispersal hyminuspotheses proposed by other workers Nevertheless the geominusgraphic distributions of alvarezsaurids are not entirely ranminusdom with respect to their phylogeny because there appearsto be more sympatry than expected from random data

The results produced by these analyses should be interminuspreted with caution First as discussed by Upchurch et al(2002) there are numerous reasons why a cladistic biogeominusgraphic data set might fail to produce statistically significantresults In particular it is important to recognize that statistical

failure means that there is an absence of evidence for one ormore biogeographic events not evidence of absence Thusalvarezsaurid distributions may actually have been shaped byvicariance andor dispersal but statistically robust evidencethat demonstrates this is lacking at present A related point isthat cladistic biogeographic analyses cannot distinguish beminustween true sympatry (speciation within a geographic area reminussulting from ecological or behavioral partitioning) and ldquowithinareardquo allopatry (ie vicariance that occurs as a result of barrierformation within the geographic area) Thus some or all of thesympatric events such as the divergences between derivedAsian alvarezsaurid taxa might actually represent vicariancebetween different parts of Asia

Second our results suggest that the biogeographic historyof alvarezsaurids is likely to be somewhat more complex thanpreviously realized This is supported by the recent discoveryof Heptasterornis a derived alvarezsaurid from the Maasminustrichtian of Romania (Naish and Dyke 2004) This taxonshares some derived character states of the femur with adminusvanced Asian alvarezsaurids hinting at a close phylogeneticrelationship (Xu et al 2011a supplementary information) It isinteresting to note that Heptasterornis potentially clusterswithin a monophyletic Laurasian clade that also includesNorth American and Asian forms This pattern could have



Haplocheirus sollers

Achillesaurus manazzonei

Alvarezosaurus calvoi

Patagonykus puertai

Linhenykus monodactylus

Albertonykus borealis

YPM 1049

Xixianykus zhangi

Parvicursor remotus

Tugriken Shireh alvarezsaur

Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia

Fig 13 The two trees produced by resolving the basal trichotomy in the alvarezsaurid clade (see text for details) Tree A is used in data set 1 and Tree B indata set 2 (see SOM 2) The geographic regions occupied by each terminal taxon are shown A Asia N North America S South America

been produced by a preminusAptian vicariance event as proposedby Novas (1996) Alternatively regression of the Turgai seaduring the AptianndashAlbian might have produced a temporarylandbridge between Asia and Europe (Upchurch et al 2002)and therefore have enabled the ancestor of Heptasterornis todisperse from the former area to the latter at this time Unfortuminusnately Heptasterornis is known from very fragmentary reminus

mains which do not permit its phylogenetic position to be esminustablished precisely (Dyke and Naish 2011 Xu et al 2011b)Thus discoveries of additional alvarezsaurid material fromthe Late Cretaceous of Europe and Central Asia and indeedEarly Cretaceous specimens from other continents will beneeded before these competing scenarios can be tested morerigorously


Fig 14 Life reconstruction of two individuals of Linhenykus monodactylus Artwork by Julius T Csotonyi

Finally the current data set is very small (12 ingroupalvarezsaurids) compared to those that have successfully reminuscovered vicariance and dispersal patterns using cladistic methminusods (eg 150 dinosaurian genera (Upchurch et al 2002) 29crocodyliform genera (Turner 2004) over 1000 extant species(Sanmartin and Ronquist 2004) There are many Late Cretaminusceous dinosaur groups with close relatives distributed acrossboth North America and Asia (eg ornithomimosaurs tyranminusnosaurs ceratopsids (Weishampel et al 2004)) and some ofthese also have representatives in South America (eg drominusmaeosaurids Makovicky et al 2005) It is conceivable that alarger data set will provide evidence of a consistent biominusgeographic pattern repeated across many of these clades prominusviding statistically significant support for vicariance andordispersal events

ConclusionLinhenykus (reconstructed in Fig 14) is an advanced alvarezminussaurid from the Late Cretaceous of Asia and is referable to theParvicursorinae This new genus provides the first example ofmonodactyly among dinosaurs Clearly as well as the majortrends in manus evolution that occurred during the theropodminusbird transition there must have been much ldquoexperimentationrdquowith hand structure in several of the ldquoside branchesrdquo such astyrannosauroids and alvarezsauroids

