AMERICAN MUSEUM Novltates

AMERICAN MUSEUM

NovltatesPUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORYCENTRAL PARK WEST AT 79TH STREET, NEW YORK, N.Y. 10024

Number 3133, 23 pp., 15 figures, 2 tables May 25, 1995

The Osteology of Concornis lacustris(Aves: Enantiomithes)

from the Lower Cretaceous ofSpain and a Reexamination of its

Phylogenetic Relationships

JOSE L. SANZ,1 LUIS M. CHIAPPE,2'3AND ANGELA D. BUSCALIONI

ABSTRACT

Additional preparation of the holotype of Con-cornis lacustris from the Lower Cretaceous (Bar-remian) deposits of Las Hoyas (Spain) has pro-vided abundant new osteological data. This spec-imen is superbly preserved and anatomical detailsare available for most skeletal regions except theskull and neck. The forelimb has "modem" pro-portions but it retains large claws. The sternum iscarinate and deeply notched. The carina, however,does not reach the cranial margin but is developedin the caudal halfofthe sternum. The dorsal, syn-sacral and caudal vertebrae are amphicoelous. The

dorsal vertebrae have deep lateral grooves andparapophyses in a central position. The ischiumhas large obturator and dorsal processes; the pubesform a distal symphysis. The hind limb is gracile,elongate, and longer than the forelimb.

Cladistic analysis of various ornithothoracinetaxa supports the allocation of Concornis lacustriswithin the Enantiornithes, a major clade of Cre-taceous birds mostly recognized from continentaldeposits. This new interpretation is substantiallysupported by 12 synapomorphies shared by C.lacustris and the remaining enantiornithines.

I Unidad de Paleontologia, Departamento de Biologia, Facultad de Ciencias, Universidad Aut6noma de Madrid,Cantoblanco 28049, Madrid, Spain.2Department of Vertebrate Paleontology, American Museum of Natural History.3To whom correspondence should be addressed.

Copyright © American Museum of Natural History 1995 ISSN 0003-0082 / Price $2.40

AMERICAN MUSEUM NOVITATES

Within the Enantiornithes, tarsometatarsal datasupport a sister-group relationship between C. la-custris and the Avisauridae.

Concornis lacustris is the oldest enantiornithinebird from the Western Hemisphere. The interpre-

tation of this species as an enantiornithine em-phasizes the wide geographical and stratigraphicaldistribution of this clade during the Cretaceous,and its large taxonomic diversity.

RESUMEN

La nueva preparacion del holotipo de Concornislacustris del Cretacico inferior (Barremiano) de LasHoyas (Espania) ha provisto nueva y abundanteinformacion osteologica. Este ejemplar, que se en-cuentra excepcionalmente preservado, provee de-talles anatomicos para la mayor parte del esque-leto con la excepcion del craneo y cuello. El miem-bro anterior tiene proporciones de aspecto "mod-erno" si bien retiene garras bien desarrolladas. Elesternon es aquillado y profundamente escotado.Curiosamente, la quilla esternal no alcanza el mar-gen craneal sino que se encuentra desarrollada enla mitad caudal del esternon. Las vertebras dor-sales, sinsacrales y caudales son anficelicas. Lasvertebras dorsales poseen surcos laterales profun-dos y parapofisis ubicadas en el centro del cuerpovertebral. Los procesos obturador y dorsal del is-quion son grandes, y los pubis forman una sinfisisdistal. El miembro posterior es gracil, elongado, ymas largo que el miembro anterior.

El analisis cladistico de los distintos taxones deOrnithothoraces sustenta la asignacion de Con-cornis lacustris a los Enantiornithes, un impor-tante clado de aves cretacicas registrado princi-palmente en depositos continentales. Esta nuevainterpretacion esta sustentada por 12 sinapomor-fias compartidas entre C. lacustris y los restantesEnantiornithes. Dentro de los Enantiornithes, lainformacion del tarsometatarso sugiere que C. la-custris es el grupo hermano de Avisauridae.

Concornis lacustris es el Enantiornithes mas an-tiguo del hemisferio occidental. La asignacion deesta especie a los Enantiornithes enfatiza nueva-mente la amplia distribucion geografica y estrati-grifica de este grupo durante el Cretacico. A suvez, esta interpretacion resalta la notable diver-sidad taxonomica alcanzada por los Enantior-nithes.

INTRODUCTION

The Las Hoyas fossil site is located in theSerrania de Cuenca (Iberian Ranges), prov-ince of Cuenca, Spain (fig. 1). Stratigraphi-cally it belongs to the Lower Cretaceous (Bar-remian) "Calizas de La Huergina" Forma-tion of lacustrine and palustrine origin (Fre-genal-Martinez, 1991; Fregenal-Martinez andMelendez, 1993). The fossil site ofLas Hoyashas provided thousands of fossil remains ofplants, invertebrates, and vertebrates (Sanzet al., 1988, 1990). Among the latter, artic-ulated skeletons of salamanders, frogs, liz-ards, crocodiles, nonavian dinosaurs, and inparticular birds, are bringing to light the fossilrecord of these groups during the Early Cre-taceous. The limestones of Las Hoyas are anexample of the so-called Konservat-lager-stdtten, in which even delicate structures suchas feathers and insect wings, are preserved.

In the past decade an increasing number

ofMesozoic birds have provided critical newevidence for interpreting the early history ofbirds (Walker, 1981; Kurochkin, 1985; Sanzet al., 1988a; Alvarenga and Bonaparte, 1992;Chiappe, 1991, 1992a; Sanz and Buscalioni,1992; Zhou et al., 1992; Sereno and Rao,1992; Perle et al., 1993, 1994; Chiappe andCalvo, 1994). In this sense, the fossil site ofLas Hoyas has contributed the important dis-coveries ofIberomesornis romerali (Sanz andBonaparte, 1992) and the well-preservedConcornis lacustris (Sanz and Buscalioni,1992) (fig. 2). Additional preparation of thelatter specimen has revealed unexpected newanatomical information that supports a newhypothesis of its phylogenetic relationships.The aim of this paper is to redescribe theholotype specimen ofConcornis lacustris anddiscuss its phylogenetic relationships in thelight of the new anatomical information.

2 NO. 3133

1995 SANZ ET AL., CONCORNIS LACUSTRIS FROM LOWER CRETACEOUS OF SPAIN 3

Fig. 1. Map indicating the area of Serrania de Cuenca (Spain) in which the quarry of Las Hoyas issituated.

