archosaur adductor chamber evolution: integration of

Archosaur Adductor Chamber Evolution: Integrationof Musculoskeletal and Topological Criteriain Jaw Muscle Homology

Casey M. Holliday1,2* and Lawrence M. Witmer1

1Department of Biomedical Sciences, College of Osteopathic Medicine, Ohio University, Athens, Ohio 457012Department of Biological Sciences, Ohio University, Athens, Ohio 45701

ABSTRACT The homologies of jaw muscles amongarchosaurs and other sauropsids have been unclear, con-founding interpretation of adductor chamber morphologyand evolution. Relevant topological patterns of muscles,nerves, and blood vessels were compared across a largesample of extant archosaurs (birds and crocodylians) andoutgroups (e.g., lepidosaurs and turtles) to test the utilityof positional criteria, such as the relative position of thetrigeminal divisions, as predictors of jaw muscle homol-ogy. Anatomical structures were visualized using dissec-tion, sectioning, computed tomography (CT), and vascularinjection. Data gathered provide a new and robust view ofjaw muscle homology and introduce the first synthesizednomenclature of sauropsid musculature using multiplelines of evidence. Despite the great divergences in ce-phalic morphology among birds, crocodylians, and out-groups, several key sensory nerves (e.g., n. anguli oris, n.supraorbitalis, n. caudalis) and arteries proved useful formuscle identification, and vice versa. Extant crocodyliansexhibit an apomorphic neuromuscular pattern counter tothe trigeminal topological paradigm: the maxillary nerveruns medial, rather than lateral to M. pseudotemporalissuperficialis. Alternative hypotheses of homology necessi-tate less parsimonious interpretations of changes in to-pology. Sensory branches to the rictus, external acousticmeatus, supraorbital region, and other cephalic regionssuggest conservative dermatomes among reptiles. Differ-ent avian clades exhibit shifts in some muscle positions,but maintain the plesiomorphic, diapsid soft-tissue topo-logical pattern. Positional data suggest M. intramandibu-laris is merely the distal portion of M. pseudotemporalisseparated by an intramuscular fibrocartilaginous sesa-moid. These adductor chamber patterns indicate multipletopological criteria are necessary for interpretations ofsoft-tissue homology and warrant further investigationinto character congruence and developmental connectiv-ity. J. Morphol. 268:457–484, 2007. � 2007 Wiley-Liss, Inc.

KEY WORDS: archosaur; feeding; crocodilian; avian;reptile; homology; mycology; jaw muscles; trigeminalnerve

Tracking the evolution of jaw adductor musclesthrough the diversity of amniote evolution has beenone of the most perplexing pursuits in cephalic evo-lutionary morphology. Jaw muscles are integralcharacters in the development and epigenesis of theskull, promoting the formation and ossification of

the bones they contact (de Beer, 1937; Moss, 1968;Hurov, 1988; Herring, 1993; Kiliardis, 1996; Tarsi-tano et al., 2001). Jaw muscles power the feedingapparatus via complex, coordinated movements, andare critical to investigations of kinematic (Cleurenand DeVree, 1992; Zweers et al., 1994; Gusseklooand Bout, 2005), motor (Zweers, 1974; Van Dronge-len and Dullemeijer, 1982; Gans and De Vree, 1986;Busbey, 1989; Cleuren et al., 1995), and bone strain(Ross and Metzer, 2004) patterns of the head duringprey acquisition, handling, and ingestion. Jawmuscles are crucial to the success of an organismand are important adaptive characters that can beused to interpret feeding function in birds and croc-odylians, as well as numerous extinct taxa (e.g.,sauropods, ceratopsians, ornithopods) that are notrepresented by living groups today (e.g., Anderson,1936; Haas, 1955, 1969; Ostrom, 1964; Rayfieldet al., 2001).

Many broad, comparative myological studieswere conducted in the early 20th century (Lakjer,1926; Lubosch, 1933; Edgeworth, 1935; Kesteven,1942–1945) to identify comparative patterns of thetrigeminal musculature in Amniota, including arch-osaurs. Numerous data exist on the cephalic myol-ogy of crocodylians, primarily that of Alligator mis-sissippiensis (Iordansky, 1964, 2000; Schumacher,1973; Busbey, 1989), but also caimans (van Dronge-len and Dullemeijer, 1982; Cleuren and De Vree,1992) and longirostrine crocodylians (Endo et al.,

Contract grant sponsor: National Science Foundation; Contractgrant numbers: IBN-0407735, IBN-9601174, IOB-0343744; Contractgrant sponsors: The Jurassic Foundation; UCMP Sam Welles Fund;Society of Vertebrate Paleontology; OU Student EnhancementAward; OU Graduate Student Senate; OU Departments of Biologi-cal and Biomedical Sciences.

*Correspondence to: Casey M. Holliday, Department of Anatomyand Pathology, Joan C. Edwards School of Medicine, Marshall Uni-versity, 1542 Spring Valley Dr., Huntington, WV 25704-9398.E-mail: [email protected]

Published online 19 March 2007 inWiley InterScience (www.interscience.wiley.com)DOI: 10.1002/jmor.10524

JOURNAL OF MORPHOLOGY 268:457–484 (2007)

� 2007 WILEY-LISS, INC.

2002). Myological data of birds have been describedin comparative works (e.g., Hofer, 1950; Stark andBarnikol, 1954), whereas other studies have focusedon particular clades, such as Tinamiformes (Elza-nowski, 1987), Anseriformes (Goodman and Fisher,1962), Galloanserae (Zusi and Livezey, 2000), andColumbiformes (Bhattacharyya, 1980, 1989), or var-ious species, such as ostrich (Webb, 1957), grebes(Zusi and Storer, 1969), cranes (Fisher and Goodman,1955), cormorants (Dullemeijer, 1951), humming-birds (Zusi and Bentz, 1984), and numerous others(see Zusi and Livezey, 2000 for a review). However,previous investigations of archosaur jaw muscle evo-lution have been incomplete in one or more tests ofhomology (e.g., connectivity, correspondence, congru-ence) or have focused on only one system (e.g.,muscles or nerves). Additionally, despite their taxo-nomic diversity, these studies have yet to be synthe-sized in a modern phylogenetic context, and havegenerated a challenging nomenclature for compara-tive biologists.

Topological criteria involve correspondences inthe relative positions of adductor chamber compo-nents. Among the criteria used to identify musclehomology, none is more commonly cited than thetrigeminal topological paradigm first implementedby Luther (1914). The relative positions of the oph-thalmic, maxillary, and mandibular divisions of thetrigeminal nerve discriminate different groups ofjaw muscles, separating them into M. constrictorinternus dorsalis, M. adductor mandibulae inter-nus, M. adductor mandibulae externus, and M. ad-ductor mandibulae posterior (Fig. 1). Despite someminor variations, these homology criteria were fur-ther elaborated by Lubosch (1933), Lakjer (1926),Edgeworth (1935), Save-Soderbergh (1945), and manyothers [see Haas (1973), Iordansky (2000), and Zusiand Livezey (2000) for reviews]. This scheme is notonly generally accepted as a robust criterion formuscle identification, but as the basis for terminol-ogy as well.

However, relatively few studies (e.g., Oelrich,1956; Haas, 1973; Rieppel, 1987, 1988, 1990) haveincorporated more than one suite of anatomicalstructures (e.g., nerves, muscles, vessels). For exam-ple, Poglayen-Neuwall (1953a) and Bubien-Walus-zewska (1981) illustrated the branches of the tri-geminal nerve without any topological reference toother tissues. Although blood vessels are intimatelyassociated with the developing musculature(Ruberte et al., 2003), the paths of vascular struc-tures relative to other adductor chamber soft tis-sues have only been noted in certain taxa [e.g.,Podarcis, Rieppel (1987); Charadriiformes, Dzerz-hinsky and Yudin (1982); Struthio, Anas, Alligator,Sedlmayr (2002)]. Despite the historical importanceof the trigeminal topological paradigm, recent myo-logical literature has generally ignored neurologicalcriteria and simply relied on the relative position ofother muscles, their aponeuroses, and their bony

attachments (e.g., Zusi and Bentz, 1984; Zusi andLivezey, 2000). Focusing on musculoskeletal crite-ria may be critical in functional investigations (vanDrongelen and Dullemeijer, 1982; Busbey, 1989;

Fig. 1. The trigeminal topological paradigm. Neuromusculartopological organization of the adductor chamber in left dorsalview, as shown in schematic of Sphenodon. A: Major musclecompartments defined by their positions relative to the trigemi-nal divisions. B: Individual muscles within each compartment.gMM, maxillomandibular ganglion; gV1, ophthalmic ganglion;mAMI, Musculus (M) adductor mandibulae internus; mAME,M. adductor mandibulae externus; mAMEM, M. adductor man-dibulae externus medialis; mAMEP, M. adductor mandibulaeexternus profundus; mAMES, M. adductor mandibulae externussuperficialis; mAMP, M. adductor mandibulae posterior; mCID,M. constrictor internus dorsalis; mLPt, M. levator pterygoideus;mPPt, M. protractor pterygoideus; mPSTp, M. pseudotemporalisprofundus; mPSTs, M. pseudotemporalis superficialis; mPTdþv,M. pterygoideus dorsalis and ventralis; mTP, M. tensor periorbi-tae; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibularnerve.

458 C.M. HOLLIDAY AND L.M. WITMER

Journal of Morphology DOI 10.1002/jmor

Cleuren and De Vree, 1992), but comparisons ofneuromuscular function and behavior across evenclosely related taxa may prove difficult without theintegration of muscle-independent hypotheses ofhomology. Finally, many earlier authors did notframe their analyses within a phylogenetic context,often subjecting Archosauria to the vagaries of par-aphyly (i.e., excluding birds).

The adductor chambers of extant crocodylians,birds, and their closest extant outgroups (lepido-saurs and turtles) were investigated to test the tri-geminal topological paradigm as a muscle homologycriterion and to pursue other topologically and evo-lutionarily informative soft- and hard-tissue pat-terns. These patterns will serve as the basis for adiscussion of soft-tissue homologies, the utility oftopology criteria, and regional evolution. Theresults of this study are integrated with data fromthe fossil record in complementary analyses (Hol-liday, 2006).

MATERIALS AND METHODS

Figure 2 provides the relationships of the major clades ofamniotes including many taxa used in this study. This analysisused a conservative consensus phylogeny for character analysis,subjectively collapsing tenuous nodes (particularly among neoa-vians). This subjectivity does not hamper the conclusions of thestudy in that we are not reanalyzing archosaur phylogeny, onlytracking particular characters in the consensus tree (e.g., Hutch-inson, 2001a,b). The phylogeny is based on Gauthier (1986), Cra-craft (1986), Benton and Clark (1988), Brochu (1999), Cracraftand Clarke (2001), and Mayr and Clarke (2003). Crocodyliansand neornithine birds are the two surviving clades of Archosau-ria, a group that includes non-avian dinosaurs, pterosaurs, andother extinct groups.Extant taxa studied include both intact heads and skeletal

specimens (Appendix 1). Numerous specimens prepared byWitmer (1995) and Sedlmayr (2002) were also available. Amongthe avian sample, the following received the most attention inthat at least six individuals of each were studied: commerciallyraised domestic breeds of chicken (Gallus gallus), duck (Anasplatyrhynchos), goose (Anser anser), and ostrich (Struthio cam-elus). Additional bird species representing most avian orderswere obtained from zoos, wildlife rehabilitation centers, and localsources for comparison. Among crocodylians, three speciesreceived the greatest attention: (1) American alligator (Alligatormississippiensis), collected from the Rockefeller Wildlife Refuge,southwestern Louisiana, (2) saltwater crocodile (Crocodylusporosus), and (3) New Guinea freshwater crocodile (C. novaegui-neae). Both Crocodylus species were obtained from Papua NewGuinea. In addition, single individuals of the common caiman(Caiman crocodilus), false gharial (Tomistoma schlegelli), andgharial (Gavialis gangeticus) were studied for comparison. Onto-genetic series of embryos of Gallus, Anas, and Alligator alsowere available (see Witmer, 1995 for details). Most data onextant nonarchosaurian amniotes were obtained from the exten-sive literature, but were confirmed via dissection of the followingspecies: (1) Squamata: Varanus exanthematicus, V. niloticus,Iguana iguana, Hydrosaurus amboinensis, Agama stellio; (2)Testudines: Chrysemys picta, Chelydra serpentine, and Mala-clemys terrapene; and (3) Mammalia: Ceratotherium simum andHomo sapiens.Four major anatomical techniques were used: (1) gross dissec-

tion; (2) serial gross sectioning; (3) latex and barium/latex vascu-lar injection, and (4) X-ray-computed tomography (CT) and mag-netic resonance (MR) imaging. Specimens were CT-scannedat O’Bleness Memorial Hospital, Athens, Ohio, on GE Hi Speed

FX/i and LightSpeed Ultra Helical CT scanners, and the Univer-sity of Texas CT Lab, Austin, TX, and MR-imaged at O’BlenessMemorial Hospital using a 1.0T GE Signa Short-Bore MRI sys-tem. In some cases, more than one technique was performed onthe same specimen (e.g., cephalic arteries were injected with con-trast media, followed by CT scanning, serially sectioning, anddissection). All specimens were obtained fresh, frozen, or fixed in10% neutral buffered formalin and stored in 70% ethanol. All dis-sections were documented via photography, and occasionallydrawings were made with a camera lucida mounted on a NikonSMZ-U microscope. Most unfixed specimens were skeletonizedby dermestid beetles, enzymatic digestion (Terg-a-Zyme, FisherScientific Inc.), or cold-water maceration. Several specimens ofeach focal taxon were frozen and serially sectioned using a band-saw, hacksaw, or scalpel. All new specimens were accessionedinto the Ohio University Vertebrate Collections (OUVC). Ad-ditional osteological specimens were studied at the CarnegieMuseum of Natural History (CMNH), Field Museum of Natural

Fig. 2. Cladogram depicting phylogenetic relationships offocal taxa and outgroups used in this study. Numerical codesindicate nodal taxa: 1, Amniota; 2, Sauropsida; 3, Diapsida; 4,Lepidosauria; 5, Archosauria; 6, Neornithes; 7, Paleognathae; 8,Neognathae; 9, Neoaves. Topology follows Gauthier (1986), Cra-craft (1986), Brochu (1999), Cracraft and Clarke (2001), andMayr and Clarke (2003).

ARCHOSAUR ADDUCTOR CHAMBER HOMOLOGY 459


History (FMNH), and the University of California Museum ofPaleontology (UCMP).