The study of alvarezsauroid historical biogeographyserves as an example of the dangers associated with interminuspreting macroevolutionary patterns based on literal readingsof the fossil record Several workers have generated scenarminusios to explain the observed geographic and stratigraphicranges of alvarezsauroids without first ruling out the possiminusbility that these distributions are effectively random becauseof the paucity of the fossil record Cladistic biogeographicanalysis indicates that there is some signal in the alvarezminussauroid record but this relates to elevated levels of sympatry(or ldquowithin continentrdquo vicariance) rather than evidence fordispersal or continentminuslevel vicariance Inferences of biogeominusgraphic history derived from small palaeobiological datasetsshould be treated with extreme caution especially when thenull biogeographic hypothesis has not been rejected In thefuture discovery of new fossils will play a vital role in imminusproving our understanding of the biogeographic history ofalvarezsauroids but the application of appropriate analyticalmethods will be equally important

AcknowledgementsWe thank Wang Jianming (Department of Land and ResourcesHohhot Inner Mongolia) Liu Jinsheng Li Zhiquan and Zhao Xinquan(all Department of Land and Resources Linhe Inner Mongolia) for cominusordinating the project Wang Haijun (IVPP) for assistance in the fieldand Yu Tao and Ding Xiaoqing (both IVPP) for preparing the speciminusmen This work was supported by the National Natural Science Founminus

dation of China the Department of Land and Resources Inner Mongominuslia China the Gloyne Outdoor Geological Research Fund of the Geominuslogical Society of London National Science Foundation Office of Inminusternational Science and Engineering grant 0812234 the Jurassic Founminusdation and the Robert Weintraub Program for Systematics and Evoluminustion at the George Washington University

ReferencesBarsbold R and Osmoacutelska H 1990 Ornithomimosauria In DB Weisminus

hampel P Dodson and H Osmoacutelska (eds) The Dinosauria (First ediminustion) 225ndash248 University of California Press Berkeley

Bonaparte JF 1991 Los vertebrados foacutesiles de la Formacioacuten Riacuteo Colominusrado de la ciudad de Neuqueacuten y cercanias Cretaacutecico Superior Argenminustina Paleontologiacutea 4 (3) 17ndash123

Bonaparte JF and KielanminusJaworowska K 1987 Late Cretaceous dinosaurand mammal faunas of Laurasia and Gondwana In PJ Currie and EHKoster (eds) Fourth Symposium on Mesozoic Terrestrial EcosystemsShort Papers 24ndash29 Occasional Papers of the Tyrell Museum of Palaeminusontology Drumheller

Butler RJ Upchurch P Norman DB and Parish JC 2006 A biogeominusgraphical analysis of the ornithischian dinosaurs In PM Barrett and SEEvans (eds) Ninth International Symposium on Mesozoic TerrestrialEcosystems and Biota 13ndash16 The Natural History Museum London

Carrano MT and Hutchinson JR 2002 Pelvic and hindlimb musculatureof Tyrannosaurus rex (Dinosauria Theropoda) Journal of Morphology253 207ndash228

Chatterjee S 1998 Counting the fingers of birds and dinosaurs Science 280355a

Chiappe LM Norell MA and Clark JM 1998 The skull of a relative ofthe stemminusgroup bird Mononykus Nature 392 275ndash282

Chiappe L Norell MA and Clark JA 2002 The Cretaceous shortminusarmed Alvarezsauridae Mononykus and its kin In LM Chiappe andLM Witmer (eds) Mesozoic Birds Above the Heads of Dinosaurs87ndash120 University of California Press Berkeley

Choiniere J 2010 Anatomy and Systematics ofCoelurosaurian Theropodsfrom the Late Jurassic of Xinjiang China with Comments on ForelimbEvolution in Theropoda 1015 pp George Washington UniversityWashington DC

Choiniere JN Xu X Clark JM Forster CA Guo Y and Han FL2010 A basal alvarezsauroid theropod from the early Late Jurassic ofXinjiang China Science 327 571ndash574