MATERIALS AND METHODS

The original description of Concornis la-custris (Sanz and Buscalioni, 1992) was basedon the unprepared specimen, so as not todestroy the wing feathers. The unpreparedspecimen comprised two main slabs of the

same block. After Sanz and Buscalioni's(1992) publication, the smaller slab (thoracicgirdle) was glued to the main one in its orig-inal position. The matrix was then removedby mechanical and chemical techniques.With a few exceptions, most enantiorni-

thine taxa are known from fragmentary re-


4~~~~~~At

Fig. 2. Concornis lacustris (LH-28 14) from the Lower Cretaceous of Spain. Specimen after prepa-ration. Scale bar = 10 mm.

mains (see for example Chiappe, 1993). Morecomplete new specimens are not yet de-scribed. This fact complicates our under-standing of the phylogenetic relationshipswithin this group. Chiappe (1992a, in press

a; see also Chiappe and Calvo, 1994) diag-nosed the Enantiornithes on the basis of sev-eral derived characters shared by all speci-mens then known (see these papers for thegeographical and stratigraphical distribution

4 NO. 3133


of this clade). The present cladistic analysis(appendix 2) combines some of these withother characters used in a recent study of thephylogenetic relationships ofthe major groupsofCretaceous birds (Chiappe, 1992b, in pressa, b; Chiappe and Calvo, 1994). The remain-ing diagnostic characters of Enantiornitheswere not included because they are uncertain(not preserved) in Concornis lacustris and ab-sent or uncertain in the remaining ingrouptaxa. In the data matrix of appendix 2, char-acter states scored for the Enantiornithes de-note the condition present in all previouslyknown enantiornithine taxa. Several char-acters of Iberomesornis romerali were re-scored after further preparation of the holo-type specimen. Both data matrices were an-alyzed using the implicit enumeration (ie)command ofthe "Hennig 86" program (Far-ris, 1988). Derived characters exclusive to asingle ingroup taxon were excluded from thedata matrices in order to avoid their influenceon the consistency indices (Carpenter, 1988;Wiley et al., 1991). The species concept usedin this paper follows the phylogenetic speciesconcept ofNixon and Wheeler (1990, 1992).Hence, the diagnosis of C. lacustris is basedon a unique combination of characters thatdifferentiates this taxon from all others.The anatomical nomenclature used here

generally follows Baumel and Witmer (1993)using the English equivalents ofthe Latin ter-minology.

SYSTEMATIC PALEONTOLOGYAVES LINNAEUS, 1758

ORNITHOTHORACES CHIAPPE AND CALVO, 1994ENANTIORNITHES WALKER, 1981

Concornis lacustris Sanzand Buscalioni, 1992

HOLOTYPE: LH-2814 (LH refers to the col-lection from Las Hoyas), housed in the Un-idad de Paleontologia, Universidad Auton-oma de Madrid, Spain. The specimen in-cludes most postcranial elements in articu-lation, including the right forelimb andthoracic girdle; the sternum; some dorsal,synsacral, and caudal vertebrae; pubis; is-chium; and hind limbs (figs. 2, 3).

LocALirrT AND HORIZON: Las Hoyas, LaCierva township, Province of Cuenca, Spain(fig. 1). "Calizas de La Huerguina" Forma-tion, Barremian, Lower Cretaceous.

DLAGNosIs: Enantiornithine bird with thefollowing autapomorphies: ribbonlike ischi-um, transverse ginglymoid articulation oftrochlea of metatarsal I, and strongly curvedand laterally excavated distal end of meta-tarsal IV. The latter two features make thistaxon different from all other enantiorni-thines for which the tarsometatarsus is known[e.g., Lectavis bretincola (Chiappe, 1993),Yungavolucris brevipedalis (Chiappe, 1993),Soroavisaurus australis (Chiappe, 1993),Neuquenornis volans (Chiappe and Calvo,1994), Avisaurus archibaldi (Brett-Surmanand Paul, 1985), and A. gloriae (Varricchioand Chiappe, 1995); see Chiappe (1993) forautapomorphic characters of these taxa]. Inaddition, Concornis lacustris differs from theArgentine Enantiornis leali (Walker, 1981)in that the latter species is very robust andlarge, and has a large and shallow transverseligamental groove in the humerus. C. lacus-tris is clearly separable from N. volans by thepresence of a long hypocleideum in the fur-cula, a sternal carina that develops only inthe caudal half of the sternum, a straightmetatarsal I, much stronger lateral grooveson the dorsal centra, a major metacarpal morerobust than the minor one, and a muchbroader intermetacarpal space. The last char-acter also distinguishes C. lacustris from theChinese enantiornithine Cathayornis yandi-ca, in which the intermetacarpal space is ex-tremely narrow. C. lacustris also differs fromthis species by having larger claws in the fore-limb and by the dorsal centra, which are notexcavated by two small depressions (cranialand caudal).

DESCRIPTIONThe holotype and single specimen of Con-

cornis lacustris is fairly complete, providingosteological details from most skeletal regionsexcept the skull and neck (figs. 2, 3). Thespecimen lies in ventral aspect with its limbspartially flexed. The preservation is excellentas exemplified by the presence of feather im-pressions (Sanz and Buscalioni, 1992). The

6 AMERICAN MUSEUM NOVITATES NO. 3133

FAC

ULN~~~ ~ ~ ~ ~ ~~~U

III

IVX

Iv~~~~~~~~~~~~~~~~~~~~~~~~~'

II

Fig. 3.= 10 mm.

Diagram of Concornis lacustris (LH-28 14). Stippled areas indicate broken portions. Scale bar


skeletal elements have suffered a slight dor-soventral compression with some areas (e.g.,thorax and pelvis) sloping toward the left sideof the specimen. No other major displace-ments are present.

AxIAL SKELETON

The new preparation has revealed frag-ments offour dorsal vertebrae, the most cau-dal portion of the synsacrum, and the firsttwo caudal vertebrae.DoRSAL VERTEBRAE: The preserved trunk

elements are regarded as midposterior dor-sals on the basis of the absence of ventralprocesses. The four vertebrae are well pre-served: the anterior three are exposed in ven-tral view and the last one in lateral view. Thearticular surfaces are amphicoelous. The ver-tebral bodies are laterally compressed. As inother enantiornithines (e.g., Neuquenornisvolans, Cathayornis yandica, El Brete speci-mens), these bodies are excavated by deeplongitudinal depressions and their par-apophyses (costal foveae) are developed in acentral position (figs. 4, 5).SYNSACRAL VERTEBRAE: The last two ver-

tebrae of the synsacrum are preserved, andbear elongate transverse processes, which re-tain their individuality and are caudally di-rected (fig. 3). The last process is narrowerthan the preceding one. This trend continuesin the caudal vertebrae.CAUDAL VERTEBRAE: The caudal centra

seem to be amphicoelous. These vertebraehave elongate and narrow transverse pro-cesses that, like those of the synsacral ver-tebrae, are directed caudally (fig. 3). At thecranial border of the second caudal there isa bifurcated, ventrally projected ossificationthat probably represents the hemal arch, asin some extant birds (e.g., Accipitridae).

RIBS: Fragments of several ribs are pre-served, although none ofthem are articulatedwith the vertebrae. Three of these ribs arefairly complete, and lack uncinate processes.

