8. Discussion of Character Analysis of the Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids­Variation and Ontogeny

MICHAEL K. BRETT-SURMAN AND

JONATHAN R. WAGNER

Abstract

Over 50 morphological characters commonly have been used for defining and diagnosing the hadrosaurids, and these are reevalu­ated in terms of data from ontogeny, paleopathology, and postcra­nial studies. Features once used to define supraspecific taxa are also reevaluated in the light of population and ontogenetic variation.

The hadrosaur postcranium becomes more robust with age, ex­pressed as more rugose muscular attachments and greater deposi­tion of bone on articular surfaces. Most ridges and bumps on the long bones are smooth, short, and do not project far from the shafts in small specimens. In the largest animals, ridges may be­come exaggerated, and trochanters and muscular insertion scars become large, pitted, and rugose, with a tendency to look patho­logical.

Four size classes in hadrosaurids are identified on the basis of gross morphology that may reflect individual age. The first class (hatchling) represents hadrosaurids with little or no expansion of the muzzle, small and smooth articular surfaces, five or six sacrals, and less than 20 tooth rows. The second class (juvenile) has a no­

135

8. Discussion of Character Analysis of the Appendicular Anatomy in Campanian and Maastricht!an North American Hadrosaurids— Variation and Ontogeny

MlCHAEL K. BRETT-SURMAN AND JONATHAN R. WAGNER

Abstract

Over 50 morphological characters commonly have been used for defining and diagnosing the hadrosaurids, and these are devalu- ated in terms of data from ontogeny, paleopathology, and postcra- nial studies. Features once used to define supraspecific taxa are also reevaluated in the light of population and ontogenetic variation.

The hadrosaur postcranium becomes more robust with age, ex- pressed as more rugose muscular attachments and greater deposi- tion of bone on articular surfaces. Most ridges and bumps on the long bones are smooth, short, and do not project far from the shafts in small specimens. In the largest animals, ridges may be- come exaggerated, and trochanters and muscular insertion scars become large, pitted, and rugose, with a tendency to look patho- logical.

Four size classes in hadrosaurids are identified on the basis of gross morphology that may reflect individual age. The first class (hatchling) represents hadrosaurids with little or no expansion of the muzzle, small and smooth articular surfaces, five or six sacrals, and less than 20 tooth rows. The second class (juvenile) has a nu

I"

ticeable mu zzle, 20 to 40 tooth rows , articulations with some ru­gosities on the long bon es, six or seven sacrals, and in t he Lam ­beosaurinae, th e appearance of inc ipient cranial crests and an is­chial foot. (Loss of the ischial "foot" is a derived co ndition .) The third size class (adult) is th e size ra nge in which all of the features useful for diagn osis of ha drosaurs are full y developed. The fourth class (old age ) is characteri zed by 10 or more sacrals, all bony pro­jections grea tly increased in size an d rugosit y, fusion of the dista l femora l condyles anteriorly, pitting of th e articular ends of the long bones, frilling or pitting of the ends of th e ungu als, and clubbing of the ap ices of the neura l spi nes. Some of these fea tures may appear pathologically in younger speci mens .

Introduction

In the summer of 1973, whi le visi ting the Field Museu m of N atural History in Chicago, one of us (M . B.-S.) found th e postcran ial re­mains of a hadrosaur from th e Rio Chico area of Arg entina (Brett­Surman 1979). Brett-Surm an set out to compare th e specime n to known had ro saur s. Despi te the ub iquitous nature of hadrosau rian postcrania in relevant pa leonto log ica l collections, littl e attempt had been mad e to use them in hadro saur systematics becau se of the er ­ron eous belief that they were of littl e systematic val ue. Brett ­Sur ma n (1975, 1989) summarized th e results of his invest igat ion s an d included a deta iled d iscussion of cha rac ters useful in dis tin­guishing groups of ha drosaurs fro m postcran ia a lone. Davies (1983) and Wagner (2001) expanded Brett-Surman 's work on hadrosaurs from Trans-Pecos Texas, w ith Wag ner incl udi ng taxa outside of H adrosauridae.

We analyze the taxon omic ch aracters of hadrosaurids on the basis of their appendicular elements and demonstrate fea tu res that represent ontogenetic stag es in taxa and valid taxonomic syn apo­morphies for other genera. Where possible, only articulated ske le­tons were used in this study to ensur e the accuracy of ap pendicul ar ra tios and other associations . Indi vidual elements were also drawn from ph otogra phs using a camera lucida. We review the ea rlier work on had ro saurid pos tcra nia by Brett-Surman (1975, 1989) and Wagner (200 1). For a mo re complete list of character sta tes, see Sere no (1999), Norma n (2002), and H ead (200 1); an d for dis­cussions of resultant ph ylogenies, see Norman (2004), Horne r et al. (2004), H ead (2001), an d Godefroi t et al. (2000) . A co mp lete list­ing of all scientific papers on had rosaurids ca n be found at http: //www.nmnh.si.edu /paleo/collect ion s.html. For simplicity, the informal term "igua no do ntoids " refers to ta xa mor e closely related to Iguanodon than to H adrosauridae.

Institutional Abbreviations. AMNH: American M use um of Natural Hi story, N ew York; CM : Carnegie Museu m of N atural History, Pitt sb ur gh , Pennsy lvania; DMNH: Denver M useu m of Nature & Science (formerly Denver Museum of Natural H isto ry), Denver, Colorado; FMNH: Field Museum of N atural H istor y,

136 • Michael K. Brett-Surman and Jonathan R. Wagner

Chicago, Illinois: Palaeoan th ropolog Museum of N ature Ottawa; ROM : ? Memorial M use" versity of Ca lifora ural History ( fo~~

ton, D. C. ; YP:\ ~:

Unive rsi ty, N ew i

Analysis of

Pectoral Girdle

T he hadros....­as a so urc e of s:.-s:,t pretati on of PC\."": ~­

cause these elem against th e ribcz g tion, and (especi'=':1 of elements mav ."­

Scapula. In ~

8.1A) show rnor; pr oximal pa rt li"".: deltoid ridge , w;- -­postero vent ra l <£-. is the deltoid fossa nar ro west part, c: th e scapula consist rectan gul ar in sf:.::. parallel to th e ve rte

senting a mo rp the larnbeosaurine

In the ha dros longer (an reroposre pared to the c (N MC 8893 ) is ~­longest of any ~.

dorsovent rallyI is width ratio gr ez:L sca pula (Fig. 8.i . :: natural posirioc had rosaur ines. ."~

tocristatus (see O r

and ro bust, with . genus. The only ~

. . one specimen 0 , ~

resembles the ha ~

The scapula .., even wi thi n ge,".

Appen ~

ticeable muzzle, 20 to 40 tooth rows, articulations with some ru gosmes on the long hones, six or seven sacrals. and in the Lam heosaunnae, the appearance of incipient cranial crests and an is cbial foot, (loss of the ischial "foot" is a derived condition.) The third size class iadult) is the size range in which all of the features useful tor diagnosis of hadrosaurs are fully developed. The fourth class (old agcl is characterized by 10 or more sacrals, all bony pro jections greatly increased in size and rugosity, fusion of the distal femoral condyles anteriorly, pitting ol the articular ends of the long hones, frilling or pitting of the ends of the unguals, and clubbing of the apices of the neural spines. Some of these features may appear pathologically in younger specimens.

Introduction

In the summer ol 1973, while visiting the field Museum of Natural History in Chicago, one of us (M. B.-S.) found the postcrama I re- mains of a hadrosaur from the Rio Chico ana of Argentina (Brett- Surman 1979;. Brett Surman set out to compare the specimen ro known hadrosaurs. Despite the ubiquitous nature of hadrosaurian postcrama in relevant paleontological collections, little attempt had been made to use them in hadrosaur systematics because of the er- roneous belief that they were or little systematic value. Bret I Surman [1975, 1989) summarized the results of his investigations and included a detailed discussion of characters useful in distin- guishing groups of hadrosaurs from postcrama alone. Da vies (1983) and Wagner (2001) expanded Brett-Surm.urs work on hadrosaurs from Trans-l'ecos Texas, with Wagner including taxa outside of Hadros.iundac.

We analyze the taxonomic characters of hadrosaunds on the basis ot their appcndicular elements and demonstrate tearures that represent ontogenetic stages m taxa and valid taxonomic synapo- morphics for other genera. Where possible, only articulated skele- tons were used in this study to ensure the accuracy of appcndicular ratios and other associations. Individual elements were also drawn from photographs using a camera lucida. We review the earlier work on hadrosaurid postcrania by Brett Surman (1975. 19S9) and Wagner (20011, for a more complete list of character states, KeScreno(l999), Norman (2002), and Head (2001): and fordia- cussions of resultant phylogenics, see Norman (2004), I lorncr et al. ,2H04), Head (2001). and Godefroii et al. (20(Xi). A complete list- ing of all scientific papers on hadrosaurids can be found at http://www.nmnh.si.edu/paleo/collcctions.html. lor simplicity, the informal term "iguanodonroids" refers to taxa more closely related to Iginwodon than to Hadro.sjurid.ic.

Institutional Abbreviations. AMNH: American Museum of Natural History, New York; CM: Carnegie Museum ol Natural History. Pittsburgh. Pennsylvania: DMNII: Denver Museum of Nature & Science (formerly Denver Museum of Natural I listou i, Denver. Colorado; IMNI I: Field Museum ol Natural History,

136 • Michael K Brett-Surman and |onathan R. Wagnei

- \'..ith some ru­nd in the Lam­

crests and an is-condi tion .) The of the features

_ d . The fourth 5 , a ll bony pro­

. n of the distal ends of the long a nd clubbing of

ures ma y appear

zseum of Natural ,:. ostcranial re­

o rgent ina (Brett­: e specimen to i ha dro saur ia n

-:-le attempt had · .:ause of the er­

value. Brett­r ; investigations useful in distin­

, alone. Davies 1 . an 's work on er including taxa

lrosaurids on the - 2-e fea tures that onorruc syna po­-rriculared skele­of appendicular "ere also drawn i ew the earlier

(1975, 1989) .: aracter sta tes,

· 1); and for dis ­• , H orn er et al.

_ .\ co mplete list­- be found at r sim plicity, the

· :-e closely related

- -n Museum of seu rn of Natural

ever Museum of _.arur al History), _-arural History,

Chicago, Illinois; [vPP Institute of Vertebrate Palaeontology and Palaeoanthropology, Academic Sinica, Beijing; NMC: Canadian Museum of Nature (formerly the National Museum of Canada ), Ottawa; ROM: Royal Ontario Museum, Toronto; TMM: Texas Memorial Museum, Austin; UCMP: Museum of Paleontology, Uni­versity of California, Berkeley; USNM: National Museum of Nat­ural History (formerly Un ited States N ational Museum), Washing­ton, D.C.; YPM: Peabody Museum of Natural History, Yale University, New H aven, Connecticut.

Analysis of Hadrosaurid Appendicular Elements

Pectoral Girdle

The hadrosaur p ectoral girdle has only rarely been considered as a source of systematic information (Brett-Surman 1989). Inter­pretation of pectoral girdle morphology is difficult, esp ecially be­cause these elem ents, although essentially flat, ar e curved to fit against the ribcage. Differing perspective in illustrations, distor­tion, and (especially in the scapula ) damage to the thinned margins of elements may all exacer bate interpretation.

Scapula. In hadrosaurids, two regions of the scapula (Fig. 8.1A) show morphological variations across a continuum. The proximal part includes the suture for the coracoid, as well as the deltoid ridge, which arises on the dorsal margin and an gles in a posteroventral direction to the ventral border. Anterior to this ridge is the deltoid fos sa, the posterior border of which terminates at the narrowest part, or scap ular neck. The second or posterior part of the scapula consists of the blade, which is laterally flattened and rectangular in shape. In correct anatomical position, th e scapula is parallel to the vertebral column. Two morphs are recognized repre­senting a morphological continuum, the hadrosaurine scapula and the lambeosaurine sca pula (Fig. 8.1 ).

In the hadrosaurine scapula (Fig. 8.1), the blad e is rela tively longer (anteroposteriorly) and not as wide (dorsoventrally) as com­pared to the condition in lambeosaurines. Brachylopbosaurus (N M C 8893 ) is the mo st extreme in th at the blade is relatively the lon gest of an y known hadrosaurid, while the width (measur ed dorsoventrally) is relatively the smallest . This results in a length/ width ratio gre ater than in any other genus. In the lambeosaurine scapula (Fig. 8.1), the blade is relatively shorter (anterop osteriorly in natural position) and wider (dorsoventrally) than in most hadrosaurines. An extreme of this is seen in Parasaurolophus cyr­tocristatus (see especiall y Ostrom 196 3), where the blade is short and robust, with a length/Width ratio smaller than in any other genus. The onl y exception to this morphological variety is seen in one specimen of Lambeosaurus (ROM 1218), which more closely resembles the hadrosaurine variety.

The scapula displays considerable variation within clades and even within genera . There is much overl apping of shap es among

Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids • 137

Chicago, Illinois; IVPP Institute of Vertebrate Palaeontology and Palacoanthropology, Academic Sinica, Beijing; N.VIC: Canadian Museum of Nature (formerly the National Museum of Canada). Ottawa; ROM: Royal Ontario Museum, Toronto; TMM: Texas Memorial Museum, Austin; UCMP: Museum ot Paleontology, Uni- versity of California, Berkeley; liSN'M: National Museum of Nat uraI History (formerly United States National Museum), Washing- ton, DC; YPM: Pea body Museum of Natural History, Yale University, New Haven, Connecticut.

Analysis of Hadrosaurid Appendicular Elements

Pectoral Girdle

The hadrosaur pectoral girdle has only rarely been considered as a source of systematic information (Brett-Surman 1989). Inter- pretation of pectoral girdle morphology is difficult, especially be- cause these elements, although essentially fiat, arc curved to lit against the ribcagc. Differing perspective in illustrations, distor don, and (especially in the scapula) damage to the thinned margins of elements may all exacerbate interpretation.

Scapula. In hadrosaunds, two regions of the scapula (Pig. 8.1 A) show morphological variations across a continuum. The proximal part includes the suture for the coracoid, as well as the deltoid ridge, which arises on the dorsal margin and angles in a postcroventral direction to the ventral border, Anterior to this ridge is the deltoid fossa, the posterior border of which terminates at the narrowest part, or scapular neck. Tin. second or posterior part of the scapula consists of the blade, which is laterally flattened and rectangular in shape. In correct anatomical position, the scapula is parallel to the vertebral column. Two morphs are recognized repre tenting a morphological continuum, the hadrosaunne scapula and the lambeos.uirinc scapula (Fig. K.I).

In the hadrosaurine scapula (Fig. S.I), the blade is relatively longer (anteropostenorlyl and not as wide idorsoventrally) as com- pared to the condition in larnbeosaurines. Brachylophosaurns (NMC 8893) is the most extreme in that the blade is relatively the longest of any known hadrosaurid, while the width (measured dorsovcntrally'i is relatively the smallest. This results in a length/ width ratio greater than in any other genus. In the lambeosaurinc scapula (Fig. 8.1), the blade is relatively shorter (anrcropostenorly in natural position) and wider (dorsovenrrally) than in most hadrosaurines. An extreme of this is seen in ParOSOlirolOphus cyr- tocristJlus (see especially Ostrom 1963), where the blade is short and robust, with a length/width ratio smaller than in any other genus. The only exception to this morphological variety is seen in one specimen of LambeosatwltS i.ROM 1218), which more closely resembles the hadrosaurine variety.

The scapula displays considerable variation within clades and even within genera. There is much overlapping of shapes among

Appendicular Anatomy in Campanian and M.i.istnchtian North American Fladrosaiiruls • 13?

OLF OLA ~............_ S CBL

_.~ ..<;.: ~~ POLCRF

~~

, j CRR

4- ' - ' ;:£'"

.~~ i ------------- ~.. A ~

HNO

CRF

I. , ~:( .':: : ':.~.;

B Dc

Figure 8.1. Com parison of the left genera; therefore, it is im possible to diagnose scapulae to the scapulae and coracoids in lateral generic level or to use length/w idth ratios to separa te taxa. As view: (A) Kritosaurus AMNH

no ted above, Brachylophosaurus stands out from th e other gene ra 5465 (length = 792 111m); (B)

with its uniq ue length/wi dth ra tio of th e bla de, bu t a sa mple size o f Camptosaurus USNM 4282 (length =482 111m). Left sterna ls less than five does not just ify the delineation of its own morpho­in ventr al view : (C) Tsinta osaurus type. It is on ly possible to state tha t a scapula is eithe r long and th in IVPP V-125 (length = 420 mm); (most likely a hadrosaurine) or it is sho rter and wider (most lik ely (D) Shan tungo sau rus (length =

a lambeosaurine). Some blad es expand a bru ptly just di stal to th e 705 111m); redrawn from Hu

neck a nd have parall el dorsal a nd ventra l borders, or co nvex dorsal (l973). See Appendix 8.1 for abbr eviations. bo rder s (e.g., Prosaurolophus, ROM 787) .

Durin g o ntogeny, th e delt o id ridge becomes elon gate and more ro bus t until it fina lly reach es the lower scapular border in adults . In juveniles, th e a rea of th e gleno id is larger th an th e area fo r th e su ­ture w ith the coracoi d (e.g ., Bactrosaurus AMNH 6577), w hi le in adults (e.g. , Anatotitan copei AMNH 5730) th e area of the glenoid and the area of the coracoid suture are about eq ual in extent. In ju­veni les th e suture for th e coracoid is slightly ro ughened, but in ad ults it is ex treme ly ru gose and massive (e.g., USNM 2413, a coracoid show ing the scapular sutural area ). In ma ny juveniles, the dorsal and vent ra l borders of the blade diverge slightly posterior (as in the igu anodontids), w hereas in ad ults the dorsal and ventra l borders are pa ra llel so tha t the blade a ppears rect an gul ar in later al aspect .

The scapular blade of ad ults is p ro porti onally longer than that of juveniles. T he major area of ossi fication is at th e distal en d of the blade, an area for the ossifica tion of the suprasca pula r cartilage, possibly to increase the area for m uscle a ttachme nt in support ing a larger mass. Hadrosaurids of the sa me age gro up mainta in th e sa me scapular prop ortions relative th e body th ro ugh out life. It would appear that this is accomplished so lely by increasing the length and wi dth of th e sca pular blade relative to th e pr oximal por­

138 • Mic hael K. Brett -Sur man and Jonath an R. Wagner

tio n of the scapc

­

­exceptional well .; lae described bv Y casuarius. It is q sca pulae referred trated sca pu lae c ;

1942 ) and E dn :IJ

Parks 1935) sho th e scapula, an.;

(Pr ieto-Marq uez .: It is impossi

th a t may be rere.. quate sample size.

Coracoid. Iz ally by a low, ar c:

of the ventral h of the bone. "acromion proces 2000; Norm an ;,3 Ia teroven tra lly, -;-:: nounced dorsal. ­gin, but whic h 0 "

opment of this coracoid a sligh: ~

th e lateral cone ­In hadrosauri

a large hooklike r '.

that for ms the a;:; ­

tur e with the docs on the dorsal sur:' extends poste riorlj the base of the '::0:

sha ped depressioz ro unded knobs 2D

is three to four --' ­Midway

co id and anterio r The posit ion of ± able (Davies 19 S.3 it is generally f~

ulation in ha dros coracoid in ad u.: in some juve

Appen

(Parks 1919, 1';L pad tha t p res u eventually closed co raco id foramen ang le, op ening : is so me times presopens into th e --­

C"=F

PEL

MID

POL

0

Figure S.I. Comparison of the left s, .:/•:<!.ir .:•:.! . .naeoidf ill Literal tiar: IAI Kriiwauras AMNII

Omptosaums USKM -t2fi2 ilenglli : 4S2 nun). Left slenials in ventral view: IC) Tsiwaos.iurus IVIT VI25 (length = 420 mm); iDi Shanningonumt {length • 70S mm): redrawn from Ha

abbreviations.

genera; therefore, n is impossible ro diagnose scapulae to the generic level or to use length/width ratios to separate taxa. As noted above, Brachylophosaurus stands out from the other genera with its unique length/width ratio of the blade, but a sample size of less than five does not justify the delineation of its own morpho- type. It is only possible to stare that a scapula is either long and thin (most likely a hadrosaurinc) or it is shorter and wider (most likely a lambeosaurine). Some blades expand abruptly just distal to the neck and have parallel dorsal and ventral borders, or convex dorsal borders (e.g., ProaMfm/on/w. ROM 78"*).

During ontogeny, the deltoid ridge becomes elongate and more robust until it finally reaches the lower scapular border in adults. In juveniles, the area of the glenoid is larger than the area for the su- ture with the coracoid (e.g., Bactrosaurits AMNH 6577), while in adults (e.g., Amtotitan cOpei AMNH 5730) the area of die glenoid and the area of the coracoid suture are about equal in extent. In |u- venilcs the suture for the coracoid is slightly roughened, but in adults it i» extremely rugose and massive fe.g^ I'SNM 2413, a coracoid showing the scapular sutural area). In many juveniles, the dorsal and ventral borders of the blade diverge slightly posterior (as in the iguanodontids), whereas in adults the dorsal and ventral borders are parallel so that the blade appears rectangular in lateral aspect.

The scapular blade of adults is proportionally longer than that of juveniles. The major area of ossification is at the distal end of the blade, an area for the ossification of the suprascapular cartilage, possibly to increase the area for muscle attachment in supporting a larger mass. Hadrosaunds ot the same age group maintain the same scapular proportions relative the body throughout life. It would appear that this is accomplished solely by increasing the length and width of the scapular blade relative to the proximal por-

138 • Michael K. Biett-Surman and Jonathan R. Wagner

r ,/

PDLb.·

HND

o

scapulae to the ~. arate taxa. As - e other genera a sample size of

its own morpho­er long and thin

ri er (most likely . st distal to the o r convex dorsal

gate and more rrc er in adults. In ~ area for the su­o 6577 ), while in

of the glenoi d in ext ent . In ju­ghened, but in

CSN M 24 13 , a any juveniles, the . igh tly posterior

rsal and ventral - gular in lateral

nger than that . ista l end of the ::2 ular cartilage, .: in supporting a

• maintain the ughout life. It increasing the proximal por ­

tion of the sca pula. Distal flaring of the scapula also occurs, and is exceptiona l well developed in Parasaurolophus, some of the scapu­lae described by Young (1958), and the paratype of Corythosaurus casuarlus. It is quite reduced in the typ e of Brachylophosaurus, an d scapulae referr ed to Shantung osaurus an d Tanius sinensis. Illus­trated scapulae of Corythosaurus (Brown 191 6; Lull and Wrig ht 1942) and Edmontosaurus (Gilmo re 1924a; Lull and Wright 1942; Parks 1935) show significant variability in the degree of flaring of the scapula, and this is reported for Bracbylopbosaurus as well (Prieto-Marquez 2000, and this volume) .

It is imposs ible at th is time to distinguish morphological types that may be referred to specific genera or clad es because of inade ­quate sample size.

Coracoid. In Iguanodon, the coracoid is crossed ventromedi­ally by a low, arc uate, dor sally convex ridge ex tending from the tip of the ventral hook to a lateral process on the cra nio do rsa l margin of the bone. This angular, slightly everted rugosity is the coracoid "a cromion process" ; its cra nial end is the biceps tubercle (Dilkes 20 00; Norman 1986). In hadrosaurids, the biceps tubercle is folded latero vent rally, producing an overh anging ridge that is most pro ­nounced dorsally where it is continuous with the cra niodorsa l mar­gin, but whi ch overhangs and scallo ps that margin ventrally. Devel­opment of this process gives the dorsocranial margin of the coracoid a slight concavity, which is accentua ted in lateral view by the lateral concavity of th is region in ar ticulated specimens.

In hadrosaurids, the coracoid (Fig. 8.1A) terminates anteri orly in a large hooklike process. Arising from this hook is the coracoid r idge that forms the anterior border of the coracoid and ends at the junc­ture with the dorsal borde r. Th e coracoidal ridge is deflected medially on the dorsal surface where the acromion ridge originates. Th is ridge extends posteriorly to the suture with the scapul a. The glenoid is at the base of the coracoid-scapular suture and has a hyperbolic cone­shaped depression at its center. The scapular sutu re is rugose with ro unded knobs and deep depre ssions. The main body of the coracoid is three to four times as thick as the coracoid ridge and hook.

Midway between the dorsal and ventral bord ers of the co ra ­coid and ante rior to the scapular suture is th e coracoid for amen. T he position of the coracoid foram en appea rs to be somewhat vari­able (Davies 1983; Brett-Surman 1989), but with a few excepti ons, it is generally farther th an its own diameter from the scapular artic­ulation in hadrosaurids. The foramen is entirely enclosed by the cora coid in adult hadrosaurids, but it lies partly open to the suture in some juveniles and in iguano dontids as a "coracoid notch " (Parks 191 9, 1920). Pro gressive ossification of the cartilaginous pad that presumably connected the coracoid and scapula ma y have eventually closed the notch during hadrosaurid ontogeny. The coracoid fora men (Fig. S. l A) passes through the bone at a slight angle, op ening toward the scapula on both sides . A sha llow trou gh is sometimes present th at approaches (Young 1958, fig. 21-1 ) or opens into the scapular suture (UCMP 94475).

Appe ndicular Anatomy in Campa nian and Maastrichtian North America n H adrosaurids • 139

non of the scapula. Distal flaring of the scapula also occurs, and is exceptional well developed in Parasaurolophus, tome of the scapu- lae described by Young (I9.5K), and the paratype of QiryfAoafwrMs casuarius. It is quite reduced in the type of Brachyhphosaurus, and scapulae referred to Shaulimgusiiiirus and T.inius sitwnsis. Illus- trated scapulae of Corytbosaurus (Brown 1916; Lull and Wright 1942) and Edmontosaams (Gilmore 1914a; Lull and Wright 1942; Parks 1935) show significant variability in the degree of flaring of the scapula, and this is reported for Brachylophosakrus as well (Prieto-Marquez 2000, and this volume).