Christiansen P and Bonde N 2002 Limb proportions and avian terrestriallocomotion Journal of Ornithology 143 356ndash371

Christiansen P and Farintildea RA 2004 Mass prediction in theropod dinominussaurs Historical Biology 16 85ndash92

Chure DJ 2001 The wrist of Allosaurus (Saurischia Theropoda) with obminusservations on the carpus in theropods In JA Gauthier and LF Gall(eds) New Perspectives on the Origin and Early Evolution of Birds122ndash130 Yale University Press New Heaven

Dyke G and Naish D 2011 What about European alvarezsauroids Prominusceedings of the National Academy of Sciences of the United States ofAmerica 108 E147

Gatesy SM 1991 Hind limb scaling in birds and other theropods implicaminustions for terrestrial locomotion Journal of Morphology 209 83ndash96

Gauthier J 1986 Saurischian monophyly and the origin of birds Memoirsof the California Academy of Sciences 8 1ndash55

Gayet M Rage JminusC Sempere T and Gagnier PY 1992 Modaliteacutes deseacutechanges de verteacutebreacutes continentaux entre lrsquoAmeacuterique du Nord et lrsquoAmeacuteriminusque du Sud au Creacutetaceacute supeacuterieur et au Paleacuteocegravene Bulletin de la SocieteacuteGeologique de France 163 761ndash791

Hu DminusY Li L Hou LminusH and Xu X 2010 A new sapeornithid birdfrom China and its implication for early avian evolution Acta Geominuslogica Sinica 84 472ndash482

Hutchinson JR 2001 The Evolution of Hindlimb Anatomy and Function inTheropod Dinosaurs 357 pp University of California Berkeley



Hutchinson JR and Chiappe LM 1998 The first known alvarezsaurid(Theropoda Aves) from North America Journal of Vertebrate Paleonminustology 18 447ndash450

Hutchinson JR and Padian K 1997 Arctometatarsalia In PJ Currie andK Padian (eds) Encyclopedia of Dinosaurs 24ndash27 Academic PressSan Diego

Irmis RB 2007 Axial skeletal ontogeny in the Parasuchia (ArchosauriaPseudosuchia) and its implications for ontogenetic determination inArchosaurs Journal of Vertebrate Paleontology 27 350ndash361

Jerzykiewicz T Currie PJ Eberth DA Johnston PA Koster EHand Zheng JminusJ 1993 Djadokhta Formation correlative strata in Chiminusnese Inner Mongolia an overview of the stratigraphy sedimentary geminusology and paleontology and comparisons with the type locality in thepreminusAltai Gobi Canadian Journal of Earth Sciences 30 2180ndash2190

Jones TD Farlow JO Ruben JA Henderson DM and HilleniusWJ 2000 Cursoriality in bipedal archosaurs Nature 406 716ndash718

Karhu AA and Rautian AS 1996 A new family of Maniraptora (Dinominussauria Saurischia) from the Late Cretaceous of Mongolia Paleontominuslogical Journal 30 583ndash592

Kessler E Grigorescu D and Csiki Z 2005 Elopteryx revisitedmdasha newbirdminuslike specimen from the Maastrichtian of the Hateg Basin (Romaminusnia) Acta Palaeontologica Romaniae 5 249ndash258

Kundraacutet M Seichert V Russell AP and Smetana K 2002 Pentadactylpattern of the avian wing autopodium and pyramid reduction hypotheminussis Journal of Experimental Zoology (Mol Dev Evol) 294 152ndash159

Le Loeuff J 1997 Biogeography In PJ Currie and K Padian (eds) Enminuscyclopedia of Dinosaurs 51ndash56 Academic Press San Diego

Longrich NR and Currie PJ 2008 Albertonykus borealis a new alvarezminussaur (Dinosauria Theropoda) from the Early Maastrichtian of AlbertaCanada implications for systematics and ecology of the Alvarezminussauridae Cretaceous Research 30 239ndash252

Longrich NR and Currie PJ 2009 Albertonykus borealis a new alvarezminussaur (Dinosauria Theropoda) from the Early Maastrichtian of AlbertaCanada implications for the systematics and ecology of the Alvarezminussauridae Cretaceous Research 30 239ndash252