THORAcIC GIRDLE AND STERNUMCORACOID: Both coracoids are preserved,

exposed in ventral view (figs. 6, 7). The ad-ditional preparation has revealed the shoul-der end of the right coracoid, while that of

the left one is missing. The coracoid of Con-cornis lacustris is elongate and lacks lateraland procoracoid processes (Sanz and Bus-calioni, 1992). The sternal half is broad andit has a convex ventral surface. Missing partsof the right coracoid show that the sternalhalf is thin. This fact and the convexity ofthe ventral surface suggest a dorsal fossa likethose ofother enantiornithines (Chiappe andCalvo, 1994), although this is not verifiablefrom the present preparation. As pointed outby Sanz and Buscalioni (1992), the lateralmargin of the sternal half of the coracoid isremarkably convex, as in the Enantiornithes(figs. 6, 7). The medial margin is concave.The sternal border is slightly concave, butlacks the remarkable concavity shown by cer-tain enantiornithines (Walker, 1981; Chiappeand Calvo, 1994). The shoulder area is veryslender, and its ventral surface forms a ridge.Medially, the supracoracoid nerve foramenopens into a medial groove and is separatedfrom the dorsal surface by a thick bar, a con-dition also found in Enantiornithes (Chiappe,1992a) (fig. 7). Laterally, the shoulder endslopes dorsosternally also as in the enantior-nithines.

ScAPuLA: The new preparation has re-vealed the shoulder portions ofboth scapulaeand part of the body of the right one (fig. 3).Unfortunately, the preserved areas containlittle significant information. The scapularshaft is straight, lacking a sagittal curvature.On the shoulder end of the left scapula thereis a subrectangular articular area interpretedhere as the articular facet for the furcula.FuRcuLA: The furcula is almost complete

and is exposed cranially (figs. 6, 7). It is veryrobust, and as in the Patagonian enantior-nithine Neuquenornis volans (Chiappe andCalvo, 1994) is laterally excavated with thecranial margin projected externally. At theshoulder end, below the articular area, it islaterally compressed with no external projec-tion. The clavicular rami are separated fromeach other by an angle of 60° (Sanz and Bus-calioni, 1992). The hypocleideum tapers dis-tally, and its cranial face bears a weak crest.In contrast to N. volans, the hypocleideumof Concornis lacustris is very long as in theChinese enantiornithine Cathayornis yandi-ca (Zhou et al., 1992).


\'* ' ffi

a.veN-.- l? _ v w.

-.-.-. .:.'

;,1iC >

''SL'[:as,

- '. 4;s so

'*4t!:'

ts. ,s

J- f

S8

-'- £f.So}.. , n;.fI_wb

h!,,.,,, o 4SE

vf F;>V'*A,

;

f }k:.,,ef/ .:-;;

b

',* ,

' .I

i4.

/ 'I..I

sY.4I

Fig. 4. Dorsal vertebrae of Concornis lacustris (LH-28 14). Scale bar = 1 mm.

8 NO. 3133

i4Fi.


STERNUM: In addition to the cranial areadescribed in the original study, the new prep-aration has revealed the caudal part of thesternum. The sternum is carinate (figs. 6, 7).The carina, however, is not developed in thecranial part of the sternum, but starts nearthe midpoint (Sanz and Buscalioni, 1992).Anterior to the carina, the sternum has abroad, flat surface, although the extension ofthis area might be exaggerated by postmor-tem compression (Sanz and Buscalioni, 1992).The cranial margin of the sternum is round.Lateral crests are well developed in front ofthe carina. As mentioned by Sanz and Bus-calioni (1992), three foramina perforate theright side of the sternal body, and two per-forate the left one. The caudal half of thesternum is deeply notched, exhibiting lateraland medial processes. The lateral process isstout, elongate, and has a remarkable distalexpansion (figs. 6, 7). The morphology ofthissternal process is very similar to those ofNeu-quenornis volans and Cathayornis yandica.The medial process is only preserved at itsbase. The latter process, however, appears tobe less robust than the lateral one (Sanz andBuscalioni, 1992). At the level of the originofthe carina, there is another process branch-ing off craniodorsally from the base of thelateral process. Cranial to this process thereis a short notch interpreted as the articulararea of the ribs (fig. 7). If this is correct, thisarea was very short, representing approxi-mately 15% of the sternal length.

THoRAcIc LIMBHUMERUS: Additional preparation has re-

vealed the right humerus, exposed in cranialview (fig. 6). It is as long as or slightly shorterthan the ulna (figs. 2, 3). Although the hu-merus is crushed, some extent of disphase-ment between the axial planes of both ex-tremities is visible, a primitive conditionknown to occur in Enantiornithes and otherbasal birds (Chiappe, 1992b). At the proxi-mal end, the pectoral crest does not curvecranially. In proximal view, the head is cra-nially concave and caudally convex. Distalto the head, in cranial aspect, there is a cir-cular depression also found in Enantiornithesand the Patagonian bird Patagopteryx defer-rariisi (Chiappe, 1 992b) (fig. 7). Of interest

FCO

STE

1'1LAG

Fig. 5. Diagram of dorsal vertebrae of the en-anthiornithine Concornis lacustris (LH-2814).Stippled areas indicate broken portions. Scale bar= 1.2 mm.

is the strong development and cranioventralprojection ofthe bicipital area, a derived con-dition ofthe Enantiornithes (Chiappe, 1 992a).On the ventral face of the distal portion ofthis crest, there is a small, shallow depressionalso characteristic of the latter taxon (fig. 7).The proximal border of the bicipital crestbears a small but deep, subtriangular trans-verse ligamental groove. The development ofthis groove is considerably smaller than inthe El Brete enantiornithines. In caudal as-pect, the ventral tubercle is directed caudally.At the distal end, the condyles are not pre-served, although it is evident that they werelocated on the cranial aspect. The distal ex-tremity is craniocaudally compressed, and itscranial surface does not show any evidenceof the fossa for the brachial muscle.ULNA: Only the proximal half of the right

ulna is preserved, and it is exposed cranio-


Fig. 6. Thoracic region of Concornis lacustris (LH-2814). Scale bar = 10 mm.

ventrally (figs. 2, 3). The ulna is significantlybroader and more robust than the radius, withan approximate proportion of 2:1 with re-spect to the radial shaft (Sanz and Buscalioni,1992). In the proximal articular area, the dor-sal cotyla is convex, although less so than in

the El Brete enantiornithines. The dorsal co-tyla is separated from the ventral cotyla andthe olecranon by a shallow depression (figs.8, 9), a condition also shared by other en-antiornithines (Walker, 1981). The dorsal co-tyla is strongly projected dorsocaudally. The

lo NO. 3133

AF


STR

Fig. 7. Diagram of thoracic region of Concornis lacustris (LH-28 14). Scale bar = 1.2 mm.

ventral cotyla is concave, and in its devel-opment the olecranon resembles that of theEl Brete enantiornithines. The cranial surfaceimmediately distal to the articular surface isnot well preserved, although there is a dis-tinct, round bicipital tubercle (fig. 9).