Ir is impossible at this time to distinguish morphological types that may be referred to specific genera or clades because of inade- quate sample size.

Coracoid. In Igtianodon, the coracoid is crossed vemromedi- ally by a low, arcuate, dorsally convex ridge extending from the rip of the ventral hook to a lateral process on the craniodorsal margin of the bone. This angular, slightly everted rugosity is the coracoid "acrnmion process"; its cranial end is the biceps tubercle (Dilkes 2000; Norman 1986). In hadrosaurids, the biceps tubercle is folded larcrovcntrally, producing an overhanging ridge that is most pro- nounced dorsally where it is continuous with the craniodorsal mar- gin, but which overhangs and scallops that margin ventrally. Devel- opment of this process gives the dorsocranial margin of the coracoid a slight concavity, which is accentuated in lateral view by the lateral concavity of this region in articulated specimens.

In hadrosaurids, the coracoid (Fig. 8.1 A) terminates anteriorly in a large hooklikc proccas. Anting from this hook is the coracoid ridge that forms the anterior border of the coracoid and ends at the junc- ture with the dorsal border. The coracoidal ridge is deflected medially on the dorsal surface where the acromion ridge originates. This ridge extends posteriorly to the suture with the scapula. The glenoid is at the base of the coracoid scapular suture and has a hyperbolic cone- shaped depression at its center. The scapular suture is rugose with rounded knobs and deep depressions. The mam body of the coracoid is three to four times as thick as the coracoid ridge and hook.

Midway between the dorsal and ventral borders of the cora- coid and anterior to the scapular suture is the coracoid foramen. The position of the coracoid foramen appears to be somewhat van- able (Davics 1983; Brett-Surman 1989), but with a few exceptions, it is generally farther than its own diameter from the scapular artic- ulation in hadrosaurids. The foramen is entirely enclosed by the coracoid in adult hadrosaurids, but it lies partly open to the suture in some juveniles and in iguanodonrids as a "coracoid notch" (Parks 1919, 19201. Progressive ossification of the cartilaginous pad that presumably connected the coracoid and scapula may have- eventually closed the notch during hadrosaurid ontogeny. The coracoid foramen (Fig. 8.1 A) passes through the hone at a slight angle, opening toward the scapula on both sides. A shallow trough i* sometimes present thai approaches (Young I9S8, hg. 21-1) or open* into the scapular suture (UCMP 94475).

Apprndicul.ir Anatomy in Campaman and Maastrichtian North American Hadrosaurids • 139

The coracoid in hadrosaurids appears to be elongate relative to tha t of nonhad ro saurids as measured perpendi cular to the scap ular facet in lateral view. The coracoid hook, where the bone articulate s wi th the sternals, is longer and or iented sub parallel to th e margin of the co racoi da l glenoi d. The dorsocran ial region of the coracoid below the sca pular facet ap pears to be slightly longer in specimens that possess a plesiomo rphically cranially directed acromion process of the scapula.

T he coracoid does not display sufficient variation among the lineages to ma ke it diagn osti c at th e spec ies level. M ost coracoids ar e cru shed flat, which ma kes it d ifficult to ascerta in their or iginal shape. It appears, ho wever, that in hadrosaurines th e co raco id is longer (anreroposterio rly) and lower (dorsovent ra lly) th an in lam­beosaurines, but the sample size is too small to justify th is proposed dich otomy. It is possible th at thi s featur e ma y be end po ints on a morphological co nti nuum.

The only difference between the early forms, such as Bac­trosaurus (AM N H 6577) and Gilmo reosaurus (AMN H 658 1), and th e mo re derived late Cre taceo us forms, is that th e coracoidal hook is smaller and pointed more ventrally th an anteroventrally.

O ntoge nt ically, the co racoida l hook is sma ll and points ven­trally in juveniles. The coracoid forame n is entirely enclosed w ithin the coracoi d, unl ike some no nhadrosaurid ornithopods. Dorsally, the kn ob at the end of the acromion ridge is sma ll and ro unded and generally smooth. The glenoid is deep as in ad ults but the depres­sion is symmetrica lly sha pe d. The scapu lar suture is slightly ru gose. The main bod y of th e co racoid is slightly th icker th an that of the coracoida l hook. One gro wth feature of interest is th at with on­togen y, the co racoid a br up tly becomes thi cker (Iatero rned ially) and more ru gose, especially at the suture with th e sca pula .

Sternals. Stern al ar e the most difficult elements to study. They are rarely preserved or prepa red and are virt ua lly ign ored in the lit ­eratur e (see Parks 19 20 for the first complete description of a had rosaurid sternal). There is no literature on the ontogenetic or population varia tio n of thi s element .

In hadrosaur ids , th e ventral segments of th e sternal ribs ap par­enrly coa lesced into a single, ca udo lat era lly di rected prong th at is fused to th e caudolateral margin of th e sterna l plate. Part of th is process may be ana logous or homologous with a portion of the "xiphist ern um " o f Croco dylus (Ro mer 1956 ); th e neutral term "handle" is pr eferred by us (Fig. 8.1 C, D). T he shaft of the handle tap ers somewha t distally and th en flares to the tip. T he tip of the handle is usually ro unded and has an everted rim for the jo int cap­sule that articula tes it wi th the stern al ri bs. The remaining port ion of the ste rn um, the " pa dd le" (Fig. 8.1 C, D ) is ap parently homolo­go us w ith the sternal plat e of no nhadrosaurids and has an everted rim for th e join t ca psule that art iculates it with the sternal rib s.

Of the sterna l elements examined, th ere appear to be two morphs that correspond to the two hadrosaurid subfamilies (Fig. 8.1C, D ). T he "paddle" is rel atively longer (dorsoventra lly) in lam­

140 • Michael K. Brett-Surman and Jonathan R. Wagner

'. ~.ff~g? j

. < . ' : ; " (' . ,r:·:.";.,:.,,, · :'~~ ..· · ·t::. .<r1!\' .'

.....

\ 'I

A

beos au rines. Th,:: ­hadrosaurines. T:, :: but the sa mple size Cur va ture of the ~- -,

Sternals of ts than th ey are to ~

stored by Dollo ~ .

between the cora; This may also have .

Humerus. T;:,;: ~

The dorsal parr _ formed by the . humeral head, ri-.:: ~

ally to form the '::: is pa rallel to the :::;;.;:-. shaft. At midshan... shaft narrows t c

humeru s, th e elera ar tic ulat ion with '­

As with the s ­guished (Fig. v ._ .

co ntinuo us morp z useful to disrin z had rosaurines I FiEJ than three time s as ally). The asvmp: the m idp oint of . '­long and gracile _ beosaurine hurne: type is seen in ty?=

In lambeosau:

The coracoid in hadrosaurids appears to lie elongate relative ro that of nonhadrosanrids as measured perpendicular to the scapular facer in lateral view. The coracoid hook, where the hone articulates with the sternals, is longer and oriented sub-parallel to the margin of the coracoidal glcnoid. The dorsocranial region of the coracoid below the scapular lacer appears to be slightly longer in specimens that possess a plesiomorphically cranially directed acromion process of the scapula.

The coracoid does not display sufficient variation among the lineages to make it diagnostic at the species level. Most coracoids are crushed Hat, which makes it difficult to ascertain their original shape. It appears, however, that in hadrosaurines the coracoid is longer lanteropostcriorlv) and lower (dorsoventrally) than in lam- beosaurincs, but the sample size is too small to justify this proposed dichotomy. It is possible that tins feature may be end points on a morphological continuum.

The only difference between the early forms, such as line- fWMMnw (AMNH 6577) and Cf/wiurrowMnu (AMNH 6.SH I), and the more derived late Cretaceous lorms. is that the coracoidal hook is smaller and pointed more vcntrally than antcroventrally.

Ontogentically, the coracoidal hook is small and points ven rrally in juveniles. The coracoid foramen is entirely enclosed within the coracoid, unlike some notihadrosaurid ormthopods. Dorsally, the knob ar the end of the acromion ridge is small and rounded and generally smooth, The glenoid is deep as in adults but the depres- sion is symmetrically shaped. The scapular suture is slightly rugose. The main body of the coracoid is slightly thicker rhan that ol the coracoidal hook. One growth Feature of interest is that with on- togi-in, the coracoid abruptly becomes thicker (lateromedially) and more rugose, especially at the suture with the scapula.

Sternals. Sternal are the most difficult elements to study. I'hey are rarely preserved or prepared and are virtually ignored in the lit erature (see Parks 1920 for die first complete description of a hadrosaund sternal). There is no literature on the onrogeneric or population variation of this element.

In hadrosaunds, the venrral segments of the sternal ribs appar enrly coalesced into a single, caudolaterally directed prong that is fused to the caudolateral margin ol the sternal plate. Parr of this process may be analogous or homologous with a portion of the "xiphisternum" ol Crotodylns (Romer 1956); the neutral term "handle" is preferred by us (Fig. 8.1C, D). The shaft of the handle tapers somewhat distally and then flares to the tip. The tip of the handle is usually rounded and has an everted rim lor the |oint cap- sule that articulates it with the sternal ribs. The remaining portion of the sternum, the "paddle" (Fig. S.IC, D) is apparently homolo- gous with the sternal plate of nonhadrosanrids and has an everted run lor the joint capsule that articulates it with the sternal ribs.

Of the sternal elements examined, there appear to be two morphs that correspond to the two hadrosaund subfamilies (Fig. S.IC. D). The "paddle" is relatively longer (dorsoventrally) in lam-

M0 • Michael k. HrettSurm.ui and ]on.irhan R. Wagner

gate relati ve to to the scapular

bone articulates . : 0 the margin of of the coracoid

. rion am ong the . M ost coracoids d ' the ir original s rhe coracoid is .lU~· ) than in lam­.'i:' rhis proposed - end points on a

is, such as Bac­ill 6581), and

oe coracoidal hook -enrrally. and points ven­

- enclosed within o pods. Dorsally, and rounded and

:5 but the depres­. .s slightly rugose.

than that of the - ' 5 that with on­rreromedially) and Mula. nts to study. They

'~ ored in the lit­description of a ;: ont ogenetic or

•zernal ribs appar­red prong that is

-;rate. Part of this a portion of the :n e neutral term

it of the handle ' :- . The tip of the ror the joint cap­

remaining portion ~. ' rently homolo­- -l has an everted ..- sterna l ribs. !;:'?t:ar to be two

subfamilies (Fig. renrrally) in lam-

HH

': "',·1" ", '.~ " ~. .;

" .>' ., .

~ A B c D

beosaurines. The " handle" is relatively lon ger and thinner in hadrosaurines. The handle appears to lengthen during ontogeny, but the sample size is to o small at pr esent for a definitive statement. Cur vatur e of the handle may represent postmortem distortion .

Sternals of Iguan od on are much closer in shape to hadrosaurids than they are to camptosaurids in whi ch the handle is absent. As re­sto red by Dollo (1883a, 1883b) and Norman (19 80), there is space between the coracoids and sternaIs for a cartilaginous xiphisternum. This may also have been true for hadrosaurids.

Humerus. The humerus (Fig. 8.2B, C) is divided into tw o parts. The dorsal part contain s the head, which fits into the glenoid for med by the scapula and coracoid. Immediately below the humeral head, the proximal part of the shaft exp ands anterolater­ally to form the deltopectoral crest. The lateral border of thi s crest is parallel to the medial border of the proximal part of th e humeral shaft. At midshaft, the deltopector al crest ab ruptly ends and the shaft narrows to its minimum width. At the distal end of the humerus, th e element expa nds into lateral and medial condyles for articulation with the radius and ulna .

As with the scapula, tw o morphot ypes of humeri can be distin­guished (Fig. 8.2), but there are many intermediate forms a long a continuous morphocline. Only the most extreme morphotypcs are useful to distinguish hadrosaurines from lambeosaurines. In hadrosaurines (Fig. 8.2 C), the deltopectoral crest is slightly less than three times as lon g (dorsoven tra lly) as it is wide (lateromedi­ally). The asymptote of the ventrolateral border is at or just a bove the midpoint of the humerus. The overall aspect of the humerus is long and gracile compared to the thi cker and more robust lam­beosaurine humerus. The best example of the "gracile" morpho­type is seen in type spec imen of Brachylopbosaurus canadensis .

In lambeosaurines (Fig. 8.2B), the deltopectoral crest is about

E F G

Figure 8.2. Com parison of the left humeri in posterior views. (A) Cam ptosaurus USN M 4282 (length =360 rnm]; (B) Par asaurolophus ROM 768 and AMN H 5893 (length =52 0 mm}; (C) Kritosa urus A MNI-l 5465 (length =764 nun}, Compariso n of the right ulnae in lateral view. (D) Camptosaurus USNM 4282 (length = 262 mm} ; (E) Parasaurolophu s FMNI-l P2739 3 (length =665 m m). Com parison of the right radii in lateral view. (F) Camptosaurus USN M 4282 (length = 232 rnm]; (G) Parasaurolophus FMNH P27293 (length =585 111111) .

Appendicular Anatom y in Campa nian and Maastrichtian North American Hadrosaur ids • 141

B C

1

' umiiL*. II": luiiuL" i, fcUuvuly limycr ^iW ihiimu n; lu.lr.i^urincy (he liiihlli: 4pp^r> i„ l.ngilicii .liiriiig „iii„i:Lii,. l'U« #Hmpk ''^ '' :"" )"'''" -i' pici:ii ,„r .1 .Idimmc »uimiL,ii. i ,;.\.,l.irL ul iliu li.iiiJk nwy rLprDciii |i,^iii„ii(i:iu ,li»u,rliun.

YLU...I, ..i ;,,;,.;„, J„„ 4ru muU, Ll„»ur m >lupu lu liHk(f„WHF# ihji; rhpy AM WHrnprnwun* in wh# ik hj'"H' ^ ii^m. A^.. »,„ruj in I i.,l|p iiDNij,, imb) @nd Nwm4n I iwu,, .hue „ ,|u.L l.uiw jun ihi L,,r.ku,J* 4H.I kknulk lur 4 u jmUgiHuii!, xinlii%pfmHm: Thi* my 4l»" i"^ h@m FFwe fw (Mdr'w„n,L..

HMWFK* I lw luiiiiuni* ihg:», JH, i 11> JiMuLiI ;nw (*A puir. Ihy ii'irsii |wn /'imam* (hf hc.ui, ^hifh hi* inw ih# nlmwd

hun^^l l«Mt!, ihf pmwiwsi |Mn ,n ihc ,ium expand* @n*Fnl*f#F: ill; ;,, Iwuii rlu .li:li,,|iLU,,r.ikrus;. Iliu Ufural h,.r.l,.i ,,, ihitcroi '» IMF*IW m rl't: !"^lHl IWW "' 'h^ |imwi;;.il |^fi ul :l,u huiiiLi.il ,lwu. Ai midkh^n, fhP (WmpKTPMl in*' ahrupd) mJi, uiii) ;hP ,iuii „.,,:,,\\, i.i „., uiiiiiiiHim wi.lih. Ai iliu Uw.il Lii.l ,n ihc hHnwFw*: (he uiLiiwii uxpqmd* mm |Nfpf#| @nii mcii^i gomlylp* for .uu. ;:l.uiu,i w,;|, (hu Milm, au.l uliu.

A, wiih ih. ,. ..iul.i, iv ,, „„,rpl,„u pu:. ul humuii. .,n k Ji»ii,i pui!,lic,l il i;:. W.i), ^; rliurr 4n mun miLriHLUi.iiL fwrm:, ^luin; -i L,,;,iniii„ii» iiiurphuJiiiu. Duly iht: must uxirumu m,,rpli„iyp,» .iru „w:lnl i„ .iDUiiguul, h.iJru^iiiin,:. In,m |.,mh:„wnrii;L*, |,| h@(JrnwHFinp> ii'i; ^ -^ '= 'h^ u«nu,,.\i.,rq| grp^ ^ di*hili ' ^ .i.,n ,i,i.. lime* # long (tjnw^nirallyl 4b ii is wide Ib^mm-ii .ill) I. I l,r .„un|H„iu ul iliu u:,iu„UiLrHl liualcr i* 4f uf |,w .,hu\c ihc mi,i|),:n,[ „f the hwme FHt Thp wwrnll wpw pf ;h@ hwmw">'» Ipf1@ AIK* WnNlc u,ni|^ruU i„ ihc ihiyku .i,;,l muru mh,M hm Ix.,,:,.;...,;.. ii.:i.,L;u>. Il,u K&, ,\^mplc „r f|%: "gf^ilc" m,,r|iln, [ypu i» fucii Hi lypu *|„:uiiiuii ul Hi.iJ -J,,/'Ac,.;:iii,, ..;;.;./. ,;^..

|i, kmhcwuriui., ,h#. NJH,, rhc Julinpcurjl .ru^, i, .,h,,i,r

••....;.

.&AI.\ii :a: , , ,:.;:, :..' . :

.''kill ri..i,f-#pf . :,." '. -: ,.:.. , .„:(""'" "/^: '- ' - ' -

, ..:- : j..:.:. ,

I ' • ' • ... ., . .

' - -" ^

: ...• I|| ...

l',:^„.,..;:,.,U.\|:.;i;.: :-, . ..... - . .

, ...Ikulir .\i,.n„i:i: .;. i 4,::,:.. I Mj.,m^|Hm Nimh AWPfk i- IW:,w,ir;.;, , |4|

.

twice as long as wide. The asymptote of the ventrolateral border is at or just below the midpoint. The overall aspect is that of a thicker and more robust element than in hadrosaurines. Because the deltopec­toral crest is relati vely wider in lambeosaurines, the length/width ratio of the crest is generally smaller than in hadrosaurines. The best example of the "robu st" morphotype is in the type specimen of Parasaurolopbus walkeri.

Several growth characteristics of the humerus are observed. In juveniles, the lateral tuberosities at the proximal end are poorly de­veloped, if present at a ll. The head is relati vely smaller compared to adults, and the deltopectoral ridge is thinner measured anteropos­teriorly in art iculation . The distal condyles are relatively smaller, and there is little difference in size between the lateral and medial tubero sities. In adults, the med ial (ulnar) distal condyle enlarges as the condyles increase in size (Brett-Surma n 19 75).

The relative length of the humerus, as compared with the femur, shows a wid e range of variability. The range for all hadrosaurids is from 47 % to 63 % , with hadrosaurines in general having a longer relative humeral length. The relative thickness of the shaft is also greater in lambeosaurines than in hadrosaurines (Brett-Surma n 1975; Egi and Weishampel 2002). The ratio of the deltopectoral width to length ranges from 19 % to 33 % with lam­beosaur ines , in gen eral, having a wider deltopectoral crest. The rel­ative size of the hadrosaurid distal condyles, as compared to mini­mum shaft diameter, was used by Sereno (19 99) to diagn ose hadrosaurids (condyle width 120% to 130 % ). H adrosaurids range from 160 % to 200%; therefore, this character is not used in this stu dy. In juveniles, th e radial and ulnar condyles ar e similar in de­gree of development, but the ulnar condyles are far more developed in adults. Extreme caution mu st be taken when considering thick­ness as a taxonomic cha racter. As in th e coracoid, thickness is more a product of ontogen y than of taxonomy.

Radius. The radius (Fig. 8.2G) is lon g and stra ight, with a cir­cular cross section at midshaft. The pro ximal end, which articu­lates with th e medial-distal condyle of th e humeru s, abruptly ex­pands int o a circular, cup-shaped process, and resembles the top of a Doric column in anterior view. The distal end is flattened and is recta ngular in anterior view.

Brown (1913a) and Lull and Wri ght (1942) stated that the ra­dius/humerus length ratio is diagnostic at the subfamily level, the humerus being longer in the hadrosaurines and the radius longer in lambeosaurines. This hypothesis is no lon ger valid (Brett-Surman 1975; Egi and Weishampel 2002 ). The radius (Fig. 8.2) is longer than th e humerus in the hadrosaurines Saurolophus osborni (AM N H 5220 ), Edm ontosaurus edm ontoni (NMC 2288), and Anatotitan copei (AMNH 5730) and in the lambeosaurine Hy­pacrosaurus altisp inus (N MC 8501). The radius is short er th an the humerus in the hadrosaurines Kritosaurus (=Gryposaurus?) in­curuimanus (ROM 764), Prosaurolophus maximus (ROM 787), and Edmontosaurus annectens (YPM 2182) and in the lam­

142 • Michael K. Brett -Surman and Jonathan R. Wagner

beo saurine Paras radius increases humerus. The res lengths of lambeos cm. The ra dius .a and is therefore .' : ~

Ulna. The triangular bone. forms a three-prozs gre es from the processes form c. ­radius. The pro: with the posrerioc condyles, which ..:.. end of the ulna is

The tra nsi . ment of the olecrs ness of the ole..-:--­19 89). The latera show any grea :: . but the shaft un ~

to its circ umfer tively thinner wi:::t

The onl y vazi is that most ha .1

more slender ar. .: same size, bur Lb=-' nomic validity.

Carpals. IG.= ho ckey puck. T:,~

as tall (dorsovena been reported as ­and as ovoid (Pri

Hadrosauri d th ey are rarely :­may go uncolle pr eparation. He .. arran gement a t tion of the man Anatotitan (Fig. one another, thosaurus casu.: are mounted nc.., over a different ~

Parks 1920) 5

th ese elements :..... 1988; Norman a radiale, interme·... form a large r ~ _

form a smaller i,

twice as long as wide. The asymptote of the ventrolatcral border is at or jtisi below the midpoint. The overall aspect is that of a thicker and more robust element than in hadrosaunnes. Because the deltopec- toral crest is relatively wider in lambeosaurines, the length/width ratio of the crest is generally smaller than in hadrosaunnes. The best example of the "robust" inorphotypc is in the type specimen of Parasaurolophus walkerL

Several growth characteristics of the humerus are observed. In juveniles, the latei.i! tuberositics at the proximal end are poorly de- veloped, if present at all. The head is relatively smaller compared to adults, and the deltopectoral ridge is thinner measured anteropos- teriorly in articulation. The distal condyles are relatively smaller, and there is little difference in size between the lateral and medial rtibcrosities. In adults, the medial (ulnar) distal condyle enlarges as the tondyles increase in size (Hrett-Surman 1975).

The relative length ol the humerus, as compared with the lemur, shows a wide range of variability. The range for all hadrosaunds is from 47% to 63%, with hadrosaunnes in general having a longer relative humeral length. The relative thickness of the shaft is also greater m lambeosaurines than in hadrosaunnes (Rrctt-Surman 1975; Egi and Weishampel 2002). The ratio of the deltopectoral width to length ranges from 19% to 3.3% with lam- beosaurines, in general, having a wider deltopectoral crest. The rel- ative size of ihe hadrosaurid distal condyles, as compared to mini- mum shaft diameter, was used by Sercno (1999) to diagnose hadrosaurids (condyle width 120% to 1.50%). Hadrosaurids range trom 1h0% to 200%; therefore, this character is not used in this Study. In juveniles, the radial and ulnar condyles are similar in de- gree of development, but the ulnar condyles are far more developed in adults. Extreme caution must be taken when considering thick- ness as a taxonomic character. As in the coracoid, thickness is more a product of ontogeny than of raxonomv.

Radius, The radius (Fig. 8.2G) is long and straight, with a cir- cular cross section at midshaft. The proximal end, which articu- lates with the medial-distal condyle of the humerus, abruptly ex- pands into a circular, cup-shaped process, and resembles the top of a Doric column in anterior view. The distal end is flattened and is rectangular in anterior view.

Brown (1913a) and Lull and Wright (1942) stated that the ra- dius/humerus length ratio is diagnostic at the subfamily level, the humerus being longer in the hadrosaunnes and the radius longer in lambeosaurines. This hypothesis is no longer valid (Brett-Surman 1975; Egi and Weishampel 2002). The radius I Fig. N.2) is longer than the humerus m the hadrosaunnes Saiirolophus osborni (AMNH 5220), &AHonfo*dMn« edmoMfowi (NMC 22HS), and AnatOtitan COpei (AMNH 5730) and in the lambeosaurine lly- pOCTOSaurus altispinus (NMC 8501). The radius is shorter than the humerus in the hadrosaunnes Kritosaurus (-CtypDsaimis}) in- curvimcinus (ROM 764). I'rosuttrolophus maximus (ROM 787), and Edmanlosaurus arviectcns (YI'M 2182) and in the lam

142 • Michael K. Bieu-Surman and Jonathan R. Wapici

areral border is at at of a thicker and

ra use the delropec­the length/width

s, urines. The best rype specimen of

" ar e observed. In end are poorly de­

er compared to a sured anteropos­relatively smaller, areral and medial

. ndyle enlarges as

pared with the e range for all

surin es in general arive thickness of

- in hadrosaurines . The ratio of the

: 0 33 % with lam­ral crest. The rel­mpared to mini­

99) to diagnose - sdrosaurids range :s not used in this - are similar in de­:;:.r more developed : onsidering thick­

i, thickness is more

zraight, with a cir­e d , which articu­om s, abruptly ex­

.: mbles the top of is flattened and is

tared that the ra­:-farnily level, the ~ radius longer in id (Brett-Surman

Fig. 8.2) is longer olopbus osborni

. -~ lC 2288), and : beosaur ine Hy­' " shorter than the G .1Josaurus?) in­

,:lS (ROM 787), and in the lam­

beosaurine Parasaurolophus walkeri (ROM 768). The length of the radius increases with ontogeny relative to the length of the humerus. The result is that hadrosaurine radii approach the relative lengths of lambeosaurines when the femur length approaches 140 ern. The radius/humerus ratio is not consistent within any clade and is therefore useless for taxonomic purposes.