Lucas SG and Hunt AP 1989 Alamosaurus and the sauropod hiatus inthe Cretaceous of the North American western interior Geological Sominusciety of America Special Paper 238 75ndash85

Makovicky PJ Apesteguiacutea S and Agnoliacuten FL 2005 The earliestdromaeosaurid theropod from South America Nature 437 1007ndash1011

Makovicky PJ Kobayashi Y and Currie PJ 2004 Ornithomimosauria InDB Weishampel P Dodson and H Osmoacutelska (eds) The Dinosauria(second edition) 137ndash150 University of California Press Berkeley

Martinelli AG and Vera E 2007 Achillesaurus manazzonei a new alvarezminussaurid theropod (Dinosauria) from the Late Cretaceous Bajo de la CarpaFormation Riacuteo Negro Province Argentina Zootaxa 1582 1ndash17

Naish D and Dyke G 2004 Heptasteornis was no ornithomimid troodontiddromaeosaurid or owl the first alvarezsaurid (Dinosauria Theropoda)from Europe Neues Jahrbuch fur Geologie und Palaumlontologie Monatminusshefte 2004 (7) 385ndash401

Novas FE 1996 Alvarezsauridae Cretaceous basal birds from Patagoniaand Mongolia Memoirs of the Queensland Museum 39 675ndash702

Novas FE 1997 Anatomy of Patagonykus puertai (Theropoda Maniminusraptora Alvarezsauridae) from the Late Cretaceous of Patagonia Jourminusnal of Vertebrate Paleontology 17 137ndash166

Page RDM 1991 Random dendrograms and null hypotheses in cladisticbiogeography Systematic Zoology 40 54ndash62

Page RDM 1993 Component vers 20 The Natural History MuseumLondon

Page RDM 1995 TreeMap for Windows vers 10a Available from httptaxonomyzoologyglaacukrodtreemaphtml

Perle A Chiappe LM Barsbold R Clark JM and Norell M 1994 Skelminusetal morphology of Mononykus olecranus (Theropoda Avialae) from theLate Cretaceous of Mongolia American Museum Novitates 3105 1ndash29

Perle A Norell MA Chiappe L and Clark JM 1993 Flightless birdfrom the Cretaceous of Mongolia Nature 362 623ndash626

Ronquist F 1998 Threeminusdimensional costminusmatrix optimization and maximinusmum cospeciation Cladistics 14 167ndash172

Russell DA 1993 The role of Central Asia in dinosaurian biogeographyCanadian Journal of Earth Sciences 30 2002ndash2012

Salgado L Coria RA Arcucci AB and Chiappe LM 2009 Remainsof Alvarezsauridae (Theropoda Coelurosauria) in the Allen Formation(CampanianndashMaastrichthian) in Salitral Ojo de Agua Riacuteo Negro Provminusince Argentina Andean Geology 36 67ndash80

Sanmartin I and Ronquist F 2004 Southern Hemisphere biogeographyinferred by eventminusbased models plant versus animal patterns Systemminusatic Biology 53 216ndash243

Senter P 2005 Function in the stunted forelimbs of Mononykus olecranus(Theropoda) a dinosaurian anteater Paleobiology 31 373ndash381

Sereno PC 1999 Dinosaurian biogeography vicariance dispersal and reminusgional extinction National Science Museum Monographs 15 249ndash257

Sullivan RM and Lucas SG 2000 Alamosaurus (Dinosauria Saurominuspoda) from the late Campanian of New Mexico and its significanceJournal of Vertebrate Paleontology 20 400ndash403

Suzuki S Chiappe LM Dyke GJ Watabe M Barsbold R andTsogtbataar K 2002 A new specimen of Shuvuuia deserti Chiappe etal 1998 from the Mongolian Late Cretaceous with a discussion of therelationships of alvarezsaurids to other theropod dinosaurs Contribuminustions in Science 494 1ndash18

Tamura K Nomura N Seki R YoneiminusTamura S and Yokoyama H2011 Embryological evidence identifies wing digits in birds as digits 12 and 3 Science 331 753ndash757