RADIus: The right radius is missing its dis-tal end (figs. 2, 3). It is significantly moreslender than the ulna. Unfortunately, it is notpossible to confirm the presence of the axialgroove typical ofthe radial shaft ofother En-antiornithes.METACARPALS AND MANUAL PHALANGES:

The hand of Concornis lacustris is slightlyshorter than the forearm (figs. 2, 3). The ma-jor metacarpal is the only one preserved, al-though some information about the other el-ements is available from their impressions.The major metacarpal is the most robust. Itis straight, of uniform width, and its ventralsurface is convex. On the basis of its im-pression, the dorsal surface ofthe minor met-acarpal was sharp. The intermetacarpal spacebetween these metacarpals is narrow (Sanzand Buscalioni, 1992), although wider thanin Neuquenornis volans. The major and mi-


HUM

- cvs

-TUB

ULN

Fig. 8. Distal end of humerus and proximalend of ulna-radius of Concornis lacustris (LH-2814). Scale bar = 1 mm.

nor metacarpals appear to be fused distally.The proximal area, however, is damaged andthe fusion between metacarpals and distalcarpals remain uncertain. The alular meta-carpal is not preserved. The distance betweenthe proximal phalanx of this digit and theproximal end of the metacarpals suggests,however, that it must have been short, as inother enantiornithines and more advancedbirds.The alular digit is preserved as an impres-

sion and bears two phalanges (fig. 3). Theproximal phalanx is slender and elongate; thedistal phalanx is a claw. The major digit hasthree phalanges, all preserved as impressions.The proximal phalanx seems to be the largest,as in Cathayornis yandica; it differs from theprimitive condition present in Archaeopteryxlithographica and Patagopteryx deferrariisi in

Fig. 9. Diagram of distal end of humerus andproximal end ofulna-radius ofConcornis lacustris(LH-2814). Stippled areas indicate broken por-tions. Scale bar = 1.2 mm.

which the intermediate phalanx is the largestof the three. This phalanx, however, is slen-der, lacking the craniocaudal expansion typ-ical ofmodem birds. A longitudinal ridge ofmatrix suggests the presence ofan axial grooveon its dorsal surface. The distal phalanx ofthis digit (fig. 3) is also a claw as in C. yandica.In the minor digit, only a single phalanx ispreserved, also as an impression. This pha-lanx is slender and tiny. Hence, in contrastto previous considerations (Sanz and Bus-calioni, 1992), the hand of Concornis lacus-tris has claws and its phalangeal formula is2-3-1 ?-x-x.

PELVIC GIRDLE

PUBIS: Both pubes are preserved in ventralview (fig. 10). Their most proximal portionsare missing. The pubis of Concornis lacustrisis slender, elongate, and suboval in cross sec-tion. It was probably oriented posteriorly, al-though it is not possible to ascertain the de-gree of reversion. Distally, the apices of bothpubes contact each other, forming a shortsymphysis (fig. 10). The distal end ofthe sym-

12 NO. 3133


K..1'I .

/ - /.

./ ./ .d

. /

4

I)r . .A,= .

7

.f'

Fig. 10. Pelvic girdle of Concornis lacustris (LH-2814). Scale bar = 5 mm.

physis is missing, hence the presence or ab-sence of a pubic foot remains uncertain.

IsCHIUM: The left ischium is fairly com-plete, while the right one is only representedby a distal fragment. The ischium is approx-imately 25% shorter than the pubis. It is lam-inar and laterally compressed. The proximalportion is dorsoventrally broad. Distally, theischium narrows and in its distal two-thirdsit has a ribbonlike aspect (fig. 10). Like thepubis, the ischium seems to be externallyconvex along almost all its length. The pre-served proximal portion is difficult to inter-pret. There is an ample ventral process witha sharp end that is interpreted as a prominentobturator process. Internal to the dorsal bor-der, at the proximal end, there is a fragmentofbone representing an ischiadic dorsal pro-cess. The existence of such a process is notunexpected considering its presence in theischium ofthe El Brete enantiornithines. Dis-tally the terminal processes ofboth ischia arenot in contact.

ILIUM: Unfortunately, there is no availableinformation.

PELVIC LIMB

FEMUR: Both femora are preserved (fig. 3).The right is exposed in medial view and theleft one craniomedially. The femur is robustand considerably shorter than the tibiotarsus.The femoral shaft is slightly curved, with thecranial surface convex. The femoral head isoval and well projected medially. A depres-sion for the insertion of the round ligamentappears to be absent. The lateral margin ofthe most proximal end of the left femur in-clines medially as in the El Brete enantior-nithines. This morphology may suggest thepresence of the enantiornithine hypertro-phied posterior trochanter (Chiappe, 1992a;Chiappe and Calvo, 1994), although this con-dition is not exposed. The distal end lacks aprominent patellar groove. The medial sur-face of the distal end is excavated by a sem-


A-

A. -It. t

l '*~~~~~~~3"iIts

-,ALIft

rs ^_1N

'NCPR

I5

CFII

"I'~,1

L:

Fig. 11. Distal end ofright femur and proximalend oftibiotarsus ofConcornis lacustris (LH-28 14).Note detail of caudal projection of lateral marginof femur, and the fibular crest. Scale bar = 1 mm.

ilunate depression with the concave borderoriented proximally. The lateral margin ofthe distal end, proximal to the lateral con-

dyle, projects caudally (figs. 11, 12) as in theEl Brete enantiornithines and Neuquenornisvolans.TIBIOTARSUS: Both tibiotarsi are preserved,

and exposed in craniomedial view. The tibio-tarsus is gracile (fig. 3). Its fibular crest is shortand strongly developed (figs. 1 1, 12). The de-gree of lateral projection ofthis crest increas-es distally, ending abruptly in a stout tuber-cle. The proximal surface of the tibiotarsusis round as in most enantiornithines [withthe exception of the Argentine Lectavis bre-tincola (Chiappe, 1993)]. The articular sur-

face of the proximal end slopes laterocrani-

Fig. 12. Diagram of distal end of right femurand proximal end of tibiotarsus of Concornis la-custris (LH-2814). Stippled areas indicate brokenportions. Scale bar = 1.2 mm.

ally. In the proximal end there is a single,weak, mediocranial cnemial crest as in theAustralian enantiornithine Nanantius eos(Molnar, 1986), Neuquenornis volans, and theforms from El Brete.

Distally, the proximal tarsals are complete-ly fused to the tibia. There is no evidence ofa distinct ascending process or sutures be-tween the tibia and the proximal tarsals. Themedial condyle is transversely broad (fig. 13)as in other Enantiornithes (Walker, 1981;Molnar, 1986; Chiappe, 1993), although notstrongly projected cranially as in Lectavis bre-tincola. In both tibiotarsi there is a proxi-modistally elongate depression on the cranialsurface of the distal end. In the right elementit forms a sort of fossa, although this may bean artifact ofpreservation. The midshaft sec-tion is subcircular and not craniocaudallycompressed as in L. bretincola.TARSOMETATARSUS: The right tarsometa-

14 NO. 3133


Fig. 13. Tarsometatarsus of Concornis lacustris (LH-2814). Scale bar = 5 mm.