Ulna. The ulna (Fig. 8.lF) has the overall shape of an inverted, triangular bone, with a conical cross section. The proximal end forms a three-pronged process, with each prong radiating at 90 de­grees from the other prongs. The medially and anteriorly projecting processes form a cradlelike structure (radial notch) that receives the radius. The proximally projecting olecranon process articulates with the posterior side of the humerus between the two distal condyles, which sit in the olecranon notch of the ulna. The distal end of the ulna is flattened like that of the radius.

The transition from juvenile to adult results in a great enlarge­ment of the olecranon notch of the ulna and an increase in robust­ness of the olecranon process (Dilkes 2001; Brett-Surman 1975, 1989). The lateral process becomes thicker. The distal end does not show any great increase in width compared to its circumference, but the shaft undergoes a tremendous increase in length compared to its circumference. This results in the forelimb becoming rela­tively thinner with increasing age.

The only variation observed among the ulnae of hadrosaurids is that most hadrosaurines appear to have limb elements that are more slender and less thick than those of lambeosaurines of the same size, but this feature is not consistent enough to have taxo­nomic validity.

Carpals. The blocklike carpals (Fig. 8.3B) resemble an ice­hockey puck. They are circular in cross section and about one-half as tall (dorsoventrally) as wide (lateromedially). They have also been reported as tetrahedral (Davies 1983; Prieto-Marquez 2000) and as ovoid (Prieto-Marquez 2000).

Hadrosaurid carpals are rare in collections, probably because they are rarely recognized because of their irregular appearance, may go uncollected, or may be inadvertently destroyed during preparation. Hadrosaurid carpals have been restored in a colinear arrangement above metacarpal III since Brown's (1912) redescrip­tion of the manus in Edmontosaurus annectens. In a specimen of Anatotitan (Fig. 8.3B, AMNH 5886), they are mounted on top of one another, both over the third metacarpal. In a specimen of Cory­thosaurus casuarius (CM 9461), the radiale and the intermedium are mounted next to each other, both under the radius and each over a different metacarpal, and other specimens (Lambe 1913; Parks 1920) suggest a side-by-side arrangement. Homologies of these elements are uncertain. The recent consensus (e.g., Erickson 1988; Norman and Weishampel1990; Rasmussen 1998) is that the radiale, intermedium, carpal I (if present), and metacarpal I fuse to form a larger block, while the ulnare and distal carpal five fuse to form a smaller block.

Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids • 143

F-Wiiw mt!fM»H wtih Hiuii#Ktiy ft-kuvt: w (k Iwwh I'f «lw hymL..;, 11,. HJi,i|f |»ihw WffiMWFin@Hdhapp; <'- -'^ m-laii^ iHigllk nf lAmlv#iMHfmp4 whwi ihe IVHIHF leiwil' n -'— ''"

" rhe FatHH^wiiK-fw Mfiw k iiiH twwiH!! within am% t'-' ami i* ih^'-h'N ii^i^ i,.; iin@HiiiimK pwFpiwt

' - \ I hp ulna ihg= N^lu ha» il^ iivmll 4^i- ' — inveMpd, IFMI#ldF Nlll*:, Will) H Ulllltal #1** *klln|l: 11)*: pMI*illMl Mid WnR*(llFH:=pFIHI|Wl ' 4Y!fhMyhpHIIIHHtll4(im#!^ ^'' '- BfW* fFlim (hf llfllPF l^nHgl: I IIP IHWiallf AHd HI"-' i -' nl'lttfliW pn,f;- i "i a yNtllWikt: WFHtfiiFf iFadial nnith) thai Fftt-nw fht= '':-' IW pHl^illHlly pFI#flllg HlttFamm pi-': ...;i...|..;uc

wiih ihf pwwniif witJt: nf Hi^ hHmwfiw hwwwi ihe iwn t!'" -' timd*lp», whiwh *if m (he nh^HRim nnffh nf fhe iiliM: fhif til "' Plld Of fhp WlM4 :* -Hfd llkf fhjf (if ihf HdlWt

TIlH IMII^IIIHII IfflHI |HYH# III Wllllll FHIlll» III a MFHI t=HlaFMi:: m*miiir;hHil*HmmHwwhnhhH!'i"' "ii".".-.'.,,, ..iiy*: m* nf fhf iitrnFaiinii piw^ (Bilk* AWiili Aim t,HFnian I-' ^ muni: rhpMimlpniAv,h*nmAihwkM: I hedi»ialHiddnwm* ^hiiw am= Mfwi iihfMi* m widfh mmpah-d - " -'Himipfpnee, i...i ii,. ,i,.,ii itiigyF^iw 4 (FEmmluu.-; .- :.. i. .fi iiW in ii* fiFtiimWk:p: flii# Fwili; in & i.u.,..i. i.^nmiiv '- nn4f ihimnpf wiih imtPMiinn ***=:

The wily HFHiitwi ni^ryLd ammw the Hhwp nf hadniwwHd» i* '.ha; ninm hadfiiMHnHw appwF in ha^ hmh tHMiitini* Hi*! - mnFH »kiitkF amtl Itv ihitk ihaw ihiw nf l4mliMAAHFinh( "f '- = -, '"!! lhi4 IHIHFP h Hill t=nil$Wm HIHHWh III hm I-'

HIIIHW WIlMlfh rdlfWk: [Mi: Nwkllk^ WffMl* iMM: MMl ^*HHhlt: HH w-=

hllkkl^ (Wk: TllH 4F^ HKWlKF III KflW ^ '= ' '' Hit: half a* tail dliiFwimiiFallfi a» wide llaimiiHt^iallyi: II11.&: km- ^i kmi ffpnffwl a» iWFahwlHl iUavii4 iWNi; PfMn MamiHW 4088) and an nwwl iPFwnMaiwiiw A"

HadRNHiFiil tafiwU arn HFt= im trnll^uniM, pMihahly hoaH^ (hey an; HFfly fmwiifHl rvtaiw nf (hm; ifwulaf appwf^" " ' #' HHknilRlPtl: HF may ^ IIHHfHlHI'h :^:in»Al lIllflHg , i iraiiiiH: I Wfnwwnd afp#k IwYt: Ivim FwrnW hi a mlimeaF aFMf#memf aho^ mwaaiFMil III nmn; MinwiA i MU' FwW"r finn llf fhi: HMHIh III f=^MnHf«^MriH dM*fWM: Im a', " -' f\NWfnffww 11%: tilt, AMNH ^NNhi, ih^ aw imwiifw nn fwp nf urn; annihef: ln*fh nvt=f ;h# ihiml iHtH4taFM!: In 4 ^mtiHKH nf f - fAuMMHi* ^#Mfm (f*l ^hli, dip Hiliale and (he iiiKfrnmliiim @f@ mnHHiwI iip^i in mh nfhpf, hnih HiWt=F the Ndm 4nW ^ -'- nm 4 tlilWif mirHK4F|Ml, ^iW n;ht, ;i - - ' "'^ i^i*i Mark; I9AIM wi#N a titWn t'di: aFF*mi*mFHi: Hn!Hn!,4;'^ -' ihi-v Wei"-"' "- iiiimiaiH: 71* FH*m ^nHviwii W:K:, ^FitkwHi IWNN; Nwfman and #i#fMmwl IwwOj Ka^Hwwn '-'"' '-"" -'- Hn'ali:, infmiwdiHRi, HFp4l I lif pfevml, and meMMfMi IIW m form a l@Wf hh#;k, whilf ihe wliiafi: and dwml Mfpul hip fw* fn hifmawMllwhlnfk:

AM^i«ikiFAMRm^inWm|MmHmaMdkl!HWfiWmaHNHFAAiHHh^HlWmMWFid4 " 14*

CRL

"'

MC 3

1II c8 MT3

11\ 11\

D F 1\1

\ \

Figure 8.3. Comparison o( the right mani in ante rior views: (A) Camptosaurus USN M 4277 (me tacarpal III length =78111m); (B) Iguanod on (redrawn (rom N orman 1986); (C) Anatotitan AMNH 5886 (length o( metacarpal III =31 0 mrn}, Com parison o( the right pedes in anterior view : (D) Camptosa urus USNM 4277 (metatarsal III length = 234 mm); (E) Iguanodon (redraw n (rom No rman 1986); (F) Sauroloph us AM NH 5270 (length of me tatarsal III =34 0 mm),

The sa mple of carpals is insufficient to determine the range of variation present in hadrosaurids. The intermedium (provisional designation) is slightly lar ger th an th e ra d iale (pro visio nal designa­tion), which, in one specimen (AM N H 5886) , has an ove rha nging lateral process that fit over the second metacarpal. Only two ca rpa ls are consistently seen among hadrosaurs, and what is termed th e "intermedium" here could very w ell be th e uln are.

Little varia tio n in morphology of the carpals was noted among th e genera. Variations in relative len gth s of metacarpal s II a nd III vary widely between 76% to 91%. Measurements of th ese ele­ments did not prove to be of diagnostic value . It is common for th e hand not to be preserved in most specimens. There appears to be a taphonomic pattern such that the head, manus, and pes ar e the first part s to become separa ted from the main portion of the ske leto n (Dodson 197 1). None of the bones of the manus or pes is fused to

another or to the limbs, thus making separa tion quite easy before final burial.

Metacarpals. The metacarpals (Fig. 8. 3B) ar e pencil shaped an d ha ve smooth, featureless articular ends. Metacarpal I is lost

144 • Michael K. Brett-Surman and Jonath an R. Wagner

and meraca roai . wi th metacarcs ca rpal IV, whi: :: _

As noted ;:.-­somewhat diverae is associated wi : th e metacar lation is l ate r~ :·- .:.­paren tly al lowe d ­of th e first pk :""­m ost in the ~

1912} an d H~:: ~­

IV is slightly Gj\-.O'. of a ll spec im ens .J.

In IglU11iOi and was apr=.;-:, manus . In lgu.: may have beer. -; pho logy see- :­pr essed to me=-,-..:.- ·

Lehman , perso na 1984: plate 20- ~ :

Dig it \ . is 2. ~ _

of ha drosa ur, ::: 200 1!.

Wewer e ~

the genera or 1:-. .=. -.

a complete maz.ns at the level or ;:'..: 7:: :

Phalanges. -:­Iguanodon (t c::- ~

than wide (1",: ::- :-- ­0-3-3-3-3. ThO' .-­and laterallv i =.;--:>

we dge-sha ped -;':.. ­ence exists be zwees and larn beosauri langes (distally z: ally consists oi large as the . row aenerallv cI::> •

and often and Ouranosuc rank of th e three phalanx 1. Thi;: propriate derive.; or less of the iL:"5:

In Saurolop ': IV-1 is unusu

Apper

rigmr S.i. Comparison of the right mam in aiiterw news: (A) Campiosaurus {'SNA! •>2"" imetacarpal 111 length = ?S mm): (H) Igu.inodon In-drawn !-<HI:

N«Mm## IM6;; fO AnaMwmm

metaoarpal 111-310 mm). Comparison nf the right pales m anttriot neu. tl)) Camptosaurus UWAI 4277 (Wwuwrwf ff/ length - 2JJ mini: if.) Imi.inodon /redrawn from Nrnman 1986); If) Sawrokpbut ^MNH »70 ft*** of metatarsal 111 = 3-10 mnt.i.

The sample ol carpals is insufficient to determine the range of variation present in hadrosaurids. The intermedium (provisional designation) is slightly larger than the radiale (provisional designa- tion), which, in one specimen (AMNH 5886), has an overhanging lateral process that fit over the second metacarpal. Only two carpals arc consistently seen among hadrosaurs, and what is termed the "intermedium" here could very well be the ulnarc.

Little variation in morphology of the carpals was noted among the genera. Variations in relative lengths or metacarpals II and III vary widely between 76% to 91%. Measurements of these ele- ments did not prove to be of diagnostic value. It is common for the hand not to be preserved in most specimens. There appears to be a taphonomic pattern such that the head, manus, and pes are the first parts to become separated from the main portion of the skeleton (Dodson 197] |. None of the bone: of the manu; or pes is fused to another or to the limbs, thus making separation quite easy before final burial.

Metacarpals. The metacarpals (Fig. S.3B) are pencil shaped and have smooth, featureless articular ends. Metacarpal I is lost

144 • Michael K. bre:t-Siirnian and Jonathan K. Wagner

Me 3

c MT3

'" F

:<mme the range of . - ium (pro visiona l

rovisional designa­. - an overhanging

~ .:arp a l. Only two . ~ ~5 , and wha t is

oe the ulnare. : was noted among

e;2 carpals II and III ems of these ele­

l: :- common for th e -= :-e appears to be a

, znd pes are the first o n of the skeleton , or pes is fused to

-: quite easy before

are pencil shaped _ _eracarpal I is lost

and metacarpal V is reduced. The manus is fun ction ally tridactyl, with metacarpal III slightly longer than metacarpal II and meta­carpal IV, which are equal in length.

As noted by Maryariska and Osmolska (198 4), digit IV is somewha t divergent in hadrosaurids. In mos t cases, thi s divergence is assoc iated with an axial twisting in the first phalanx, where by the metacarpal joint is convex in tw o dim ensions. Th e distal ar ticu­lati on is lat erally compressed and arcuate in extensor view, and ap­parently allowe d the digit to splay laterally. Th e dista l articulation of th e first ph alan x is twi sted as well, such that its hinge line is al­most in th e same plane. In Edmontosaurus anne ctens (Os born 1912) and Hypacrosau rus altispinus (Brown 1913b), me tacarpal IV is slightly divergent laterally, a ltho ugh th is is not a ppa ren tly tru e of all specimens of Edm ont osaurus (Lambe 1913; Marsh 1892).

In Iguanod on , metacarp al V is lat erally compressed proxi ma lly an d was apparently somewha t divergent fro m th e rest of the manus. In Iguan odon and had rosaurs, this digit is elonga te and may hav e been prehensile (N orma n 1980). In contrast to th e mo r­phology seen in Iguanodon, metacarp al V may have been ap­pressed to metacarpal IV in at least some hadrosaurs (Wagner and Lehman, personal communication; q.v. M aryariska an d Os m6 lska 1984: plate 20-8; Parks 1920).

Digit V is absent in two well-preserved, articulated spec imens of hadrosaurs from the Aurisino Limestone of Italy (Dalla Vecchia 2001).

We were unable to detect any differences in me tacarpa ls among th e genera of hadrosaurids. The sample size of th e spec imens with a complete manus is too small at thi s tim e to detect any trend s even at th e level of subfamily.

Phalanges. The phalanges of had rosaurids are most similar to Iguanodon (fide N orman 1980). They are longer (proxima l distally) tha n wide (late ro medially) and flattened. The genera l formul a is 0-3-3-3-3. The unguals are hoof shaped but relat ively less develop ed and lat erally flared than the pedal phalanges. Some had rosaurs had wedge-shaped phalanges, wh ereas others did not. One slight differ­ence exis ts between the ph alanges (Fig. 8.3B) of som e had rosaurines and lambeosaurines. In the hadrosaurines, the secon d row of ph a­langes (distally from the met acarpals) of digits II, III, an d IV gener ­a lly consists of well-de veloped elements that are proport ion ally as large as the proximal row of ph alanges. In th e lambeosaurines, th is row generally consists of elements th at are considerably reduced and often pebblelike. Sereno (199 9 ) allied Iguanodon, hadrosaurs and Ouranosaurus on the presence of manual digits in the second rank of the three middle digits that are more than twice the length of pha lanx 1. This is probably a typographical error because the ap ­pro priate derived state is for the second-rank pha langes to be 50% or less of th e first rank in length.

In Saurolophus angustirostris (Roz hdestvensky 195 7), pha lanx IV-1 is unusually short, and its proxim al and distal ar ticular sur­faces are set at an angle to one an other. Slight asymme try in pha-

Appendicular Ana tomy in Cam panian and Maastrichtian North American Hadrosaurids • 145

,„i ....... .....ilVisru.i... ,.i n.L .....;...,:, ru;..:;.,::.ill) iriil.Kul, ..i. ;;,u.,L.,rpal HI Uightly lmi*F 'hail u,Li.iu,p.,l II an,l muia

urp.il IV ^l.i.l, .irL jijual m Wgih, A, m.iLLl hy Maryan,ka ,;.. i^l.: ,l^"-4,, .i,^, |y j,

^,inL. I..i .!...:....! .1. l....l:..>..unJ». Ill 111,^1 L\l»Ub,lh^Jlu,gi:ilLL: W ^,..,i..i ^i;h am **wl iw:»m^ in (hu ln»i phalanx, wlicrchy ihLmLUL.irp^liwuiu, i; , . n.. .ii,;..larum l.iu.,1, :> in. r.iih . .,ni|]rL»w:(| g^ Aftualc ii| c\icn*.,i . i. \i. a,..l ap paimrly jlluuLii ih. .i:,;:i :,: ,play laiur^lly. I he Jwul .irikula;i„n „l ihc i:r»i |,lul.i,i\ :» i\.i»iuJ .1* wull. nth ihai n» lun^u Imu i, il nnwi m ;l. plane. In K.!,';,.,Uu,.,;„;,, .iu,;.Lf,,..^ iW„,rn |U|J, .;,..! ;h,..;.:,.h;;„m .,/;;^f;i,f li;.,.'ii Ivliln. imi.i..i:p.;l |\ i»,li(ih!l) Jivcfguni l.uurally, altm,,*!. ihi»,*,.„i .i|'|iarui,il, i,... urj|l)iUL;:,.j:;,.,i; ; ..:,»,lainlx. rUi;Mar»lilhV.',.

In fi,;<.i,;,..f,„;. ii.u.i.a.pal \ i» In. rail) Lumpm^cLl pi II in.i -..,) .;jY.,n;;il:. >,.i;iL.i,.;i .iiu^ni Ir.nii liiu run ,,i ii.. ' '-' 'v'l""'"" "iJ l,aJ:,.....,,: ihi* dl@H K Llui^.nu ...U -" '- i ...,-'--;'- lvS,li.l:;..,:i;r.;>.;.,;hL.:..; H.ulugy ,ii:u in fg,u»,.J,m. niLl.K.npal V may haic hwn i|. iru^cJ i., niLU.arpai IV in ai lew b I, ..i.....mr, .Wagner ami .I.....I:.. r^,:.,.,l ...:.,niiiniL.m,,,;: .,.-.. Mar).,n,k iJ ll»mm>k.i

iVhl:,,,.,,, J„,i:l'.,rL> IVJ,„. int;n \ i, ahxruf ,n rw" wt=n rtw^rml, ^niyui"^* ,mu„,ui„

,,l hj.lr„wuM m,m (he Ami,m„ I imuxum. „l li.il) ,Uall.i Vu.. i.,, : '

,. ,.i,.,i:i. i...iLi.Li.„n .liiiLiL-nLW in mu....u,.,I, .uu,,ng ihc KciiLiri „l i;...i,.,^,:r;.l>. Fhu simple »iz.: ui iliu »|,uuiHLiii* *i;h .. w.mpluii: nwn..» i» l,„, »mill 41 ihi» umu t" JuiLLi an) in„.i, .' ,:.. iLvJ.Huihnwnly.

)V:.J.i';sL, II,. ,' -' '^» "' hi.liusuiiiJ, .ir. m,,n fiiiiil.u i„ fyMi;.J,.,iiliJL Nuniwn i 'Ml,. I he) arul.,n^r,p,.,Mi:i.ilJwHll)i man wide ib^.:.. .niihi 4n»i ".ni.-i H.. LunLf.,i fummi., :> ''^ = ' n,. ..;.,,. u..,.. i i^i, .iL.ihuirLlam.lU.-.1. u.., ..i 4R(I laiurall) HarcJ ihan ihc p..l.,l pluUinn,. ",„nu l,.„l:. s.iui> li.iJ

i.j. ,i,.,|*.l phaiangci, whu, j, .,il,i> .iu :„,i. Wnu*l;gh| dl"^ ., jxi»i, Iwv.uu, il;. , i. .1 i;,.. ,. th, ul M,:„L h.,Lln„.,uiiii.:*

#Rd Uinhcus.iniinL^. In HiL hj.lnu^iiru.. ^ ,i I ... .1 plw lanku, ,.li,:.,l| !,,.,:, il, ;,... ...,r|Ml„ „i Jigii, II, |||, ami IV pun., .,li ..,;;,i,i, ,.l .icilJL Ll.,;:^! _k:;,.j:i:, ihai .Hi pr.,pu,u.mally a, biy 4» 'he pmxiinul niw m plwl.ingc, I,; (Rf |4mhcii,.iiinn,». ihi» n,w ^i..:.,il' o,n>w» „f JuiiLi,,, Hui aru unwiiur-ilily rL\l;,...i Md Hum pLhl&likc. ^rn,.. .ivy'j, .Wind fi;i,.;„..J,,». luJn„.ui,, aii.l I ';,..;,;,.,.;;„«„ „:. ,hu prL^i^u uf nunua) (jlgllli IP '"*: w=."uJ

,,1 ihu ilirui: „i:Jul^ ili^ii* ihai arL muru ilian iwiff (hf Wn@lh "I pn.,lanx I. Il.i, i» pr„h.,l,|y a h|X,|tr.ipl,iLal um,r huLau^ il.i .,, pruprwiL JunyuJ wak i, l.„ ih. „v.,:J rank phalantiu» n, lw iUHc .,i L,.;.,,, liiu Ii,,; i.uik in Icnglli.

I;. ",.;;«;. J,#„i .i,:<,x,n,,^(r» ,K„;ii,iLi,iLLn»k) l^i-|, pli4lan«( IV-1 ,» „,i„»„all) »hnri, ami n» pr,,xu:,al a„.i .ii^.i Aiiiuilar „„- ULL, aru ,u ai .;: i .Muil.u ",1,,: , n ph.;

Apj:-:..'..-:l"^..""lliy:n* , I I....;...:.;:..:. ..,. ..._.:...:. lUJiiKHMfMl*: H»

0

p op

Figure 8.4. Comparison of the left ilia in lateral views . (A) Edmontosaurus DMNH 1493 (right in reverse; length = 1160 m m}; (B) Paras au rolophus FM NH P27393 (length = 975 mrn}; (C) me dial view of Krito saurus USNM 8058 (length = 900 mm}; (D) Lambeosaurus R OM 121 8 (length =1003 mm} ; (E) Tanius (length = 920 mm) (redrawn from W iman 1929); (F) Bactrosaurus A MNH 6577 (length = 690 mm}; (G) Gilmor eosaurus AMNH 6551 (length =650 111m); (H) Camptosaurus USN M 5473 (length =642 mm).

c

t. , . . . , · L " :::)~"" ' ·~ · ~ ·~E

G

. "~"

B

.,,' . .. ;::.:.: .....,;.- . ­. ' - ' .~.', . -. . .. .'~:::JD

~~ F

~~. ,..,.•,.c~: ,..~.."·.·· ..;:"?::'

H

lanx IV-2 is evident in specimens of Kritosaurus incuruimanus (Par ks 1919) and Edmontosaurus ann ectens. Phalanx V-4 has been recovered with specimens of Kritosaurus incuruimanus (Parks 1920) and Brachylopbosaurus goodtoini (Prieto-Marquez 2000). Unguals are generally smooth along the anterior edges. In some very large animals, the unguals are pitted, have roughened edges, and have deep gouges at th e leading an terior edge resulting in a "frilling" (Bactrosaurus, AMNH 6553; Gilmore 193 3). The latter is a gr owth-related feature and is not systematically useful (Brett­Surman 1975, 1989).

Pelvis and Hindlimb

Ilium. The ilium (Fig. 8.4) is composed of three distinct sec­tions. The preacetabular process curves ventrally and is a flat, ver­tical blade in all but the oldest individuals (see below) . The midsec­tion of the ilium contains the acetabulum, and the dorsal margin bears the "antitrochanter" (or, more properly, the suprailiac crest, fide Rowe 1986). The middle area between the dorsal and ventral margins contains the concave depression for the M . iliofemoral is (see Dilkes 1993). The medial side of the ilium contains the articu­lations for the sacral ribs. The postacetabular process is paddlelike and rectangular in lateral view. Iliac length is isometr ic during on­togeny and is approximately 93% of femur length. In iguanodon­tids, thi s ratio is nearer to 100 %. Iliac height compared to length is between 16 % and 25 %, with much variation between taxa.

The preacetabular process appears to be deeper, as.measured at its minimum depth nearest to midshaft, relative to the height of the ilium at the pubic peduncle in some iguanodontians th an in others.

146 • Michael K. Brett-Surman and Jonathan R, Wagner

H owever, the iiu for hadrosaur; , changes with 2" eta bular pr ocess :;;: in taxa with p 2.';-'

trosaurus, Parssa 20-27 degr ees. z., discussed below. ~

some ta xa mav ~ ­

H owever, the di'="" imal may be 5 c=. ation is also hig:'.. Maiasaura (0 degrees in C.o,,:. Gilmore 1946 , process i s m U CG ;:

degree of deBe.::i- ­In lguanodcs

lowly em bayec' : deeply em bayed ve sloping caudai :=..3.'

sally. This process The iliac acetal­the base of the -;" medially as we" as lateral margin s 0 : .

nearl y at the 5.= '74.~

on ly slightly ar6 ilium contacts ~

art icular surface nosaurus, Bactrns and possibly is the lateral margin rends only a bou: ­la r process, a;}': ­

In early Dc. ,:,,:- ­lar, with a n e2.' ~ . what pendant 2.- ­

ranee from th o: .: rocess in l a r e =-~

roun ded ve iliac blade, F~: c.