RESULTSOrganization of the Analysis of AdductorChamber Similarity

The adductor chamber contents were analyzedproceeding in a medial (deep) to lateral (superficial)direction. Using musculoskeletal criteria, threegeneralized regions were identified: the palatal,temporal, and orbitotemporal regions. Complement-ing these regions, trigeminal topological criteriapartition the muscles into four separate groups(Figs. 1, 3C): M. constrictor internus dorsalis, M.adductor mandibulae internus, M. adductor mandi-bulae externus, and M. adductor mandibulae poste-rior (Luther, 1914; Lakjer, 1926). This paper focusesprimarily on the adductor musculature proper (i.e.,M. adductor mandibulae) found within the palataland temporal regions, whereas the orbitotemporalregion (housing the protractor musculature andadnexa) is taken up elsewhere (Holliday, 2006).Nonetheless, several orbitotemporal structures arementioned as landmarks in several figures, namelythe ophthalmic nerve, the motor branch to M. con-

strictor internus dorsalis, and the protractor mus-culature (M. protractor pterygoideus and M. levatorpterygoideus). Tables 1 and 2 list the hypotheses ofmuscular homology and nomenclature for eachmajor muscular group in reference to past workson sauropsid adductor muscles. Cranial (Fig. 4,Table 3) and mandibular (Table 4) attachments arereported for each muscle in each major diapsidclade. These data are followed by specific and in-formative neurovascular patterns associated withthe musculature.

The Palatal RegionM. adductor mandibulae internus. Defined

by its relationship medial to the maxillary nerve(Luther, 1914) and lateral to the palatal bones, M.adductor mandibulae internus has classicallyincluded both the pterygoideus musculature (e.g.,M. pterygoideus dorsalis and ventralis) and M.pseudotemporalis (e.g., Luther, 1914; Lakjer, 1926;Hofer, 1950; Schumacher, 1973). In contrast, Edge-worth (1935) categorized M. pseudotemporalis asM. adductor mandibulae medius based on its earlyseparation from the M. adductor mandibulae inter-nus anlage during development and its disparate

Fig. 3. Overview of adductor chamber and its contents. A: Reference image of axial section through Iguana iguana depictingthe location of the slice schematized in B–D. B–D: Schematics of axial sections of plesiomorphic sauropsid adductor chamber in leftcaudal view. B: Major regions and muscular subunits of adductor chamber using musculoskeletal criteria. C: Major muscular subu-nits using trigeminal topological paradigm criteria. D: Major muscles of interest discussed in this paper. bs, basisphenoid; ept, epi-pterygoid; gV, trigeminal ganglion; ls, laterosphenoid; mAMI, Musculus (M) adductor mandibulae internus; mAME, M. adductormandibulae externus; mAMEP, M. adductor mandibulae externus profundus; mAMEM, M. adductor mandibulae externus medialis;mAMES, M. adductor mandibulae externus superficialis; mAMP, M. adductor mandibulae posterior; mCID, M. constrictor internusdorsalis; mLPt, M. levator pterygoideus; ml, membrane limitans; mn, mandible; mPPt, M. protractor pterygoideus; mPSTp, M.pseudotemporalis profundus; mPSTs, M. pseudotemporalis superficialis; mPT, M. pterygoideus; mTP, M. tensor periorbitae; pa, pa-rietal; po, postorbital; pt, pterygoid; pr, prootic; sq, squamosal; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve.



TA

BL

E1.

Syn

onym

sof

M.

ad

du

ctor

ma

nd

ibu

lae

inte

rnu

sa

mon

gd

iap

sid

su

sed

inth

isst

ud

ya

nd

sele

cted

lite

ratu

re

Author

(Top

row,crocod

ylians;

Bottom

row,birds)

M.adductor

mandibulaeintern

us(m

AMI)

This

Study

M.pterygoideu

sdorsa

lis

M.pterygoideu

sven

tralis

M.pseudotem

poralisprofundus

M.pseudotem

poralissu

perficialis

Lakjer,1926

þþ

�M.adductor

mandibulaeex

tern

us

profunduspars

IIIa

þþ

þþ

Edgew

orth,1935

��

��

�M.adductor

mandibulaemed

ialis

Kesteven

,1942–1945

M.pterygoideu

sintern

us

andanterior

M.pterygoideu

sinferior

M.pterygoideu

smed

ius

Sch

umach

er,1973a

M.pterygoideu

sintern

us

M.pterygoideu

smed

ius

M.pterygoideu

sex

tern

us

M.pterygoideu

santerior

M.pterygoideu

sposterior

M.pseudotem

poralis

Iord

ansk

y,2000a

M.pterygoideu

santerior

M.pterygoideu

sposterior

M.adductor

mandibulaeinterm

edius

Busb

ey,1989a

M.pterygoideu

santerior

M.pterygoideu

sposterior

M.pseudotem

poralis

M.pseudotem

poralis

Hofer,1950b

M.pterygoideu

sdorsa

lis

M.pterygoideu

sven

tralis

M.quadratomandibularis

M.adductor

mandibulaeex

tern

us

profunduspars

anterior

Zwee

rs,1974b

��

þþ

Elzanow

ski,1987b

M.pterygoideu

slateralis

M.pterygoideu

smed

ialis


þVanden

BergeandZwee

rs,1993b

M.pterygoideu

slateralis

M.pterygoideu

smed

ialis

þ/M

.adductor

mandibulaecaudalis

þOelrich

,1956c

M.pterygom

andibularis

þþ

þHaas,

1973c ;Wu,2003c

M.pterygoideu

satypicus

M.pterygoideu

stypicus

þ

�,Not

men

tion

edor

muscle

not

present;þ,sa

meterm

inolog

yasthis

study.

aCrocodilianterm

inolog

yon

ly.

bAvianterm

inolog

yon

ly.

c Lep

idosaurianterm

inolog

yon

ly.



TA

BL

E2.

Syn

onym

sof

M.

ad

du

ctor

ma

nd

ibu

lae

exte

rnu

sa

nd

pos

teri

ora

mon

gd

iap

sid

su

sed

inth

isst

ud

ya

nd

sele

cted

lite

ratu

re

Author

(Top

row,

crocod

ylians;

Bottom

row,birds)

M.adductor

mandibulaeex

tern

us(m

AME)

M.adductor

mandibulae

posterior

(mAMP)

This

Study

M.adductor

mandibulae

extern

usprofundus

M.adductor

mandibulae

extern

usmed

ialis

M.adductor

mandibulae

extern

ussu

perficialis

M.adductor

mandibulaeposterior

Lakjer,1926

M.adductor

mandibulae

extern

usIIIb

M.adductor

mandibulae

extern

usIIIb

M.adductor

mandibulae

extern

usIIa,b

M.adductor

mandibulaeex

tern

usIIIc

M.adductor

mandibulae

extern

usII,Ia,b,c,

M.adductor

mandibulae

extern

usII

M.adductor

mandibulae

extern

usIIIrostralis

þ,M.adductor

mandibulae

extern

uspars

IIIcaudalis

Edgew

orth,1935

M.adductor

mandibulaeex

tern

us

M.adductor

mandibulaeex

tern

us

M.adductor

mandibulaemed

ius

Adams,

1919

M.capitim

andibularismed

ius

M.capitim

andibularis

superficialis

M.capitim

andibularismed

ius

M.capitim

andibularisprofundus

M.capitim

andibularis

superficialis

Kesteven

,1942–1945

M.pterygoideu

sex

tern

us

M.temporo-massetericus

M.pterygoideu

smed

ius

M.temporalis

M.quadrato-m

andibularis

M.massetericus


Sch

umach

er,1973a

M.pseudotem

poralis

þþ

þ,part

ofmAME

profundus

Iord

ansk

y,2000a

þM.adductor

mandibulae

extern

ussu

perficialis

þþ

Busb

ey,1989a

M.adductor

mandibulae

extern

usprofunduspars

posterior

þþ

þ

Hofer,1950b

M.adductor

mandibulae

extern

uscaudolateralis,part

ofM.adductor

mandibulaeex

tern

us

rostromed

ialis

M.adductor

mandibulae

extern

usrostromed

ialis

M.adductor

mandibulae

extern

usprofundus

þ

Elzanow

ski,1987b

M.adductor

mandibulaeex

tern

usprofundus

þþ

þVanden

Bergeand

Zwee

rs,1993b

M.adductor

mandibulaeex

tern

usven

tralis

M.adductor

mandibulae

extern

usrostralis

M.adductor

mandibulaeex

tern

us

pars

profunda,M.adductor

mandibulae

extern

usossisquadrati

Zusi

andLivezey,

2000b

M.adductor

mandibulaeex

tern

us

coronoideu

s,M.adductor

mandibulae

extern

uszy

gom

aticu

s,M.adductor

mandibulaeex

tern

ussu

perficialis

M.adductor

mandibulaeex

tern

us

articularisex

tern

us

þ,M.adductor

mandibulae

extern

usarticularisintern

us

Oelrich

,1956c

þþ

þ,M.adductor

mandibulae

extern

uslevatorangulioris

þ

Haas,

1973c

þþ

þ,M.adductor

mandibulae

extern

uslevatorangulioris

þ

þ,Sameterm

inolog

yasthis

study.

aCrocodilianterm

inolog

yon

ly.

bAvianterm

inolog

yon

ly.

c Lep

idosaurianterm

inolog

yon

ly.



fiber direction compared to the pterygoideusmuscles. Nevertheless, based on position and inner-vation, the pseudotemporalis muscles are bestincluded within M. adductor mandibulae internus,and we join other workers in doing so.

Partitioning of the pterygoideus musculature hasoccurred in varying ways in the different sauropsidclades, obscuring general patterns. For example,Lakjer (1926) and Iordansky (1964) proposed ante-rior (rostral) and posterior (caudal) parts in croco-dylians, Zusi and Storer (1969) and Vanden Bergeand Zweers (1993) identified lateral and ventralparts in birds, and Schumacher (1973) used dorsaland ventral constructs in turtles. The present studyfailed to identify any criteria independent of the re-spective bony attachments of the muscles that dis-criminate subdivisions of the ptergyoideus muscu-lature. Thus, this study defers to the dorsalis/ven-tralis nomenclature based on the dorsal and ventralattachments of the muscles to the palate and theirmandibular insertions in crocodylians and birds.However, it appears that other reptiles (i.e., turtlesand Sphenodon) also have more than one distinctpterygoideus belly, suggesting it may be a sharedfeature that was elaborated by archosaurs andpotentially lost in squamates (Witmer, 1995).

M. pterygoideus dorsalis—Crocodylia. The dorsalpterygoideus [M. pterygoideus anterior of Iordan-sky (1964)] occupies a substantial portion of the

heads of crocodylians, particularly the dorsal sur-face of the palate and suborbital space (Figs. 5E, 8).The cranial attachments of the muscle include thecaviconchal fossa (Witmer, 1995) of the maxilla/pal-atine articulation, the caudolateral surface of thepostconchal nasal cartilage, the dorsomedial sur-face of the palatine, the ventrolateral surface of thelacrimal, the dorsomedial surface of the maxilla/ectoptergyoid articulation, the suborbital fenestra,the cartilaginous interorbital septum, the lateralsurface of the cultriform process, and the ascendingprocess of the pterygoid (Figs. 5, 8). The muscleruns caudally through the postnasal fenestra, ven-tral to M. tensor periorbitae and M. depressorauriculae inferioris in the suborbital space, medialto the maxilla, jugal, ectoptergyoid, and ventral toM. pseudotemporalis profundus and M. adductormandibulae posterior, medial to M. intramandibu-laris, and lateral to M. pterygoideus ventralis in thetemporal region.

Musculus pterygoideus dorsalis attaches to theventromedial surface of the angular and articularof the lower jaw, just ventral to the jaw joint (Figs.5E, 8). The medial surface of the muscle attaches asa strong tendon to the ventromedial edge of themedial mandibular fossa, just caudal to the ptery-goid flange. The lateral surface of the muscleattaches as a tendon to the dorsomedial edge of thearticular, just medial to the jaw joint and retroartic-

Fig. 4. Skulls of representa-tive sauropsid taxa in left lat-eral view indicating attachmentareas of important muscles. A:Sphenodon punctatus. B: Alli-gator mississippiensis. C: Stru-thio camelus. D: Ardea hero-dias. mAMI, Musculus (M)adductor mandibulae internus;mAME, M. adductor mandibu-lae externus; mAMEP, M.adductor mandibulae externusprofundus; mAMEM, M. adduc-tor mandibulae externus medi-alis; mAMES, M. adductormandibulae externus superfi-cialis; mAMP, M. adductor man-dibulae posterior; mCID, M.constrictor internus dorsalis;mLPt, M. levator pterygoideus;mPPt, M. protractor pterygoi-deus; mPSTp, M. pseudotem-poralis profundus; mPSTs, M.pseudotemporalis superficialis;mPT, M. pterygoideus; mTP, M.tensor periorbitae.



TA

BL

E3.

Ori

gin

s(d

orsa

la

tta

chm

ents

)of

hom

olog

ous

ad

du

ctor

cha

mber

mu

scle

sof

Lep

idos

au

ria

,C

roco

dyl

ia,

an

dN

eorn

ith

es

Muscle

Origin

Lep

idosauria

Crocodylia

Neo

rnithes

M.adductor

mandibulae

intern

us

M.pterygoideu

sdorsa

lis

Caudalsu

rface

ofinterorbitalseptum.

Dorsa

lsu

rface

ofpterygoidand

ectopterygoid

Dorsa

lsu

rface

ofpalatine,

pterygoid,

ectopterygoid,ven

tralsu

rface

ofinterorbitalseptum

Dorsa

lsu

rface

ofpalatineand

pterygoid

M.pterygoideu

sven

tralis

Ven

traled

geof

pterygoid,and

quadrate,separate

apon

eurosis

Caudom

edialandcaudolateral

edgeof

pterygoid

Ven

tralsu

rface

ofpalatineand

pterygoid

M.pseudotem

poralisprofundusLateralsu

rface

ofep

ipterygoidand

prootic

Ven

trolateralsu

rface

oflaterosp

hen

oidlateralbridge

Rostralsu

rface

ofquadrate

orbital

process

M.pseudotem

poralis

superficialis

Lateralsu

rface

ofprootic,mem

brane

limitans,

rostrolateralsu

rface

ofparietal

Caudod

orsa

lsu

rface

oflaterosp

hen

oid

Rostrod

orsolateralsu

rface

oflaterosp

hen

oid

M.adductor

mandibulae

extern

us

M.adductor

mandibulae

extern

usprofundus

Caudolateralsu

rface

ofparietal,

rostromed

ialsu

rface

ofsq

uamosal

Ven

trolateralsu

rface

parietal

Lateralsu

rface

ofparietal,sq

uamosal,

andlaterosp

hen

oid

M.adductor

mandibulae

extern

usmed

ialis

Rostrolateralsu

rface

ofsq

uamosal

Rostrom

edialsu

rface

ofquadrate

Lateralsu

rface

ofsq

uamosal

M.adductor

mandibulae

extern

ussu

perficialis

Rostrolateralsu

rface

ofsq

uamosal,

ven

trom

edialsu

rface

ofsq

uamosal

andpostorbital

Rostrolateralsu

rface

ofquadrate

andquadratojugal

Lateralsu

rface

ofsq

uamosal,

dorsolateralsu

rface

ofquadrate

otic

process

M.adductor

mandibulae

posterior

M.adductor

mandibulae

posterior

Ven

trolateralsu

rface

ofquadrate

Rostralsu

rface

ofquadrate

Rostrolateralsu

rface

ofquadrate

TA

BL

E4.