tarsus is preserved and exposed in dorsal view(fig. 13). The left one is only represented byits most proximal end. The tarsometatarsusis significantly shorter than the tibiotarsus.Metatarsals II-IV are straight throughout theirlength. They are firmly united to each otherbut only fused proximally (Sanz and Busca-lioni, 1992). At the proximal end, the inter-cotylar eminence is not developed. The prox-imal articular surface does not slope dorsallyas in several avisaurid enantiornithines (e.g.,Soroavisaurus australis, Avisaurus archibaldi;Chiappe, 1993). Metatarsals II-IV are locat-ed in the same plane, although metatarsal IIIis located slightly forward to the other two.Metatarsal III is the longest (fig. 13). The cen-tral portion of the shaft of metatarsal III isstrongly convex, as in Avisauridae (Chiappe,1992a, 1993). The trochlea of metatarsal IIIis badly preserved; nevertheless, it is evidentthat it is narrower than that of metatarsal IIas in other enantiornithines. In dorsal view,metatarsals II and III are the most conspic-

uous. Metatarsal IV is slightly more slender.At the distal end, its lateral surface is some-what excavated. The trochlea of metatarsalIV is significantly more reduced than thoseof metatarsals II and III, and it is stronglycurved laterally. Unfortunately, the presenceor absence ofthe dorsolateral tubercle, whichis present in the metatarsal II of other En-antiornithes (Chiappe, 1993), is not discern-ible. Distally, metatarsals II-IV are firmlyunited in broad contact, unlike those in S.australis in which the distal end ofmetatarsalII is widely separated from metatarsal III. InConcornis lacustris, the distal end of meta-tarsal II is nearly straight, lacking the strongmedial curvature found in some avisaurids(Chiappe, 1993). As in the latter taxa, meta-tarsal I is not reverted but articulates withthe medial margin of metatarsal II. Metatar-sal I is straight, lacking the J-like morphologypresent in S. australis and Neuquenornis vo-lans (Chiappe, 1993). The dorsal portion ofthis metatarsal is not completely preserved.


The plantar area of its trochlea is broad andis separated from the dorsal area by a con-striction. This morphology defines a trans-verse ginglymoid articulation.PEDAL PHALANGES: Only the right foot pre-

serves its phalanges. Digits I and IV are com-plete and the I is reverted (fig. 13). It is robustand possesses a well-developed distal (un-gual) phalanx. The intermediate phalanges ofdigit IV are relatively long. The distal pha-langes lack a pronounced flexor tubercle (Sanzand Buscalioni, 1992).

PHYLOGENETIC DISCUSSION

Sanz and Buscalioni (1992) recognized thatConcornis lacustris was different from andmore derived than Iberomesornis romerali,but their study only arrived at a tentativehypothesis of relationship. This preliminarystudy interpreted C. lacustris as the sistergroup of Ornithurae, although the inclusionoftwo more taxa (Ambiortus dementjevi andEnantiornithes) in the cladistic analysis re-

sulted in a large number of different topol-ogies (Sanz and Buscalioni, 1992). The newosteological information allows a more de-tailed analysis of the phylogenetic relation-ships of C. lacustris within the context of themost relevant ornithothoracine taxa.Appendix 2 shows the 84 characters used

to analyze the relationships of the Ornith-othoraces (see Chiappe, 1992b, in press a, b;Chiappe and Calvo, 1994). Three most par-simonious trees (consistency index 0.84; re-tention index 0.87; length 101) resulted fromthe analysis of this data matrix. These threetopologies vary in the relationships among

the ornithurine taxa used in the analysis(Hesperornithiformes, Ichthyornithiformes,and Neornithes), and therefore they are ir-relevant for establishing the phylogenetic re-

lationships of Concornis lacustris.The strict consensus cladogram of figure

14 shows that Concornis lacustris is closelyrelated to all other enantiornithines. All thesetaxa share the following 12 unambiguoussynapomorphies: humerus with a prominent,cranioventrally projected, bicipital crest(character 72), and a small fossa for muscularattachment on the lateral surface ofthis crest(character 83); ulna with convex dorsal co-

tyle, separated from the olecranon by a groove

(character 73); coracoid with convex lateralmargin (character 74), and with the supra-coracoid nerve foramen opening into an elon-gate medial furrow and separated from themedial margin by a thick bony bar (character75); dorsal vertebrae with strong lateralgrooves on the centra (character 76), and theparapophyses (costal foveae) located in thecentral part of the centrum (character 77);lateral border of the distal end of the femurprojecting caudally (character 78); wide andbulbous medial condyle of the distal end ofthe tibiotarsus (character 79); tarsometatar-sus with metatarsal IV significantly smallerthan metatarsals II and III (character 80), andtrochlea of metatarsal II broader than thoseof metatarsals III and IV (character 81); fur-cula laterally excavated (character 84). Thepresence of a round articular surface at theproximal end ofthe tibiotarsus (character 82)supports this relationship as well, althoughthis character has an ambiguous optimiza-tion in the present analysis. In addition tothese synapomorphies, C. lacustris shares thedistal expansion of the lateral process of thesternum with the enantiornithines Neuquen-ornis volans and Cathayornis yandica. Thischaracter was not included in the cladisticanalysis because the sternum is poorly pre-served in several ingroup taxa (e.g., Ibero-mesornis romerali, Patagopteryx deferrariisi)and because its ancestral condition is uncer-tain as well.As mentioned above (see Materials and

Methods), the interrelationships within theEnantiornithes are not well understood, a re-ality that stems from the fact that most en-antiornithine species are known from isolat-ed bones or fragmentary specimens (Walker,1981; Molnar, 1986; Chiappe, 1993), or thatformal and detailed descriptions are stillpending (Hutchinson, 1993; Dong, 1993).Besides Concornis lacustris, Neuquenornisvolans and Cathayornis yandica are the onlyenantiornithines known (or at least pub-lished) from fairly complete specimens.Chiappe (1993) presented the first attempt tounderstand the interrelationships among dif-ferent species of Enantiornithes using datafrom the tarsometatarsus. In the present pa-per the available information from the tar-sometatarsus of C. lacustris was incorporatedinto these data (Appendix 3). Three most par-

16 NO. 3133


Fig. 14. Strict consensus cladogram (consistency index 0.84; retention index 0.87; length 101) illus-trating the phylogenetic relationships of Concornis lacustris to other Mesozoic taxa (see Appendix 2 forlist of characters). Characters diagnosing different nodes are: Node 1, Ornithothoraces: 13, 14, 19, 24,28, 32*, 34, 40. Node 2: 1*, 2*, 3*, 5*, 8*, 16*, 17, 18, 22*, 25, 26*, 27*, 30*, 31*, 36*, 37, 38, 39*,41*, 42*, 43*, 46*, 49*, 55*, 57*, 58*, 60, 64*, 65, 68*. Node 3: 10*, 11*, 29, 33, 48, 50, 52*, 53*, 59,62, 66, 71. Node 4, Ornithurae: 4*, 6*, 7*, 9, 12, 15, 20*, 21, 23, 35, 44, 45, 47*, 51, 54*, 56, 58*, 61,63, 67*, 69, 70. Node 5: 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82*, 83, 84. Asterisked characters havean ambiguous optimization.

simonious cladograms (consistency index0.83; length 12) resulted from this analysis,and their strict consensus tree is illustratedin figure 15. The three fundamental clado-grams depict the alternative relations amongthe Argentine Lectavis bretincola and Yun-gavolucris brevipedalis, and a clade formedby C. lacustris and the Avisauridae. The sis-ter group relationship of the latter two taxais supported unambiguously in these threecladograms by the strong transverse convex-ity of the dorsal surface of the midshaft ofmetatarsal III (see Appendix 3, character 8).