:\1. ilio fernora lis Claosaurus, S!:',: ._

illustr a ted by Y w est is more ,::­ ­ischial ruberosiry the cranial

hadrosauri.is .trochan ter - is _

figure8.4. ' tmifiartsiin of the left ilia tit htcr.il uitWS. I A) Edmonwwuru* OMNH HVJ Inghr in reverse! length - 1160 mm}; (111 Parasaurotoplms I'MNll M7JW |kq|*& - 97* wW; fQ medial view of Kfitosautus USNM WI* flm*f& - *00 """/; //)/ Lamkowufu* KOAI IJJ* fln%lt = 1001 nun); ilil Tar.li)> (length = 920 mini (redrawn (ram Winian 1929/; (# BmenowunM AMNH 6J77 (tw#* - 690 "Wi fG/ Gilmoreattunu AMNII 655/ (length = 650 mini; (II)

(length - 642 mm)

laDX IV-2 is evident in specimens of Kritosattrus incutVitnanus (Parks 1919) and Edrnontosattrus annsclens. Phalanx V-4 has been recovered with specimens of KritOSaurtiS fneurvinumus (Parks 1920) and Brachylophosaurus goodwint (Pneto-Marqucz 2000). Unguals are generally smooth along the anterior edges. In some very large animals, the unguals are pitted, have roughened edges, and have deep gouges at the leading anterior edge resulting in a "frilling" (BdcfroawnM. AMNH 6553; Gilmore 1933). The Litter is a growth-related feature and is not systematically useful (Brett- Surman 1975, 1989).

Pelvis and Hindlimb

Ilium. The ilium (Fig. 8.4) is composed of three distinct sec lions. The preacciabular process curves ventrally and is a flat, ver- tical blade in all but the oldest individuals (see below). The midsec- tion of the ilium contains the acetabulum, and the dorsal margin bears the "anntrochaiuer" (or, more properly, the suprailiac crest, fide Rowe 1986). The middle area between the dorsal and ventral margins contains the concave depression for the M. iliofemorahs (see Dilkes 1993). The medial side of the ilium contains the articu- lations for the sacral ribs. The postacetabular process is paddlelike and rectangular in lateral view. Iliac length is isometric during on- togeny and is approximately 93% of lemur length. In iguanodon- tids, this ratio is nearer to 100%. Iliac height compared to length is between 16% and 15%, with much variation between taxa.

The preacciabular process appears to be deeper, as measured at its minimum depth nearest to midshaft, relative to the height of the ilium at the pubic peduncle in some iguanodontiaiis than in others.

146 • Michael k Krctt-Surman and Jonathan R. Wagner

.is incuru im anus : [anx V-4 has been turuimanus (Parks

Marquez 2000). edges. In some

e roughened edges, - lge resulting in a

":: 1933). The latter ically useful (Brett­

f - ee distinct sec­. and is a flat, ver­- m y ). The midsec­-- e dorsal margin ~ suprai liac crest,

:.orsal and ventral -~ ),-1. iliofemoralis ... main s the articu­

cess is paddlelike " erric during on­

_ -~ . In iguanodon­[;:;- ared to length is .. "cen taxa. . --. as measured at -0 meheight of the : - 5 than in others.

However, the limited data available suggest that the avera ge ratio for hadrosaurs is around 33% ± 7%. It is unclear if this ratio changes with growth. Among hadrosaurs, the angle of the preac­etabular process in lateral view relative to the main body is greater in taxa with particularly deep ilia, -35 degrees or more (e.g., Bac­trosaurus, Parasaurolophus, Hypacrosauru s), and it is shallower, 20-27 degrees, in taxa with shallow ilia (e.g., Edmontosauru s], As discussed below, apomorphic reduction of the pubic peduncle in some taxa may incre ase this angle to near 40 % (e.g., Kritosaurus ). However, the difference between ilia from two sides of the same an­imal may be 5 degre es or more (e.g., Morris 1978). Ind ividual vari­ation is also high, ran ging from 20 to over 30 degrees in juvenile Maia saura (Carpenter 1999: fig. 12.29), and from 25 to over 35 degrees in Corythosaurus (Brown 1916; Parks 1935; possibly Gilmore 1946). Th e degree of deflection of the pr eacetabular process is much great er than in any other iguanodontians, but the degr ee of deflection is too variable to be of taxonomic use.

In Iguanodon, the medial surface of the ilium is relatively shal­lowly embayed for the acetabulum, while the late ral surface is deepl y embayed vent rally by an acetabular fossa having a steeply slop ing caudal margin and a slightl y produced marginal " lip" dor­sally. This process is more properly termed the "antitrochanter. " Th e iliac acetabular fossa extends laterally, almost to the level of the base of the preacetabular process, and "cups" the femoral head medially as well as dorsally. In hadrosaurids, both the medial and lateral margins of the acetabular fossa are lower such that they are nearly at the same level, and the aceta bular margin of the ilium is only slightly arched in lateral view. The acetabular sur face of the ilium contacts the femoral head only dorsally, and the acetabular arti cular surface is only barely visible in lateral view. In Oura­nosaurus, Bactrosaurus, Altirhinus, Probactros aurus gobiensis, and possibly Iguanodon , an intermediate state is observed where the lat eral margin of the acetabular surface is lower such that it ex­tends only about halfway to the level of the base of th e preacetabu­lar pr ocess, and the medial margin is barely arched.

In early hadrosaurids, the suprailiac crest is distinctly triangu­lar, with a nearl y vertical caudal margin in lateral view. It is some ­what pendant and elongate , often traversing nearl y half of the dis­tance from the dorsal to the ventra l border of the postacetabular process in lateral view. The suprailiac crest in these taxa has a low, rounded ventral buttress extending from the lateral surface of the iliac blade, probably representing the intermuscular line between M. iliofemoralis and M . iliofibularis (Dilkes 2000). In Barsboldia, Claosaurus, Secernosaurus, "Kritosaurus " australis, unn amed ilia illust rated by Young (1958), and all hadrosaurids, the suprailiac crest is more cranially located such th at its caudal edge is over the ischial tuberosity on the lateral face of the ischiadic peduncle, and the cranial edge is nearly over the center of the acetabulum, as it is in hadrosaurids. In all of these ta xa except Claosaurus, the "an­titrochanter" is stro ngly developed, the postacetabular process is

Appendicular Anatomy in Campanian and Maast richtian North American Hadrosaur ids • 147

However, the limited data available suggest that the average ratio for hadrosatirs is around .53% = 7%. It is unclear if this ratio changes with growth. Among hadrosaurs, the angle of the preac- ctabular process in lateral view relative to the main body is greater in taxa with particularly deep ilia, -35 degrees or more (e.g., Bac- trosqurus, I'arasaurolophns, Hypaarosaurus), and it is shallower. 20-27 degrees, in taxa with shallow ilia (e.g., Ediiumlosjurus). As discussed below, apomorphic reduction ol the pubic peduncle in some taxa may increase this angle to near 40% (e.g.. Kiitas.iuriis). However, the difference between ilia from two sides of the same an- imal may be 5 degrees or more (e.g., Morris 1978). Individual vari- ation is also high, ranging from 20 to over 30 degrees in juvenile Mjusaiira (Carpenter 1999: fig. 12.29), and from 25 to over 35 degrees in Carythosuttrtts (Brown 1916; Park* 1935; possibly Giimore 1946). The degree of deflection of the preacetabular process is much greater than in any other iguanodontians, but the degree of deflection is too variable to be of taxonomic use.

In Iguatiadon, the medial surface of the ilium is relatively shal- lowly embayed for the acetabulum, while the lateral surface is deeply embayed ventrall) by an acetabular fossa having a steeply sloping caudal margin and a slightly produced marginal "lip" dor- sally. This process is more properly termed the "anntrochanter." The iliac acetabular fossa extends laterally, almost to the level of the base of the preacetabular process, and "cups' the femoral head medially as well as dorsally. In hadrosaunds, both the medial and lateral margins of the acetabular fossa are lower such that they are nearly at the same level, and the acetabular margin of the ilium is only slightly arched in lateral view. The acetabular surface of the ilium contacts the femoral head only dorsally. and the acetabular articular surface is only barely visible in lateral view. In Ouni- Hosaurits, Rachosjuriii, Allirhiitus, Prubacirosaurui gobieasis, and possibly IguauoJon. an intermediate state is observed where the lateral margin of the acetabular surface is lower such that it ex- tends only about halfway to the level of the base of the preacetabu- lar process, and the medial margin is barely arched.

In early hadrosaunds, the supraihac crest is distinctly triangu- lar, with a nearly vertical caudal margin in lateral view. It is some what pendant and elongate, often traversing nearly half of the dis- tance from the dorsal to the ventral border of the postacetabular process in lateral view. The supraihac crest in these taxa has a low, rounded ventral buttress extending from the lateral surface ol the iliac blade, probably representing the intermuscular line between M. iliofemoralis and M. iliofibulans (Dilkcs 2000). In Barsbuldu, ClaasJiitHS, Seceninsiiurtis. "Kritosanrits" australii, unnamed ilia illustrated by Young (1958). and all hadrosaunds, the supraihac crest is more cranially located such that its caudal edge is over the ischial tuberosity on the lateral face of the ischiadic peduncle, and the cranial edge is nearly over the center of the acetabulum, as it is in hadrosaunds. In all of these taxa except Cfaqsaurus, the "an- titrochanicr" i* wrongly dcvcliiped, ihc powacdaniilar pfneess it

Appeniliuil.it Anatomy in i .uiip.inuti .mil Maastnclnian North American I iadrosaurids • 147

TABLE 8.1. Relative compari son of the hadrosaurid ilium

Ch aracter istic

Preacetab ular process rela­tively deflected vent rally

LIH ra tio of ilium, ave rage (SD)

Postaceta bular process LIH ratio, averag e (SD)

Iliot ibialis process extends to meet peduncle

Ischial and pu bic peduncl e

Hadrosaurine

Less (average, 32 degrees)

5.41 (0.05)

2.46 (0.14)

Ra rely

Less ro bust

morphocline.

Lamb eosaur ine

More (average, 42 degrees)

4.43 (0.5)

1.88 (0.25)

Often

Relativel y more ro bust

~ 2ae ..

slightly displaced ventrally, and th e dorsal margin of the ilium is strongly " no tched" accordingly. The development of the supra iliac crest is size related with a stro ng overprin t of individual variation (Brett-Surman 1989; Lull and Wright 1942).

Five morphotypes of ilia are recognized am ong the five clad es of hadrosaurids (" tribes" fide Brett-Surm an in Cha pman and Brett ­Surman 199 0). Th ese morph otypes grade into one another alo ng a morphocline and are not useful taxonomically (Morris 1978) . The two extremes on the morphocline are the ilium of Edmontosaurus and that of Parasau rolophus. Both display hete rochronic and onto­genetic tr ends (Table 8.1).

The first morphotype is represented by Bactrosa urus and Gilmoreosaurus (Fig. 8.4F, G). Th e pr eacetabular process is ven­trally deflected as in later had rosaur ids. Both ischial and pubic pe­dun cles are structured in typical hadrosaurid fashion, as is the M. iliofemoralis depression. Th e postacerabular process has been greatly modified from the camptosa urid (Gilmore 190 9) and Iguanodon (Norman 1980) condition where this process is rudi ­mentary (if pr esent at all), but it is not as well developed as in the more advanced had rosaurids Ed m ontosauru s and Parasaurolo­ph us. The suprai liac crest is rud imentary and is more of a laterally projecting lip on the dorsal margin of the ilium as in Iguanodon. In ot her hadrosaurids, it is a mass ive process (compare Bac trosaurus in Fig. 8.4F to Parasaurolophus in Fig. 8.4B). The do rsal margin is curved in the typical hadrosaurid fashion.

A second mo photype is exemplified by Edmontosaurus (Fig. 8.4A) and A na totitan . In these genera, the ilium is greatly elon­gated and not relati vely as high as in other genera, resulting in a length/height (LIH) ratio that is grea ter th an in any oth er iliac type. In lat eral view, the ilium is th in and greatly elongated, in co ntras t to th e mo re robust app earance as in Parasaurolophus. The postac­

148 • Michael K. Brett-Surman and Jonathan R . Wagn er

IAHI E R.I. Relaiivc comparison of the hadrosaurid ilium morphoclmc.

Characteristic

Prcacetabular process rela- tively deflected vcnirally

I/l-l ratio of ilium, average »SD)

Postacetabulai process I/H ratio, average (Sl)i

lliotibialts process extends to meet peduncle

Ischial and pubic peduncle

Hadrosaurme

l*% (averao. U degree*)

S.4I (0.05)

2.4f, 1O.H1

Rarely

Less robust

1 amheosaurinc

More (average, 42 degrees)

4.43 (OJ)

I.NN (0.25,

Often

Relatively more robust

slightly displaced vcntrally, and the dorsal margin of the ilium is

strongly "notched" accordingly. The development ol the suprailiac

crest is size related with a strong overprint ol individual variation

(Brerr-Surman 1989; Lull and Wright 1942).

Five morphotypes ol ilia are recognized among the five clades

of hadrosaunds ("tribes" fide Brett Surman in Chapman and Brett-

Surman 1990;. These morphotypes grade into one another along a

morphoclmc and are not useful taxonomically (Morris 1978). The

two extremes on the morphoclmc are the ilium ol F.dimmlosaurus

and that of Parasaurolopbus, Both display hcterochronic MU\ onto

genetic trends (Table 8.1).

The first morphotype is represented by Baclrnsaurus and

(iiii)ioreosiitiriis (Fig. S.4F, G). The prcacctabulai process is vcn-

trally deflected as in later hadrosaurids. Both ischial and pubic pe-

duncles are structured in typical hadrosaurid fashion, as is the M.

iliofemoralis depression. The postacetabular process has been greatly modified from the camptosaund (Gilmorc 1909) and

Iguanodon (Norman 1980) condition where this process is rudi-

mentary (if present at all), but it is not as well developed as id the

more advanced hadrosaurids Ldtiinntosaitrns and Parasaurolo-

pbus. The suprailiac crest is rudimentary and is more ol a laterally

projecting lip on the dorsal margin of the ilium as in Igiuwodoit. In

other hadrosaurids, it is a massive process (compare Rnctros.iunis

m Fig. X.4F to Parasaurolopbus in Fig. 8.4BJ. The dorsal margin is

curved in the typical hadrosaurid fashion.

A second mophotype is exemplified by Edntotttosaurus (Fig.

S.4A) and Atmlntitiiii. In these genera, the ilium is greatly elon-

gated and not relatively as high as in other genera, resulting in a

length/height (L/H) ratio that is greater than in any other iliac type.

In lateral view, the ilium is thin and greatly elongated, in contrast to

the more robust appearance as in Parastiurnluphus. The postac-

I4S • Michael K Brett-Surman and Jonathan R. VPagnci

rphocline.

Lambeosaurine

M ore (average, 42 degrees)

4.43 (0.5)

1.88 (0.25)

Often

Relatively more robust

::-'n of the ilium is , : of the suprailiac _,: \ 'idual variation

g the five clades - d man and Brett­, - ~ another along a _{orris 1978). The

: Edm ontosaurus - ronic and onto­

.ictrosaurus and at process is ven­

.ai al and pubic pe­";"l ion, as is the M. ;--ocess has been more 1909) and ' 5 process is rudi­

- ~ \'elop ed as in the and Parasaurolo­

ore of a laterally .:.: in Iguanodon. In

-;ar e Bactrosaurus .=. dorsal margin is

.ontosaurus (Fig. <:1 is greatly elon­era, resulting in a

-- :- other iliac type. - .ed, in contrast to _ tu s. The postac­

etabular process also has the greatest LlH ratio of any clade, and the body is relatively shallow.

The third morphotype is seen in Kritosaurus (Fig. 8.4C), where the ilium is somewhat similar to the second morphotype in lateral view but has a relatively taller body, and the ventral deflection of the preacetabular process is greater. This deflection is more pro­nounced than in any other clade. The postacetabular process also appears to be deflected dorsally, giving the overall aspect of the ilium in this clade an accentuated "sigmoidal" curve. This feature is common to this clade and to the saurolophs.

A fourth type of ilium is exemplified by Corythosaurus and Saurolophus (Fig. 8.4D), in which the LIH ratio is smaller than that of the pr evious types. The preacetabular process is usually more ventrally deflected as in the Kritosaurus clade. The postacetabular process has a smaller LIH ratio than in an y other hadrosaurine dade.

The fifth morphptype of ilium is represented only by Parasauro­lophus and Tsintaosaurus (Fig. 8.4B). As in the previous type, the preacetabular process is strongly deflected ventrally, but it is thicker and wider than in all other clades . The body is the deepest of any lin­eage, and the postacetabular process has the smallest LIH ratio of any lineage. The suprailiac crest is relatively more massive and ex­tends farther laterally than in any other clade.

In summary, in morphotypes 2 through 5, the ilium becomes progressively more massive. The LIH ratio of the element as a whole decreases while the suprailiac crest and the postacetabular process increase in robustness. This is a consequence of an increase in the relative and absolute size of the suprailiac crest. The result is increased surface area for the attachment and size ofthe M. iliotib­ialis, M. iliofemoralis, and M. caudifemoralis externus complexes in the lambeosaurines. This would make the legs relativel y more powerful than those of hadrosaurines with the same femoral length, but it does not imply greater speed.

Finally, ontogenetic characters among the ilia are surprisingly few considering the diversity of form s. The ilium maintains its gen­eral shape and dimensions throughout life. Juveniles can be recog­nized only by their absolute size. In very old individuals, for exam­ple USNM 3814, the preacetabular process has thickened, as has its dorsal rim and the medial ridge that extends onto the preacetab­-ular bar (Fig. 8.5C, shown in medial view) for articulation with the sacrum and the origin of M. puboiliofemoralis internus. This has proceeded to such an extent that pre acetabular bar has changed from a narrow cross-sectioned blade to a thickened bar with .a T­shaped cro ss section.

Pubis. The hadrosaur pubis is one of the thinnest, most fragile bones in the skeleton, and an undamaged specimen is very rare. The thin, frail prepubic margin and the shaft of the postpubis are rarely preserved intact. Postmortem distortion and .perspectivedif­ferences exaggerate an apparent high degree of individual variation in this element. Fortunately, the small sample of juvenile and

Appendicular Anatomy in Campanian and M aastrichtian N orth American Hadrosaurids • 149

euhular pr,k.c» al,o ha), ihe greaiew I/I I ram, of au> cLdc. ui.d ihe hotly i> relaiiveb shallow.

ITie third m,irph(,i)pr is x.en m KrifowwrH* iHig. 8.4LI, where ihe ihum i» >omewhai similar i„ the *evuud morphotype in lateral view hiii ha» 4 relatively uller biidy, and (he vciiiral dcllection (if ihe preaceiaUular pnxess i» greater. "ITii» delleciion i» more pfl^ mum.., J man in any mher clade. Ihi piw,iacetabular pr,,.es» ulw appear* in he deflected dur&ally, giving the overall aspect of ihe ihum in thi> vlade an avCLiuuated "»igmoidar curve. Thi* feature :» uumnon to ihi» clade and lo ihe wuroloph,.

A luurih type of ilium i^ exemplified hy (:oiyf/;i,fjnni$ and S.IKMI/U/'/'.'I^ (pjg. 8.41)1. in which the I./I I ram: i» smaller than thai ut ihe pre\iuu» typc». The prcacetahular prnce»» w u&ually mure venirally dellecied a* in the knnwjm:^ clade. The pn#iacclabular proce%> ha> a smaller Ull rano ihan m any other hadrnsaurine Jade.

The litih morphpiype of ihurr. i% repmented only hy Pm..j<»m> Aip/iw# and fawf.H,i.mnia ihig. N.4H|. A» m tin prevmuk tv|X, ihi prcaceiahuiar pruce*> i» tinmgly di-AccicJ vcntrally, hui it i* thicker and wuler ihan in all other i la Jet. Ihe body i* ihe dec;ic»i ui air. I;;, cage, and the postaceiabular prnca» hat the smallest 1 'II ran,, or any lineage. Ihi ,upiaileiL cre*i i» relatively mure ma»wve and ex- tend* lanher laterally than in any other clade,

In nummary, in murphoiype* 2 through i, ihe ilium hecume* progressively mure mawive. I he LI I ratio id ihe element a» a whiile decreases while the supradiac iresi and the pii*i.iCL tabular pr.ke." uiLiea^: m robusi;ies.\. Ihis 1, .1 consequence ot an mcie.iM in lhe relative and absolute size of the »uprailiac uesi. The result is mi rented -.uiface aiea liu ihe ituclimcnt and ti/e id the M. ilioiih- 1 alls, \l. iiiofemoralis, and M. caiiddemurali* cxtcrmis complexes m the lambeosaurmo. I hi* would make the leg* relatively more powerlul ihan thnw. of hadrusaurmc* with the wine lemnrjl length, bin n duo mn imply greater speed.

I inally, ,,m,,genenc character* among the ilia are siupri>i.... lew u,n*iJenng the diverkiiy uf lurm&. The ilium mjinram* it* gen- eral shape and dimcn»mn» throughmu life. Juveniles can be recog- tiued imly by their ahsoluiL >i/c. In Mry u|J individuals, for exain- ple I SN\| 3NH, the pre.iLCubular proccw hj:, thickened, a:, ha, ii^dorwl rim aiidlliemedul riilgelh.it extends 01,u the pre.i^eiah ular bar (Fig. N.f(_, shown in medial wew 1 tor ammlaiion with the sacrum mid the origin f)| M. puboilioiemorali* inicrnus. This has prikteded lu »uch an extern lhai preacetahular bar ha» changed from a iwrruw &;»**» »eviiuncJ Wade m a thickened hir with .1 T ,haped cr(»» Motion.

f'n/'iy I he hadro^iur pubi» 1, one ol the thmnew, mo*i fragile bones in the skeleton, and an undamaged specimen is very r.,n The thin. Irail prepubi*. margin and ihe *hali of ihc pu»ipubi» are rarely preserved intact. Ho»tinnricm disiortion and perspective dil lerence» exaggerate an apparent high degree or individual varimum in ih,» element. Kornmaiely, the wnall sample of ;uvemlc and

Appcndicular Anaiour. ;n | .irapaiuan aiul M.i.;>;.- un;i.in North Amen...;. :i. . .-.. ..ovls • 14'*

I,~ , PUN

ACT_ ;( "-·

'" ~...;;~'.v~ ·~':-" ; 1 ~. A I

PUB

F

Figure 8.5. Comparison of righ t subadult prepubes ava ila ble suggests th at prepubic mo rpohology in pubes ill lateral view. (A) juveniles is con sistent w ith that in adults. Unfortunately, specime ns Parasaurolophus FMNH P2 73 93 on exhi bit have a substantial amo unt of restora tion of the pubis (lengt h of prepubic blade =430

and may easily mislead researchers (Brett-Surman 1975, 1989; mm); (B) Corythosau rus AMNH 5240 (leng th of prepubic blade = D ilkes 1993). 490 mm); (C) ProsauroJophus The p ubi s (Fig. 8.5) is divided into three portions. The most ROM 787 (length of prepubic posterior sect ion co ntains the postpubic process, which extends blade =533 m m); (D) Kritosaurus posterioventrally alon g the ventral margin of th e ischium. It is rod­RO M 764 (length o f prepubic

like and often vestigia l. The middle section forms the acetabular blade =534 m rn}; (E) Edmont osaurus D MNH 1493 margin, wi th the posteriorly projecting isch ial peduncle and the (length of prepubic blade =630 dorsally pro jecting iliac peduncle. Th e most anterior portion is the mm}; (F) Ca mptosaurus YPM prepubic process, comprising a "neck" and "blade" (Fig . 8.5A). 1880 (length of prepubic blade = The p repubis is likely the origin of hypaxial mu sculature and 336 m m}.

M. ambiens (Hutchinson 20 01a; Romer 1927). Carr ier and Farmer (2000) proposed that the pr epubic process had a role in mechanical respi ration among orn ithisc hians, a connect ion previously pro­posed by Brett-Sur ma n (19 75 ) for had rosaurs . Crocodiles use th eir M. rectus ab dominis to push the liver toward th e lungs as a pseudo d iaphragm , and if hadrosaurs breathed in a sim ilar man ner, then a correlation may exist between complexity of the narial apparatus and size of th e M. rectus abdominis and its origin on th e prepubis. This correlati on is exactly what is found in hadrosaurids. T he Parasaur olophus cla de has the most co mplex nares and the thi ck­est, shortest , and widest prepubic blade. The orientat ion of th e blade is directl y in line with th e ribcage so that the for ce on th e pre­pubis from muscular contraction is tr ansmitted along its longitudi­nal axis, thereby minimizing lateral components of fo rce. The lat­

150 • Mich ael K. Brett-Surman and Jonathan R. Wagner

s the oth er da.

Parass ur ~ ~ b is (Fig. wider man in ~

rectan gle wi±. ­~e dorsal and ~:e ri o r bo r.i er ~__ d is r ela j\­ ~ : .

eral forces rha : ::-:J

the puboischio re rib s and carti lage

The length ilium in hadross rnontosau rus, ~

prepubic length . 62%. Hadrosaur larnbeosaur ines _. deli neat e clades,

There arc .: sponding to b -:: the refore, corn~ 2J

along a mo rpho The first ~r­

also by the he ~

longer and expanded in G'

The surface area : est of any clade .