Inse

rtio

ns

(ven

tra

la

tta

chm

ents

)of

hom

olog

ous

ad

du

ctor

cha

mber

mu

scle

sof

Lep

idos

au

ria

,C

roco

dyl

ia,

an

dN

eorn

ith

es

Muscle

Origin

Lep

idosauria

Crocodylia

Neorn

ithes

M.adductor

mandibulae

intern

us

M.pterygoideu

sdorsa

lis

Caudom

edialsu

rface

ofangular

andarticular,

Caudom

edialsu

rface

ofangular

andarticular

Rostrod

orsolateralsu

rface

ofmed

ialmandibularprocess

M.pterygoideu

sven

tralis

Caudov

entrolateralsu

rface

ofangularandarticular,lateral

surface

ofjugal,postorbital

Caudov

entraled

geof

angular,

caudolateralsu

rface

ofangularandsu

rangular

Rostoven

trom

edialsu

rface

ofmed

ialmandibularprocess,

lateralsu

rface

ofmandible

M.pseudotem

poralisprofundus

Dorsomed

ialsu

rface

ofsu

rangularandcoronoid

Caudod

orsa

led

geof

angular

Ven

trom

edialsu

rface

ofcoronoidprocess,med

ial

mandibularfossa

M.pseudotem

poralis

superficialis

Med

ialsu

rface

ofcoronoid

process

Cartilagotransilien

sMed

ialsu

rface

ofcoronoid

process

M.adductor

mandibulae

extern

us

M.adductor

mandibulae

extern

usprofundus

Dorsolateralsu

rface

ofcoronoid

process

Dorsomed

ialsu

rface

ofcoronoid

eminen

ceof

surangular

Dorsolateralsu

rface

ofcoronoid

process

M.adductor

mandibulae

extern

usmed

ialis

Dorsolateralsu

rface

ofsu

rangular

Dorsolateralsu

rface

ofcoronoid

eminen

ceof

surangular

Dorsolateralsu

rface

oflower

jaw

M.adductor

mandibulae

extern

ussu

perficialis

Dorsolateralsu

rface

ofsu

rangular

Dorsa

lsu

rface

ofsu

rangular

Lateralsu

rface

oflower

jaw,

lateralmandibularprocess

M.adductor

mandibulae

posterior

M.adductor

mandibulae

posterior

Med

ialmandibularfossa,

dorsomed

ialsu

rface

ofangularandarticular

Med

ialmandibularfossa,

dorsomed

ialsu

rface

ofangularandarticular

Caudod

orsomed

ialand

caudod

orsa

lsu

rface

oflower

jaw



ular process. Generally, the muscle excavates alarge fossa on the caudomedial surface of the man-dible, caudal to the medial mandibular fossa andventral to the articular and retroarticular process.

M. pterygoideus ventralis—Crocodylia. The ven-tral pterygoideus tendinously attaches via Iordan-sky’s (1964) ‘‘U’’ tendon to the caudal rim of thepterygoid flange (Fig. 5D) and the caudolateral sur-face of the ascending process of the pterygoid (Fig.5E). It then passes lateral to the cervical muscula-ture and conspicuously wraps around mPTd andthe retroarticular process to attach to the caudolat-

eral surface of the angular (Fig. 8). The muscle isgenerally parallel fibered with several large intra-muscular tendons contributing to its internal archi-tecture.

M. pterygoideus dorsalis—Neornithes. The dorsalpterygoideus of birds consistently attaches to thedorsal and lateral surfaces of the palatine and pter-ygoids. In most cases, separate medial and lateralbellies individually attach to the surfaces of thepterygoids and palatines, respectively (e.g., Lakjer,1926; Hofer, 1950; Goodman and Fisher, 1962; Zusiand Bentz, 1984; Figs. 6, 9). In palaeognaths (e.g.,

Fig. 5. Schematics of serial, horizontal sections of the left adductor chamber of Alligator mississippiensis in dorsal view. A–E:Sections proceed dorsal to ventral. F: Denotes location of sections in head. a, vIO, infraorbital artery and vein; a, vJU, jugal arteryand vein; a, vM, mandibular artery and vein; a, vMM, maxillomandibular artery and vein; an, angular; ar, articular; aTO, tempor-oorbital artery; ce, cerebrum; ch, choana; ct, cartilago transiliens; dvs, dural venous sinus; ec, ectopterygoid; gMN; harderian gland;gV, trigeminal ganglion; h, hypophysis; ju, jugal; lb, lateral bridge of the laterosphenoid; ls, laterosphenoid; mAMEP, M. adductormandibulae externus profundus; mAMEM, M. adductor mandibulae externus medialis; mAMES, M. adductor mandibulae externussuperficialis; mAMP, M. adductor mandibulae posterior; mBM, M. branchiomandibularis; mDM, M. depressor mandibulae; mIRA,M. intramandibularis; mOD, M. obiquus dorsalis; mPSTp, M. pseudotemporalis profundus; mPSTs, M. pseudotemporalis superficia-lis; mPTd, M. pterygoideus dorsalis; mPTv, M. pterygoideus ventralis; mRD, M. rectus dorsalis; mRL, M. rectus lateralis; mRV, M.rectus ventralis; mTP, M. tensor periorbitae; nAO, ramus to the corner of the mouth (anguli oris) of the mandibular nerve; nPTc,caudal branch of the pterygoid ramus of the mandibular nerve; pa, parietal; pe, periorbita; pr, prootic; ptb, pterygoid buttress; qj,quadratojugal; qu, quadrate; ri, rictus; sa, surangular; t, tarsus for M. depressor auriculae inferioris; V1, ophthalmic nerve; V2,maxillary nerve; V3, mandibular nerve.



Struthio, Rhea, Eudromia), mPTd attaches to thedorsal surface of the lateral palatine lamina andpasses lateral to the ventral ptergyoideus bellies(Fig. 6D). Psittaciformes evolved a remarkable,enlarged belly of M. pterygoideus dorsalis, the M.ethmomandibularis, which attaches to the interor-bital septum rostral to the septal attachments of M.protractor pterygoideus (Zusi, 1993; Tokita, 2004).Along its path in the palate, M. pterygoideus dorsa-lis is bordered dorsally by the suborbital air sinus(Witmer, 1995) and laterally by the jugal and M.pseudotemporalis profundus. Distally, M. pterygoi-deus dorsalis attaches to the medial surface of themandible, ventral to the jaw joint, and to the rostralsurface of the medial mandibular process, borderedventrally and medially by M. ptergyoideus ventralisand laterally by M. pseudotemporalis profundus(Fig. 9). Despite slight taxonomic differences inmuscle morphology, M. pterygoideus dorsalis con-sistently runs from the dorsal surface of the palateto the medial surface of the mandible either imme-

diately rostral or ventral to the medial cotyla of thejaw joint and always runs between M. pseudotem-poralis profundus and M. pterygoideus ventralis.

M. pterygoideus ventralis—Neornithes. The ven-tral ptergyoideus attaches to the ventral surfaces ofthe palatine and pterygoid. Separate bellies oftenarise from the ventral or medial surface of the pala-tine and pterygoid. Palaeognaths evolved a complexset of ventral pterygoideus muscles (e.g., lateraland medial bellies) that attach mediolaterallyacross the ventral surface of the pterygoid andenclose the choana. Palaeognaths also evolved anovel M. pterygoideus ventralis belly, M. retractorpalatini (Webb, 1957; Burton, 1974), which attachesto the palatal mucosa rostrodorsal to the other M.pterygoideus ventralis bellies, passes dorsal to themedial mandibular process, and inserts on the par-asphenoid lamina between the internal carotid fora-men and the ala parasphenoidale. Distally, ventralpterygoideus attaches along the ventral rim of themedial mandibular process in most neognaths, bor-

Fig. 6. Schematics of serial,horizontal sections of the leftadductor chamber of Struthiocamelus in dorsal view. A–E:Sections proceed dorsal to ven-tral. F: Denotes location of sec-tions in head. ar, articular; bs,basisphenoid; dp, diploe; gLA,lacrimal gland; gMN, Harderiangland; ju, jugal, l, labyrinth; ls,laterosphenoid; mAME, muscu-lus (M) adductor mandibulaeexternus; mAMEP, M. adductormandibulae externus profun-dus; mAMES; M. adductor man-dibulae externus superficialis;mAMP;M. adductormandibulaeposterior; mDM, M. depressormandibulae; mIRA; M. intra-mandibularis; mOD, M. obli-quus dorsalis; mPPq,M. protrac-tor quadratus; mPPt, M. pro-tractor pterygoideus; mPSTp;M. pseudotemporalis profundus;mPSTs; M. pseudotemporalissuperficialis; mPTd, M. ptery-goideus dorsalis; mPTv, M. pter-ygoideus ventralis; mRD,M. rec-tus dorsalis; mRL, M. rectus lat-eralis; mRM,M. rectus medialis;mRV, M. rectus ventralis; mTP,M. tensor periorbitae; pe, perior-bita; pop; postorbital process; q,quadrate; ri, rictus; rOP, oph-thalmic rete; sf, superficial fattytissue; sq, squamosal; tc, tym-panic cavity; V2, maxillarynerve; V3, mandibular nerve.



dered dorsally by mPTd and ventrolaterally by M.serpihyoideus, a hyolingual muscle. Representa-tives of several avian clades (Lakjer, 1926; Hofer,1950; pers. obs.) including pelicaniforms, procellar-iiforms, owls, parrots, penguins, and auks are char-acterized by M. pterygoideus ventralis muscles thatattach to the lateral surface of the mandible similarto the condition found in crocodylians (Fig. 9).

M. pseudotemporalis profundus—Crocodylia. Asmall slip of muscle attaches to the lateral bridge ofthe laterosphenoid, ventral to M. pseudotemporalissuperficialis, with some fibers attaching to the ven-trolateral surface of the maxillary nerve. This mus-cle melds with the dorsal fibers of M. pterygoideusdorsalis near the caudodorsal surface of the angular(Fig. 8). Iordansky (1964) identified this muscle asM. adductor mandibulae intermedius, but Busbey(1989) identified it as M. pseudotemporalis. Topo-logical and attachment criteria (see Discussion)indicate that this muscle is most likely a rudimen-tary belly of M. pseudotemporalis profundus, and ingeneral, is indistinguishable from mPTd at its man-dibular insertion.

M. pseudotemporalis profundus—Neornithes.This muscle attaches to the quadrate orbital processin extant birds [M. quadratomandibularis, Hofer(1950) and Elzanowski (1987); M. adductor mandi-bulae caudalis, Vanden Berge and Zweers (1993)],and is generally parallel fibered with a strong tendi-nous attachment to the caudolateral edge of themuscle (Fig. 9). In Struthio and tinamous (e.g.,Nothoprocta, Tinamus, and Eudromia), a short,taut ligament (orbitoquadrate ligament; Elzanow-ski, 1987) connects the ventral margin of M. pseudo-temporalis superficialis with the dorsal aspect of M.pseudotemporalis profundus (with some blendedmuscle fibers). Dzerhinsky and Yudin (1982)regarded this shared ligament as evidence for evolu-tion of the two pseudotemporal muscles from a com-mon developmental rudiment. Musculus pseudo-temporalis profundus typically attaches dorsal tothe medial mandibular fossa, medial to the entry ofthe mandibular nerve into the Meckelian canal(Figs. 9, 10), rostroventral to M. pseudotemporalissuperficialis, ventral to M. adductor mandibulaeexternus profundus, and lateral and rostral to themandibular attachment of pterygoideus dorsalis.Neighboring soft tissues of M. pseudotemporalisprofundus are discussed with M. pseudotemporalissuperficialis below.

M. adductor mandibulae posterior. Thismuscle maintains a consistent position on the bodyof the quadrate among all sauropsids, bordered byM. pterygoideus rostromedially, M. pseudotempora-lis rostrally, M. adductor mandibulae externus pro-fundus rostrolaterally, and M. adductor mandibulaeexternus superficialis laterally. Because M. adduc-tor mandibulae posterior develops from either M.adductor mandibulae internus or M. adductor man-dibulae externus in lepidosaurs and turtles, respec-

tively (Edgeworth, 1935), Rieppel (1987, 1990) exp-ressed concern about their homology. However, theconsistent musculoskeletal and other relevant crite-ria in adult forms support the homology of thesemuscles and they will be treated as such. Adductormandibulae posterior has intimate developmentalties to Meckel’s cartilage during the development ofthe mandible (Edgeworth, 1935; de Beer, 1937),maintaining attachments on the cartilage whenpresent in adults (e.g., crocodylians) and occupyingthe majority of the medial mandibular fossa.

M. adductor mandibulae posterior—Crocodylia.This is one of the larger muscles of the adductorchamber of crocodylians (Schumacher, 1973; Bus-bey, 1989). The large quadrangular muscle attachesto most of the quadrate, medial to M. adductormandibulae externus superficialis, ventral to M.adductor mandibulae externus medialis, and lateralto M. pterygoideus ventralis (Figs. 5, 8). The muscleis composed of a number of intramuscular apo-neuroses that often leave specific crests andtubercles on the quadrate (see Iordansky, 1964; Fig.4B) and that give it a parallel-fibered orientationfrom a lateral view but a pinnate one in cross sec-tion. The thickened mandibular adductor tendonserves as the rostrolateral boundary of the muscleand its primary tendinous anchor to the quadratebody. The muscle inserts on the medial aspect of themandible, occupying the majority of the medialmandibular fossa where it attaches to the dorsalsurface of the angular, rostral surface of the articu-lar, and the medial surface of the dermis overlyingthe external mandibular fenestra (Fig. 10B). Themuscle borders M. pterygoideus dorsalis dorsolater-ally, M. adductor mandibulae externus profundusand M. adductor mandibulae externus superficialiscaudomedially, and M. intramandibularis caudallyin the medial mandibular fossa.

M. adductor mandibulae posterior—Neornithes.In ratites, M. adductor mandibulae posteriorattaches to the dorsomedial surface of the mandi-ble, caudal to the mandibular attachments of M.adductor mandibulae externus profundus, a charac-teristic similar to that found in other non-aviansauropsids (Fig. 10A–C). On the other hand, M.adductor mandibulae posterior [synonymous withM. adductor mandibulae ossis quadrati, VandenBerge and Zweers (1993); M. adductor mandibulaecaudalis, Buhler (1981)] of most neoavians typicallyattaches to the dorsal and lateral surface of themandible (Fig. 10D). In galloanserines and someneoavians, M. adductor mandibulae posterior hastwo parts, a large belly that attaches to the lateralportion of the orbital process and body of the quad-rate (here defined as M. adductor mandibulae pos-terior medialis), and a thin belly that attaches tothe otic process of the quadrate (M. adductor man-dibulae posterior lateralis) [synonymous with M.adductor mandibulae articularis internus, Zusi andLivezey (2000); Figs. 4D, 9C]. The muscle is bor-



dered caudomedially by M. pseudotemporalis pro-fundus, rostrolaterally by M. adductor mandibulaeexternus superficialis, and rostromedially by M.adductor mandibulae externus profundus. AmongGalloanserae, M. adductor mandibulae posteriormedialis is often large and pinnate, fanning outacross the lateral surface of the mandible, whereasM. adductor mandibulae posterior lateralis is slimand parallel-fibered, running from the otic tubercleto the caudodorsal part of the mandible, caudodor-sal to the attachment of M. adductor mandibulaeposterior medialis and rostrolateral to the jaw joint.In most birds, M. adductor mandibulae posteriorlateralis maintains a consistent position immedi-ately rostral to the jaw joint and quadatromandibu-lar ligament, similar to the pattern found in gal-loanserines. However, the mandibular attachmentsof M. adductor mandibulae posterior medialis varygreatly among neoavians, ranging from positions inthe caudal medial mandibular fossa to large lateralmandibular attachments (Fig. 10D).