Our phylogenetic study indicates that Con-cornis lacustris shares a large number of syn-apomorphies with other enantiornithine birds(fig. 14). Though restricted to tarsometatarsaldata, this study regards C. lacustris as thesister taxon of the Avisauridae (fig. 15), anenantiornithine clade known from the LateCretaceous ofArgentina and the United States(Chiappe, 1993). The common presence ofnumerous derived characters, along with itsnested position within defined enantiorni-thine taxa, heavily supports the identificationof C. lacustris as an enantiornithine species.


1 ENANTIORNITHES

Fig. 15. Strict consensus cladogram (consistency index 0.83; retention index 0.87; length 12) illus-trating the phylogenetic relationships of Concornis lacustris to other enantiornithine taxa (see Appendix3 for list ofcharacters). Characters diagnosing different nodes are: Node 1, Enantiornithes: 1, 4, 10. Node2: 8. Node 3, Avisauridae: 2*, 5. Node 4: 7*, 9. Node 5: 6. Asterisked characters have an ambiguousoptimization.

CONCLUSIONThe allocation ofConcornis lacustris to the

Enantiornithes is well substantiated by thepresent study. C. lacustris is therefore the old-est known enantiornithine bird from theWestern Hemisphere. Its Barremian age andthe fact that it appears to be phylogeneticallyinternested within several of the latest Cre-taceous taxa suggest a much earlier diversi-fication for these taxa.The interpretation of Concornis lacustris

as an Enantiornithes emphasizes anew the

wide geographical and stratigraphical distri-bution of this clade during the Cretaceous.This conclusion also highlights the remark-able taxonomic diversity attained by the En-antiornithes.

ACKNOWLEDGMENTSWe are especially grateful to A. Diaz

Romeral for donating the holotype of Con-cornis lacustris. We are also grateful to K.Padian (Museum ofPaleontology, University

18 NO. 3133


of California, Berkeley) and D. Unwin (Uni-versity of Bristol, Bristol) for their construc-tive reviews. We also thank G. F. Kurtz forthe photographs, J. Bonaparte (Museo Ar-gentino de Ciencias Naturales, Buenos Aires)and F. Ortega (Universidad Autonoma, Ma-drid) for the preparation ofthe specimen, andNicholas Frankfurt (AMNH) for mounting

the illustrations. This research was supportedby the Junta de Comunidades de Castilla-LaMancha, DYGICYT (Promocion General delConocimiento), the European Community(Human Capital and Mobility Program), anda Frick Research Fellowship (AMNH) to L.Chiappe.

REFERENCES

Alvarenga, H. M. F., and J. F. Bonaparte1992. A new flightless land bird from the Cre-

taceous ofPatagonia. In K. E. Campbell(ed.), Papers in avian paleontology,honoring Pierce Brodkorb, pp. 51-64.Los Angeles: Natural History Museumof Los Angeles County, Science Series36.

Baumel, J. J., and L. Witmer1993. Osteologia. In J. J. Baumel (ed.) Hand-

book of avian anatomy: Nomina ana-tomica avium (2nd. ed.). Publ. NuttallOrnithol. Club 23: 45-132.

Brett-Surman, M. K., and G. Paul1985. A new family of bird-like dinosaurs

linking Laurasia and Gondwanaland. J.Vertebr. Paleontol. 5(2): 133-138.

Carpenter, J. M.1988. Choosing among multiple equally par-

simonious cladograms. Cladistics 4:291-296.

Chiappe, L. M.1991. Cretaceous avian remains from Pata-

gonia shed new light on the early radi-ation ofbirds. Alcheringa 15(3-4): 333-338.

1992a. Enantiornithine tarsometatarsi and theavian affinity of the Late CretaceousAvisauridae. J. Vertebr. Paleontol. 12(3):344-350.

1992b. Osteologia y sistematica de Patagopte-ryx deferrariisi Alvarenga y Bonaparte(Aves) del Cretacico de Patagonia. Fil-ogenia e historia biogeografica de las avesCretacicas de America del Sur. Ph.D.diss., Universidad de Buenos Aires,Buenos Aires, 429 pp.

1993. Enantiornithine (Aves) tarsometatarsifrom the Cretaceous Lecho Formationof Northwestern Argentina. Am. Mus.Novitates 3079: 27 pp.

In press a. Late Cretaceous birds of southernSouth America: anatomy and system-atics of Enantiornithes and Patagopte-ryx deferrariisi. Rev. Geologica Chilena(n.s.).

In press b. Phylogenetic position of the Creta-ceous birds of Argentina: Enantior-nithes and Patagopteryx deferrariisi. InD. S. Peters (ed.), Proc. 3rd. Symp. Soc.Avian Paleontology and Evolution.Frankfurt: Courier Forschungsinstitut-Senckenberg.

Chiappe, L. M., and J. 0. Calvo1994. Neuquenornis volans, a new Upper Cre-

taceous bird (Enantiornithes: Avisaur-idae) from Patagonia, Argentina. J. Ver-tebr. Paleontol. 14(2): 230-246.

Dong, Z. M.1993. A Lower Cretaceous enantiornithine

bird from the Ordos Basin of InnerMongolia, Peopleaas Republic of Chi-na. Can. J. Earth Sci. 30(10-11): 2177-2179.

Farris, J.1988. Hennig 86 references. Documentation

for version 1.5. Privately published.Fregenal-Martinez, M. A.

1991. El sistema lacustre de Las Hoyas (Cre-tacico inferior, Serrania de Cuenca): Es-tratigrafia y Sedimentologia. Tesis deLicenciatura, Departamento de Estra-tigrafia, Universidad Complutense deMadrid, 226 pp.

Fregenal-Martinez, M. A., and N. Melendez1993. Sedimentologia y evoluci6n paleogeo-

grifica de la cubeta de Las Hoyas (Cre-tacico inferior, Serrania de Cuenca).Cuad. de Geol. Iberica 17: 231-256.

Hutchinson, H.1993. Avisaurus: a "dinosaur" grows wings. J.

Vertebr. Paleontol. 13(3), Suppl.: 43.Kurochkin, E. N.

1985. A true carinate bird from Lower Cre-taceous deposits in Mongolia and otherevidence of early Cretaceous birds inAsia. Cretaceous Res. 6: 271-278.

Molnar, R. E.1986. An enantiornithine bird from the Lower

Cretaceous of Queensland, Australia.Nature 322: 736-738.


Nixon, K. C., and Q. D. Wheeler1990. An amplification of the phylogenetic

species concept. Cladistics 6: 211-223.1992. Extinction and the origin of species. In

M. J. Novacek and Q. D. Wheeler (eds.),Extinction and Phylogeny, pp. 19-143.New York: Columbia Univ. Press.

Perle, A., M. A. Norell, L. M. Chiappe, and J. M.Clark

1993. Flightless bird from the Cretaceous ofMongolia. Nature 362: 623-626.

Perle, A., L. M. Chiappe, Barsbold R., J. M. Clark,and M. A. Norell

1994. Skeletal morphology ofMononykus ole-cranus (Theropoda: Avialae) from theLate Cretaceous ofMongolia. Am. Mus.Novitates 3105: 29 pp.