Turner AH 2004 Crocodyliform biogeography during the Cretaceous evminusidence of Gondwanan vicariance Proceedings of The Royal Society ofLondon Series B 271 2003ndash2009

Turner AH Nesbitt SJ and Norell MA 2009 A large alvarezsaurid fromthe Late Cretaceous of Mongolia American Museum Novitates 36481ndash14

Upchurch P 2008 Gondwanan breakminusupmdashlegacies of a lost world Trendsin Ecology and Evolution 23 229ndash236

Upchurch P Hunn CA and Norman DB 2002 An analysis of dinominussaurian biogeography evidence for the existence of vicariance and disminuspersal patterns caused by geological events Proceedings of the RoyalSociety of London Series BminusBiological Sciences 269 613ndash621

Vargas AO and Fallon JF 2005 Birds have dinosaur wings the molecularevidence Journal of Experimental Zoology (Mol Dev Evol) 304 B 86ndash90

Vargas AO Kohlsdorf T Fallon JF Brooks JV and Wagner GP2008 The evolution of HoxDminus11 expression in the bird wing insightsfrom Alligator mississippiensis PLoS ONE 3 e3325

Wagner GP and Gauthier JA 1999 123 = 234 a solution to the probminuslem of the homology of the digits in the avian hand Proceedings of Naminustional Academy of Sciences USA 96 5111ndash5116

Weishampel DB Barrett PM Coria RA Loeuff JL Xu X ZhaoXJ Sahni A Gomani E and Noto CR 2004 Dinosaur distribuminustion In DB Weishampel P Dodson and H Osmoacutelska (eds) TheDinosauria (second edition) University of California Press Berkeley

Wilson JA 1999 A nomenclature for vertebral laminae in sauropods andother saurischian dinosaurs Journal of Vertebrate Paleontology 19639ndash653

Xu X Clark JM Mo JminusY Choiniere J Forster CA Erickson GMHone DWE Sullivan C Eberth DA Nesbitt S Zhao Q Herminusnandez R Jia CK Han FminusL and Guo Y 2009 A Jurassic ceratosaurfrom China helps clarify avian digit homologies Nature 459 940ndash944

Xu X Sullivan C Pittman M Choiniere J Hone DWE UpchurchP Tan QminusW Xiao D and Tan L 2011a The first known monodacminustyl nonminusavian dinosaur and the complex evolution of the alvarezsauroidhand Proceedings of National Academy of Sciences of the United Statesof America 108 2338ndash2342

Xu X Sullivan C Pittman M Choiniere J Hone DWE Upchurch PTan QminusW Xiao D and Tan L 2011b Reply to Dyke and Naish Eurominuspean alvarezsauroids do not change the picture Proceedings of the Naminustional Academy of Sciences of the United States of America 108 E148

Xu X Wang DY Sullivan C Hone D Han FL Yan RminusH and DuFminusM 2010 A basal parvicursorine (Theropoda Alvarezsauridae) fromthe Upper Cretaceous of China Zootaxa 2413 1ndash19

Zhou ZH 1995 Is Mononykus a bird Auk 112 958ndash963


anthropology the Longhao Institute of Geology and Paleonminustology and the Bureau of Land and Resources of BayannurInner Mongolia prospected in the Upper Cretaceous dinominussaurminusbearing deposits of the Bayan Mandahu area in northminuscentral Inner Mongolia On August 14th two of the authors(MP and JC) discovered and collected some small theropodbones in the red beds at the Gate Locality (Jerzykiewicz et al1993) and they revisited the site and collected a few more elminusements belonging to the same specimen on August 18th Thespecimen was found on a low outwash plain east of the highcliffs at the Gate Locality and was fully exposed on a surfaceof extensively weathered red massive sandstone directly beminuslow a thick resistant sandstone layer (less than 1 m in thickminusness) with numerous calciteminusrich concretions Digging belowthe exposed bone failed to reveal additional elements