M. constrictor ventralis. The constrictor ven-tralis muscles include M. intermandibularis andpossibly M. intramandibularis. The intermandibu-laris spans the space between the two mandibleswith transversely oriented fibers. In birds, the twosides meet at a midline raphe, whereas in crocodyli-ans, the two sides are more continuous with oneanother. The homology of M. intramandibularis iscontentious. The two hypotheses of homology forthe M. intramandibularis are (1) it is a lateraland dorsal extension of the M. intermandibularis(Rieppel, 1990), or (2) it is part of pseudotemporalis(Elzanowski, 1987). Crocodylia, Palaeognathae(Elzanowski, 1987), Sphenisciformes, Pelicani-formes (Dzerhinsky and Yudin, 1982), and Procel-lariformes (Hofer, 1950) possess robust M. intra-mandibularis muscles that occupy the rostral por-tion of the medial mandibular fossa (Figs. 5E, 6E,8, 10). In both archosaur clades, M. intramandibu-laris connects to the ventral portion of M. pseudo-temporalis superficialis and is innervated by thesame proximal alveolar branch (Poglayen-Neuwall,1953b) of the mandibular nerve that also innervatesM. intermandibularis. In ratites (e.g., Struthio,Rhea) and other birds, a thin intertendon connectsM. pseudotemporalis and M. intramandibularis. Incrocodylians, this tendon develops a large fibro-cartilaginous sesamoid cartilage, the cartilago tran-siliens (Figs. 5, 8D), and M. intramandibularisattaches to it ventrally. The presence of sesamoidsand intertendons suggests that the M. intramandi-bularis and M. pseudotemporalis superficialis areparts of one homologous muscle rather than twoseparate muscles. Hypothesis 2 is further supportedby the topological patterns these muscle share withneighboring neurovasculature.

Neurovasculature in the palatal region. Theboundaries of the pterygoideus muscles, as well asthe adductor chamber as a whole, are well circum-

scribed by characteristic patterns of nerves andblood vessels. Among all taxa sampled, the maxil-lary nerve, the pterygoid branch of the mandibularnerve (i.e., the pterygoid nerve) (discussed with M.pseudotemporalis superficialis below), and thesphenopalatine artery (Weber, 1996; Sedlmayr,2002) pass between M. pterygoideus dorsalis and ei-ther the extraocular muscles, as in turtles, lizards,and crocodylians, or the suborbital air sac in birds.In crocodylians, the large caudal branch of ptery-goid nerve passes along the dorsal surface of M.pterygoideus ventralis, whereas among birds thepterygoid nerve merely pierces M. pterygoideusdorsalis to innervate M. pterygoideus ventralis(Figs. 8, 9). The external jugular vein, external ca-rotid artery, and cranial nerves IX–XI bound thecaudomedial surface of M. pterygoideus ventralis,running between this muscle and the cervical mus-culature. Hyolingual muscles (e.g., M. serpihyoi-deus, M. branchiomandibularis, and M. stylohyoi-deus) bound the medial and ventral portions of M.pterygoideus ventralis and pharyngeal mucosa bor-ders the ventral surface of the M. pterygoideus ven-tralis in birds. These different structures can beused to isolate the pterygoideus musculature fromsurrounding tissues. However, no other structurescan be used to separate individual bellies of themuscle.

There are no neurovascular structures that dif-ferentiate either M. intramandibularis from M.intermandibularis or M. intramandibularis from M.pseudotemporalis. Yet, both the mandibular nerveand its accompanying mandibular artery passbetween M. intermandibularis and M. adductormandibulae posterior and then subsequently lateralto M. intramandibularis within the medial mandib-ular fossa. After branching from the mandibularnerve, the proximal part of the inferior alveolarnerve runs medially through the rostral portion ofM. intramandibularis and exits the medial surfaceof the mandible to ramify across the ventral part ofM. intermandibularis. The similar topological rela-tionships these muscles share with neighboringneurovasculature further supports the homologouslink between M. intramandibularis and M. pseudo-temporalis superficialis noted above.

The Temporal RegionM. adductor mandibulae internus.M. pseudotemporalis superficialis—Crocodylia.

The homology of M. pseudotemporalis has been con-tentious in the crocodylian myological literature.We recognize the presence of M. pseudotemporalisin crocodylians (Lakjer, 1926), but disagree withthe interpretations of its attachments presented byIordansky (1964), Schumacher (1973), and Busbey(1989). Homology of M. pseudotemporalis is takenup in the Discussion, and synonymies are presentedin Tables 1 and 2. Musculus pseudotemporalis



superficialis attaches to the caudal surface of thepostorbital process of the laterosphenoid, caudal toM. tensor periorbitae (Figs. 4B, 8, 11), rostral to M.adductor mandibulae externus profundus, and ros-trodorsal to the maxillomandibular foramen (Fig.5A,B). It attaches ventrally to the dorsomedial sur-face of the cartilago transiliens, with some fibersmerging with M. adductor mandibulae externusprofundus near the medial surface of the coronoideminence (Figs. 5D, 8, 10B).

M. pseudotemporalis superficialis—Neornithes.Avian clades differ with regard to attachment siteof M. pseudotemporalis superficialis on the lateros-phenoid (Vanden Berge and Zweers, 1993; Baumeland Witmer, 1993). For example, in tinamous (e.g.,Eudromia) and ratites (e.g., Struthio, Rhea) otherthan Apteryx, M. pseudotemporalis superficialissolely excavates the dorsotemporal fossa (Webb,

1957; Elzanowski, 1987; Figs. 4C, 6, 11C) and isbordered rostrally by M. tensor periorbitae, ventro-medially by the maxillomandibular foramen, andventrolaterally by M. adductor mandibulae exter-nus profundus (Fig. 6A). Apteryx has a greatlyenlarged M. pseudotemporalis superficialis thatoccupies the rostrolateral surface of the laterosphe-noid and is bordered laterally by an expansive peri-orbital fat pad (Hofer, 1950), which is likely associ-ated with the ophthalmic rete and temporal vessels.Among neognaths (e.g., Galloanserae, Gavia, Puffi-nus, Falconiformes, Gruiformes, Sphenisciformes,Psittaciformes, Passeriformes), M. pseudotempora-lis superficialis attaches to the rostral (orbital) sur-face of the laterosphenoid just ventral to M. tensorperiorbitae, lateral to M. rectus lateralis, and ros-tromedial to M. adductor mandibulae externus pro-fundus. In some birds, (e.g., Aechmorphorus,

Fig. 7. Major features of the adductor chamber of Iguana iguana in left lateral view. Image is composite based on CT data forskeletal anatomy and dissections for soft-tissue anatomy. A: Head skeleton. B: Superficial dissection (M. levator anguli oris omit-ted). C: Intermediate depth. D: Deep dissection. aTO, temporoorbital artery; aTS, superficial temporal artery; mAMEM, musculus(M) adductor mandibulae externus medialis; mAMEP, M. adductor mandibulae externus profundus; mAMES, M. adductor mandi-bulae externus superficialis; mAMP, M. adductor mandibulae posterior; mLPt, M. levator pterygoideus; mPPt, M. protractor ptery-goideus; mPSTp, M. pseudotemporalis profundus; mPSTs, M. pseudotemporalis superficialis; mPT, M. pterygoideus; nAO, ramus tothe corner of the mouth (anguli oris) of the mandibular nerve; nCA, caudal ramus of the mandibular nerve; nCID, motor ramus tothe constrictor internus dorsalis muscles of the mandibular nerve; nFR, frontal ramus of the ophthalmic nerve; nPT, pterygoidramus of the mandibular nerve; nSO, supraorbital ramus of the maxillary nerve; pa, parietal; po, postorbital; pr, prootic; pt, ptery-goid; qu, quadrate; sq, squamosal; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve.



Anhinga, Apteryx, Ardea, Pelicanus, Phalacrocorax,and Ciconiiformes), M. pseudotemporalis superfi-cialis attaches to the lateral surface of the lateros-phenoid and shares the rostral surface of the dorso-temporal fossa with M. adductor mandibulae exter-nus profundus, which covers M. pseudotemporalissuperficialis laterally (Fig. 4D). In strigids (e.g.,Bubo, Otus) and caprimulgids (e.g., Caprimulgus)and other neoavians with enlarged eyes, M. pseudo-temporalis superficialis remains on the rostroven-tral part of the laterosphenoid and is usually dis-placed ventrally by the eye and merges with therostroventral portion of M. adductor mandibulaeexternus profundus. Pelicaniformes (e.g., Phalacro-corax, Pelicanus), Procellariformes (e.g., Phoebas-tria), and others (Dzerhinsky and Yudin, 1982)have a second caudal belly of M. pseudotemporalissuperficialis, Hofer’s (1950) ‘‘caput absconditum’’(Fig. 9C). This muscle is a small, thin fleshy bellythat attaches in the dorsal tympanic recess. Strik-ingly, the muscle is orthogonal to the main M. pseu-dotemporalis superficialis belly and merges with itscaudolateral surface as only a thin tendon. Distally,M. pseudotemporalis superficialis consistentlyattaches to the dorsomedial surface of the mandi-ble, dorsal to the medial mandibular fossa, medialand ventral to the attachments of M. adductor man-dibulae externus profundus and the coronoid proc-ess, and rostral to the attachments of M. adductormandibulae posterior (Figs. 10C,D).

M. adductor mandibulae externus. Musculusadductor mandibulae externus is the most function-ally and anatomically variable group of the jawmusculature (Table 3). The M. adductor mandibulaeexternus is defined by its position lateral and rostralto the maxillary and mandibular nerves, respec-tively (Lakjer, 1926; Fig. 1). The group representsan amalgam of variably constructed ‘‘temporal’’muscles typically partitioned into superficial (M.adductor mandibulae externus superficialis), medial(M. adductor mandibulae externus medialis), anddeep (M. adductor mandibulae externus profundus)parts in non-avian sauropsids (Fig. 7), or, amongbirds, M. adductor mandibulae externus rostralisand M. adductor mandibulae externus temporalis(Vanden Berge and Zweers, 1993). Smaller, identifi-able bellies also arise within these muscles in vari-ous clades (e.g., M. adductor mandibulae externusprofundus anterior, M. adductor mandibulae exter-nus zygomaticus, M. levator anguli oris). Generally,M. adductor mandibulae externus occupies most ofthe temporal fossa and lateral region of the adduc-tor chamber. It has broad attachments to the dorso-temporal fossa and medial surfaces of the laterallybounding dermatocranium (postorbital, squamosal;Figs. 4, 7–9, 11) and generally has a vertical fiberorientation as it inserts on primarily the dorsal andlateral surfaces of the surangular (Fig. 10).

Among outgroup taxa, the M. adductor mandibu-lae externus of turtles splits into a large M. adduc-

tor mandibulae externus profundus that fills thetemporal fossa and expands caudally into the post-temporal fenestra (Schumacher, 1973; Rieppel,1990) and smaller M. adductor mandibulae exter-nus medialis and M. adductor mandibulae externussuperficialis bellies that occupy the rostrodorsalsurface of the quadrate and medial surface of thepostorbital, quadratojugal, and jugal, respectively.These morphologies are similar to those foundamong lepidosaurs where M. adductor mandibulaeexternus profundus, M. adductor mandibulae exter-nus medialis, and M. adductor mandibulae externussuperficialis all occupy respective positions withinthe dorsotemporal fossa (Haas, 1973; Figs. 4, 7, 11).

M. adductor mandibulae externus profundus—Crocodylia. Musculus adductor mandibulae exter-nus profundus [Schumacher’s (1973) M. pseudotem-poralis] is the only muscle in the dorsotemporalfossa and attaches to the lateral surface of the pari-etal and rostral surface of the squamosal (Figs. 4B,5A, 8, 11B). The muscle is small, semicircular incross-section, and conically pinnate among brevir-ostrines including alligatorids and many Crocody-lus species. Among these taxa, M. pseudotemporalissuperficialis forms its complement such that to-gether they form a two-part circular muscle groupthat occupies the temporal region (Fig. 5B). In long-irostrine taxa (e.g., Gavialis, C. johnstoni, andTomistoma; Iordansky, 1973; Langston, 1973; Endoet al., 2002), M. adductor mandibulae externus pro-fundus has a larger, circular cross-section. The cau-dolateral portion of M. adductor mandibulae exter-nus profundus attaches to the rostromedial surfaceof the mandibular adductor tendon, and attachesdistally as a strong tendon at the rostral edge of thedorsal surface of the surangular, just caudal to therictus in all crocodylian taxa (Figs. 5C, 8, 10B). It isbordered medially by the cartilago transiliens andM. pseudotemporalis superficialis, rostrolaterallyby the rictus, and caudally by M. adductor mandi-bulae externus medialis and M. adductor mandibu-lae externus superficialis.

M. adductor mandibulae externus profundus—Neornithes. M. adductor mandibulae externus pro-fundus is characterized by a variety of differentsubunits, and, except for palaeognaths, is consis-tently responsible for excavating the dorsotemporalfossa (Figs. 4D, 11C,D). Among palaeognaths (e.g.,Eudromia, Struthio, Rhea), M. adductor mandibu-lae externus profundus attaches to the postorbitalprocess and is divisible into several, variable,smaller bellies (Webb, 1957; Elzanowski, 1987;(Figs. 4C, 6D). Despite the partitioning of M. adduc-tor mandibulae externus profundus into differentbellies (Zweers, 1974; Weber, 1996; Zusi and Live-zey, 2000), the basal M. adductor mandibulae exter-nus profundus pattern of attachments is retainedin galloanserines. The main belly of M. adductormandibulae externus profundus (M. adductor man-dibulae externus coronoideus; Zusi and Livezey,



2000) occupies the dorsotemporal fossa proper(impressio coronoideus; Zusi and Livezey, 2000),and the rest of the muscle is subdivided along ven-tral surfaces of the postorbital and zygomatic proc-esses (Weber, 1996). The medial muscles of thegroup (e.g., M. adductor mandibulae externus coro-noideus) attach to the coronoid process. The lateralmuscles of this group (e.g., M. adductor mandibulaeexternus zygomaticus) attach along the rostrolat-eral surface of the mandible and to the lateral man-dibular process, rostral to the attachments of M.adductor mandibulae externus superficialis (Fig.10). Among neoavians, two basic M. adductor man-dibulae externus profundus patterns are prevalent.