Kritosaurus .: This type is noca rral bord ers. ! :i.::

ventrally, The ~.:

tosaurus. The third :: s

contains a dorsa porti on an over border projects . ro unded, gra:::u,: Straight. The ­

adrosaurine-so Corytboss s

The bla de clos proJectmg bUlge

ent a more :::i":'

gular one. Tse abruptly as ir; 5 subparallel, a

uch short er, . ~

Figure S.S. Comparison »/ right pubes in lateral view. lAj kT*muru*ophMfMNHP27WJ (length ill prepubic blade « •! 10 mm); (D) Corythosiurus AMNH 5240 /length ol prepubic blade = •190 mm}; IC) Proiaurolopltut ROW 787 /length ofpnpublt blade - t|J mm!; IO\ KtitOSBUnU ROW 764 fkdpA ("/pwpiiAif

Ednumlosaurus DMNII 1193 (length i>l prtpubit blade - 630 mml; (F> Campimaurus VPAI I MO (length ofpnpubit blade = 336 mml.

subadulr prepubes available suggests that prepubic morpohology in juveniles is consistent with thai in adults. Unfortunately, specimens on exhibit have a substantial amount or restoration ot the pubis and may easily mislead researchers (Brctt-Surman 1975, I9NV; Dilka IV9.1).

The pubis (Fig. 8.5) is divided into three portions. The most posterior section contains the postpnbic process, which extends postenoventrally along the ventral margin of the ischium. It is rod- like and often vestigial. The middle section forms the acetabular margin, with the posteriorly projecting ischial peduncle and the dorsally projecting iliac peduncle. The most anterior portion is the prepubic process, comprising a "neck" and "blade" (Fig. 8.5At.

The prepubis is likely the origin of hypaxial musculature and M. ambiens (Hutchmson 2001a; Romer 1927). Carrier and Farmer (2000) proposed that the prepubic process had a role in mechanical respiration among ornithischians, a connection previously pro- posed by lirert-Surman (197J) tor hadrosaurs. Crocodiles use their M. rectus abdominis to push the liver toward the lungs as a pseudo diaphragm, and if hadrosaurs breathed in a similar manner, then a correlation may exist between complexity of the narial apparatus and size of the M. rectus abdominis and its origin on the prepubis. This correlation is exactly what is found in hadrosaurids. The Parasanrohphus clade has the most complex nares and the thick- est, shortest, and widest prepubic blade. The orientation of the blade is directly in line with the ribcagc so that the force on the pre- pubis from muscular contraction is transmitted along its longitudi- nal axis, thereby minimizing lateral components of force. The lar-

150 • Michael K Btett-Sutman and Jonathan R- Wagner

I

~. ,. ...>.: ·· 7.: <,:

: ~

. ":: ~: .":.- ., ~

. ._......,..•.•-.

. . . " :.

. ....;:-. ~ ...

. ic morpohology in ..mately, specimens

ration of the pubis an 1975, 1989;

rtions. The most . : 5, which extends : ischium. It is rod­

s the acetabular peduncle and the

erio r portion is the e" (Fig. 8.5A). musculature and

:=arrier and Farmer role in mechanical r; previously pro­-·ocodiles use their e ungs as a pseudo

-r manner, then a narial apparatus on the prepubis.

- drosaurids. The - s and the thick­

ientation of the 'orce on the pre­ng its longitudi­

of force. The Iat­

eral forces that exist as a result of the acti ons of the am biens and the puboischiofemoralis externus muscles are braced by the sacral ribs and cart ilage (Maryariska and Osm6lska 1981 ).

The length of the prepubis averages 50 % the length of the ilium in hadrosaurids. There is noted variability, especially in Ed­m ontosaurus, from 46% to 60% of iliac length . The ratio of the prepubic length to height is quite variable, ranging from 26% to 62% . Hadrosaurines on average have a ratio less th an 47%; that of lambeosaurines is more than 47%. There is too much overlap to delineate clades.

There are five different pubic morphologies (Fig. 8.5) corre­sponding to five clades. All these types grade int o one another; therefore, comparisons can only be mad e between the extremes along a morphocline.

The first type is exemplified by Edmontosaurus (Fig. 8.5E) and also by the hadrosaurine Gilmoreosaurus. The neck is relatively longer and thinner than in other clades. The blade is symmetrically expa nded in most cases, with a slight emphasis of the dorsal side . The su rface area of the blade (in lateral view) is relati vely the small­est of any clade.

Krit osaurus characterizes th e second type of pubis (Fig. 8.5D). Thi s type is notable for having a blade with par allel dorsal and ven­tral bo rders. The prepubic part of this element is strongly deflected ventrally. The neck is relatively shorter and wider than in Edmon­tosauru s.

The third type is typified by Saurolophus (Fig. 8.5C). The blade contains a dorsally projecting asymmetr ical bulge, which gives this portion an overall triangular shape in lateral view. The anterior border projects sharply in the ventra l dir ection. The tip is well ro unded, grading gradually into the ventral border, which is stra ight. The neck is relatively shorter and wider than in other hadrosau rines.

Co rythosaurus represents the four th type of pubis (Fig. 8.5B). The blade closely resembles that of Saurolophus but the dorsally projecting bulge is more flattened, giving this portion of the ele­ment a more trapezoidal aspect in lateral view rather than a trian­gular one. The dorsal border is deflected ventrally but not as abruptly as in Saurolophus. The dorsal and ventral borders are subparallel, almost as in Krit osaurus, but the blade is relatively much shorter. The anterior border is well rounded and longer than in the other clades, and the neck is shorter and wider.

Parasaurolophus and Bactrosaurus represent the fifth type of pubis (Fig. 8.5A ). In this type, the blade is relatively shorter and wider than in other clades. In lateral view, the blade resembles a rectangle with the longer axis of the rectangle oriented vertically. Th e dorsal and ventral bord ers are subparallel in most cases. The anterior border is perpendicular to the dorsal and ventral borders and is relatively longer than in the Corythosaurus-like pubis. The neck is very short and relatively wider than in other clades, giving it a robust appearance. One unique differ ence of this clade is the mas-

Appendicular Anatomy in Campan ian and Maastrichtian North American Hadrosaurids • 151

era! forces that exist as a result of the actions of the amblens and the puboischiotemoralis externus must lei arc braced hy the sacral ribs and cartilage iMaryariska and Osmolska 1981).

The length of the prcpubis averages 5(1% the length ol ihe ilium in hadrosaunds. There is noted variability, especially in Ed- montosattrus, from 46% to 60% of iliac length. The ratio of the prepubic length to height is quite variable, ranging from 26% to 62%. I ladrosaurmes on average have a ratio less than 47%; that of lambeosaurincs is more than 47%. There is too much overlap to delineate clades.

There are five different pubic morphologies (Fig. 8.5) corre- sponding to five chides. All these types grade into one another; therefore, comparisons can only be made between the extremes along a morphocline.

The first type is exemplified by F.dntonlosaiirus (Fig. 8.5E) and also by the hadrosaurine Ctilmoreusjitrm. The neck is relatively longer and thinner than in other chides. The blade is symmetrically expanded in most cases, with a slight emphasis ol the dorsal side. The surface area of the blade (in lateral view) is relatively the small- est of any chide.

Knfofjwnw characterize: rhc second type of pubis (Fig. 8.SD). This type is notable for having a blade with parallel dorsal and ven- tral borders. The prepubic part ol this element is strongly deflected ventrally. The neck is relatively shorter and wider than in Edition tosjiiriii.

The third type is typified by StMrolophus (Fig. 8.50). The blade contains a dorsally protecting asymmetrical bulge, which gives this portion an overall triangular shape in lateral view. The anterior border project* sharply in the ventral direction. The tip is well rounded, grading gradually into the ventral border, which is straight. The neck is relatively shorter and wider than in other hadrosaurines.

Corythosaurus represents the fourth type of pubis (Fig. 8.5B). The blade closely resembles that of Sjurolophus but the dorsally projecting bulge is more flattened, giving this portion of die ele- ment a more trapezoidal aspect in lateral view rather than a trian- gular one. The dorsal border is deflected ventrally but not as abruptly as in SauroIophliS. The dorsal and ventral borders are subparallel, almost as in Krilosaiiriis, but the blade is relatively much shorter. The anterior border is well rounded and longer than in the other clades, and the neck is shorter and wider.

I'.ir.isnniolofibus and Biictrusiiurus represent the fifth type of pubis (Fig. 8.5A). In this type, the blade is relatively shorter and wider than in other clades. In lateral view, the blade resembles a rectangle with the longer axis ol the rectangle oriented vertically. The dorsal and ventral borders are subparallel in most cases. The anterior border is perpendicular to the dorsal and ventral borders and is relatively longer than in the Corytbosaimis-Uke pubis. The neck is very short and relatively wider than in other clades, giving it a robust appearance. ( me unique difference of this clade is the mas-

Appendicular Anatomy in C.imp.ini.in and Maastriehtun North American Hadrosaurids • 151

aspA SHF

ILP ISFH

c= ~ ~

PUP ~D" " "'-'~

~.,..:..r~. . . :: : : ~ ~: .·k; ..< I~~ . "~: .. : ~.

" ." , .: ..' . . ' . '," .r;:::. , ...~ B ~

--~h~ ....

~f ~ ' " ''

.' ,

~---,~ C

:....~. ,.:";,,

Figure 8.6. Com parison of the siveness of the iliac peduncle. The articular surfa ce is deep, cuplike, right ischia in lateral view, (A) and well rounded. In other clades, the iliac peduncle is long and Parasaurolophus FMNH P27393 thin , with a shallow cup . Parasaurolophus cyrtocristatus (FMNH(length = 1078 111m) ; (B)

P27393) displays a large ridge that extends from the iliac peduncleCorychosaurus AM N H 524 0 (length =1030 mm); (C) to the ischial peduncle. This ridge is not as well developed as in the Shan cungosaurus (length = 1602 oth er genera. Th e type of Parasaurol ophus walkeri has a patholog­m m) redrawn from Hu (1973) ; ical fusion of the pubis and ilium at the pubic peduncle. (D) Anatocican AMNH 5730 In summary, the pubis increases in robustness with each of the(length = 116 0 mm); (E)

clades by increasing the surface area of the blade and enlar ging theGilmo reosaurus AMNH 6551 (length = 718 mm}; (F) attachment with the prepubic neck by increasing its width and de­Ca mpcosaurus YPM 1878 (length creasing its length . The most gracile forms (the edmontosaurs) have =553 m m }, the most elongate prepubis compared to animals of comp arable

size in other clades. The most robust forms (the parasaurolophs) have the shortest prepubis.

Ju veniles have pubes with the same morphology as the adults. It is remarkable that had rosaurids show such a uniformity of pat­tern throughout the ontogeny of the individual. The postpubis, be­cause of its lack of distinctive features and rare preservation, pro ved to be of no value for the diagnosis of sup rageneric clades.

Ischium. The ischium projects at approx imately a 45-degree angle posterio ventrally from the acetabulum and is composed of three portions. The most proximal porti on contains the acetabulum, the iliac and pubic peduncles, and the obturator process and fenes­tra. The middle portion conta ins the ischial shaft, which is columnar and featureless. The most distal porti on contains the terminal knob or "foot" (Fig. 8.6E), comp osed of the "heel" and "toe" (Fig. 8.6A).

152 • Mich ael K. Brett-Surman and Jonathan R. Wagner

C;..e ~ 91~ ~ "2.=',,: - ­

o r tne .:.;.:. ~~

.-2.:.-ger 5 ;--o...~

.=.:-ot:n~ 5:: .. --= -~ -- 0 : ~~ ­

_ : _~ _ _L . _ _._ -..,,;,..0 ::: ~.~ _

ISFH

ISFT ^ OBP OBF

-L#"

Figure S.6. Comparison of the right ischia in lateral vieu: lAI

(length? I078 mm); IH)

Conthow* AAINH M40 fkgf* « l&M mm*;; fO Sli.inlungosaurjs /length e 1602 mm) redrawn from Hu (1973); ff); Anawuwo A.WNN < : /length* 1160 mm); (h) Ctilmorcouurus AMSH 653i

CampwwunM YfM fS7» (k^A

siveness of the iliac peduncle. The articular surface is deep, cuplike, and well rounded. In other clades, the iliac peduncle is long and thin, with a shallow cup. fdrafdMro/opAwf fyrfofrw/afM* (FMNH P27393) displays a large ridge thai extends from the iliac peduncle to the ischial peduncle. This ridge is not as well developed as in the other genera. The lype of Pdrjamro/optuf ww/*fr* ha* a patholug- ical fusion of the pubis and ilium at the pubic peduncle.

In summary, the pubis increases in robustness with each of the clades by increasing the surface area of the blade and enlarging the attachment with the prepubic neck by increasing its width and de- creasing its length. The most gracile forms (the cdmontosauis) have the most elongate prepubis compared to animals of comparable size in other clades. The most robust forms (rhe parasaurolophs) have the shortest prepubis.

Juveniles have pubes with the same morphology as the adults. It is remarkable that hadrosaurids show such a uniformity of pat- tern throughout the ontogeny of the individual. The postpubis, be- cause of its lack of distinctive features and rare preservation, proved to be of no value for the diagnosis of suprageneric clades.

Ischiwn. The ischiun) projects at approximately a 4.5-degree angle postcrioventrally from the acetabulum and is composed of three portions. The most proximal portion contains the acetabulum, ihe iliac and pubic peduncles, and the obturator process and fenes- tra. The middle portion contains the ischial shaft, which is columnar and featureless. The most distal portion contains the terminal knob or "foot" (Fig. S.6F), composed of the "heel" and "toe" (Fig. S.6A).

152 • Michael K. Brctt-Surman and Jonathan R. Wagner

".....~,

. ,~ ". :'

. , i' .

0 ~-:- , - '-·---s.';:: ..' "".' ,:.',

D dlt

ce is deep, cuplike, : ~uncl e is long and - r zrista tus (FMN H c::. - e iliac peduncle ce veloped as in the eri has a pa tholog­

-= uncle. -: 5 with each of the ~ and enlarg ing the '= its width and de­

ec ontosaurs) have , - of compara ble

._ , parasauro lophs)

iogy as the adults. uniformity of pat­

_The pos tpu bis, be­rare preservation ,

__ =- rageneric clades. - ely a 45-degree

- . is composed of ~5 the aceta bulum, _ _rocess and fenes­~ \'h i ch is columnar - :be terminal knob

- '"roe" (Fig. 8.6A).

Use of the ischium in had rosaur systematics dates to Brown (1914), and the "footed" ischium is often used and misused in hadrosaur systematics, the . Th e presence of a " footed ischium " is the peromorph ic condition (Brett -Surma n 1989 ). Distal expansio n of the caudal end is found in all iguanodontians. Lack of a " footed ischium" is th e derived, neotenic condition that characterizes the hadrosaurines. Altho ugh other features of the hadrosaur ischium are frequ ently obscured or altered by distortion, damage, and in­complete ossification, th e ischium is the most resistant bone in the hadrosaur pelvis.

Th e degrees of development of the "foot" has been discussed by Brett-Surman (197 5, 1989), Godefroit et a1. (2000), and Casa novas­Cladellas (1999). With increasing size, the distal expansion acquires certain characteristics in a moderately consistent order. In graci le is­chia, the minimal thickness is approximately three-quarters down the shaft. In robu st taxa with very low length/foot ratios, the foot is very small, often appea rs flat-bottomed, and may form an acute angle with the shaft (poss ibly related to the shallowing of the ischial-vertebral angle in had rosaurs). In ischia with a ra tio of 8:1 to 6:1, the curved "ball" of the foot develops, and there may be a sma ll inflection of the dorsal ridge of the ischial shaft where it curves into the sale of the foot, forming a slight "heel" in larger specimens . In larger specimens in this range (ischial ratio 7:1 to 6:1), the " ball" of the foot may migrate ventrally, beveling the "heel" area. In ischia of adults with rat ios between 6:1 and 4:1, the " heel" may increase in pr ominence as a slight flange, but it is still relati vely sma ll. At around 5:1, the tip of the foot may extend cranioventra lly past the "ball" of the foo t as a str aight , dorsoventally narrow, elongate "toe." At his point, authors seem to be in agreement that the distal expansion is to be termed "club-shaped " as opposed to "foo ted" (e.g., Casanovas-Cladellas et a1. 1999), although this distinction is arbitrary. These ischia have a "sock, " a th ickening of the distal half of the shaft that accentuates the dorsal curvature of the bone and shifts the locat ion of minimal shaft thickness to near the middle of the shaft. In ischia with length/foot ratio of less than 4:1, the heel is expanded and protrudes dorsally and distally in lateral view, the toe is well developed, and the "sock " extends about two-third s up the shaft, such that the thinnest point is just distal to the obturator proces s, and the entire distal shaft is a thick ened club. Stem hadrosaurs and iguanodontian s therefore cover the entire range of ischial foo t size. Because the relat ive size of the ischial foot is onto­genetically varia ble and the shape of the foot is dependent on rela ­tive size, comparisons of "clubbed" versus "footed" ischia (e.g., Casanovas-Cladella s et a!. 1999) will not unequivocally separate a stem had rosaur from a nonhad rosaur.

The ischial symphysis is redu ced to a plate of bone surrounding the ischial obturato r foramen (Hutchinson 2001a) . In many taxa, including tetanuran theropods and orn ithischians, the obturator foramen was "o pened" by reduction in the ossificat ion of its outer margin; its ventral margin becam e the obtu rator process, and this is

Appendicular Anatomy in Cam panian and Maastrichtian No rt h American Had rosaurids • 153

Use of the ischium in hadcosaui systematic! dates to Brown 11914), nncl the "tooted" ischium is often used and misused in

hadcosaui systematic*, the. The presence of a "footed ischium" is the peromorphic condition (Bretr-Surman 14*9). Distal expansion

of the caudal end is found in all iguanodontians. Lack of a "footed

ischium" is the derived, ncotcnic condition thar characterizes the

hadrosaurines. Although other features of the hadrosaur ischium

are frequently obscured or altered bv distortion, damage, and in-

complete ossification, the ischium is the most resistant bone in the

hadrosaur pelvis.

The degrees of development of the "foot" has been discussed by

ItrccSurman (1975,1989), Godefroit et al. (2000), and Casanova*- Cladellas (1999). With increasing size, the distal expansion acquires certain characteristics in a moderately consistent order. In gracile is-

chia, the minimal thickness is approximately three-quarters down

the shaft, in robust ta.va with very low length/foot ratios, the foot is

very small, often appears flat-bottomed, and may form an acute

angle with the shall (possibly related to I he shallowing ol the ischial-vertebral angle in hadrosaurs). In ischia with a ratio of N: I to

6:1, the curved "ball" of the foot develops, and there may be a small

inflection ol the dorsal ridge of the ischia! shaft where it curves into I he sole of the foot, forming a slight "heel" in larger specimens. In

larger specimens in this range (ischial ratio 7:1 in 6:11, the "ball" of

the foot may migrate ventrally, beveling the "heel" area. In ischia of

adults with ratios between 6:1 and 4:1, the "heel" may increase in

prominence as a slight flange, but it is still relatively small. Ar

around 5:1, the tip of the foot may extend cranioventrally past the

"ball" ol the foot as a straight, dorsoventally narrow, elongate

"toe." At his point, authors seem to be in agreement that the distal

expansion is to be termed "club-shaped" as opposed to "footed"

(e.g., Casanovas-Cladcllas ct al. 1999), although this distinction is

arbitrary. These ischia have a "sock," a thickening of the distal half

of the shaft that accentuates the dorsal curvature of the bone and shifts the location of minimal shaft thickness to near the middle of

the shall, In ischia with length/foot ratio of less than 4:1, the heel is

expanded and protrudes dorsally and distally in lateral view, the toe

is well developed, and the "sock" extends about two-thirds up the

shall, such that the thinnest point is just distal to the obturator

process, and the entire distal shaft is a thickened club. Stem

hadrosaurs and iguanodontians therefore cover the entire range of

ischial loot size, Because the relative size of the ischial foot is onto-

gcnetically variable and the shape of the foot is dependent on rela-

tive size, comparisons of "clubbed" versus "footed" iscrua (eg., Casanovas-Cladellas et al. 1999) will not unequivocally separate a

stem hadrosaur from a nonhadrosaur.

The ischial symphysis is reduced to a plate of bone surrounding

the ischial obturator foramen (I lutchinson 2001a). In many taxa,

including tetanuran theropods and ornithischians, the obturator

foramen was "opened" by reduction in the ossification of its outer

margin; its ventral margin became the obturator process, and this is

Appendicular Anatomy in Cfcmpaman and Maasinchnan North American I ladrosaiirids • 153

separated from the ischiadic peduncle by the obturator notch (Hutchinson 2001a; Novas 1996). In iguanodontians, the obtura­tor process is somewhat variable in development, but shows such remarkable consistency in shape that one suspects a process of pro­gressive ossification of a cartilaginous or other connective tissue precursor. Ossification takes place in discrete stages: in the first stage, the obturator process is a low, triangular fin, and there is a low fin extending ventrally from the pubic peduncle (e.g., Norman 2002). Following this, the distal end of the obturator process swells ventrolaterally into a rugose "postpubic peduncle," diamond shaped in lateral view, that projects somewhat laterally from the tip (e.g., Taquet 1976) and apparently contacted the postpubis (Dilkes 1993). The lateral surface of the portion of the obturator process separating the postpubic peduncle from the ischial shaft is exca­vated by a trough parallel to the shaft. Development of a postpubic peduncle often, but not always, creates a slight notch between the ventral margin of the obturator process and the shaft of the is­chium as in some theropods. This notch is variable among individ­uals, dependent on the degree of ossification of the caudal end of the postpubic peduncle. At around the same stage of development of this structure, the ventral fin on the pubic peduncle produces a mediolaterally thin, hook-shaped spur that grows around the obtu­rator notch, partially enclosing it (e.g., Brinkmann 1984; Dilkes 1993; Gilmore 1917; Rozhdestvensky 1957). This hook is smaller in taxa where the pubic peduncle makes a shallower angle with the shaft (see above). In some cases (e.g., Sternberg 1935), the postpu­bic peduncle grows craniodorsally toward the ischiadic peduncle.

In the final stage, a thin, slightly everted, rugose bar, apparently a continuation of the postpubic peduncle, closes the obturator fora­men ventrally (e.g., Gilmore 1924b). Systematic variation in this character was not observed. Closure of the ischial obturator notch appears to have set in at smaller sizes in some specimens with more "robust" ischia (e.g., Boyd and Ott 2002), but closure also appears to progress with age, such that moderate to large specimens of un­footed ischia have closed or nearly closed obturator foramina (e.g., Brett-Surman 1989; Brown 1913a; Gilmore 1917; Hu 1973; Ma­teer 1981). However, even smaller unfooted ischia may have a com­pletely enclosed foramen (Maryariska and Osm6lska 1984), and the presence or preservation of this feature is variable within species (d. Maryariska and Osm6lska 1984 with Rozhdestvensky 1957). However, a rather small, "gracile" ischium from Big Bend has over half of the foramen encircled, with broken edges indicat­ing more bone was present in a small-size individual (TMM 42309; Davies 1983).

The ventral edge of the ischium is quite thin, probably more so in gracile ischia. In taxa with a shallow angle between the shaft and the pubic peduncle, the flange beneath the latter responsible for closing the obturator foramen is somewhat less robust. Often clo­sure is accomplished by a thin arm of bone that appears particu­larly vulne~able to damage and may be broken off in many speci­

154 • Michael K. Brett-Surman and Jonathan R. Wagner

mens (e.g.. G:,­as th e type _: which the O .:':I

nearly ocd :.:'::~

is eirher lil,\hC'

pathology ~~

orne s pec~~ ­

appears to be- -­ma n 19- 5. ~.; _

Four ~-~-=-s

by Late C :'e:.;,:"o'l

osaurus lE g.. ­tween lgusn c.i: the obturate - ­

ecurved ~.c ­

less man :h..:. : _ ::errrnna ,e; ;.;:;. ~ . ventr a l side .

Th e ;.;:-.:c.-:. 1

saunnes; ::X ::~

the relative 5:'-~

:orn1s. The {)~ _ ...l

:; long, ;::: ~ig~ ­

chium , T r.:: : -".re!y p r esen t,

. ~ ­rscm

~=2J~11~ ~ ..:: 5-~

separated from the ischiadic peduncle by the obturator notch (Hutchinson 2001a; Novas 1996). In iguanodontians, the obtura- tor process is somewhat variable in development, bin shows such remarkable consistency in shape that one suspects a process of pro- gressive ossification of a cartilaginous or other connective tissue precursor. Ossification takes place in discrete stages: in the first stage, the obturator process is a low, triangular fin, and there is a low fin extending ventrally from the pubic peduncle (e.g., Norman 20021. Following this, the distal end of the obturator process swells ventrolaterally into a rugose "postpubk peduncle," diamond shaped in lateral view, that projects somewhat laterally from the up (e.g., Taquet 1976) and apparently contacted the postpubk (Dilkcs 1993). The lateral surface of the portion of the obturator process separating the postpubic peduncle from the ischial shaft is exca- vated by a trough parallel to the shall. Development of a postpubic peduncle often, but not always, creates a slight notch between the ventral margin of the obturator process and the shaft of the is- chium as in some thcropods. This notch is variable among individ- uals, dependent on the degree of ossification of the caudal end of the postpubic peduncle. At around the same stage of development of this structure, the ventral fin on the pubic peduncle produces a mediolaterally thin, hook-shaped spur that grows around the obtu- rator notch, partially enclosing it [e.g.. Brinkmann 1984; Dilkcs 1993; Gilmore 1917; R,:xhd«rvcntky |4i7). This hook i» smaller m taxa where the pubic peduncle makes a shallower angle with the shaft (see above). In some cases (e.g., Sternbcrg 1935). the postpu- bic peduncle grows craniodorsally toward the ischiadic peduncle.