The preservation and distribution of the bones suggest thatthe specimen was originally preserved in a concretion that hadfully weathered out on the surface of the outwash plain Themajor elements of the skeleton were clustered within a 30 cmradius while some smaller pieces discovered during the initialsearch on the 14th had washed down slope to a distance of lessthan 1 m During the second site visit a wider survey revealed afew further elements up to 4 m from the skeleton A disminusarticulated and highly fragmentary turtle specimen was found~1 m from the specimen but the extensive weathering of theturtle elements suggests that they had been exposed for a muchgreater time and were not associated with the theropod Thepresence of a robust medial manual digit and a fused blockminuslike carpometacarpus identify this specimen as an alvarezminussaurid theropod but several features distinguish it from otherknown alvarezsaurids and a phylogenetic analysis places it atthe base of the derived clade Parvicursorinae (Xu et al 2011a)Furthermore only one metacarpal of this specimen possessesmanual phalanges indicating that this animal has only one phaminuslanxminusbearing digit and thus represents the first known monominusdactyl nonminusavian dinosaur This specimen was the subject of abrief description (Xu et al 2011a) however given the signifiminuscance of this taxon for understanding alvarezsauroid evolutionwe provide here a more detailed account of the osteologicalfeatures of the specimen Furthermore we provide in the presminusent paper a quantitative analysis of alvarezsauroid biogeominusgraphy incorporating information from this find

Institutional abbreviationsmdashIVPP Institute of VertebratePaleontology and Paleoanthropology Beijing China

Other abbreviationsmdashCI consistency index RI retentionindex

Systematic palaeontologyDinosauria Owen 1842Theropoda Marsh 1881Alvarezsauridae Bonaparte 1991Parvicursorinae Karhu and Rautian 1996

RemarksmdashThe Parvicursorinae were originally proposed byKarhu and Rautian (1996) but a phylogenetic taxonomic defiminusnition for the clade was first proposed by Choiniere et al(2010) which is a nodeminusbased definition (Choiniere et al2010) Linhenykus and other recently reported parvicursorinessuch as Albertonykus (Longrich and Currie 2009) and Xixiaminusnykus (Xu et al 2010) are morphologically very similar tomembers of this nodeminusbased Parvicursorinae but they are exminuscluded from the nodeminusbased clade based on the recoveredalvarezsauroid phylogeny We believe that it is not necessaryto establish a new clade and the most informative option is totreat Parvicursorinae as a stemminusbased taxon allowing for theinclusion of Linhenykus and other taxa (Xu et al 2011a b)Thus we define Parvicursorinae as the most inclusive cladeincluding Parvicursor remotus but not Patagonykus puertai

Revised Parvicursorinae diagnosismdashDorsal vertebrae opisminusthocoelous and without hypospheneminushypantrum articulationsdorsal parapophyses elevated to level of diapophyses dorsalinfradiapophyseal fossae hypertrophied posterior sacral verminustebrae with hypertrophied ventral keels anterior caudal verteminusbrae with anteriorly displaced transverse processes metacarminuspal II that is mediolaterally wider than proximodistally longand dorsoventrally shallow bears two articular facets on theproximal surface and lacks collateral ligamental fossae manminusual ungual II with ventral axial groove and without flexortubercle tibial distal end with lateral malleolus anteroposminusteriorly expanded astragalar ascending process Lminusshaped andspecialized arctometatarsalian condition with metatarsal IIIshorter than metatarsals II and IV Other possible parvicursominusrine synapomorphies include posterior dorsal vertebrae withsmall and posterodorsally oriented neural spines biconvexvertebra near posterior end of dorsal series keeled and boatminuslike sternum with median groove along midline manual phaminuslanx IIminus2 longer than IIminus1 supracetabular crest of ilium moreprominent anteriorly than posteriorly metatarsus longer thanfemur metatarsals II and IV subequal in length and pedaldigit III more slender than digits II and IV

Genus Linhenykus Xu Sullivan Pittman ChoiniereHone Upchurch Tan Xiao Tan and Han 2011aType species Linhenykus monodactylus Xu Sullivan Pittman Choiminusniere Hone Upchurch Tan Xiao Tan and Han 2011a see below

Linhenykus monodactylus Xu Sullivan PittmanChoiniere Hone Upchurch Tan Xiao Tan andHan 2011aFigs 1ndash12