In the first and most common state (e.g., Passeri-formes, Falconiformes, Columbiformes, Laridae,Caprimulgiformes, Psittaciformes), M. adductormandibulae externus profundus excavates a simpledorsotemporal fossa and attaches to the coronoideminence (Fig. 11D). The second common pattern isthat of many Pelecaniformes (Pelicanus, Phalacro-corax, Anhinga), Podicipediformes (Aechmor-phorus), and Ciconiiformes (Ardea), in which thecaudal, proper fossa is expanded rostrally into anorbital lamina formed by the laterosphenoid(Burton, 1974). In this case, M. adductor mandibu-lae externus profundus occupies the caudal part ofthe dorsotemporal fossa, and a rostral expansion of

Fig. 8. Major features of the adductor chamber of Alligator mississippiensis in left lateral view. Image is composite based on CTdata for skeletal anatomy and dissections for soft-tissue anatomy. A: Head skeleton. B: Superficial dissection. C: Intermediatedepth. D: Deep dissection. E: deepest dissection. aTO, temporoorbital artery; aTS, superficial temporal artery; ct, cartilago transili-ens; eam, external acoustic meatus; gV, trigeminal ganglion; ju, jugal; lb, lateral bridge of the laterosphenoid; ls, laterosphenoid;mAMEM, musculus (M) adductor mandibulae medialis; mAMEP, M. adductor mandibulae externus profundus; M. adductor mandi-bulae externus superficialis; mAMP, M. adductor mandibulae posterior; mIRA, M. intramandibularis; mPSTp, M. pseudotemporalisprofundus; mPSTs, M. pseudotemporalis superficialis; mPTd, M. pterygoideus dorsalis; mPTv, M. pterygoideus ventralis; nAO,ramus to the corner of the mouth (anguli oris) of the mandibular nerve; nCA, caudal ramus of the mandibular nerve; nCID, motorramus to the constrictor internus dorsalis muscles of the mandibular nerve; nFR, frontal ramus of the ophthalmic nerve; nJU, ju-gal branch of the maxillary nerve; nPTc, caudal branch of the pterygoid ramus of the mandibular nerve; nPTr, rostral branch ofthe pterygoid ramus of the mandibular nerve; nSO, supraorbital branch of the maxillary nerve; po, postorbital; ptb, pterygoid but-tress; qj, quadratojugal; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve.



this muscle occupies the lateral portion of the or-bital lamina, bordered medially by M. pseudotem-poralis superficialis and relevant neurovascularstructures (Fig. 11D).

M. adductor mandibulae externus medialis—Crocodylia. This muscle occupies an intermediateposition, lodged among the other muscles of theadductor chamber. It attaches to the trapezoidalregion of the quadrate just caudoventral to the tri-geminal foramen, dorsolateral to the attachmentfor M. adductor mandibulae posterior and ventro-medial to M. adductor mandibulae externus profun-dus (Busbey, 1989; Figs. 4B, 8). The muscle sharesattachments to the cranial adductor tendon with M.adductor mandibulae externus profundus and M.adductor mandibulae posterior. It merges laterallywith M. adductor mandibulae externus superficialisand rostrolaterally with M. adductor mandibulaeexternus profundus to attach onto the coronoid emi-

nence (Figs. 5C, 8). The remaining fibers attach tothe surangular ventromedial to the attachments ofM. adductor mandibulae externus superficialis anddorsal to the medial mandibular fossa (Fig. 10B).

M. adductor mandibulae externus medialis—Neornithes. M. adductor mandibulae externusmedialis is not sufficiently distinct to be reliablyidentified in birds. The problems underlying thisassessment are elaborated on in the Discussion.

M. adductor mandibulae externus superficialis—Crocodylia. Rather than attaching within the dor-sotemporal fossa as in other non-avian sauropsids(Figs. 4A, 7), M. adductor mandibulae externussuperficialis attaches to the ventrolateral surface ofthe quadrate and quadratojugal (Figs. 4B, 8). Itthen descends caudally as a parallel-fibered muscle.The lateral margin of the M. adductor mandibulaeexternus superficialis is generally free of the der-matocranium, and its investing fascia has only

Fig. 9. Major features of the adductor chamber of Phoebastria immutabilis in left lateral view. Image is composite based on CTdata for skeletal anatomy and dissections for soft-tissue anatomy. A: Head skeleton. B: Superficial dissection (jugal omitted). C: In-termediate depth. D: Deep dissection. aOC, occipital artery; aTO, temporoorbital artery; aTS, superficial temporal artery; eam,external acoustic meatus; mAMEP musculus (M) adductor mandibulae externus profundus; M. adductor mandibulae externussuperficialis; mAMP, M. adductor mandibulae posterior; mIRA, M. intramandibularis; mPSTp, M. pseudotemporalis profundus;mPSTs, M. pseudotemporalis superficialis; mPTd, M. pterygoideus dorsalis; mPTv, M. pterygoideus ventralis; mPPq, M. protractorquadratus; mPPt, M. protractor pterygoideus; mPSTa, M. pseudotemporalis superficialis pars absconditum; mPSTp, M. pseudotem-poralis profundus; mPSTs, M. pseudotemporalis superficialis; mPTd, M. pterygoideus dorsalis; mPTv, M. pterygoideus ventralis;nAO, ramus to the corner of the mouth (anguli oris) of the mandibular nerve; nCA, caudal ramus of the mandibular nerve; nCID,motor ramus to the constrictor internus dorsalis muscles of the mandibular nerve; nFR, frontal ramus of the ophthalmic nerve;nJU, jugal branch of the maxillary nerve; nPT, pterygoid ramus of the mandibular nerve; nSO, supraorbital branch of the maxil-lary nerve; qu, quadrate; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve.



minor attachments to the dermis and the perios-teum of the medial surface of the jugal (Fig. 5). Themedial surface of the muscle has numerous fibersthat attach to the lateral lamina of the mandibularadductor tendon and M. adductor mandibulae pos-terior. It then attaches along the dorsal surface ofthe surangular and rostral to the quadratomandib-ular ligament and synovial capsule (Figs. 8B, 10B).

M. adductor mandibulae externus superficialis—Neornithes. This muscle is thin in palaeognathsand attaches to the temporal fascia dorsomedial tothe jugal with only minor attachments on the post-orbital process (Figs. 4C, 6C). Among most neo-gnaths, M. adductor mandibulae externus superfi-cialis attaches to the subtemporal fossa of the squa-mosal (i.e., suprameatal process; Zusi and Storer,1969) as a long pinnate muscle with tendinousattachments to the zygomatic process and nuchalcrest (Figs. 4D, 9, 11D). A separate muscle slip ofM. adductor mandibulae externus superficialis alsooccasionally attaches to the quadrate otic process[synonymous with M. adductor externus pars pro-fundus, Hofer (1950); M. articularis externus,Weber (1996), Zusi and Livezey (2000); Fig. 4D].However, among many birds, the lateral surface ofthe quadrate is devoid of muscle attachment. In cor-morants and possibly other Pelicaniformes, a sesa-moid develops in the medial portion of M. adductormandibulae externus superficialis as it wraps overthe lateral quadrate cotyla of the squamosal at therostrolateral margin of the subtemporal fossa. Themuscle then attaches to the dorsolateral surface ofthe mandible rostral to the mandibular attach-ments of M. adductor mandibulae posterior and

caudal to the attachments of M. adductor mandibu-lae externus profundus (Fig. 10D).

Neurovascular structures of the temporalregion.

Maxillary nerve. The position of the maxillarynerve has received the most attention (Luther, 1914;Lakjer, 1926; Lubosch, 1933; Edgeworth, 1935) as acriterion of the trigeminal topological paradigm thatdistinguishes the adductor mandibulae internusgroup (e.g., M. pseudotemporalis superficialis) medi-ally from the adductor mandibulae externus (e.g.,M. adductor mandibulae externus profundus) grouplaterally (Figs. 1, 7–9, 12). This criterion holds for allbirds, such that even in palaeognaths, where M.pseudotemporalis superficialis occupies the dorso-temporal fossa somewhat caudolaterally to the max-illomandibular foramen, the nerve maintains a ple-siomorphic course, wrapping around the caudolat-eral surface of the muscle, and passes between M.pseudotemporalis superficialis and M. adductormandibulae externus profundus before entering thesuborbital region (Fig. 6B). Among crocodylians, themaxillary nerve has a short path through the tempo-ral region, exiting the rostral edge of the maxillo-

Fig. 11. Musculoskeletal patterns of the right dorsotemporalfossa in representative diapsid clades in dorsal view. A: Lepido-sauria. B: Crocodylia. C: Paleognathae. D: Neoaves. ls, laterosphe-noid; mAMEM, musculus (M) adductor mandibulae externusmedialis; mAMEP, M. adductor mandibulae externus profundus;M. adductor mandibulae externus superficialis; mPSTs, M. pseudo-temporalis superficialis; mTP,M. tensor periorbitae; pa, parietal; po,postorbital; pr, prootic; sq, squamosal.

Fig. 10. Musculoskeletal patterns of right mandible in repre-sentative diapsid clades in medial view. A: Lepidosauria. B: Croc-odylia. C: Palaeognathae. D: Neognathae. Note: pterygoideusmuscles are not figured for sake of clarity. cp, coronoid process;ct, cartilago transiliens; mAMEM, musculus (M) adductor mandi-bulae externus medialis; mAMEP, M. adductor mandibulae exter-nus profundus; M. adductor mandibulae externus superficialis;mAMP, M. adductor mandibulae posterior; mIRA, M. intramandi-bularis; mmf, medial mandibular fossa; mmp, medial mandibularprocess; rap, retroarticular process.



Fig. 12.


mandibular foramen, running between the lateralbridge of the laterosphenoid medially and M. pseu-dotemporalis superficialis laterally and dorsomedialto M. pterygoideus dorsalis and M. pseudotempora-lis profundus (Figs. 5, 8). Therefore, based on theseand other data analyzed in the Discussion, the jawmusculature of crocodylians does not conform to thetrigeminal topological paradigm as originallydescribed (e.g., Luther, 1914; Lakjer, 1926; Edge-worth, 1935).

Ramus supraorbitalis of the maxillary nerve.The supraorbital nerve is the first branch of the max-illary nerve [Oelrich, 1956; Webb, 1957; Soliman’s(1963) ramus frontalis; Bubien-Waluszewska, 1981]and typically branches dorsally from the maxillarydivision and passing lateral to M. pseudotemporalissuperficialis and M. tensor periorbitae at the bound-ary between the orbital and temporal regions (Figs.7–9, 12). The nerve then ramifies throughout thecaudolateral part of the orbit, lacrimal gland, andskin. Being a branch of the maxillary nerve, thesupraorbital nerve usually maintains a positionallyequivalent intermuscular course, passing betweenM. adductor mandibulae internus and externus inbirds and lepidosaurs (Fig. 12A,C). However, in croc-odylians, the supraorbital nerve uncouples from themaxillary division, instead of arising from the tri-geminal ganglion within the caudodorsal region ofthe trigeminal fossa (Figs. 8, 12B). While the maxil-lary nerve passes rostrally just lateral to the lateralbridge of the laterosphenoid and ventromedial to M.pseudotemporalis superficialis, the supraorbitalnerve runs dorsomedially, passing either through thecaudal part of the laterosphenoid or within thesuture between the laterosphenoid and the quadrate,often running through a separate foramen. Thenerve enters dorsally into the dorsotemporal fossabetween the parietal and M. adductor mandibulaeexternus profundus (Fig. 8), and then passes betweenM. adductor mandibulae externus profundus and M.pseudotemporalis superficialis (Fig. 12B), before join-ing the temporoorbital artery and the maxillarynerve and ramifying across the back of the orbit andlacrimal gland.

Ramus jugalis of the maxillary nerve. Boundingthe rostrolateral margin of the adductor chamber isthe second branch of the maxillary nerve, here called

the jugal branch of the maxillary nerve (i.e., jugalnerve) (Figs. 7–9, 12). Although Oelrich (1956) foundthis nerve in Ctenosaura to have orbital branchessimilar to those of the supraorbital nerve, this studyfailed to find equivalent branches among the sample.The nerve runs rostral to M. adductor mandibulaeexternus profundus and M. adductor mandibulaeexternus superficialis, dorsal to M. pterygoideus dor-salis, and ramifies across the skin below the orbit. Ingeneral, this nerve lies rostral to the orbitotemporalboundary and dorsal to the pterygoideus muscles,therefore outside of the adductor chamber. In doingso, the nerve forms one of the boundary structures ofthe adductor chamber.

Ramus ptergyoideus of the mandibular nerve.The pterygoid branch of the mandibular nerve (i.e.,pterygoid nerve) is the primary motor nerve of theadductor mandibulae internus muscles. In crocody-lians, the pterygoid nerve splits into rostral and acaudal branches near its separation from the man-dibular nerve (Fig. 8). The rostral branch apo-morphically passes between M. pseudotemporalissuperficialis and profundus and then passesbetween M. pseudotemporalis profundus and M.pterygoideus dorsalis, subsequently ramifyingacross the dorsal surface of M. pterygoideus dorsa-lis (Fig. 12B). The caudal branch of the pterygoidnerve passes medial to M. adductor mandibulaeposterior, often giving off motor rami to this muscle(Poglayen-Neuwall, 1953b; pers. obs.), and ramifiesbetween M. pterygoideus dorsalis and M. pterygoi-deus ventralis in the caudal portion of the adductorchamber. The path of the pterygoid nerve is welldocumented among birds (Hofer, 1950, Dzerzhinskyand Yudin, 1982; Vanden Berge and Zweers, 1993;Weber, 1996). The nerve always passes rostroven-trally across the quadrate between the otic and or-bital processes and then between M. pseudotempor-alis profundus and M. adductor mandibulae poste-rior lateral to the quadrate and between M.pseudotemporalis profundus and M. pterygoideusdorsalis medial to the quadrate (Figs. 9, 12C). Itsends off branches to M. pseudotemporalis profun-dus and M. pseudotemporalis superficialis and thenramifies across the dorsal surface of M. pterygoi-deus dorsalis to innervate M. pterygoideus ventra-lis. The pterygoid nerve also often carries motor

Fig. 12. Topological patterns of muscles, nerves, and vessels in the adductor chamber of sauropsids. Orientation as in Figure 1.A: Plesiomorphic condition common to Lepidosauria and Testudines. B: Extant crocodylian condition. C: Typical extant avian condi-tion. ?, unclear presence of mAMEM; aPR, profundus branch of the temporoorbital artery; aSP, sphenopalatine artery; aST, stape-dial artery; aTO, temporoorbital artery; aTR, rostral trigeminal artery; gMM, maxillomandibular ganglion; gV1, ophthalmic gan-glion; mAMEM, musculus (M) adductor mandibulae externus medialis; mAMEP, M. adductor mandibulae externus profundus;mAMES, M. adductor mandibulae externus superficialis; mAMP, M. adductor mandibulae posterior; mLPt, M. levator pterygoideus;mPPt, M. protractor pterygoideus; mPSTp, M. pseudotemporalis profundus; mPSTs, M. pseudotemporalis superficialis; mPT, M.pterygoideus; mTP, M. tensor periorbitae; nAO, ramus to the corner of the mouth (anguli oris) of the mandibular nerve; nCA, cau-dal ramus of the mandibular nerve; nCID, motor ramus to the constrictor internus dorsalis muscles of the mandibular nerve; nFR,frontal ramus of the ophthalmic nerve; nJU, jugal branch of the maxillary nerve; nPT, pterygoid ramus of the mandibular nerve;nPTc, caudal branch of the pterygoid ramus of the mandibular nerve; nPTr, rostral branch of the pterygoid ramus of the mandibu-lar nerve; nSO, supraorbital branch of the maxillary nerve; V1, ophthalmic nerve; V2, maxillary nerve; V3, mandibular nerve; vSP,sphenopalatine vein.



fibers to M. adductor mandibulae posterior thatbranch off near the rostroventral surface of themuscle as the nerve travels across the quadrate.