In the final stage, a thin, slightly everted, rugose bar, apparently a continuation of the postpuhk peduncle, closes the obturator lora- men ventrally (e.g., Gilmore 1924b). Systematic variation in this character was not observed. Closure of the ischial obturator notch appears to have set in at smaller sizes in some specimens with more "robust" ischia (e.g., Boyd and Ott 2002), but closure also appears to progress with age, such that moderate to large specimens of un footed ischia have closed or nearly closed obturator foramina (e.g., Brett Surman 1989; Brown 1913a; Gilmore 1917; Hu 1973; Ma teer 19X1). However, even smaller un footed ischia may have a com- pletely enclosed foramen (Maryanska and Osmolska 1984), and the presence or preservation of this feature is variable within species |cf. Maryanska and Osmolska 1984 with Rozhdcstvensky 1957). However, a rather small, "gracile" ischium from Big Bend has over half of the foramen encircled, with broken edges indicat- ing more bone was present in a small-size individual (TMM 42309; Davies 198.1).

The ventral edge of the ischium is quite thin, probably more so in gracile ischia. In taxa with a shallow angle between the shaft and the pubic peduncle, the flange beneath the latter responsible for closing the obturator foramen is somewhat less robust. Often clo- sure is accomplished by a thin arm of bone that appears particu- larly vulnerable to damage and may be broken off in many speci-

154 • Michael K. r»re:i-Stirm.in and Jonathan R. Wagner

(L rurator notch .> s, the obtura­

, ut shows such a pro cess of pro­co nn ective tissue =g:es: in the first

- . and there is a '. :: (e.g., Norman :0· process swells

'1 le," diamond ra lly from the tip

r srpubis (Dilkes _rur ator process

'"J shaft is exca­~ . of a postpubic - h between the shaft of the is­

~ among individ­ae caudal end of .: of development uncle produces a around the obtu­-,.i1 1984; Dilkes . hook is smaller er angle with the -.) -), the postpu­- iadic peduncle. .. ar,apparently - obtu rator fora­variation in this obturator notch

. ens with more sur e also appears

cimens of un­r foramina (e.g., : H u 1973; Ma­

av have a com­sxa 1984), and variable within Rozhdestvensky . from Big Bend

edges indicat­: ITMM 42309;

robably more so :xn the shaft and

responsible for ~ st. Often clo­~ _pears particu­_. - . -:: ill many speer­

mens (e.g., Gilmore 1924b). Even in truly robust specimens, such as th e type of Parasaurolophus cyrtocristatus (Ostrom 1963) in which the obturator process is a large, rectangular flange that nearly occludes the obturator notch, the bar enclosing the foramen is either un ossified or unpreserved. It is also conceivable that pathology may have caused early ossification of this structure in some specimens. However, closure of the ischial obturator notch appears to be so variable as to be systematically useless (Brett-Sur­man 1975, 1989).

Four types of ischia can be recognized. Three are represented by Late Cretaceous forms. The first type is exemplified by Gilmore­OSatO'Us (Fig. 8.6E) and displays an intermediate morphology be­tween Iguan odon (Norman 1980) and hadrosaurines. In this type, the obturator notch is op en in adults. The shaft is relatively more decurved than in other hadrosaurids, but the degree of curvature is less than that of mo st specimens of Iguanodon. Distally, the shaft terminates in a rounded knob that protrudes asymmetrically to the ventral side.

The second type of ischium is represented by the hadro­saurines; exemplified by Anatotitan copei (Fig. 8.6D). In this type, the relative size of th e peduncles is considerably reduced in most forms. The obturator notch is open, except in old adults. The shaft is long, stra ight, and relatively thinner than in any other type of is­chium. The distal end usually tapers to a rounded point. A knob is rarely present. ]f present, however, it is rudimentary and always considerably smaller than in other types.

Corytbo saurus and Lambeosaurus exemplify the third type of ischium (Fig. 8.6B), where the pubic and ischial peduncles are equal in size or the iliac peduncle is slightly larger. The shaft is long, straight, and gen erally much thicker than in hadrosaurines of the same size. Posteriorly, th e diameter of the shaft gradually increases distally, where it terminates in an abrupt expansion into a structure resembling a foot in lateral view. This expansion is totally in a ven­tral direction when the ischium is viewed laterally in natural articu­lation. There is no "heel," but there is a slight tapering of the foot at the "toes " (see Fig. 8.6A).

The fourth typ e of ischium is represented by Hypacrosaurus and Parasaurolophus (Fig. 8.6A), which is most distinctive for its robustness. All landmarks previously cited are relatively larger in size and thicker than in any other type of ischium. The iliac pedun­cle displays a prominent "lip" that projects posteriorly on its artic­ular surface. The pubic peduncle is a large roughened process with a broad articulation. The obturator notch is closed in adults, but this area is frequently damaged during fossilization, creating the impression that the notch is normally open . The shaft is long and straight, and it does not increase in diameter for the proximal half of its length, The distal half gr adually increases in diameter until the abrupt expansion into a large footlike process that is up to 50% larger than in the other forms displaying th is stru ctur e. This foot ha s a distinctive "heel" and " to es" (Fig. 8.6A) that are more

Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids • 155

men* (e.g., Gilmore I924h). liven in truly robust specimens, such ,is the type of Parasaurolophus cyrtocristatus (Ostrom 1963) in which the obturator process is B large, rectangul.ir flange that nearly occludes the obturator north, the bar enclosing the foramen is either unossified or unpreserved. It is also conceivable that pathology may have caused early ossification of this structure in some specimens. However, closure of the ischial obturator notch appears to be so variable as to he systematically useless (Brett-Sur- maii 1973, 1989).

Four types of tSChifl can be recognized. Three are represented by Late Cretaceous forms. The first type is exemplified by Cilmore- osanrus. (Fig. N.6I1) and displays an intermediate morpholog\ be rween Iguantjctuii (Norman 1980) and hadrosaurmes. In this type, the obturator notch is open in adults. The shaft is relatively more ilecurvcd than in other hadrosaurids, but the degree of curvature is less than that of most specimens of Igiuinodon. Distally, the shaft terminates in a rounded knob that protrudes asymmetrically to the ventral side.

The second type of ischium is represented by the hadro saurines; exemplified by Atutotilan copei (Fig. 8.60). In this type, the relative size of the peduncles is considerably reduced in most forms. The obturator notch is open, except in old adults. The shaft is long, straight, and relatively thinner than in any other type of is- cliium. The distal end usually tapers to a rounded point. A knob is rarely present. If present, however, it is rudimentary and always considerably smaller than in other types.

Cotythosaurus and Latnbeosaiirns exemplify the third type ot ischium (big, 8.6B), where the pubic and ischial peduncles ate equal in size or the iliac peduncle is slightly larger. The shaft is long, straight, and generally much thicker than in hadrosaurines of the same size. Posteriorly, the diameter of the shaft gradually increases distally, where it terminates in an abrupt expansion into a structure resembling a foot m lateral view. This expansion is totally m a ven- tral direction when the ischium is viewed laterally in natural articu- lation. There is no "heel," but there is a slight tapering of the foot at the "roes" (see Fig. 8.6A).

The fourth type of ischium is represented by HypacroSOUfUS and I'anisauroloplms (Fig, 8.6A), which is most distinctive tor its robustness. All landmarks previously cited are relatively larger in Size and thicker than in any other type of ischium. The iliac pedun- cle displays a prominent "lip" thai projects posteriorly on its amc ular surface. The pubic peduncle is a large roughened process with a broad articulation. The obturator notch is closed in adults, but this area is frequently damaged during tossilization. creating the impression that the notch is normally open. The shaft is long and straight, and it does not increase in diameter for the proximal half nl us length. The distal hall gradually increases in diameter until the abrupt expansion into a large footlike process that is up to 50% larger than in the other forms displaying this structure. This fool has a distinctive "heel" and "toes" (Fig. X.hA) that are more

Appi'iidicular Anatomy in Campanian and M.mstricluian North American Hadros.niiicls • Is5

prominent and massive, and that proj ect farther posteri orly than related forms. A ridge extends from the pubic peduncl e across the shaft to the posterior side and continues onto the foot. Thi s ridge has no t been noted in other ischial types, but it may be a growth featur e rather than a generic characteristic.

One growth feature is noticeable in the ischium. The distall y enlarged knob or "foot " is on togenetic ally vari able. Excellent ex­ampl es of this are present in the Two Medicine faunal collection (Ca mpanian) housed in the U.S. National Mu seum (lot number USNM 358593). In this faun a, as well as the Iren Dabasu fauna (e.g., specimens of Bactrosaurus and Gilmoreosaurus hou sed at the AMNH), there are thr ee size or age gro ups. The first may be termed hatchlings and are individuals of hindlimb length less than 30 cm. The second age group may be called juveniles and have a hindl imb length of 30-90 em. The adult age gro up consists of the largest animals with hindlimb lengths of more than 90 em. There is no indication of an ischial foot in any of the hatchlings. The juve­niles have a distal enlargement that is not quite a foot but that is definitely too large to be from a hadr osaurine. Th e tru e foot ap­pears sudd enly in this group-within the time it takes the ischiu m to grow another 5 em in length (USNM lot number 35 8593). Per­haps th is signals the attai nment of sexual maturity. Thi s is the only ma jor growth featu re noted in an appendicular element that ap­pears after the hatchling stage.

In summary, only pelvic elements have clear characterist ics that can be of diagnostic value to the level of subfamily and clade. All othe r postcranial elements display either a complete intergradation of form with only the extremes exhibiting subfamilial features (such as th e hum erus as discussed above), or the postcran ia are too similar in all the genera to be of taxonomic use.

Femur. The femur (Fig. 8.7) is a long, cylindrical element . Th e head of the femur is cylindrical and relatively larger than in other or­nithopods. The greater troc hanter is massive and covers the entire lateral aspect of the proximal part of the femur. The lesser trochanter is relatively much smaller and often fuses to the greater trochanter in old adults (Fig. 8.7G). The greater and lesser trochanters are usually separated from each other by a small cleft. The shaft of the femur is st raight and circular in cross section, with the fourth trochanter at midpoint. The distal condyles are larger and more ro bust than in any ornithopod clade and project posteriorly. The ant erior cond yles may fuse to form an anterior cond ylar canal in old adults (e.g., USNM 7582, 7948).

The lesser tro chanter displays considerable variation in size, orientation, and degree of fusion to the greater trochanter. This is indi vidual var iation, for virtually every skeleton exa mined shows differences between the right and left sides. Variab ility of the lesser trochanter thus is of no value in the diagnosis of species. The lesser trochant er is sometimes fused to the grea ter along the dorsal mar ­gin of the femur in Large, evidently older specimens. The two are also fused in the modest-size type specimens of Kritosaurus incurvi­

156. Michael K. Brett-Surman and Jonathan R. Wagner

A

~ --

y(f-­

lr

... l"

Ii ~

­

...

prominent and massive, and thai project farther posteriorly than related loans. A ridge extends I mm the pubic peduncle across the shaft to the posterior side and continues onto the foot. 1'his ridge has not been noted in other ischial types, but it may be a growth feature rather than a generic characteristic.

One growth feature is noticeable in the ischium. The distally enlarged knob or "foot" is ontogcnctically variable. Excellent ex- amples of this are present in the Two Medicine fauna! collection (Campanian) housed in the U.S. National Museum (lot number USNM 358593). In this fauna, as well as the Iren Dabasu fauna (e.g., specimens of Btictmsiiurus and Gilinoreosaurus housed at the AMNH), there arc three si/e or age groups. The first may be termed hatchhngs and are individuals of hindlimh length less than 30 cm. The second age group may be called juveniles and have a hindlimh length of 30-90 cm. The adult age group consists of the largest animals with hindlimb lengths of more than 90 cm. There is no indication of an ischial foot in any of the hatchhngs. The juve- niles have a distal enlargement that is not quite a foot but that is definitely too large to be from a hadrosaurine. The true foot ap- pears suddenly in this group—within the time it takes the ischium to grow another 5 cm in length (USNM lot number 358593). Per- haps this signals the attainment of sexual maturity. This is the only- major growth leature noted in an appendicular element that ap- pears after the hatchlmg stage.

In summary, only pelvic elements have clear characteristics that can be of diagnostic value to the level of subfamily and cladc. All other postcranial elements display either a complete intergradation of form with only the extremes exhibiting subfannlial features I such as the humerus as discussed above), or the postcrania are too similar in all the genera to be of taxonomic use.

Femur. The femur (Fig. 8.7) is a long, cylindrical element. The head of the femur is cylindrical and relatively larger than in odicr or- nuhopods. The greater trochanter is massive and covers the entire lateral aspect of the proximal parr of the femur. The lesser trochanter is relatively much smaller and often fuses to the greater trochanter in old adults (Fig. S.7G). The greater and lesser rrochamers are usually separated from each other by a small cleft. The shaft of the femur is straight and circular in cross section, with the lourth trochanter at midpoint. The distal condylcs are larger and more robust than in any ornitbopod cladc and project posteriorly. The anterior condyles may fuse to form an anterior condylar canal in old adults (e.g., USNM 75S2, 7948).

The lesser trochanter displays considerable variation in size, orientation, and degree of fusion to the greater trochanter. This is individual variation, for virtually every skeleton examined shows differences between the right and left sides. Variability of the lesser trochanter thus is of no value in the diagnosis of species. The lesser trochanter is sometimes fused to the greater along the dorsal mar- gin of the femur in large, evidently older specimens. The two are also fused in the modest-size type specimens of Krirmaiinii incurvi-

156 • Michael K. Brctt-Siirman and Jonathan R. Wagner

~ steriorly than .l.i. d e across the

< :OOt . This ridge y be a growth

_ . The distally e. Excellent ex­

:aunal collection m (lot number Da basu fauna

s housed at the e first may be ength less than iles and have a

onsists of the 90 em . There is lings. The juve­

.:. :oot but that is e true foot ap­

- - 'es the ischium er 358593). Per­

,_. This is the only etement that ap­

racteristics that . :. and clade. All .e im ergradat ion :ami.l ia l features

rcrania are too

. - I element, The _ - an in other or­. .:overs the entire

esser trochanter arer trochanter in ~ ters are usually '. of the femur is rth trochanter at

. bust than in any . r condyles may

' L$ (e.g., USNM

anat ion in size, hanter. This is

examined shows '-iliry of the lesser

ies , The lesser ::- :h e dorsal mar­

s. The two are osaurus incurvi-

A

D

8'B

' fb-. .r . .

F : ":r' : ',: .;, , ~., . . '.':,"

~ ' ' ~ : . ' ,- .

-~

GRT

FMH

. .

. " ; '

GE

Figure 8.7. Com parison of hindlimb elements. D istal end of the tibiae, fibulae, and calcanea for (A) Camptosaurus USN M 2210 in left lateral view; (B) Anat ot itan A M N H 5730 in lateral view and (B') in anterio r view; (C) Larnbeosaurus ROM 1218; (D) Parasaurolophus FMNH P27393 (without calcaneum); (E) posterior view of the femu r of Anarotitan AMNH 5730 (length = 1135 mm); (F) proximal view and distal view of the femur of Gilmoreosaurus AMNH 6551; (G) lateral view of the femur of Corythosallrus CM 9461 (length = 760 mm).

Appendicular Anatomy in Campanian and Maastrichtian North American H adrosaurids • 157

AST

LRT

FMH

Flgurt S.~. Comparison of' hmdlimh elements. Distill end of the tibiae, fibulae, and calcanea for I A) ( jmptosaurus ' \.\ \l 12IK in left literal new: tP) An.notii.it- AMNH "30 m lateral view and (B I in anterior view; (C) l.jmliros-jurus RO.M HIS; ID) Pjraviiirolopl'.u> I-MSH P27593 (without ealcaneuml; (E) posterior t fan "/ the lemur (>/ Aiutoiil.ui AMNH S730 (length = 113S mm); irj proximal view and distal vmv of the femur o/'CflrpoROftunu AMXIi 6551; Kit lateral view of the femur o/'C.orythos.iurus CM W6! /length = 760 mini.

Appendiailar Anatomy in Camp.inian and Maasinchtian North American Hadrosaurids • 157

manu s and Had rosaur us, although Pinn a (1979) describes a speci­men of Kritosaurus notabilis in which th e tw o trocha nte rs ar e co m­pletely sepa rat e. Dilkes (19 93) reports no ontogenetic fusion in the trochanters of Maiasaura, and Davies (19 83) described the pres­ence and absence of this fusion over a broad size range in femora fro m Big Bend.

The deg ree to which the distal condyles appro ach each other on th e cranial side of th e femur appea rs to be size related (P. Gal­to n, per sonal communicati on). Weishampel and Horner (1990) suggest that closure of the condylar groove ma y be species depend­ent, not ontogenetic. The condyles are repo rt ed to approach one another during gro wth in Maiasaura (Dilkes 1993 ), but th ey are joined even in small indiv idua ls of Telm atosaurus (Weishampel et al. 1993) and Bactrosau rus (Godefro it et al. 19 98).

Generally, smaller hadrosaur femora appear concave cranially. With growth, th e cranial margin becomes straighter, and some femora even appear somewha t bowed crani all y, a ltho ugh the shaft is straight to slightly con vex caudally. These chang es are caused in part by the development of a low, rounded ridge, the linea interrnu­culari s crania lis of Hutchinson (200 1b), that passes mediolaterall y dow n th e femoral sha ft. This ha s been interpreted as the int ermus­cular line between the M. femorotibialis extern us and int ern us (H utc hinson 20 01b ), and sometimes a dist inct scar for the latter appears on the lateral side of bone caudal to thi s str uctu re (Dilkes 1993; Young 1958, fig. 30). This ridge is only promin ent in some hadrosaur femo ra, often those possessing a stro ngly developed four th trochanter, and no systema tic pattern is evident in its distri­bution.

In smaller hadrosaurs, th e fourt h trochanter is low and ro unded, and the vent ra l margin of the four th trochanter tapers into the femoral shaft. With incr easing body size, the fourth tr och anter is farther from the femoral sha ft (Dilkes 2001), and the ventral ma rgin for ms an abrupt angle with the shaft, eventually develop ing a hori­zonta l ventral margin distall y. Sereno (1999) lists a "s ubtriangular," or nonpend ant, fourt h trochanter as an apomorphy of Hadrosa uri­dae, wh ereas Norm an (2002 , character 63) dist ingu ishes between a triangular trochanter and a rounded tro chanter, with the latter diag­nosing hadrosaurids and Bactrosaurus. In Iguanodon, th e fourth trochanter is a caudoventrally dir ected prong associated with the seco ndary tend on o f the M. caud ifemor alis longus (H utchinson 200 1a).

Individual var iation is manifest in th e hadrosaurian femora, but the genera l morphology is identica l in all clades (Fig. 8.7) . O n­togenet ic va riat ion is relatively sma ll and is confined to th ose fea­tures associated with incr easin g size, such as muscle scars, exp an­sion of the base of the fourth trochanter, and in some cases, clo sure of the anterior condyla r ca na l (Brett-Surman 1989 ).

Tibi a. The tibi a (Fig. 8.7) is long and straight, wi th a propor­tionally larger cnemial crest th an in other orni thop ods. The distal end is notched for the as trag alus. Th e tibia is remarkabl y consistent

158 • Michael K. Brett-Surman and Jonathan R. Wagner

in shape o;;. :o~ =

hibirs posie -s sha ft length :c' ­

tr an sforms ;;.~

tr iangular ;~L.5

individu al s. ~

pro ducing a c:2­

cranial process ; N o ' -2.r1.;­

saur ids; howeve nose hadrosaurs down the s r,-". ~

large ornirbc p Fibu/...

Iguanodon ~ ~

roximal eli': ~

ons of th e ::ro: anded in:o : ~ -

. A .,:tr.lg.l.:.cs.

ir to Ig:t';1;v ';'""'

mama and Hadrosaitrus, although Pinna (197%) describes a sped men of KritOSdUrHS not.ibilis in which the two trochantc-rs are com- pletely separate. Dilkes (19931 reports no ontogeneric fusion in the trochanters of h&aiasawra, and Davies (19S3) described the pres- ence and absence of this fusion over a broad size range in femora from Big Bend.

The degree to which the distal condyles approach each other on the cranial side of the femur appears to be size related (P. Gal- ton, personal communication). Weishampel and Homer (1990) suggest that closure of the condylar groove may be species depend- ent, not ontogciKiic. The condyles are reported to approach one another during growth in Matasaura (Dilkes 1993). but they are joined even in small individuals of TebiiaiOSaurtli (Weishampel et al. 1991) and Bjcfrosjunw (Godefroit ei al. 1998).

Generally, smaller hadrosaur femora appear concave cramally. With growth, the cranial margin becomes straighrcr, and some femora even appear somewhat bowed cranial I >*, although the shaft is straight to slightly convex caudally. These changes are caused in part by the development of a low, rounded ridge, the linea intermu- culans cranialis of Hutchinson (2001b), that passes mcdiolaterally down the femoral shaft. This has been interpreted as the intermus- Ctllar line between the M. femorotibialis cxtcrnus and interims (Hutchinson 2001b). and sometimes a distinct scar for the latter appears on the lateral side of bone caudal to this structure (Dilkes 1993; Young I9.S8. fig. 30). This ndgc is only prominent in some hadrosaur femora, often those possessing a strongly developed fourth trochanter, and no systematic pattern is evident in its distri- bution.

In smaller hadrosaurs, the fourth trochanter is low and rounded, and the ventral margin of ihe fourth trochanter tapers into the femoral shaft. With increasing body size, the fourth trochanter is farther from the femoral shaft (Dilkes 2001), and the ventral margin forms an abrupt angle with the shaft, eventually developing a hori- zontal ventral margin distally. Srreno (1999) lists a "subtriangular." or nonpendant, fourth trochanter as an apomotphy of I ladrosauri- dae, whereas Norman (2002. character 63) distinguishes between a triangular trochanter and a rounded trochanter. with the latter diag- nosing hadrosaurids and liactrusaurus. In Iguanndun, the fourth (rochannrr is a caudovenirally directed neon* associated with ihc secondary tendon of the M. caudifcmorahs iongus (Hutchinson 200lal.

Individual variation is manifest in the hadrosaunan femora, but the general morphology is identical in all clades (Fig. 8.7). On- togeneric variation is relatively small and is confined to those fea- tures associated with increasing size, such as muscle scars, expan- sion of the base of the fourth trochanter. and in some cases, closure of the anterior condylar canal (Brett Surman 1989).

I'tbui. The tibia (Fig. S.?) is long and straight, with a propor- tionally larger cnemial crest than in other ornithopods. The distal end is notched for the astragalus. The tibia is remarkably consistent

158 • Michael K. Brett-Siirman and Jonathan R. Warner

es .ri bes a speci­. - znrers are com­0:: ': rus ion in the 0: d th e pres­

range in femora

ach eac h other - relared (P. Ga l­

Horner (1990) ~ 50 cies depend­: ~ a pproach one :: . bur th ey are

- "\\eishampel et

;"..:ave cra nia lly. :: rer, an d some zhough the sha ft

;;r:s a re ca used in a e linea int erm u­es mediolatera lly .:: th e interrnu s­us and internus

-2Z" for th e latte r :::uctu re (D ilkes min ent in some . gly developed

o .: enr in its dist ri­

:e:- is low a nd - 'mer ta pers into

rth tr ochanter is ::- ventral m ar gin

e .eloping a hori­o - subtr iangular,"

- of H ad rosau ri­.shes between a

·- - the latt er diag­-: i on, the fourth - ria ted with the ys (H utchinson

. ur ian femora, · -: (Fig. 8.7) . On­· ned ro th ose fea­. ' .e sca rs, ex pa n­me cases, clos ure

- vith a propor­s. The d istal

--.,a bly consistent

in shape ontogene tica lly, exce pt for the cnemial cres t. The cre st ex ­hib its positive allo metry in its extent down th e sha ft relative to sha ft len gth (althou gh qu antifying this relati on ship is d ifficult), an d transforms fro m a sma ll triangular nubbin in small ind ividual to a tri an gul ar flange, th en to a tr apezo idal wing in pr ogressively larger individuals. The en ds of th e fibula expand somewha t with growth, producin g a crania lly direct ed bulbous process distally and a short cranial process proxi ma lly.

No va riation was observed between the gen era of hadr o­saurids; however, Go defroit et a1. (1998) and Xu et al. (2000) diag­nose hadrosaurs on th e widening of the proximal head extending down the sha ft o f th e tibi a. The proximal ends of the t ibiae of all large ornithop od s are slightly flared.

Fibula. The fibula (Fig . 8.7 B) is extremely simila r to th at of Iguanodon (fide Norman 1980). It is long and st ra ight, and the proximal end forms a con cave cup that receives th e lateral pro jec­t ions of th e p ro xim al tib ial crest. The distal end of th e fibul a is ex­panded into a knob th at sits in th e dorsally projecting cup of the ca lca ne um.

Two morphs are o bserved. The first morph (Fig. 8.7B) is typ i­fied by all genera except Parasaurolophus. In thi s type, th e distal end of th e fibul a is mod erately expanded into th e shape of a ball. The second morph (Fig. 8.7C) is represented by th e typ e specimen of Parasaurolo phus cyrtocristatus in which the distal en d is greatly expanded co mpared to all other hadrosaurids, possibl y becau se of the loss of th e calc aneum (see discussion below) .

In general , fibul a length is highly correlated with tibia len gth (averagin g 95%), appears to be isometric with respect to t ibia length across the hadrosauri ds a nd shows no systema tic dist ribu­tion of variation. Ex pan sion of th e distal end occurs wi th inc reas ed size in orn itho po ds, as is clearly evidenced in Iguanodon (Norma n 19 80 , 1986) and Hypacrosaurus (e.g ., Bro wn 1913b; H orn er an d Cur rie 1994 ). The dist al end of th e fibul a is expande d in a ll adult hadrosaurs, a nd th ere is no evidence that it is any more pro­nounced in any taxon rela tive to an y other.