Sanz, J. L., Bonaparte, J. F., and A. Lacasa1988a. Unusual Early Cretaceous bird from

Spain. Nature 331: 433-435.Sanz, J. L., S. Wenz; A. Yebenes, R. Estes, X.

Martinez-Delclos, E. Jimenez-Fuentes,C. Dieguez, A. D. Buscalioni, J. L. Bar-badillo, and L. Via

1988b. An Early Cretaceous faunal and floralcontinental assemblage: Las Hoyas fos-sil site (Cuenca, Spain). Geobios 21(5):611-635.

Sanz, J. L., C. Dieguez, M. A. Fregenal-Martinez,X. Martinez-Delclos, N. Melendez, andF. J. Poyato-Ariza

1990. El yacimiento de f6siles del Cretacicoinferior de Las Hoyas, Provincia deCuenca (Espania). Com. Reunion de Ta-fonomia y Fosilizacion, Madrid, pp.337-355.

Sanz, J. L., and A. Buscalioni1992. A new bird from the Early Cretaceous

of Las Hoyas, Spain, and the Early ra-diation of birds. Paleontology 35(4):829-845.

Sereno, P. C., and Rao, C.1992. Early evolution of avian flight and

perching: new evidence from LowerCretaceous of China. Science 225: 845-848.

Varricchio, D. J., and Chiappe, L. M.1995. A new enantiornithine bird from the

CretaceousTwo Medicine Formation ofMontana. J. Vertebr. Paleontol. 15(1).

Walker, C. A.1981. New subclass of birds from the Creta-

ceous of South America. Nature 292:51-53.

Wiley, E. O., D. Siegel-Causey, D. R. Brooks, andV. A. Funk

1991. The compleat cladist. Univ. KansasMus. Nat. Hist., Spec. Publ.19: 1-158.

Zhou, Z., Jin, F., and Zhang, J.1992. Preliminary report on a Mesozoic bird

from Liaoning, China. Chinese Sci. Bull.37(16): 1365-1368.

APPENDIX 1Anatomical abbreviations used in figures 2-12

alular digitcarinadorsal cotyle of the ulnacircular depression of proximal humerusfibular crestmedial condyle of the femurcostal margin of the coracoidcoracoidcaudal projection of the lateral margin ofthe femurcaudal vertebraeventral cotyle of the ulnadepression ofbicipital area ofthe humerusdorsal vertebraefurcular facet of scapulacostal fossae (parapophysis)femurfemoral headforaminafurcula

HUMHYPISC

IDOLAGLCRLGRLPRMACMADMIC

MIDMCO

MPROPRPECPUBRADRIBSCA

humerushypocleidiumischium

dorsal process of ischiumcentral groove of the vertebral centrumlateral crests of the sternal carinaligamental groove of the humeruslateral process of the sternummajor metacarpalmajor digit of manusminor metacarpalminor digit of manusmedial condyle of tibiotarsusmedial process of the sternumobturator process of ischiumpectoral crestpubisradiusribsscapula

ALDCARCDOCIRCFICMECOMCORCPR

CVECVSDEPDVEFACFCOFEMFEHFORFUR

20 NO. 3133


supracoracoid nerve foramensternumgroove separating the CDO and CVEsynsacral vertebrae

TIT tibiotarsusTMT tarsometatarsusTUB bicipital tubercle of ulnaULN ulna

APPENDIX 2

List ofcharacters and data matrix (0, primitive;1, 2, derived; ?, missing or uncertain; N, not com-parable) used in the parsimony analysis. The the-ropod dinosaurs Dromaeosauridae and Archae-opteryx lithographica are used as outgroups.

1. Fusion of premaxillae in adults: absent (0),present (1).

2. Maxillary process ofpremaxilla forming mostof the facial margin: absent (0), present (1).

3. Frontal process of premaxilla extends cau-dally to the level of lacrimals: absent (0), present(1).

4. Premaxillary teeth absent: absent (0), present(1).

5. Osseous external naris considerably larger thanantorbital fenestra: absent (0), present (1).

6. Loss ofmaxillary fenestrae: absent (0), present(1).

7. Cranial jugal border very close to the caudalmargin of the osseous external naris: absent (0),present (1).

8. Ectopterygoid: present (0), absent (1).9. Orbital process of quadrate sharp and point-

ed: absent (0), present (1).10. Quadratojugal cotyle in lateral face ofman-

dibular process ofquadrate: absent (0), present (1).11. Quadrate pneumaticity: absent (0), present

(1).12. Articular pneumaticity: absent (0), present

(1).13. Prominent ventral processes on cervicodor-

sal vertebrae: absent (0), present (1).14. Dorsal vertebral count more (0) or less (1)

than 13-14 elements.15. Dorsal vertebral count more (0) or less (1)

than 11 elements.16. Wide vertebral foramen in dorsal series, with

ratio between the vertical diameter ofthe vertebralforamen and the articular cranial facet greater than0,40: absent (0), present (1).

17. Synsacrum with less (0) or more (1) than 8vertebrae.

18. Heterocoelous cervical vertebrae: absent (0),present (1).

19. Pygostyle: absent (0), present (1).20. Caudal vertebral count more (0) or less (1)

than 15 elements.

21. Ossified uncinate processes: absent (0), pres-ent (1).

22. Broad, firm articulation between the cora-coid and scapula: present (1), absent (0).

23. Procoracoid process: absent (0), present (1).24. Strutlike coracoid: absent (0), present (1).25. Scapulocoracoid articulation well below the

shoulder end of coracoid: absent (0), present (1).26. Supracoracoid nerve foramen (incision) lo-

cated in the medial margin ofcoracoid: absent (0),present (1).

27. Coracoid forming a sharp angle with thescapula at the level of the glenoid cavity: absent(0), present (1).

28. Scapula with sharp caudal end: absent (0),present (1).

29. Sagittally curved scapular shaft: absent (0),present (1).

30. Large, rectangular ossified shaped sternum:absent (0), present (1).

31. Ossified keel of sternum: absent (0), present(1).

32. U-shaped furcula, interclavicular anglesmaller than 900: absent (0), present (1).

33. Proximal and distal ends of humerus ex-panded roughly in the same plane: absent (0), pres-ent (1).

34. Humerus longer (0) or shorter than or nearlyequivalent (1) to ulna.

35. Craniocaudally convex, spherical head ofhumerus: absent (0), present (1).

36. Ventral tubercle of humerus projecting dis-tinctly caudally, separated from the humeral headby a deep capital incision: absent (0), present (1).

37. Humerus with well-developed transverseligamental groove: absent (0), present (1).

38. Distal condyles of humerus located mainlyon cranial aspect: absent (0), present (1).

39. Dorsal condyle of ulna developed as a sem-ilunar ridge: absent (0), present (1).

40. Shaft of radius considerably thinner thanthat of ulna, ratio of diameter of radius to ulnasmaller than 0.70: absent (0), present (1).

41. Carpometacarpus: absent (0), present (1).42. Extensor process on carpometacarpus: ab-

sent (0), present (1).43. Fused pelvic elements: absent (0), present

(1).