Type localitymdashCoordinates in decimal degrees N4174392E10674436Bayan Mandahu ldquoGate Localityrdquo Wulansuhai Formation

Type horizonmdashCampanian Late Cretaceous (Jerzykiewicz et al 1993)

MaterialmdashThe taxon is only represented by the holotypespecimen IVPP V17608 a partial postcranial skeleton includminusing 4 partial cervical vertebrae 3 partial dorsal vertebrae 5partial sacral vertebrae 13 caudal vertebrae a fragmentary anminusterior or middle chevron part of the left scapulocoracoid a







Axial skeleton














ventral keel

parapophysis

pneumaticforamen

epipophysis


ventral groove5 mm








neuralspine

postzygapophysis

parapophysiscondyle


ventral keel

neural spine

condyle

condyleventral keel

condyle

diapophysis

condyle5 mm

pneumatic foramen







5 mm







neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)















There is however a third interpretation of the observeddistributions of alvarezsaurids that is the null biogeographichypothesis The latter states that there is no detectable biominusgeographic pattern because the data are inadequate or beminuscause genuine distribution signals have been obscured bymissing data The null biogeographic hypothesis should berejected (via analytical and statistical approaches) before weaccept the possibility of a genuine biogeographic pattern(Page 1991 Upchurch et al 2002 Sanmartin and Ronquist2004 Upchurch 2008) but no previous study has attemptedthis for the alvarezsaurids Below we use the new alvarezminussauroid phylogeny presented in Xu et al (2011a) to test comminuspeting biogeographic hypotheses utilizing a cladistic biogeominusgraphic method and statistical evaluation based on randomminusization tests




































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia





















































































Axial skeleton














ventral keel

parapophysis

pneumaticforamen

epipophysis


ventral groove5 mm








neuralspine

postzygapophysis

parapophysiscondyle


ventral keel

neural spine

condyle

condyleventral keel

condyle

diapophysis

condyle5 mm

pneumatic foramen







5 mm







neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia
























































































ventral keel

parapophysis

pneumaticforamen

epipophysis


ventral groove5 mm








neuralspine

postzygapophysis

parapophysiscondyle


ventral keel

neural spine

condyle

condyleventral keel

condyle

diapophysis

condyle5 mm

pneumatic foramen







5 mm







neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia






















































































neuralspine

postzygapophysis

parapophysiscondyle


ventral keel

neural spine

condyle

condyleventral keel

condyle

diapophysis

condyle5 mm

pneumatic foramen







5 mm







neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia





















































































5 mm







neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia





















































































neural spine

prezygapophysis

prezygapophysis

prezygapophysis

ventral keel

neural spine

postzygapophysis

neural canal

5 mm

condyle






5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia




















































































5 mm






Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia




















































































Forelimb





sternalkeel

glenoid lipcoracoid

sternalkeel

groove

5 mm

















radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia































































































radialemedial facet

radiale

radiale

radiale

radiale


extensor fossa


centralfacet


central facet

medial facet

median ridge

notch

ventral groove

ventral groove

5 mm

5 mm 5 mm

(E) (F)



Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia

















































































Pelvic girdle




Hindlimb




acetabular fossa


5 mm




femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia


















































































femoral head

pneumaticforamen


fibularcrest

ascendingprocess

ascendingprocess

10 mm

















median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia































































































median ridge

notch

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

II

meta-tarsal

II

meta-tarsal

IV

meta-tarsal

IV

pedalphalanx

IV-1

pedalphalanx

II-1

pedalphalanx

II-2

pedalphalanx

IV-15 mm

5 mm(A)



















































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia

































































































































Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

Asia

South America

North America




Patagonykus puertai



YPM 1049

Xixianykus zhangi

Parvicursor remotus


Mononykus decranus

Shuvuuia deserti

South America

South America

Asia

North America

Asia

Asia

Asia

Asia

Asia

South America

South America

South America

Asia

Asia

North America

North America

Asia

Asia

Asia

Asia

Asia
















































































osteology of the late cretaceous alvarezsauroid linhenykus ... · 2023 g street nw, washington, dc...

Documents