Astragalus . T he as tragalus of hadrosaurids (Fig. 8.7D) is simi­lar to Iguanodon (fide N orman 1980), but has a re la tively hig her anterior ascending pr ocess. This process is proportionally larger and distinctly tri an gul ar in juveniles of hadrosaurids. A juv en ile hadrosaurid astragalus from Big Bend (part ofTMM 42534) sho ws an unusual indentati on and a cha nge in the character of th e bone between the interc ondyla r po rtion below the ascending pro cess an d the cap over the medial malleolus of the tibia. Others a ppea r to have smaller, perhaps inco mplete ly for med cranial edges (Di lkes 1993; H orner and Currie 1994). It is possibl e that ex pansion of th e as traga lus ove r th e med ial co ndy le occ urs during postnatal on­togeny in had rosaurs.

O ne so urce of a pparent variatio n in th e as tragalus involves po­sitioning in mounted skeletons. Bro wn's (19 13 b, fig. 7) compara­tive illu stration shows seeming ly different astragali w ith th e left in

Appendicular Anatomy in Campanian and Maastricht ian North American H adrosaurids • 159

111 shape ontogenetically, except tor the cnemial crest. The crest ex- hibits positnc allometry in its extent down the shaft relative to shaft length (although quantifying this relationship is difficult), and u.ms forms from a small triangular nubbin in small individual to a triangular flange, then to a trapezoidal wing in progressively larger individuals. I he ends of the fibula expand somcwh.it with growth, producing a cramally directed bulbous process distally and a short cranial process proxunally.

No variation was observed between the genera of hadro saurids; however. Godefroit et al. (1998) and Xu et al. (2000) diag- nose hadrosaurs on the widening of the proximal head extending down the shah of the tibia. The proximal ends ol the tibiae of all large omithopods are slightly flared.

l-iliula. The fibula (Fig. 8.~B) is extremely similar to that of lgiunodoH (fide Norman 1980). It is long and straight, and the proximal end forms a concave cup that receives the lateral protec- tions of the proximal tibia! crest. The distal end of the fibula is e\ paneled into a knob that sits in the dorsally projecting cup ol the calcancum.

Two morphs are observed. The first morph (Fig. 8.7B) is typi- fied by all genera except I'arasauroluphus. In this type, the distal end of the fibula is moderately expanded into the shape of a ball. The second morph (Pig. 8.7C) is represented by the type specimen of Panisaiiiuluphus cyrtOCristattiS in which the distal end is greatly expanded compared to all other hadrosaunds, possibly because of the loss of the calcancum (see discussion below).

In geucr.d. libul.i length is hidi',\ &.oin-l.ued with ubia Ici'.mh (averaging 95%), appears to be isometric with respect to tibia length across the hadrosaunds and shows no systematic distribu- tion of variation. Expansion of the distal end occurs with increased size in omithopods, as is clearly evidenced in Igiunodon (Norman 1980. IV86) and Hy/wrrowwrw (e.g.. Brown 1913b; Homer jnj Curne 1994), The distal end of the fibula is expanded in all adult hadrosaurs, and there is no evidence that it is any more pro- nounced in any taxon relative to any other.

Astragalus. The astragalus of hadrosaurids (Fig. X.7D) is simi- lar to Igiianodon (fide Norman 1980), but has a relatively higher anterior ascending process. This process is proportionally larger and distinctly triangular in juveniles of hadrosaurids. A juvenile hadrosaurid astragalus from Rig Bend (part ol TMM 42534] shows an unusual indentation and a change in the character of the bone between the intercondylar portion below the ascending process and the cap over the medial malleolus of the tibia. Others appear to have smaller, perhaps incompletely formed cranial edges [Dilkes 1993; Ilorner and Currie 1994). It is possible that expansion of the astragalus over the medial condyle occurs during postnatal on- togeny in hadrosaurs.

One source of apparent variation in the astragalus involves po- sitioning in mounted skeletons. Brown's (1915b, fig. 7) compara- tive illustration shows seemingly different astragali with the left in

Apprudiciilar Anatomy in ( .unp.inian and Maas'.richtian North American Hadrosauoiis • I s9

articulation with the tibia, whereas th e right is displa ced down and is viewed from a different an gle. The astragalus of all hadrosaurs is notched on the lateral side to receive a peglike process from the ca l­can eum (contra Chatterjee 1982 ).

In th e hadrosaurids, two types of ast ragali were observed . The first type (Fig. 8.7B) is repre sented by a ll the genera except Parasaurolophus. In thi s type, th e astragalu s is trian gular in shape when viewed ant eriorly, but skewed late rally int o a 25-5 0-105 de­gree configuration rather than the equilateral shape of Parasaurolo­phus. The outer malleolus of the tibi a is rela tively smaller and less rugose in th is type. The second type (Fig. 8.7D ) is represented solely by th e type specimen of Parasaurolophus cyrtocristatus. In th is type, the astra ga lus is also trian gular in sha pe when viewed an ­teriorl y, but it is mu ch reduced in size and thickness compared to other hadrosaurids, and it does not extend as fa r medially under the inn er malleolus. To compensa te for the reduced astragalu s in Parasaurolopbus, the inn er malleolu s of the tibi a is re latively more rugose and expa nded th an in th e other genera. This increase is re­lated to th e possible loss of the ca lcan eum in Parasaurolopbus as discussed below.

Sma ll, presumably juvenil e, spec imens of hadrosaurines and lambeosaurines , especially Corythosauru s, with tibiae 60 ern or less in length, tend to have as tragali resembling the first type but with a more tri an gular ascending proc ess. Except in Parasaur olop hus, the astragali of larger individuals enlarge and exp and medially. This suggests that the form of the astragalus cha nges with growth. The sa mple size is too small a t this tim e for any definitive co nclusions.

Calcaneum . In pr ofile, the calcaneum (Fig. 8.7B) forms a qu ar­ter sectio n of a circle. The arc of the circle forms th e articula tion for the metatarsal s and faces anter iorly in natural position. The two ra dii marking this quarter section form tw o cups, one facing prox­imally to receiv e the fibula and one facin g posteriorly to receive the tib ia. The ca lca neum is sma ll and does not ex tend ventroposteri­orly beneath th e tibi a to th e posterior side . The distal end of the tibi a fits in a sligh t concavity on the posterior side of th e calca­neum. In Camptosauru s (USN M 4282 ), the ca lcaneum forms the entire surface of articulatio n for th e distal ta rsals on th e lat eral side . In hadrosaurids, the ca lcaneum and the posterior side of the tib ia both art iculate w ith the distal tarsals on the lateral side . A peglik e process inserts into a notched on the lat eral side of the as­tragalu s. Th is condition is oppos ite of what ha s been reported else­where for orn ithopo ds (Cha tterjee 1982) .

The possible loss of the ca lcane um in P. cyrtocristatus deserves comment. The enlarged kn ob at th e distal end of th e fibula ma y rep resent the fusion of the calcaneum to the fibula. The ca lcaneum may also have failed to ossify and the fibula has enlarged to com­pensate. Becau se th e sa mple is based only on th e type of P. cyr­tocristatus, there is a distinct poss ibility th at th e calcaneum simply was not fossilized. But thi s is doubtful becaus e of the lack of space

160 • Michael K. Brett-Sur man and Jon athan R. Wagner

for th e ca!~:'C

rure on the ~

~Jewt~r5 . r ,

of all hadrosaes varia tion was for complete r ­for any de fini -

Sereno l l 9'::5 an apo morp hy ian s" have c

appear to have four instances ! 979; Lull anc ! 99()}, Bec2:'..:S<: .. :orms (e.g.. . . e, this .~own tax

Sever.

use ~~

articulation with the tibia, whereas the right is displaced down and is viewed from a different angle. The astragalus of all hadrosaurs is notched on the lateral side to receive a peglike process from the cal- caneum (contra Chattcqec 19X2).

In the hadrosaurids, two types ot astragali were observed. The first type I Fig. B.?B) is represented by all the genera except Parasaurolophits. In this type, the astragalus is triangular m shape when viewed anteriorly, but skewed laterally into a 25 50-105 de- gree configuration rather than the equilateral shape ot Parasaitrolo- phut. The outer malleolus oi the tibia is relatively smaller and less rugose in this type. The second type (Fig. 8.7D) is represented solely by the type specimen of Parasnurvlitftbus cyrtocristams. In this type, the astragalus is also triangular in shape when viewed an- teriorly, but it is much reduced in size and thickness compared to other hadrosaurids, and it does not extend as far medially under the inner malleolus. To compensate tor the reduced astragalus in Parasaurolopbus. the inner malleolus of the tibia is relatively more rugose and expanded than in the other genera. This increase is re- lated to the possible loss of the calcaneum in Parasaurnlopbus as discussed below.

Small, presumably juvenile, specimens of hadrosaurines and lambeosaunnes. especially Corylbosaurus. with tibiae 60 cm or less in length, tend to have astragali resembling the first type but with a more triangular ascending process. Excepi in fiiMMwro/op&M*. mc astragali ol larger individuals enlarge and expand medially. This suggests that the form of the astragalus changes with growth. The sample size is too small at this time for any definitive conclusions.

Calcaneum. In profile, the calcaneum (Fig. 8.7R) forms a quar- ter section of a circle. The arc of the circle forms the articulation for the metatarsals and faces anteriorly in natural position. The two radii marking this quarter section form two cups, one facing prox- imal ly to receive the fibula and one facing posteriorly to receive the tibia. The calcaneum is small and does not extend ventroposteri- orly beneath the tibia to the posterior side. The distal end of the tibia fits in a slight concavity on the posterior side of the calca- neum. In Camptasaurtis (USNM 4282), the calcaneum forms the entire surface of articulation tor the distal tarsals on the lateral side. In hadrosaurids, the calcaneum anil the posterior side of the tibia both articulate with the distal tarsals on the lateral side. A peglikc process inserts into a notched on the lateral side of the as- tragalus. This condition is opposite of what has been reported else where fol ornithopods (Chatterjec 1982).

The possible loss of the calcaneum in P. i-yrtocristatu< deserves comment. The enlarged knob at the distal end of the fibula may represent the fusion of the calcaneum to the fibula. The calcaneum may also have failed to ossify and the fibula has enlarged to com- pensate. Because the sample is based only on the type of /'. cyr- locrislatus, there is a distinct possibility that the calcaneum simply was not fossilized. But this is doubtful because of the lack of space

160 • Michael K Brott-Surmnn and Jonathan R Wagner

:' -ced down and ~ hadrosaurs is

: O ~ 5 from the cal­

~ observed . The e genera except aagular in shape .:. _ --50-105 de­

: Parasaurolo­smaller and less

is represented c: iocristatus. In

. t en viewed an­0: 0 compared to medially under

eu astragalus in -- relat ively more ' " increase is re­ss i urolophus as

i! rosaur ines and ia e 60 em or less

_- :ype but with a ssurolophus, the

medially. This irh growth. The ive conclusions.

forms a quar­- 0 a rticulation for

sirion . The two one facing prox­

I}~: :: to receive the ventroposteri­

.r ; sral end of the side of the calca­: ~ eum forms the - s on the lateral zerior side of the

- - lateral side. A • 0 side of the as ­- n reported else­

cristatus deserves : the fibula may

' . The calcaneum enlarged to com­~ type of P. cyr­-~ caneum simply - - e lack of space

for the calcaneum to occupy and the absence of any articular fea­ture on the astragalus.

Metatarsals and Distal Tamils. The metatarsals (Fig. 8.3C, D) of all hadrosaurids resemble one another closely. No consistent variation was observed among the genera; however, the sample size for complete and articulated metatarsals is insufficient at this time for a ny definitive statement.

Sereno (1999, character 96) lists loss of distal tarsals 3 and 4 as an apornorphy of hadrosaurids + Ouranosaurus. " Iguanodon­tians" ha ve two or three ossified distal tarsals, and hadrosaurids appear to have onl y the fourth. This bone has been reported in only four instances (Brett-Surman 1975; Godefroit et al. 1998; Horner 1979; Lull and Wright 1942; see also Weishampel and Horner 1990). Because any inference for the loss of dist al tarsals in other forms (e.g., Ouranosaurus nigerensis) relies heavily on negative ev­idence, this character sho u ld be scored onl y for particularly well­known ta xa and is probably of limited use .

Several changes in pedal form occurred during the evolution of hadrosaurs. These appear to be related to max imum adult body size because they are broadly paralleled in ceratopsians, but some appear even in juvenile hadrosaurs (Dilkes 1993). These include splayed metatarsals II and IV, the pres ence of axially foreshortened phalanges (up to four times wider than long), dorsoventral compression of the phalanges, and axial shortening of the unguals. Some of these cha nges may represent scoreable characters, but they are not dis­cussed in detail her e because of difficulties in quantification.

The repeated failure to identify a metatarsal I in articulated had rosaurids, including " mummies" (e.g., O sborn 1912) and the Auri sino Limestones stem hadrosaurid (Dalla Vecchia 2001), is problematic. Continued use of this character in attempting to re­solve the relationships of hadrosaurs and close outgroups is inad­visabl e.

Phalanges. In hadrosaurids, the pedal phalanges (Fig. 8.3) are of similar shape except in Gilmoreosaurus, in which the unguals ar e more clawlike and are therefore similar to the camptosaurid pattern. This is peculiar in light of the fact that Bactrosaurus john­soni, a species sympatric with Gilmoreosaurus, has typical hadrosaurid phalanges with more hooflike unguals. The most proximal row of phalanges is longer (proximodistall y) than wide (late romedially). The reverse condition is true for all other pha­langes of the pes. The proximal row is noticeably larger and more robust than the more distal rows. The gen eral ph alangeal formul a is 0-3-4-5-0 (?). See Gregory (194 8) for a discussion of whether the fifth digit is present. (The formula for th e Aurisino Limestone hadrosaur is 0-3-4-4-0 [Dalla Vecchia 2001J.) The lack of any ma­terial unquestionably articulated and possessing a fifth digit re­qu ires that the fifth digit be regarded as absent. The pedal unguals of Ma iasaura and Brach ylophosaurus have an axial keel on the plantar face of the bone (Fiorillo 1990; Prieto-Marquez 2000).

Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids • 161

tor the calcaneum tO occupy and the absence of any articular fea- ture on the astragalus.

.VlfWfar*;/* JM«/ L)»fa/ 7ar*j/*. The metatanak (Fig. 8.3C, D) of all hadrosaurids resemble one another closely. No consistent variation was observed among the genera; however, the sample size tor complete and articulated mctararsals is insufficient at this time lor any definitive statement.

Sereno (1999, character 96) lists loss of distal tarsals 3 and 4 as an apomorphy ot hadrosaurids + Ouninosanriis. "Iguanodon- rians" have two or three ossified distal tarsals, and hadrosaurids appear to have only the fourth. This bone has been reported in only lour instances (Brett Surman 1975; Godefroit ct al. 1998; Homer I'''*': lull and \\u..hi I'".'; ,ee alsu \\ei\hampel and ll'Tiiei 1990). Because any inference for the loss of distal tarsals m other forms (e.g., OttranosaurttS nigerensis) relics heavily on negative ev- idence, this character should be scored only for particularly well- known taxa and is probably of limited use.

Several changes in pedal torm occurred during the evolution ot hadrosaurs. These appear to be related to maximum adult body size because they arc broadly paralleled in ceratopsians. but some appear even in juvenile hadrosaurs (Dilkcs 1993). These include splayed nietatarsals II and IV, rite presence ot axially foreshortened phalanges (tip to lour times wider than long), dorsovenrral compression ol the phalanges, and axial shortening of the unguals. Some of these changes may represent scorcable characters, but they are not dis- cussed in detail here because of difficulties in quantification.

The repeated failure to identify a metatarsal I in articulated hadrosaurids, including "mummies" (e.g., Osborn 1912) and the Aurisino limestones stem hadrosaurid (Dalla Vecchia 2001), is problematic. Continued use ol this character in attempting to re- sol\e the relationships of hadrosaurs and close ourgroups is inad- visable.

Phalanges. In hadrosaurids, the pedal phalanges (Fig. 8.3) are of similar shape except in (iilniorcosaiirns, in which the unguals are more clawlike and are therefore similar to the camprosaurid pattern. This is peculiar in light of the fact that Uactrosaurus John- snni, a species sympatric with Gilmorcosaiiriis, has typical hadrosaurid phalanges with more hoollike unguals. The most proximal row of phalanges is longer (proxunodistally) than wide (lateromedially). The reverse condition is true for all other pha- langes ol the pes. The proximal row is noticeably larger and more robust than the more distal rows. The general phalangeal formula is 0-3-4-5-01?). See Gregory 11948) for a discussion of whether the fifth digit is present. (The formula for the Aurisino Limestone haunwaur k 0-3-4-4-0 (Dalla Vecchia 1001 j.) The lack of any ma tcnal unquestionably articulated and possessing a fifth digit re- quires that the filth digit be regarded as absent. The pedal unguals of Mmasciiira and Btcith\lopbo$auri4$ have an axial keel on the plantar face of the bone (Fionllo 1990; I'rieto-Marquez 2000).

Appendicular Anatomy in l.amp.inian and Maastriclitian North American Hadrosaurids • W»'l

Dilkes (1993 ) noted that Maiasaura has similar pr ominent extensor bumps on th e phalanges in that taxon.

Altho ugh it appears th at th e pes is not diagnostic to subfamily level, differences have been noted (Brown 19 13a). In a comparison of the types of "Diclon ius mirabilis" (=Anato titan) and Hy­pacrosaurus, Brown (1913b, fig. 7) ob served that the hadrosaurine pes has relat ively less robust metatarsals but wider, flatter, and sometimes lon ger ph alanges. Lambeosaurines, in co ntr ast, have rel­at ively more robust met atarsals but less robust ph alanges. This re­sults in a hadrosaurine foot th at is shorter and wider than th e more elongated lambeosaurine foot. Perhaps the y represent an ad apta­tion of the hadrosaurines to th eir more hea vily overgrown lowland environments , in contrast to th e more open lambeosaurin e enviro n­ments where cursorial abilities would be at a premium.

Acknowledgments. We thank the followin g people for their help during the long gestation of this study (in alphabetical or der): Rich ard Ashmore, Momchil Atanassov, Don Baird , James Barrick, Da vid Berm an, John Bolt, Christopher Brochu, Peter Buchh olz, Ken­neth Carpenter, Ralph Chapma n, Sankar Cha tterjee, Chip Clark, Jennifer Clark , Fred Co llier, Alan Couls on, Phillip Currie, Fabio Marco Dalla Vecchia, Kyle Davies, David Dilkes, Gordon Edmund, Eugene Gaffney, Peter Galton, Joseph Gregory, Ja ime Headden, Pat Holroyd, Thomas R. H oltz, John H orner, Nicholas H atton III, Erie Kauffman, Richard Kissel, Harold Koern er, Wan n Langston, Dou­glas A. Lawson, Guy Leahy, Thomas Lehman, Robert Long, Robert Makela, Alessandro M arrissa, Je ffery Martz, Christopher M e­Gowan, John S. McIntosh, William M orri s, M ickey M ortimer, Dar­ren Naish, George Olshevsky, H alsk a Osm6lsk a, John Ostrom, Pamela Owen, Kevin Padi an, Gregory Paul , Robert Purdy, Timothy Rowe, Dale Ru ssell, Raymond Rye, Judith Schieb out, Richard Stra uss, Hans-Dieter Sues, Denny Suratt, Jann Thompson, David Trexler, Samuel Welles, C. C. Young, and the Science and Resource M anagement division of Big Bend National Park.

This study represents a syno psis and elaboration of work con ­ducted towa rd ma ster' s degrees for both authors, doc toral work for M. K. B.-S., and ongoing work by both authors. Research wa s funded by gran ts and scho larships to J. R. W. fro m the Department of Geo logica l Sciences at Texas Tech University and Sigma Xi. All figures are by Douglas A. Lawson , except Figure 2C and D, whi ch are by Gregory S. Paul.

References Cited

Boyd, C, and C ]. Ott. 2002. Probabl e lambeosau rine (Ornith ischia, H ad rosau ridae) specimen from the Late Cre taceo us Hell Creek For­mation of Monrana. Journal of Vertebrate Paleontology 13(3, Suppl.): 32A.

Brett-Surma n, M . K. 1975. The ap pendicular anato my of hadrosau rian di­nosaurs. M.S. thesis. University of Californ ia, Berke ley.

___. 1979. Phylogeny and paleobiogeography of hadr osauri an di­nosau rs. N ature 277: 560-562.

162 • Michael K. Brett -Surman and Jon ath an R. Wagner

--- : 9 S.~ ~ :__...,

:r:-O\ \ J1 .. B.. ~ ~< .:-

~Or&:l:1= :: ~

0 ,-- _,-~. - ­

---_ ! 9 : : v..­

dermis, - :)9-- ~ .:; .

~ :..u...~"1G.riL rw­

.:.!: S .:.:=­:.:-eilli:' r :

_ , _'-:i ,-"_-.

_..2.:..~ ~e :.. ;' · ~ .:. .:. ~

_ ,-~ _ .-_ .1. .I..

.- - - ....- - _ :Jl ::- _ .. · : .:.1...l":'U . J:. 1t.. _ _

. .045 0 ::' -- ..

C"'''::s. ..

_ F? : '7:­- :::t ;-: ::-.e .. S.. 1 ~L _- ·

Dilkcs (1993) noted that Maiasaura has similar prominent extensor

bumps on the phalanges m that raxon. Although it appears that the pes is not diagnostic to subfamily

level, differences have been noted (Brown 1913a). In a comparison

of the types of "Didonitis utirahilis" l=AnatoiitJii) and Hy-

pacrosaurus, Brown (1913b, fig. 7) observed that the hadrosaurine

pes has relatively less robust mecatarsals but wider, flatter, and

sometime* longer phalange-., lambcosaurinea, in contrast, have rcl atively more robust metararsals but less robust phalanges. This re-

sults in a hadrosaurine foot that is shorter and wider than the more

elongated lambeosaurme foot. Perhaps they represent an adapta

tion of the hadrosaurines to their more heavily overgrown lowland

environments, in contrast to the more open lambeosaurme environ-

ments where cursorial abilities would be at a premium.

Acknowledgments. Wr thank the following people for their help during the long gestation of this study (in alphabetical order):

Richard Ashmore, Momchil Atanassov, Don Baird. James Barnck,

David Herman. John Bolt, Christopher Brochu, Peter Buchholz. Ken-

neth Carpenter, Ralph Chapman, Sankar Chatterjcc, Chip Clark,

Jennifer Clark, Fred Collier. Alan Coulson. Phillip Curne, Pa bio

Marco Da I la Vcccbia, Kyle Da vies, David Dilkcs, Gordon Edmund,

Eugene Gaffney, Peter Galron, Joseph Gregory, Jaime Headden, Pat

Holroyd, Thomas R. Holtz, John Homer, Nicholas Hotton III, Erie K.iuffman, Richard Kissel, Harold Koerncr, Wann Langsron, Dou-

glas A. Lawson, Guy Leahy, Thomas Lehman, Robert Long. Robert

Makela, Alcssandro Marnssa, Jeffcry Marc/., Christopher Mc- Gowan. John S. Mclntosh, William Morris, Mickey Mortimer, Dar-

ren Naish, George Olshevsky, Halska Osmolska, John Ostrom,

Pamela Owen, Kevin Padian, Gregory Paul, Robert Purdy, Timothy

Howe, Dale Russell, Raymond Rye, Judith Schiebout, Richard

Strauss. Hans-Dieter Sues, Denny Suratt, Jann Thompson, David

Trexler, Samuel Wcllea, C C. Young, and the Science and Resource Management division of Big Bend National Park.

This study represents a synopsis and elaboration of work con-

ducted toward master's degrees for both authors, doctoral work for

M. K. B.-S.. and ongoing work by both authors. Research was funded by grants and scholarships to J. R. W. from the Department

of Geological Sciences at Texas Tech University and Sigma Xi. All

figures are by Douglas A. Lawson, except Figure 2C and D, which

are by Gregory S. Paul.

References Cued

Bovd, C, and C.j. Oil. 2002. Probable l.nnbcosaurine (Omiihischia, Hadrosaundac) specimen from the Late Cretaceous Hell Creek For- mation of Montana, jiuirnal Of Verlrbmte Paleontoln%y 131.3, Suppl.): 32A.

Brett-Surman, M. K. 197.5. The .lppendicul.ir anatomy of hadrosaurinn di- nosaurs. M.S. thesis. University of California. Berkeley. 1979. Phylogeny and paleobiogeography of hadrosaunan di-

nosaurs. S.ilme I": 5ri0-562.

162 • Michael K. Brett-Surman and Jonathan R. Wagner

ent extensor

sric to subfamily a comparison

- titan i and Hy­t ; e h adrosaurine

ider, flatter, and rontrast, have rel­

alanges, This re­. "'r than the more resent an adapta ­e-grown lowland

- saurine en viron­m.

ple for their help . iabetical order):

. James Barrick, er Buchholz, Ken­

"ee, Chip Clark, _: Cur rie, Fabio Gordon Edmund,

e Headden, Pat Hott on III, Erie

- La ngston, D ou­rt Long, Rob ert

Christo pher Mc­ev M ortimer, Dar­ca. John Ostrom,

Pur dy, Timothy - 0 ebout, Richard , . ompson, David

ce and Resource

~ "on of work con­doct o ral work for :-5. Research was

he Department and Sigma Xi. All _C and D, which

_ One (Orn ithischia, s Hell Creek For­

? z leontology 13(3,

orhadrosaur ian di­o keley,

»: hadrosaurian di­

___. 1989. Revision of the Hadrosauridae (Reptilia: Ornithischia) and their evolution during the Campanian and Maastrichrian. Ph.D. diss. George Wash ington University. Available at http://www.nmnh. si.edu/paleo/bib/hadrobib.htm.