SNVSTRSVLSYV


44. Acetabulum small, ratio of acetabulum toilium less than or equal to 0.11: absent (0), present(1).

45. Ilium-ischium-pubis more-or-less parallel:absent (0), present (1).

46. Pelvis with prominent antitrochanter: ab-sent (0), present (1).

47. Iliac brevis fossa: present (0), absent (1).48. Iliac fossa for M. cuppedicus (= M. iliofe-

moralis internus): present (0), absent (1).49. Ischiadic terminal processes not in contact:

absent (0), present (1).50. Apices of pubis not in contact: absent (0),

present (1).51. Shaft ofpubis laterally compressed through-

out its length: absent (0), present (1).52. Pubic foot: present (0), absent (1).53. Retroverted pubis, separated from the main

synsacral axis by an angle smaller than 450: absent(0), present (1).

54. Femur with distinct fossa for capital liga-ment: absent (0), present (1).

55. Trochanteric crest offemur: absent (0), pres-ent (1).

56. Femur with deep patellar groove: absent (0),present (1).

57. Popliteal fossa of femur bounded distally bya transverse ridge: absent (0), present (1).

58. Tibiofibular crest in the lateral condyle offemur: absent (0), poorly developed (1), prominent(2).

59. Posterior trochanter of femur: present (0),absent (1).

60. Complete fusion of calcaneum, astragalus,and tibia: absent (0), present (1).

61. Cranial cnemial crest of tibiotarsus: absent(0), present (1).

62. Extensor canal on tibiotarsus: absent (0),present (1).

63. Iliofibularis tubercle of fibula caudolaterallyor caudally directed: absent (0), present (1).

64. Fibula greatly reduced in length: absent (0),present (1).

65. Free distal tarsals in adults: absent (0), pres-ent (1).

TAXA 5 10 15 20 25 30

DromaeosauridaeArchaeopteryx

IberomesornisEnantiornithesPatagopteryxHesperornithiformesIchthyornithiformesNeornithesConcornis

66. Distal tarsals and metatarsals fused com-pletely to form a tarsometatarsus: absent (0), pres-ent (1).

67. Metatarsal fusion starting distally: absent(0), present (1).

68. Metatarsal V: present (0), absent (1).69. Proximal end of metatarsal III plantarily

displaced with respect to metatarsals II and IV:absent (0), present (1).

70. Intercondylar eminence of tarsometatarsuswell developed: absent (0), present (1).

71. Distal vascular foramen oftarsometatarsus:absent (0), present (1).

72. Prominent bicipital crest of humerus, cran-ioventrally projecting: absent (0), present (1).

73. Convex external cotyle of ulna, separatedfrom the olecranon by a groove: absent (0), present(1).

74. Convex lateral margin of coracoid: absent(0), present (1).

75. Supracoracoid nerve foramen of coracoidopening into an elongate furrow medially and sep-arated from the medial margin by a thick bonybar: absent (0), present (1).

76. Strong lateral grooves on the bodies ofdorsalvertebrae: absent (0), present (1).

77. Costal fossae (parapophyses) located in thecentral part of the bodies of dorsal vertebrae: ab-sent (0), present (1).

78. Caudal projection of the lateral border ofthe distal end of femur: absent (0), present (1).

79. Wide and bulbous medial condyle of tibio-tarsus: absent (0), present (1).

80. Metatarsal IV significantly smaller thanmetatarsals II and III: absent (0), present (1).

81. Trochlea metatarsal II broader than thetrochlea ofmetatarsals III and IV: absent (0), pres-ent (1).

82. Round articular surface in the proximal endof tibiotarsus: absent (0), present (1).

83. Lateral face of the humeral bicipital crestwith a small fossa for muscular attachment: absent(0), present (1).

84. Furcula laterally excavated: absent (0), pres-ent (1).

CHARACTERS35 40 45 50 55 60 65 70 75 80 84

00000 00000 00000 00000 00000 00000 0?000 00000 00000 00000 00000 00000 00000 00000 00000 00000 000?00000 00000 0?000 ?0000 00000 0000? 0000? 00000 00000 000?0 000?0 0?000 00?00 00000 00?00 000?0 ?0?0

????? ????? ??110 ?0010 0??10 ??10? ?1?10 ?00?1 ???00 ????? 0?0?? 0???0 0???0 00?00 00?0? 00??0 010?11101 ??10? ?01?? 1111? 01011 11101 11010 11111 1110? 110?? ???11 01101 00?11 00100 01111 11111 1111????? ???01 10110 111?? 01011 1111? ??100 ??1?1 1?100 10111 011?? 01?11 01011 1?100 10000 00000 00??11111 11111 01111 11111 11111 10011 ONNO1 00100 ??111 1111111111112111111111111100000000000001?1?? ???11 11111 11011 ?1111 11111 11111 11111 11111 1?111 11111 11211 11?11 17111 10000 10000 001?11111 11111 11111 11111 11111 11111 11111 11111 11111 11111 11111 11211 11111 11111 10000000000000?????777?? ????? ???? 01011 11?01 11010 1 111 ?????????10 0??0? 0???1 0??11 00?00 01111 11111 1111

22 NO. 3133


APPENDIX 3

List ofcharacters and data matrix (0, primitive;1, derived; ?, missing or uncertain) used in thecladistic analysis of tarsometatarsal characters(after Chiappe, 1993; see Varricchio and Chiappe,1995, for Avisaurus gloriae). Mononykus olecran-us (Perle et al., 1993, 1994) and Patagopteryx de-ferrariisi (Alvarenga and Bonaparte, 1992) are usedas outgroups.

1. Metatarsal IV: large (0); considerably smaller(1).

2. Medial rim of the trochlea of metatarsal IIIwith a strong plantar projection: absent (0); present(1).

3. Plantar surface oftarsometatarsus excavated:absent (0); present (1).

4. Tubercle on the dorsal face of metatarsal II:absent (0); present (1).

5. J-shaped metatarsal I: absent (0); present (1).6. Trochlea of metatarsal IV medially concave:

absent (0); present (1).7. Proximal articular surface dorsally inclined:

absent (0); present (1).

TAXA CHARACTERS

Mononykus olecranus 00000 00000Patagopteryx deferrariisi 00100 00000

Yungavolucris brevipedalis 1001? ?0001Lectavis bretincola 1?11? ?00??Avisaurus archibaldi 11 1? 11111Soroavisaurus australis 11111 01111Neuquenomis volans 111 ?1 ??101Avisaurus glonae 1101? 11111Concomis lacustris 1???0 00101

8. Strong transverse convexity ofthe dorsal sur-face of the midshaft of metatarsal III: absent (0);present (1).

9. Distal end of metatarsal II strongly curvedmedially: absent (0); present (1).

10. Trochlea of metatarsal II broader than thetrochleae ofmetatarsal III and IV: absent (0); pres-ent (1).

Recent issues of the Novitates may be purchased from the Museum. Lists of back issues of theNovitates, Bulletin, and Anthropological Papers published during the last five years are availablefree of charge. Address orders to: American Museum of Natural History Library, Department D,Central Park West at 79th St., New York, N.Y. 10024. TEL: (212) 769-5545. FAX: (212) 769-5009. E-MAIL: [email protected]

® This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

AMERICAN MUSEUM Novltates

Documents

Transcript of AMERICAN MUSEUM Novltates