Ramus anguli oris of the mandibular nerve. Thenerve to the corner of the mouth (i.e., rictal nerve)is a relatively large sensory nerve that has previ-ously been identified among several sauropsids(Poglayen-Neuwall, 1953a,b; Barnikol, 1954; Oel-rich, 1956; Bubien-Waluszewska, 1981; McDowell,1986). The rictal nerve passes rostrolaterally fromthe mandibular nerve between the muscles of M.adductor mandibulae internus and externus andramifies across the rictus (i.e., the corner of themouth; Figs. 5C, 6E, 7–9, 12). Along its course, therictal nerve joins with communicating branches ofthe palatine ramus of the facial nerve that havecrossed laterally over the dorsal surface of M. ptery-goideus dorsalis. With few exceptions, the rictalnerve in birds and lepidosaurs always startsbetween M. pseudotemporalis superficialis and M.adductor mandibulae externus profundus and endsbetween the same muscles at the corner of themouth. However, in crocodylians, the nervediverges from this pattern, passing between M.adductor mandibulae externus profundus and M.adductor mandibulae externus medialis, and subse-quently ramifies at the rictus (Figs. 8, 12).

Ramus caudalis of the mandibular nerve. Thecaudal branch of the mandibular nerve (i.e., caudalnerve) maintains a position rostral to M. adductormandibulae posterior among non-archosaurian sau-ropsids (Poglayen-Neuwall, 1953a; Oelrich, 1956;Soliman, 1963; Haas, 1973; Dzerzhinsky and Yudin,1982; Rieppel, 1990; Figs. 7–9, 12). The caudalnerve is the motor supply to M. adductor mandibu-lae posterior and M. adductor mandibulae externussuperficialis in crocodylians and birds (Hofer, 1950;Poglayen-Neuwall, 1953b), and terminally ramifiesas a sensory nerve across the rostral edge of theexternal acoustic meatus. Among lepidosaurs, thecaudal nerve follows a relatively consistent pathbetween M. adductor mandibulae externus and M.adductor mandibulae posterior muscle groups. Incrocodylians, the nerve variably passes laterallythrough M. adductor mandibulae posterior near thecaudal border of M. adductor mandibulae externusprofundus and M. adductor mandibulae externusmedialis (Figs. 8, 12). The nerve subsequently pier-ces the caudodorsal portion of M. adductor mandi-bulae externus superficialis and enters the overly-ing dermis near the dorsolateral surface of thequadratojugal. Among birds, the nerve maintains aconstant position rostral to M. adductor mandibulaeposterior and caudal to M. adductor mandibulaeexternus superficialis (e.g., M. adductor mandibu-lae externus articularis externus) and typicallyruns with the superficial temporal artery, a lateralbranch of the temporoorbital artery.

Temporoorbital artery. The most topologically in-formative vascular structures in the adductor

chamber are the temporoorbital artery [temporalartery of Oelrich (1956) and Haas (1973); externalophthalmic artery of Baumel (1993)] and the oph-thalmic rete, which are both rostral continuationsof the stapedial artery. The temporoorbital artery(Sedlmayr, 2002) plesiomorphically shares the samepath as the maxillary nerve (Figs. 6A, 7, 9, 12). Incrocodylians, the rostrodorsal shifting of the oticregion led the temporoorbital artery to enter thedorsotemporal fossa at a point dorsal to M. adductormandibulae externus profundus rather than ven-tromedial to it (Figs. 8, 12). Despite this shift, thevessel maintains a position between M. adductormandibulae externus profundus and M. pseudotem-poralis superficialis. In all birds investigated,except Galloanserae, the temporoorbital arteryruns between the muscles of M. adductor mandibu-lae internus and externus. However, in Galloan-serae, the temporoorbital artery departs from themaxillary nerve and passes between M. adductormandibulae externus profundus pars coronoideusmedially and pars zygomaticus laterally (Fig. 12D).Despite these changes, the temporoorbital arterystill reunites with the maxillary nerve on the ros-trodorsal margins of these muscles.

Occipital and superficial temporal arteries. Incrocodylians, the occipital artery branches from theexternal carotid artery (Sedlmayr, 2002) outside ofthe adductor chamber and the skull in general, yetmaintains a position dorsomedial to M. depressormandibulae. On the other hand, the artery originatesfrom several different regions in birds. In ducks, theoccipital artery branches from the internal carotid ar-tery caudal to the middle ear cavity, whereas inchickens, the artery arises from the external carotidartery outside of the middle ear cavity (Baumel,1993). However, in flamingos (Holliday et al., 2006),gulls (Midtgard, 1984), and cormorants, the arterybranches off the stapedial artery at the rostral por-tion of the middle ear cavity. The occipital arterythen passes lateral to M. pseudotemporalis superfi-cialis caudodorsally into a bony canal or the dorsaltympanic recess. In the latter case, the artery is oftencoupled with M. pseudotemporalis superficialis parsabsconditum (Fig. 9). Despite this variation in origin,the vessel typically exits the skull between the cau-dal margin of M. adductor mandibulae externussuperficialis and the rostrodorsal edge of M. depres-sor mandibulae, dorsal to the external acoustic mea-tus and deep to M. splenius capitus.

In crocodylians, the superficial temporal arteryenters a canal in the postorbital at the rostrolateralcorner of the dorsotemporal fenestra and exits ros-tral to the external acoustic meatus, dorsal to M.adductor mandibulae externus superficialis. Inbirds, the superficial temporal artery [temporal ar-tery of Sedlmayr (2002)] consistently runs betweenM. adductor mandibulae externus superficialis andM. adductor mandibulae posterior. The vesselbranches off the temporoorbital artery, passes later-



ally, caudal to the squamosal attachments of M.adductor mandibulae externus superficialis andramifies across the lateral surface of the head ros-tral to the external acoustic meatus. The mostlikely lepidosaurian homolog of the superficial tem-poral artery is the auricular artery (Oelrich, 1956),which, although branching off of the mandibular ar-tery rather than the temporoorbital artery, alsopasses between M. adductor mandibulae externussuperficialis and M. adductor mandibulae posteriorto ramify across the rostral border of the externalauditory meatus.

The topological relationships between the caudalnerve and superficial temporal artery suggest thatthe quadrate belly of the avian M. adductor mandi-bulae externus superficialis (i.e., M. adductor man-dibulae externus articularis internus; Zusi andLivezey, 2000) is more topologically similar to M.adductor mandibulae posterior, and may simply bea laterally displaced belly of this deeper muscle.Indeed, these two muscles have continuous attach-ments across both the lateral surface of the quad-rate and the dorsolateral surface of the lower jaw.Alternatively, the space constraints developed by asmall M. adductor mandibulae externus superficia-lis belly on the quadrate may be responsible for adorsal shift in the path of the caudal nerve.

DISCUSSIONThe Apomorphic Adductor Chamberof Crocodylians

Lepidosaurs best approximate the plesiomorphicsauropsid condition for the topological relationshipsof adductor chamber constituents (Fig. 12A), andthese patterns are fundamentally similar in turtles,as well. Divergences from these patterns—forexample, a nerve running medial rather than lat-eral to a muscle—likely reflect apomorphic shifts inadductor chamber construction. Parsimony (Pat-terson, 1982) suggests that the best-supported hy-pothesis of muscle homology is that which exhibitsthe fewest number of derived character statechanges in topology. For example, the adductorchambers of palaeognaths are more or less topologi-cally similar to the plesiomorphic lepidosaur condi-tion (Fig. 12C) despite the obvious morphologicaldifferences between these clades. Among neo-gnaths, several neurovascular structures exhibitvariation in their topological positions to muscles(Fig. 12D). In Galloanserae, both the rictal nerveand the temporoorbital artery deviate laterally,running through M. adductor mandibulae externusprofundus rather than medial to it. In some Psitta-ciformes (e.g., Ara ararauna), the rictal nervepasses through M. pseudotemporalis superficialisrather than lateral to it. Because neither neurovas-cular structure completely shifts its intermuscularcourse, it is difficult to regard the positionalarrangement as a full character transition. None-

theless, the adductor chambers of turtles, lepido-saurs, and neornithines are remarkably similar andconservative, particularly compared to the patternsfound in crocodylians (Figs. 12, 13).

The muscles of the crocodylian dorsotemporalfossa (i.e., M. pseudotemporalis superficialis vs. M.adductor mandibulae externus profundus) have themost variable interpretations of homology (Lakjer,1926; Iordansky, 1964; Haas, 1973; Busbey, 1989;Fig. 13). Sauropsid dorsotemporal musculature istypically dominated by two muscles: (1) a rostralbelly that attaches to the front of the neurocranium(e.g., prootic, laterosphenoid) and (2) a caudal bellythat attaches to the parietal and squamosal (andabuts M. adductor mandibulae externus medialis),typically occupying much of the caudal part of thedorsotemporal fossa (Figs. 4, 12, 13). This studysubjected alternative crocodylian homology hypoth-eses to a topological similarity test to determine thenumber of steps from the plesiomorphic conditionthat each hypothesis would require (Fig. 13). Thecrocodylian homology hypothesis with the fewestcharacter state changes from the plesiomorphiccondition (Fig. 13A) is ‘‘crocodylian homology hy-pothesis A’’ (CHHA; Fig. 13). This hypothesis identi-fies the rostral belly of the dorsotemporal region asM. pseudotemporalis superficialis and the caudalbelly as M. adductor mandibulae externus profun-dus (Fig. 13B), requiring only three changes in neu-romuscular topology: positional switches betweenM. pseudotemporalis superficialis and the maxil-lary nerve, the rictal nerve and M. adductor mandi-bulae externus profundus, and the pterygoid nerveand M. pseudotemporalis superficialis. On the con-trary, the most common previous interpretation ofhomology—‘‘crocodylian homology hypothesis B’’:both bellies are M. adductor mandibulae externusprofundus (Lakjer, 1926; Iordansky, 1964; Busbey,1989)—requires changes in six character suites(i.e., the mandibular, pterygoid, rictal, and supraor-bital nerves and temporoorbital artery) and the for-mation of two separate M. adductor mandibulaeexternus profundus bellies (pars rostralis and parscaudalis). Two alternative homology hypotheses(crocodylian homology hypotheses C and D) alsoresult in more steps than crocodylian homologyhypotheses A. Hypothesis C interprets the musclesin the dorsotemporal fossa and on the laterosphe-noid to be M. pseudotemporalis (e.g., Schumacher,1973) and M. adductor mandibulae externus pro-fundus to attach to the quadrate (i.e., homologousto M. adductor mandibulae externus medialis inother hypotheses). While this scenario supports therictal nerve- M. adductor mandibulae externus pro-fundus criterion, other relevant criteria fail (i.e.,maxillary, mandibular, and pterygoid nerves, tem-poroorbital artery etc.), and a neomorphic muscle,M. intermedius (Iordansky, 1964), is required.

Crocodylian homology hypothesis D, a novel hy-pothesis proposed here, has only one extra step



Fig. 13.


compared to homology hypothesis A. This hypothe-sis suggests that the small slip of muscle on the lat-erosphenoid lateral bridge is a small, neomorphicbelly of M. pterygoideus dorsalis (M. pterygoideusdorsalis minimus) instead of M. pseudotemporalisprofundus, which in turn is considered to be elimi-nated. Except for this switch in terminology, whichpresumes a nonhomology between the two muscles,Crocodylian homology hypothesis D has the samecharacter-state changes as Crocodylian homologyhypothesis A, thus supporting the positional inter-pretation of M. pseudotemporalis superficialis andM. adductor mandibulae externus profundus incrocodylian homology hypothesis A. Despite thefailure of the M. pseudotemporalis superficialis–maxillary nerve character suite (i.e., the failure ofthe trigeminal topology) in both homology hypothe-ses A and D, M. pseudotemporalis superficialis stillmaintains a position medial to the mandibularnerve, both within the temporal region, and in themedial mandibular fossa via its connection with M.intramandibularis. Moreover, although previoushypotheses relied on the maxillary nerve as the pri-mary criterion (e.g., M. pseudotemporalis superfi-cialis lies medial to the maxillary nerve), the M.pseudotemporalis superficialis–maxillary nerve cri-terion would require the muscle on the laterosphe-noid to be M. adductor mandibulae externus pro-fundus, in turn breaking the M. adductor mandibu-lae externus profundus–mandibular nerve criterionthat the paradigm assumes. Thus, not only do par-ticular characters in crocodylians (e.g., rictal nerve,maxillary nerve) depart from the plesiomorphiccondition, it is impossible to homologize crocodylianjaw muscles with those of other reptiles withoutviolating the assumptions of the trigeminal topolog-ical paradigm.

The adductor chamber of crocodylians has divergedfrom the classic trigeminal topology most likelybecause of the suturing of the palate to the brain-case—specifically, the characteristic evolution andeventual elimination of the epipterygoid in earlyeusuchians (see below and Holliday, 2006). Previousstudies have also recognized the fallibility of the tri-geminal paradigm, noting its susceptibility to devel-opmental and phylogenetic perturbations. Haas(2001) recognized that not all amphibians fit the com-mon topological pattern. Presley (1993) describedhow the route of the maxillary division shifts due todevelopment of new bony elements (e.g., the alisphe-

noid of mammals). Rieppel (1988) illustrated perhapsthe most the extreme violation of the paradigm inthe amphisbaenian Trogonophis in which the maxil-lary division of the trigeminal nerve actually passesmedial to the epipterygoid, within the cavum epipter-icum. Therefore, it may not be so surprising thatsuch an apomorphic reptilian group as crocodyliansalso violates this paradigm.

Topology, Development, and Dermatomes:The Need for Additional Nerves

The inability to rely on the main trigeminal divi-sions requires the use of additional topological crite-ria, such as secondary nervous structures (e.g.,pterygoid, caudal, and rictal nerves) that werefound to be almost as consistent as the ophthalmic,maxillary, and mandibular nerves. Indeed, certainstructures exhibited static patterns among allclades, including the paths of the caudal and man-dibular nerves lateral (rostral) to M. adductor man-dibulae posterior, and some of the contents of thecavum epiptericum (e.g., M. protractor pterygoi-deus, ophthalmic nerve, and the motor branch toM. constrictor internus dorsalis; Holliday, 2006).The relationships between the caudal and mandibu-lar nerves and M. adductor mandibulae posteriorare consistent among the sauropsid groups investi-gated despite the muscle’s developmental originfrom either M. adductor mandibulae internus orexternus rudiments (Rieppel, 1987, 1990). The cau-dolateralmost soft-tissue structure in the adductorchamber is M. adductor mandibulae posterior,which may simply occupy the periphery of the topo-logically informative structures. This peripheryconstruct holds true for the medialmost structuresas well (e.g., M. pterygoideus dorsalis, M. protractorpterygoideus, pterygoid nerve, and the motorbranch to M. constrictor internus dorsalis), whichalso exhibit relatively little variation in topologicalpatterns compared to those in the temporal region.

Although adult trigeminal topologies may be sim-ilar, differing ontogenetic trajectories of the charac-ter complexes may cloud the assessment of homol-ogy (Rieppel, 1988). Overall, however, conservativedevelopmental mechanisms are responsible for theconsistent patterns found in the sauropsid adductorchamber. Predominantly sensory and some motorrami of the trigeminal nerve proved informative asboundaries of the adductor chamber and as criteria

Fig. 13. Parsimony analysis of muscle homology hypotheses within the dorsotemporal fossa of crocodylians [i.e., CrocodylianHomology Hypotheses (CHH) A–D]. A: Plesiomorphic condition common to lepidosaurs, turtles, and birds. Color codes are the sameas in Figure 12. B: Four separate hypotheses of muscle homology listing the number of topological and muscular character statechanges (n ¼ x) away from the plesiomorphic condition. CHHA is the most parsimonious homology hypothesis. See Figure 4 formuscle attachments. aTO, temporoorbital artery; dtf, dorsotemporal fossa; gV, trigeminal ganglion; mAMEM, musculus (M) adduc-tor mandibulae externus medialis; mAMEP; M. adductor mandibulae externus profundus; mAMEPc, caudal belly of mAMEP;mAMEPr; rostral belly of mAMEP; mINT, M. intermedius; mPTd, M. pterygoideus dorsalis; mPTdm, M. pterygoideus dorsalis parsminimus; nPT, pterygoid ramus of the mandibular nerve; nSO, supraorbital branch of the maxillary nerve; V1, ophthalmic nerve;V2, maxillary nerve; V3, mandibular nerve.



for muscle homology. In general, the major sensorycomponents (e.g., mandibular, supraorbital, andfrontal nerves) project to their target dermal fieldsvia chemoattractant signals from the integumentprior to the development of significant muscular or-ganization (Covell and Noden, 1989; Kuratani andTanaka, 1990; Scott, 1992), motor rami and inner-vation (Vogel, 1992; Song and Boord, 1993), andmuscular attachment (Edgeworth, 1935; de Beer,1937; McClearn and Noden, 1988).

These data suggest that the target integumentaryregions of the supraorbital, rictal, and caudalnerves—the orbitotemporal boundary, the corner ofthe mouth, and the ototemporal boundary, respec-tively—are developmentally conserved dermatomes(Fig. 14). In addition, the two motor rami (motorbranch to the M. constrictor internus dorsalis andpterygoid nerve) also maintain consistent somato-topic relationships between their nuclei and theirmuscular targets, M. constrictor internus dorsalisand M. adductor mandibulae internus, respectively(Barnikol, 1951; Song and Boord, 1993; Figs. 7–9,12). However, other rami, such as the main motorbranch to M. adductor mandibulae externus(Poglayen-Neuwall, 1953a,b), project via intramus-cular rather than intermuscular routes, and offer lit-tle resolution of muscle homology. Additional motorrami hitchhike along the large sensory branches(e.g., caudal and rictal nerves) and innervatemuscles along their intermuscular paths (Fig. 12D).Thus, many muscle groups receive dual innervationfrom anatomically different nerves (Barnikol, 1951;Poglayen-Neuwall, 1953a,b). For example, motorrami to the sauropsidM. adductor mandibulae exter-nus superficialis typically travel along one or twomain rami as well as the caudal and rictal nerves.Motor rami to M. adductor mandibulae posteriororiginate from the pterygoid and caudal nerves, andindividually from the mandibular nerve. Therefore,not only the main trigeminal divisions but also sen-sory and some motor branches all contribute as validtopological criteria for testingmuscle homology.

Homology and the Requirementof Multiple Testing Criteria

The atomistic breakdown of organisms intosmaller subunits is necessary for comparisons ofidentifiably similar parts, and in many cases, thereis little disagreement as to the commonality ofshared features among different taxa (Rieppel andKearney, 2002; Hall, 2003). For example, the homol-ogy of the adductor chamber as a whole among sau-ropsids and even within Amniota is unquestioned.The space is consistently surrounded by other ho-mologous parts including the orbital, otic, ence-phalic, pharyngeal, and integumentary componentsof the head. However, the division of the adductorchamber itself into smaller parts and the testing ofhomology of these components becomes increas-

ingly nebulous. The greater morphological complex-ity and taxonomic variability of these subunitsrequire not only more rigorous homology tests butalso multiple tests using a range of criteria. Occa-sional failure of any single test should not necessar-ily falsify a hypothesis of homology, but merely onlydrive investigations into further similarity testingand probing the mechanisms underlying these dif-ferences.

Obviously, not all putative homologies are robustunder all criteria, nor would they necessarily be

Fig. 14. Postulated dermatome loci of topologically informativesensory nerves. A: Lepidosauria (Iguana iguana). B: Crocodylia(Alligator mississippiensis). C: Neornithes (Anas platyrhynchos).dtf, dorsotemporal fossa; eam, external acoustic meatus; nAD,dorsal alveolar branch of the maxillary nerve; nAO, ramus to thecorner of the mouth (anguli oris) of the mandibular nerve; nAV,ventral alveolar branch of the mandibular nerve; nCA, caudalramus of the mandibular nerve; nEC, external cutaneous branchof the mandibular nerve, nFR, frontal ramus of the ophthalmicnerve; nJU, jugal branch of the maxillary nerve; nMU, muscularbranches of the mandibular nerve; nSO, supraorbital branch ofthe maxillary nerve; or, orbit; ri, rictus.



under a different criteria set. For example, the posi-tion of M. adductor mandibulae externus medialis isoften the means to establish the homology of othermuscles (Lakjer, 1926; Haas, 1973). However, M.adductor mandibulae externus underwent signifi-cant reorganization during the evolution of birds andcrocodylians, greatly modifying the adductor cham-ber compared to that of lepidosaurs. In lepidosaurs,the M. adductor mandibulae externus complex ispartitioned by the Bodenaponeurosis, the parasagit-tally situated aponeurosis that attaches to the coro-noid process, and reliably differentiates M. adductormandibulae externus profundus and M. adductormandibulae externus medialis (Lakjer, 1926; Haas,1973; but see Rieppel, 1990). However, the Bodena-poneurosis may be an autapomorphy of Lepidosauriarather than a shared feature of Sauropsida, and, ifits homolog is still present at all, it has been greatlymodified in both crocodylians and birds.

In crocodylians, the mandibular adductor tendon(Iordansky, 1964; Schumacher, 1973) is the bestcandidate for a homolog of the Bodenaponeurosis.However, it has undergone significant folding asso-ciated with the suturing of the quadrate to thebraincase, and forms more of a shared aponeurosis(for all bellies of M. adductor mandibulae externusand posterior) rather than the anchor for two (M.adductor mandibulae externus profundus, M.adductor mandibulae externus medialis; Fig. 5B–D). Moreover, M. adductor mandibulae externusmedialis (Iordansky, 1964; Schumacher, 1973; Bus-bey, 1989) is a relatively nebulous muscle belly thathas few to no fascial boundaries separating it fromM. adductor mandibulae externus profundus, M.adductor mandibulae externus superficialis, M.pseudotemporalis superficialis, and M. adductormandibulae posterior, particularly near its mandib-ular attachment (Fig. 8). In ratites, M. adductormandibulae externus has an internal aponeurosisthat partially separates a deep belly (M. adductormandibulae externus profundus) from a more lat-eral one (M. adductor mandibulae externus medi-alis) (Webb, 1957; Elzanowski, 1987; Fig. 6C,D), butthe two bellies are generally indistinguishable sug-gesting that, if present, the Bodenaponeurosis isgreatly reduced.

Perhaps the best avian candidate for a Bodenapo-neurosis homolog is the aponeurosis paracoronoidea(Weber, 1996; Zusi and Livezey, 2000) in Galloan-serae. However, avian myologists (e.g., Hofer, 1950;Vanden Berge and Zweers, 1993; Weber, 1996; Zusiand Livezey, 2000) do not typically recognize M.adductor mandibulae externus medialis but ratherapply a different nomenclature. Overall, the avianM. adductor mandibulae externus pars ventralis(pars medialis; Van Gennip, 1986; Vanden Bergeand Zweers, 1993) may be most similar to the M.adductor mandibulae externus medialis of othersauropsids. Nonetheless, birds have partitionedtheir adductor musculature into so many functional

compartments that the identification of an unam-biguous M. adductor mandibulae externus medialisis difficult if not impossible. Finally, even if the Bod-enaponeurosis, or its possible crocodylian (mandib-ular adductor tendon) or avian (aponeurosis para-coronoidea) homologs, is a reliable homology crite-rion (Rieppel, 1990), this study failed to find anynonmuscular topological criteria (i.e., nerves, ves-sels) that separate M. adductor mandibulae exter-nus medialis from M. adductor mandibulae exter-nus profundus in crocodylians or birds, eliminatingit from accurate comparisons among taxa.

Likewise, although the pterygoideus musclesmaintain consistent bony attachments across sau-ropsids and particular neurovascular structures(e.g., pterygoid nerve and sphenopalatine artery)separate the pterygoideus muscles from othermuscles, only musculoskeletal criteria distinguishM. pterygoideus dorsalis from M. pterygoideus ven-tralis. Therefore, although the identifications of M.pterygoideus dorsalis and ventralis are robustusing various musculoskeletal and developmentalcriteria, they draw no support from neurovascularcriteria. Therefore, hypotheses of homology of thesemuscles may not be as robust as those that are sub-ject to all three testing criteria. Likewise, applic-ability of relatively few testing criteria led to thehesitant identification of the small muscle belly onthe crocodylian laterosphenoid lateral bridge as M.pseudotemporalis profundus. The muscle sharespositional qualities with M. pterygoideus dorsalisbut, if the muscle is indeed M. pseudotemporalisprofundus, it has an apomorphic position relative toramus pterygoideus of the mandibular nerve (nPT;Figs. 12, 13). Thus, additional data (e.g., develop-mental data, motor innervation) are necessary tofurther test the homology of this muscle.

Muscles may share topological similarity but, dueto evolutionary changes, may violate tests of attach-ment similarity or developmental connectivity. Theadductor mandibulae posterior attaches within themedial mandibular fossa in all non-neognath dia-psids. However, it attaches to the lateral or dorsalsurface of the mandible in many neognaths, violat-ing the musculoskeletal attachment position com-mon to other sauropsids. Likewise, M. adductormandibulae posterior may develop from either M.adductor mandibulae internus or externus rudi-ments (Rieppel, 1990) among non-avians despite itshomologous position caudal to mandibular and cau-dal nerves among diapsids. Therefore, while neuro-logical criteria support homology of M. adductormandibulae posterior among diapsids, developmen-tal and perhaps musculoskeletal criteria do not.

Finally, hypotheses of homology at an intramus-cular level of similarity can be confounded by acces-sory structures. The formation of intertendons andfibrocartilaginous sesamoids in jaw muscles is rela-tively common in sauropsids (Hofer, 1950; Schu-macher, 1973). These cartilaginous structures have



often been used as a musculoskeletal criterion toseparate two different muscles, particularly M.pseudotemporalis superficialis from M. intramandi-bularis in turtles, crocodylians, and birds (Lakjer,1926; Hofer, 1950; Iordansky, 1964; Schumacher,1973; Vanden Berge and Zweers, 1993). Increasesin compressional force on muscular tissues lead toincreased formation of fibrocartilage, in turn form-ing a thickened tendon or sesamoid that acts as afunctional enthesis (Benjamin and McGonagle,2001). It is thus more plausible that M. intramandi-bularis is merely the continuation of M. pseudotem-poralis superficialis into the Meckelian (medialmandibular) fossa and therefore a single continuousmuscle rather than two separate muscles thatshare a common sesamoid attachment. Thus, thecomponents of larger homologous cephalic struc-tures present a variety of interpretations of homol-ogy, each reliant on the relative power of severalsimilarity testing criteria, each of which must beincorporated to adequately describe the morphologyand evolution of jaw muscles, their intertwiningneurovasculature, and the adductor chamber as awhole among amniotes.

ACKNOWLEDGMENTS

The authors thank A. Biknevicius, S. Reilly, andB. Sindelar for offering guidance and supportthroughout the development and production of theresearch. Many thanks go to A. Clifford, J. Daniel,D. Dufeau, D. Ford, T. Hieronymus, R. Ridgely, J.Sedlmayr, J. Tickhill, and the students and facultyof the Ecology and Evolutionary Biology group atOhio University (OU), all of whom have offeredinsightful comments and contributed to provokingdiscussions. This manuscript benefited from inputfrom A. Haas, F. Harrison, and an anonymousreviewer. We are grateful to those who donatedspecimens to the project including: R. Elsey, Rockef-eller National Wildlife Refuge, Grand Chenier, LA;M. Staton, Mainland Holdings Pty., Lae, PapuaNew Guinea; C. Delorey, Brevard Zoo, Melbourne,FL; B. Zaun, Kauai National Wildlife Refuge,Kilauea, HI; W. Roosenburg and S. Moody, OU; andThe Wilds, Cumberland, OH. We thank H. Mayleand O’Bleness Memorial Hospital, Athens, OH, andT. Rowe and University of Texas CT for providingaccess to imaging facilities and data. Thanks to B.Livezey (CMNH), A. Resetar and D. Willard(FMNH), and P. Holroyd (UCMP) who offered accessto their extant vertebrate skeletal collections.

LITERATURE CITED

Adams LA. 1919. A memoir on the phylogeny of the jawmuscles in recent and fossil vertebrates. Annals of the NewYork Academy of Science 28:51–166.

Anderson HT. 1936. The jaw musculature of the phytosaur.Machaeroprosopus. J Morphol 59:549–587.

Barnikol HA. 1951. Uber einige gesetzmaßigkeiten im Aufbauder motorichen Trigeminusaste bei Vogeln. Anat Anz 98:217–223.

Barnikol HA. 1954. Zur Morphologie de Nervus trigeminusder Vogel unter besonderer Berucksichtigung der Accipitres,Cathartidae, Striges und Anseriformes. Zoologie 157:285–332.

Baumel JJ. 1993. Systema cardiovasculare. In: Baumel JJ, edi-tor. Handbook of Avian Anatomy: Nomina Anatomica Avium,2nd ed. Cambridge, MA: Nuttall Ornithological Society. pp407–475.

Baumel JJ, Witmer LM. 1993. Osteologia. In: Baumel JJ, editor.Handbook of Avian Anatomy: Nomina Anatomica Avium, 2nded. Cambridge, MA: Nuttall Ornithological Society. pp 45–132.

Benjamin M, McGonagle D. 2001. The anatomical basis for dis-ease localization in seronegative spondyloarthropathy atentheses and related sites. J Anat 199:503–526.

Benton MJ and Clark JM. 1988 Archosaur phylogeny and therelationships of the Crocodylia. In: Benton J, editor. The Phy-logeny and Classification of the Tetrapods, Vol. 1: Amphib-ians, Reptiles, Birds. The Systematics Association Special,Vol. 35A. Oxford: Clarendon Press. pp 295–338.

Bhattacharyya BN. 1980. The morphology of the jaw andtongue musculature of the common pigeon. Columba livia, inrelation to its feeding habit. Proc Zool Soc (Calcutta) 31:95–126.

Bhattacharyya BN. 1989. Functional morphology of the jawmuscles of two species of imperial pigeons. Duculaenea nico-barica and Ducula badia insignis. Gegenbauers Morpholo-gisches Jahrbuch 135:573–618.

Brochu CA. 1999. Phylogenetics, taxonomy, and historical bio-geography of Alligatoroidea. J Vert Pal 19(Supplement to2):9–100.

Bubien-Waluszewska A. 1981. The cranial nerves. In: King AS,McClelland J, editors. Form and Function of Birds, Vol. 2.New York: Academic Press. pp 385–438.

Buhler P. 1981. Functional anatomy of the avian jaw apparatus.In: King AS, McClelland J, editors. Form and Function inBirds, Vol. 2. New York: Academic Press. pp 439–468.

Burton PJK. 1974. Feeding and the Feeding Apparatus inWaders: A Study of Anatomy and Adaptations in the Chara-drii. London, United Kingdom: British Museum (Natural His-tory).

Busbey AB, III. 1989. Form and function of the feeding appara-tus of Alligator mississippiensis. J Morphol 202:99–127.

Cleuren J, De Vree F. 1992. Kinematics of the jaw and hyolin-gual apparatus during feeding in Caiman crocodilus. J Mor-phol 212:99–127.

Cleuren J, Aerts P De Vree F. 1995. Bite and joint force analy-sis in Caiman crocodilus. Belg J Zool 125:79–94.

Covell DA Jr, Noden DM. 1989. Development of the embryonicavian trigeminal sensory-motor complex: An immunocyto-chemical and HRP analysis. J Comp Neurol 286:488–503.

Cracraft J. 1986. The origin and early diversification of birds.Paleobiology 12:383–399.

Cracraft J, Clarke JA. 2001. The basal clades of modern birds.In: Gauthier J, Gall LF, editors. New Perspectives on the Ori-gin and Early Evolution of Birds: Proceedings of the Interna-tional Symposium in Honor of John H. Ostrom. New Haven:Yale University Press. pp 143–156.

de Beer GR. 1937. The Development of the Vertebrate Skull.Oxford: Clarendon Press. 552 pp.

Dullemeijer P. 1951. The correlation between muscle systemand skull structure in Phalacrocorax carbo sinensis (Shawand Nodder): I–III. Proceedings of the Koninklijke Neder-landse Akademie van Wetenschappen, Series C 54:247–259,400–404, 533–536.

Dzerzhinsky RY, Yudin KA. 1982. On homology of the jaw inthe tuatara and the birds. Ornithilogical Studies in the USSR[Zool. Inst., USSR Acad. Sci., Moscow] 2:408–436.

Edgeworth FH. 1935. The cranial muscles of vertebrates. Cam-bridge: Cambridge University Press. 493 pp.



Elzanowski A. 1987. Cranial and eyelid muscles and ligamentsof the tinamous (Aves: Tinamiformes). Zoologische Jahrbuchfur Anatomie 116:63–118.

Endo H, Aoki R, Taru H, Kimura J, Sasaki M, Yamamoto M,Arishima K, Hayashi, Y. 2002. Comparative functional mor-phology of the masticatory apparatus in the long-snoutedcrocodiles. Anat Histol Embryol 31:206–213.

Fisher HI, Goodman DC. 1955. The myology of the WhoopingCrane, Grus americana. Urbana, Illinois: University of Illi-nois Press.

Gans C, De Vree F. 1986. Functional bases of fiber length andangulation in muscle. J Morphol 192:63–85.

Gauthier J. 1986. Saurischian monophyly and the originof birds. In: Padian K, editor. The Origin of Birds and theEvolution of Flight. San Francisco: California Academy ofSciences. pp 1–55.

Goodman DC, Fisher HI. 1962. Functional anatomy of the feed-ing apparatus in waterfowl (Aves: Anatidae). Carbondale, Illi-nois: University of Southern Illinois Press. 193 pp.

Gussekloo WS Bout RG. 2005. Cranial kinesis in palaeognathusbirds. J Exp Biol 208:3409–3419.

Haas G. 1955. The jaw musculature in Protoceratops and inother ceratopsians. Am Mus Novit 1729:1–24.

Haas G. 1969. On the jaw musculature of Ankylosaurus. AmMus Novit 2399:1–4.

Haas G. 1973. Muscles of the jaws and associated structures inthe Rynchocephalia and Squamata. In: Gans C, Parsons TS,editors. Biology of the Reptilia, Vol. 4: Morphology D. NewYork: Academic Press. pp 285–490.

Haas A. 2001. Mandibular arch musculature of anuran tad-poles, with comments on homologies of amphibian jawmuscles. J Morphol 247:1–33.

Hall BK. 2003. Descent with modification: The unity underlyinghomology and homoplasy as seen through an analysis of de-velopment and evolution. Biol Rev 78:409–433.

Herring SW. 1993. Epigenetic and functional influences on skullgrowth. In: Hanken J, Hall BK, editors. The skull, Vol. 1: De-velopment. Chicago: University of Chicago Press. pp 153–206.

Hofer H. 1950. Zur Morphologie der Kiefermuskulatur derVogel. Zool Jahrb Abt Anat Ontog 70:427–556.

Holliday CM. 2006. Evolution and function of the jaw muscula-ture and adductor chamber of archosaurs (crocodilians, dino-saurs, and birds), PhD Dissertation, Ohio University, Athens,Ohio. 325 pp.

Holliday CM, Ridgely RC, Balanoff AM, Witmer LM. 2006. Ce-phalic vascular anatomy in flamingos (Phoenicopterus ruber)based on novel vascular injection and CT imaging analyses.Anat Rec 288A:1031–1041.

Hurov J. 1988. Growth allometry of craniomandibular muscles,tendons, and bones in the laboratory rat (Rattus norvegicus):Relationships to oromotor maturation and biomechanics offeeding. Am J Anat 182:381–394.

Hutchinson JR. 2001a. The evolution of femoral osteology andsoft tissues on the line to extant birds. Zool J Linn Soc131:169–197.

Hutchinson JR. 2001b. The evolution of pelvic osteology andsoft tissues on the line to extant birds. Zool J Linn Soc131:123–168.

Iordansky NN. 1964. The jaw muscles of the crocodiles andsome relating structures of the crocodilian skull. Anat Anz115:256–280.

Iordansky NN. 1973. The skull of the Crocodilia. In: Gans C,Parsons TS, editors. Biology of the Reptilia, Vol. 4: Morphol-ogy D. New York: Academic Press. pp 201–262.

Iordansky NN. 2000. Jaw muscle of the crocodiles: structure,synonymy, and some implications on homology and functions.Russ J Herpetol 7:41–50.

Kesteven HL. 1942–1945. The evolution of the skull and the ce-phalic muscles. Australian Mus Mem 8:1–361.

Kiliardis S. 1996. Effects of masticatory muscle function onbone mass in the mandible of the growing rat. Acta Anat155:200–205.

Kuratani S, Tanaka S. 1990. Peripheral development of aviantrigeminal nerves. Am J Anat 187:65–80.

Lakjer T. 1926. Die Trigeminus-Versorgte Kaumuskulature derSauropsiden. Auf kosten der Carlsbergstiftung, Kopenhagen.154 pp.

Langston W Jr. 1973. The crocodilian skull in historical per-spective. In Gans C, Parsons TS, editors. Biology of the Repti-lia, Vol. 4. New York: Academic Press. pp 263–285.

Lubosch W. 1933. Untersuchungen uber die Visceralmuskulaturder Sauropsiden. Gegenbaurs Morph Jb 72:584–666.

Luther A. 1914. Uber die vom N. trigeminus versorgte Musku-latur der Amphibien, mit einem vergleichenden Ausblick uberden Adductor mandibulae der Gnathostomen, und einem Bei-trag zum Verstandnis der Organisation der Anuranlarven.Acta Societatis Scientiae Fennicae 44:1–151.

Mayr G, Clarke J. 2003. The deep divergences of neornithinebirds: A phylogenetic analysis of morphological characters.Cladistics 19:527–553.

McClearn D, Noden DM. 1988. Ontogeny of architectural com-plexity in embryonic quail visceral arch muscles. Am J Anat183:277–293.

McDowell SB. 1986. The architecture of the corner of the mouthof colubroid snakes. J Herpetol 20:353–407.

Midtgard U. 1984. Blood vessels and the occurrence of arterio-venous anastomoses in cephalic heat loss areas of mallards.Anas platyrhynchos, Aves. Zoomorphology 104:323–335.

Moss ML. 1968. The primacy of functional matrices in orofacialgrowth. Dent Pract 19:65–73.

Oelrich TM. 1956. The anatomy of the head of Ctenosaura pecti-nata (Iguanidae). Misc Publ Mus Zool Univ Mich 94. 122 pp.

Ostrom JH. 1964. A functional analysis of jaw mechanics in thedinosaur Triceratops. Postilla 88:1–35.

Patterson C. 1982. Morphological characters and homology. In:Joysey KA, Friday AE, editors. Problems of PhylogeneticReconstruction. New York: Academic Press. pp 21–74.

Poglayen-Neuwall I. 1953a. Untersuchungen der Kiefermusku-latur und deren Innervation bei Schildkroten. Acta Zoologica34:241–292.

Poglayen-Neuwall I. 1953b. Untersuchungen der Kiefermusku-latur und deren Innervation an Krokodilen. Anat Anz 99:257–276.

Presley R. 1993. Preconception of adult structural pattern inthe analysis of the developing skull. In: Hanken J, Hall BK,editors. The Skull, Vol. 1: Development. Chicago: Universityof Chicago Press. pp 347–377.

Rayfield E, Norman D, Horner C, Horner J, Smith P, ThomasonJ, Upchurch P. 2001. Cranial design and function in a largetheropod dinosaur. Nature 409:1033–1037.

Rieppel O. 1987. The development of the trigeminal jaw adduc-tor musculature and associated skull elements in the lizardPodarcis sicula. J Zool 212:131–150.

Rieppel O. 1988. The development of the trigeminal jaw adduc-tor musculature in the grass snake Natrix natrix. J Zool216:74770.

Rieppel O. 1990. The structure and development of the jawadductor musculature in the turtle Chelydra serpentina. ZoolJ Linn Soc 98:27–62.

Rieppel O, Kearney M. 2002. Similarity. Biol J Linn Soc 75:59–82.Ruberte JA, Carretero M, Navarro R, Marcucio S, Noden D.

2003. Morphogenesis of blood vessels in the head muscles ofavian embryo: Spatial, temporal, and VEGF expression analy-ses. Dev Dyn 2276:470–483.

Ross CF, Metzger KA. 2004. Bone strain gradients and optimi-zation in tetrapod skulls. Ann Anat 186:387–396.

Save-Soderbergh G. 1945. Notes on the trigeminal musculaturein non-mammalian tetrapods. Nova Acta Regiae SocietatisScientiarum Upsaliensis Series IV 13:1–50.

Schumacher GH. 1973. The head muscles and hyolaryngealskeleton of turtles and crocodilians. In: Gans C, Parsons TS,editors. Biology of the Reptilia, Vol. 4: Morphology D. NewYork: Academic Press. pp 101–200.

Scott SA. 1992. The development of peripheral sensory innerva-tion patterns. In: Scott SA, editor. Sensory Neurons: Diver-



sity, Development, and Plasticity. Oxford: Oxford UniversityPress. pp 242–263.

Sedlmayr JC. 2002. Anatomy, evolution, and functional signifi-cance of cephalic vasculature in Archosauria, PhD Disserta-tion, Ohio University, Athens, Ohio. 398 pp.

Soliman MA. 1963. Die Kopfnerven der Schildkroten. Zeitschriftfut Wissenschaftliche Zoologie 168:216–312.

Song J, Boord RL. 1993. Motor components of the trigeminalnerve and organization of the mandibular arch muscles invertebrates: Phylogenetically conservative patterns and theirontogenetic basis. Acta Anat 148:139–149.

Starck D, Barnikol A. 1954. Beitrage zur Morphologie der Tri-geminusmuskulatur der Vogel (besonders der Accipitres, Ca-thartidae, Striges und Anseres. Gegenbaurs MorphologischesJahrbuch 94:1–64.

Tarsitano SF, Oelofsen B, Frey E, Riess J. 2001. The origin oftemporal fenestrae. S Afr J Sci 97:334–336.

Tokita M. 2004. Morphogenesis of parrot jaw muscles: under-standing the development of an evolutionary novelty. J Mor-phol 259:69–81.

Van Drongelen W, Dullemeijer, P. 1982. The feeding apparatusof Caiman crocodilus: A functional morphological study. AnatAnz 151:337–366.

Van Gennip EMSJ. 1986. Skull mechanics in the pigeon.Columba livia, a three-dimensional kinematic model. J Mor-phol 213:197–224.

Vanden Berge JC, Zweers GA. 1993. Myologia. In: Baumel JJ,editor. Handbook of Avian Anatomy: Nomina AnatomicaAvium, 2nd ed. Cambridge, Massachusetts: Nuttall Ornitho-logical Society. pp 189–250.

Vogel KS. 1992. Origins and early development of vertebratecranial sensory neurons. In: Scott SA, editor. Sensory Neu-rons: Diversity, Development, and Plasticity. Oxford: OxfordUniversity Press. pp 171–193.

Webb M. 1957. The ontogeny of the cranial bones, cranial pe-ripheral and cranial parasympathetic nerves, together with astudy of the visceral muscles of Struthio. Acta Zool 38:81–203.

Weber E. 1996. Das skelet-muskel-system des keiferapparatesvon Aepypodius arfakianus (Salvadori, 1877) (Aves, Megapo-diidae). Courier Forschungsinstitut Senckenberg 189:1–132.

Witmer LM. 1995. Homology of facial structures in extant arch-osaurs (birds and crocodilians), with special reference to para-nasal pneumaticity and nasal conchae. J Morphol 225:263–327.

Wu X-C. 2003. Functional morphology of the temporal region inthe Rynchocephalia. Can J Earth Sci 40:589–607.

Zusi, R. L. 1993. Patterns of diversity in the avian skull. In:Hanken J, Hall BK, editors. The Skull, Vol. 2. Chicago: Uni-versity of Chicago Press. pp 391–437.

Zusi RL, Bentz GD. 1984. Myology of the Purple-throated carib(Eulampis jugalaris) and other hummingbirds (Aves: Trochili-dae). Smithsonian Contributions Zool 385:1–70.

Zusi RL, Livezey BC. 2000. Homology and phylogenetic implica-tions of some enigmatic cranial features in galliform andAnseriform birds. Ann Carnegie Mus 69:157–193.

Zusi RL, Storer RW. 1969. Osteology and myology of the headand neck of the pied-billed grebes (Podilymbus). Misc PublMus Zool Univ Mich 139:1–49.

Zweers GA. 1974. Structure, movement, and myography of thefeeding apparatus of the mallard (Anas platyrynchos L.): Astudy of functional anatomy. Neth J Zool 24:323–467.

Zweers GA, Bout RG, Heidweiller J. 1994. Motor organizationof the avian head-neck system. In: Davies MNO, Green PR,editors. Perception and Motor Control in Birds. Berlin:Springer. pp 201–222.



archosaur adductor chamber evolution: integration of

Documents