Brown, B. 1912. The osteology of the manus in the Family Trachodonti­dae . Bulletin of the American Museum of N atural History 31: 105­108 .

___. 1913 a. The skeleton of Sauro lophus, a crested duck-bill dinosaur from the Edmonton Cretaceo us. Bull etin of the American M useum of Na tura l History 32 : 387- 393.

___. 1913b. A new trachodont dinosaur, Hypacrosaurus, from the Ed­mon to n Cretaceous of Albert a. Bulletin of the American Mu seum of Natural History 32: 395-406.

___. 1914 . Corytbosaurus casuarius, a new crested dinosaur from the Belly River Cretaceo us with provisional classification of the fami ly Trachodo ntidae. Bulletin of the American Mus eum of N atu ral His­tory 33 : 559-565.

___. 1916. Corytbosaurus casuarius, skeleto n, musculature, and epi­dermi s. Bulletin of the Am erican Museum of Na tural Hi story 35 : 709-716.

Brinkmann, W. 1984. Erster nachw eis eines hadr osauriers (Orn ithischia) aus dem unte ren Garumn ium (Maastrichtium) des Beckens von Trem p (Provinz Lerida, Spanien). Palaeontologische Zeitschrift 58: 295 - 305.

Carpenter, K. 1999. Eggs, N ests, and Baby Dinosaurs. Bloomington: Indi­ana University Press.

Carr ier, D. R., and C. G. Farmer. 2000 . The evolution of pelvic asp iration in archosa urs . Paleobiology 26: 271-293.

Casanovas-Cladelias, M. L., X. P. Pereda-Suberbiola, J. v. Sant afe, and D. B. Weishampel. 199 9. First lambeosaurine hadrosaurid fro m Eu­rop e: Paleobi ogeographical impl ications. Geo logical Magazine 136: 205 -211.

Cha pman, R. E., and M. K. Brett-Surman. 1990 . Morph ometric observa­tions on hadrosaurid dinosaurs. In K. Carpenter and P. J. Curri e (eds.), D ino saur Systematics: Approaches and Perspectives, pp. 163-177. Cambridge: Cam bridge University Press.

Chatterjee, S. 198 2. Phylogeny and classification of Thecodontian reptiles. N ature 295: 317- 320 .

Dalla Vecchia, F. M. 20 01. Terrestri al ecosystems on the Mesozoic peri­Adriati c carbonate platforms: The vertebrate evidence. Proceedings of the Seventh Symposium on Meso zoic Terrestrial Ecosystems, Buenos Aires 1999: 77-83.

Davies, K. L. 1983. Hadrosaurian dinosaurs of Big Bend National Par k. M.S. thesis. University of Texas, Austin.

Dilkes, D. W. 1993. Grow th and locom otion in the Hadrosaurian dinosaur M aiasaura from the Upper Cretaceous of Mont ana. Ph.D. diss. Uni­versity of Toronto.

___. 2000. Appendicular myology of the had rosaurian dinos aur Ma­iasaura from the Late Cre taceous (Campanian) of Montana. Trans­acti ons of the Royal Society of Edinburgh , Earth Sciences 90: 87­125.

___. 2001. An ontogeneti c perspective on locom otion in the Late Cre­taceous dinosaur Maiasaura (Ornithischia: Hadrosauridae). Cana­dian Journal of Earth Sciences 38: 1205-1227.

Appendicular Anat omy in Campanian and Ma astrichti an North Amer ican Hadrosaurids • 163

1989. Revision of (he I l.ulrosauridac i Kcptilia: Ornitlusclii.il and their evolution during the Campauian and Mj.istrichtian. Ph.D. diss. George Washington University. Available at http://ww\v.nninh. si.edu/palco/bib/hadrobib.htm.

Brown, B. 191?. The osteology of the nianiis in ihc Family Trachodonti- dae. Bulletin of the American Museum of Natural History 31: 105- 108. 1913a. I he skeleton of Saiirolophus, a crested duck bill dinosaur

from the Edmonton Cretaceous. Bulletin of the American Museum of Wmmd Hmfufy 32: 387-39 I. 1913b. A new trachodont dinosaur, Hypacrofaurut, from the I'd

monton Cretaceous of Alberta, bulletin of the American Museum of AUfimif Hwfory 32: 39^-406. 1914. Corythosaums casuanus, a new crested dinosaur I rum the

Belly River Cretaceous with provisional classification of the family Trachodontidac. Bulletin of the American Museum of Natural His fu*y #3: 5 $9-565.

1916. Corythosaums casuanus. skeleton, musculature, and epi- dermis. Bulletin of the American Museum of Natural History 35: 709? I a.

Brinkmann. W. 198-1. Erstcr nachwcis ernes badrosauncrs (Orniiluschi.il aus dem unteren Carumnium 'Maastncfitiumj des Reckons von Tremp [ProvJrrz Lerida, Spanien). Palaeontologtsehe Zeilschrift 58: 295-*U5.

Carpenter, K. 1999. Lues, Nests, ami Baby Dinosaurs. Bloomington: Indi- .m.i University Press.

Carrier, D. R., and C G. Farmer. 2000. The evolution oi pelvic aspiration in archosaurs. Paleohiology 26: 271-293.

Casanovas-Ciadellas, M. I... X. I'. Percda-Subcrbiola, J. V. Santafc, and I). B. W'cshampcl 1999. First lambeosaurinc hadrosaund from Eu- rope: Palcobiogeographic.il implications. Geological Magazine 136: 205-211.

Chapman, R. F , and M. K. Brett-Surman. 1990. Morphonu-tnc observa- tions on hadrosaund dinosaurs In K. Carpenter and P. J. Curric (eds.). Dinosaur Systematics: Approaches and Perspectives, pp. I(i3-I~~. Cambridge: Cambridge University Press.

Chatrerjce, S. 1982. Phytogeny and classification of Thecodontian reptile-. Nu*M7e29i: 117-320.

Dalla Vecchia, F. M. 2001. Tcuesinal ecosystems on the Mesozoic pen Adriatic carbonate platforms: The vertebrate evidence. Proceedings Of the Seventh Symposium on Mesozoic Terrestrial Ecosystems. BMOKW A,re* 1999: 77-X.C

Davies, K. I . 19S3 lladrosaurian dinosaurs of Big Bend National Park. M.S. thesis. University oi Texas, Austin.

Dilkes, D. W. 1993. Growth and locomotion in the lladrosaurian dinosaur Maiasaura from the Upper Cretaceous of Montana. Ph.D. diss. Uni- versity ot Toronto. 2000. Appeiidiculnr myology ot the lladrosaurian dinosaur Ma-

i.isaitra from the Late Cretaceous (Campaniati) of Montana. Trans- jfTiOMS o/ fte RoW $oewfY of &/m/w«nj/;. Fvrfk St frncf* 90: 87-

2001. An ontogenetic perspective on locomotion in the Late Cre- taceous dinosaur Maiasaura (Ornithischia: I ladrosauridael. Cana- dian Journal of Earth Sciences 38: 1205-1227.

Appcndicular Anatomy in Campanian and Maastrichtian North American I ladrosaunds • 163

Dodson, P. D. 1971. Sedim entology an d tapho nomy of the Oldman For­mati on, Din osaur Provincia l Park, Alberta. Palaeogeograph y, Palaeo climatology, Palaeoecology 10: 2 1~74 .

Dolto, L. 1883a. Troisieme no te sur les dinosau riens de Bern issart . Bulletin du Mu see Royal d'His toire N atur elle de Belgique 2: 85- 12 6.

___. 1883b . Qu arti eme note sur les dinosaurines de Bernissarr. Bulletin du Mu see R oyal d'Histoire Naturelle de Belgique 2: 223-252.

Egi, N. , and D. B. Weishampel. 2002. M orphometric analysis of hume ral shapes in hadr osaurids (Orn ithopoda, Dinosauria). Senck enb er­giana Leth aea 82: 43-58.

Erickson, B. R. 1988. Notes on the postcranium of Camptosaurus. Scien ­tific Publications o f th e Science Museum of Mi nnesota 6(4): 3-13.

Fiorillo, A..R. 1990. The first occurrence of Hadrosaur (Dinosa ur ia) remain s from the Marine Claggett Formation, Late Cretaceous of South Cen­tral Montana. Journ al of Vertebra te Paleontology 10: 515- 517

Gilmor e, C. W. 1909. Osteology of th e Jurassic reptile Cam ptosaurus, wit h a review of the species an d genus, and description of two new species. Proceedings of the U.S. National Museum 36: 197-332 .

___. 1917. Brachyceratops, a ceratopsian dinosaur from the Two Med­icine Formati on of Montana. U.S. Geological Survey Professional Pap er 103: 1-45.

___. 1924a . A new species of hadrosaurian dinos aur from the Edm on­ton Formation (Cretaceous) of Alberta, Canada. Geological Series 43. Bulletin of the Departm ent of Min es, Ge olog ical Survey o f Cana da 38: 13-26.

___. 1924b. On the skull an d skeleto n of Hyp acrosauru s, a helmet­crested dinosa ur from the Edmonton Cretaceous of Alberta . Geo­logical Series 43. Bulletin of the Department of Mines, Geological Survey of Canada 38: 49-64.

___. 193 3. On the dinosau rian fauna of the Iren Dab asu formation. Bulletin o f th e American Mu seum of N atural History 67 : 23:....78.

___. 1946. N otes on recentl y mounted fossil skeletons in the United States National Mu seum. Proceedings of th e U.S. N ational Mu­seum 96 : 201 - 20 3.

Godefroit, P., Dong Z .-M ., P. Bultynck, Li H., and Feng L. 199 8. Sino­Belgian Coope rative Program. Cretaceous dinosau rs and mammals from Inner Mongolia. 1. New Bactrosaurus (Dinosa uri a: Had ro­sauroidea) material from Iren Da basu (Inner Mongolia, P. R. Chi na). Bulletin des Institutes Royal des Science Natura lles Bel­giq ue 68(Suppl.) : 1-70 .

God efroit, P., Zan S., and Jin L. 2000. Charonosaurus jiayinensis n.g., n.sp. , a lambeosaurine dinosaur from the Late Ma astri chtian of northeastern China. Compte Renda de l'A cademie du Sciences Na­turelles, Paris, series 2, Sciences de la Terre et des planetes 330: 875-882.

Gregory, J.T. 194 8. The type of Claosaurus( 2) affinis Wieland. American Journal of Science 246 : 29-30.

Head , J.J. 2001. A rean alysis of the phylogenetic posit ion of Eolambia carolj onesa (Dinos auria: Iguanodontia ). Journal of Vertebrate Pale­ontolog y 21: 392-396.

H orner, J. R. 1979. Upper Cret aceous dinosaur s from the Bearpaw Shale (marine) of south-central M ontana with a checklist of Upper Creta­ceous dinosaur remains from marine sediments in Nort h America. Journal of Paleontology 53 : 566- 577.

164 • M icha el K. Brett -Surman and Jon athan R. Wagner

Horner, ]. R., au': l of a new s;: from ~ 1

Horner ~

brid ge: H orner, ]. R.. D. L

In D. B. U-~

nosauris. : : H u C. 1973. A ;:;,~

tung. Ac..; H utchinson, ]. R.. ::

on the line ; Linnean So,

-_.2001 b. line to e:-..-:.z. Society I j~ ­

Lambe, L. M. 19- ­monton f ·__

Lull, R. S., and ~.

America. G Marsh , O . C. IS-;2.,

can [ourn.: Maryariska, T., s:

from the Y Palaeont.

--.1984. P, comments 141.

M ateer, N . 198 1. Cretace and B. S. .. pp. 4-9-75. _

Morris, W. J. 1 9 - '~

icine Form men . JOIl.

Norman, D. B. L sartensis o; Sciences ~ .

-_.1 986. Ornitho de Belgiq

-_.2002. Probactos Lin naean ­

-_.20 04. Hv Osrnols versiry of

Norman, D. B., '" lated 0 . m61ska .e of Califo

Novas, F. E. 19% . 015)' 16: - ..~

Osborn, H . F. 1

Ap....

Dodson, V. 0. 1971, Sedimentology and tapbonomy of the Oldman Fbi nintion. Dinosaur Provincial Park. Alberta. Palaeogeograpby. Palceoclmtatology, PauKoecology 10: 21 -74.

Dollo, L. lS83a. Troisieme note sur les dinosaunens de BtflUSSart. Bulletin Ju Musee Royal d'Histoire Nalurelle de Belgtque 2: 85-126. 1883b. Quarticme note sur les dinouurino de Bcrnissart. Bulletin

tin Musee Royal d'Histoire Naturelle tie Belgtque 2: 223-252. Egi, N.. and D. B. W'cishampel. 2002. Moiphotnetnc analysis or humeral

shapes in hadrosaunds lOrnuhopoda, Duiosauna). Senckenber- giana l.etbaea 82: 43-58.

Krickson, B. It. 1988. Notes on the postcraniuin of Qmttptosaurto. Scien- tific Publications of the Science Museum o/ Minnesota 6(4): 3-13.

Fionllo. A. R. 19%). The first occurrence of Hadrosaur (Dinosauria) remains from the Marine Claggett Fonnarion. Laic Cretaceous of South Cen- tral Montana. Journal ofVertebrate Paleontology 10: 515-517

Gilmore. C. W. 1909. Osreology of the Jurassic reptile CamptOSaurus, With a review of the species and genus, and description of two new species. Proceedings of the U.S. National Museum 36: 197-332.

__^_. 1917. Braclryccratops. a ceratopsian dinosaur from the Two Med- icine Formation of Montana. U.S. Geological Surrey Professional Atprr IUJ: 1-45 1924a. A new species of hadrosaunan dinosaur from the Edmon-

ton Formation I Cretaceous) of Alberta, Canada. Geological Series 43. Bulletin of the Department of Mints, Geological Survey of Gowda 3#: 13-26. 1924b. On the skull and skeleton of Hypaaosaurus, a helmet

crested dinosaur from the Edmonton Cretaceous of Alberta. Geo- logical Series 43. Bulletin of the Department of Mines, Geological Swn*?o/Gnwd*3& 49-64. 1933. On the dinosaurran fauna of the Iron Dabasu formation.

Bulletin of the American Museum of Natural History 67: 23-78. 1946. Notes on recently mounted fossil skeletons in the United

States National Museum. Proceedings of the U.S. National Mu- Mww 96: 201-203.

Uodefroit. P.. Dong Z. XL, P. Bukynck. la H_ and Feng L 199*. Sino- Belgian Cooperative Program. Cretaceous dinosaurs and mammals from Inner Mongolia. I. New Bactrosaurus f Dinosauria: Hadro sauroidea) material from Iren Dabasu (Inner Mongolia, I'. R. China). Bulletin des Institutes Royal des Science Xatuialles Bel- *W"f 68(Suppl): 1-70.

Ctodefroit. P., Zan S., and Jin L. 2000. Charonosaitnis naymensis n.g.. tt.sp.. a lambeosaurinc dinosaur from the Late Maastrichtian of northeastern China. Compte Rendu de I'Academie du Sciences Na- turelles, Paris, series 1. Sciences de la Terre ct des planetes '•> Hi: S75-XK2.

Gregory. J.T. 1948. The type ol Claosaurus(l) affims Wieland. American /our*;/ n/ &iMff 246: 29_)Q.

Head, J. J. 2001. A rcanalysis of the phylogenetic position of Eolamhui caroljonesa (Dinosauna: Iguanodontia). Journal of Vertebrate Pale u*fo6*y 21:392-396.

I lorner, J. R. 1979. Upper Cretaceous dinosaurs from the Bearpaw Shale (marine) of south-centra! Montana with a checklist of Upper Creta- ceous dinosaur remains I rum marine sedimenrs in North America. journal of Paleontology 53: 566-5

164 • Michael K. Brett-Surruan and Jonathan R. Wagner

0: the Old man For­Palaeogeography,

. Bern issart . Bulletin 2: 85-12 6.

Bernissart. Bulletin e 2: 223-252. m alysis of humeral

--uria). Senchenber­

- nptosaurus. Scien­; inesota 6(4 ): 3-13. Dinosaur ia) remains

aceous of South Cen­c-, 10: 515- 517 ~ ri1e Camptosaurus, s rription of two new

•/111 36: 197-332. ~ from the Two Med­

" Survey Professional

'ill from the Edmon­" . Geological Series

Geological Survey of

icrosaurus, a helmet­•e us of Albert a. Ceo­:;- of iVfines, Geological

Da basu form ation. History 67: 23-78. eletons in the United " U.S. N ational Mu­

. Feng L. 1998. Sino ­- urs and mammals Dinosauria: Hadro ­

ner Mo ngolia, P. R. tience Naturalles Bel­

" " tntS jiayinensis n.g. , .2 e M aastr icht ian of "?I1lie du Sciences Na­

_ et des planetes 330:

" Wieland. American

- position of Eolambia ! of Vertebrate Pale-

the Bearpaw Shale cklist of Uppe r Creta­

., '-5 in North America.

Horner,]. R., and P.]. Currie. 1994. Embryonic and neonatal mo rphology of a new species of H ypacrosau rus (Orn ithischia, Lambeosauridae) from M ontana and Albert a. In K. Carpenter, K. F. Hirsch , and ]. R. Horner (eds.), D inosaur Eggs an d Babies, pp.312-336. Ca m­brid ge: Cambridge University Press.

Horner, ]. R., D. B. Weishampel, and C. A. Forster. 2004. Hadrosau ridae. In D. B. Weisham pel, P. Dodson , an d H. Osm 61ska (eds.), Th e Di­nosauria, pp. 438-463". Berkeley: University of Ca lifornia Press.

Hu C. 1973. A new hadros aur from the Cretaceous of Chucheng, Shan­tun g. A cta Geologica Sinica 2: 179- 206.

Hutchinson,]' R. 20 01a . The evolution of pelvic osteology and soft tissues on the line to extant birds (Neornithes). Z oological Journal of the Linnean Society 131: 123-1 68.

___. 20 01b . The evolution of femoral osteology and soft tissues on the line to extant bird s (N eorn ithes) . Z oological Journal of th e Linnean Society 13 1: 169-197.

Lambe, L. M. 1913. The manus in a specimen of Trachodon fro m the Ed ­monton Forma tion of Alberta. Ottawa N atu ralist 27: 21 -25 .

Lull, R. S., an d N. E.. Wright . 1942. Ha dro saur ian dinosaurs of North America. Geological Society of America Special Papers 40: 1- 242.

M arsh, O. C. 1892. Restorati ons of Claosaurus and Ceratosaurus.. Ameri­can Journal of Science 44: 34 3- 34 9.

Marya riska, T , and H. Osm6lska. 1981. First lambe osaurine dinosa ur from the Nemegt Formation, Upper Cretaceous, Mongolia. Acta Palaeontologica Polonica 26: 24 3-255.

_ _ _ . 1984. Postcran ial anatomy of Saurolophus angu stirostris with co mments on other had rosaurs. Palaeontologica Polonica 46: 119­14 1.

Mateer, N. 1981. The rept ilian megafau na from the Kirtland Shale. (Late Cretaceous ) of the San Juan Basin. In S. G. Lucas, ]. K. Rigby Jr., and B. S. Kues (eds .), Advances in San Jua n Basin Paleontology, pp. 49-75. Albu querque: University of N ew Mex ico Press.

Morris, W. ]. 197 8. H yp acrosaurus altispinusi Bro wn from th e Two Med­icine Formation, M ontana, a taxonomically indetermina te speci­men. Journal of Paleontology 52 : 200- 205.

Norman, D. B. 1980 , On the om ithisch ian din osaur Igu an odon bernis­sartensis of Bern issart (Belgium). Memoires de l'lnstitut Royal des Sciences. Naturelles de Belgi qu e 178 : 1-103..

___. 1986. On the anatomy of Igu anodon atberjieldensis (Ornithischia: Ornithopoda ). Bulletin de l'Institut Royal des Sciences N atur elles de Belgiqu e, Sciences. de la Terre 56 : 28 1-372.

___. 2002. On Asian orn ithopods (Dinosauria: Ornith ischia). 4 . Probactosaurus Rozhdestvensky, 1966. Zoological Journal of th e Linnaean Society 136: 113-144.

___. 2004. Basal Iguanodontia . In D. B. Weishampcl, P. Dodson, and H. Osm61ska (eds.), Th e Dinosauria , pp . 4 13-437 . Berkeley: Uni­versi ty of Ca lifornia Press.

Norman, D. B., and D. B. Weishampel. 1990. Iguanodontianide and re­lat ed Ornithopoda, In D. B. Weishampel, P. Dodson, and H. Os ­m61ska (eds.), The Dinosauria, pp. 510- 533. Berkeley: University of Ca liforn ia Pness.

Novas, F. E. 1996. Din osau r mon oph yly, Journal o f Vertebrate Paleontol­ogy 16: 723-741.

Osborn, H . F. 1912. Int egument of the iguanodontian dinosaur Tra-

Appendicular Anatomy in Campanian an d Ma astrichtian N orth American H ad rosaurids .. 165

I lorncr,). R., and P. |. Curric. 1994. Embryonic ,ind ncon.it.il morphology of a new species of Hypacrosaurus (Ornithischia, Lambeosauridae from Montana .ind Alberta. In K. I"arpenrer. K. I:. Hirsch, and J. K. Homer leds.), Dinosam Hggs and Babies, pp. 312-336. Cam- bridge: Cambridge University Press.

I lorner, I. R., D. IS. Weishampcl, and C. A. Forster. 200-1. Fladrosaundae. In U. B. Weuhampd. n Dodvm. and 11. Own,\hka (eds.;, Tk Ot- tiosauria, pp. 438-463'. Berkeley: University oi California Prrv.

Hu C. 197 v A new hadrosaur from the Cretaceous of Chuchcng. Shan- tung. Ada Geologica Sittico 1: 179-206.

liurchmson. J. R. ZOOla. The evolution of pelvic osteology and soft tissues on the line to extant birds (Neormrhcsl. Zoological Journal oftht Luntean Society 131. 123-168. 2001b. The evolution of femoral osteology and soft tissues on the

line to extant birds (Neornithes). Zoological lournal of the lannean Socwry 131: 169-197

I.ambe, L M. 1913. The m.mus in a specimen ol Vrachodon from the Ed- monton Formation of Alberta, Ottawa Naturalist !'•• 21-25.

Lull. U.S., and N. li. Wright. 1942. Iladrosaunan dinosaurs of North America. Geological Society of America Special Papers 40: 1-242.

Marsh, O. C. 1892. Restorations of ('laosauins and Ceratosaumt. Ameri- can Journal of Science 44: 543-349.

M.iryaiiska. I'., and H. Osmolska. 1981. First lambeosaurmc dinosaur from the Ncmcgt Formation. Upper Cretaceous, Mongolia. Ada Palaeontologies Polonica 26: 241-25 S. I9S4. Po-.rcrani.il anatomy of Saurolophus angustirosltis with

comments on other hadrosaurs. Palaeontologies I'olomca 46: 119- 141.

Matcer, N. 1981 The reptilian mcgafaum From the Kirtlaud Shale Hate Cretaceous] oi the San Juan Basin. In S. G. Lucas, J. K. Rigby Jr., and U.S. Kues (eds./. Advances m San Juan Basin Paleontology, pp. 49-75. iMbuquerque: University of New Mexico Press.

Morns, W J. I97S. Hypacrosaurus altispiitus! Brown from the Two Med- icine Formation, Montana, a raxonomically indeterminate Speci- mm. /owrMj/'i/ Tj/i-'wifiVrun' (2: 2(X*-205.

N'orman, D. B. 1980. On the ornithischiar. dinosaur Iguatiodon beino SOrtensis of Bernissart (Belgium). Memoires de I'Institttt Riry.il des Sciences Nalurelles <le Uelgiqiie I 78: 1 — 10.3. 1986. On die anatomy of Iguanodon atherpeldensis [Ornithischia:

Ornithopoda). Hulletin de I'hiilitul Royal des Sciences \alurelles de belgique. Sciences de la Tent 56: 2Sl-?~2. 2002. On Asian ormthopods (Dinosaun.t: Ornithischia). 4.

I'robactosaurus Rozlnlestvcnsky, 1966. Zoological journal o/ the Unnaoan Society 136: 113-144. 2004. Basal Igu.modontni. In D. B. Weishampcl. P. Dodson, and

H. Osmolska (eds.). The Dmosauna, pp. 413—137. Berkeley: Uni- versity of California Press.

Norman, D B., and D. B. Weishampcl. 1990. Iguanodontianide and re- lated Ornithopoda. In D. B. Weishampcl, P. Dodson. and II. Os- molska icds.;. The Dinosauria, pp. J10-533. Berkeley: University of California Press.

Novas. F. E. 1996. Dinosaur monophyly. Journal of Vertebrate Paleontol- ogy 16:723-741.

Osborn, HE 1912. Integument ol the iguanodontian dinosaur Tra-

Appendicular Anatomy in Campaman ami .Yl.iasirichrian North American Hadrosaunds • 165

chodon . Mem oirs of the A m erican Mu seum of Natural History 1: 33-54.

Ostrom, J. H . 1963. Parasauro lophus cyrtocristatus, a crested hadro­saurian dinosaur from New Mexic o. Fieldiana Geology 14: 143­168 .

Parks, W. A. 1919. Preliminar y descripti on of a new species of tra chod ont dinos aur of the genus Kri tosaurus, Kritosaurus incurvimanus. Tran sactions of the Royal Soc iety of Canada, ser. 3, 13: 51- 59.

___. 1920. The osteology of the rrach odont dinosaur Krit osaurus in­curuim anus. University ofToronto Studies, Geological Series 11: 1­74.

___. 1935. New species of trachodont dinosaurs from the Cretaceous form at ions of Alberta with notes on other species. University of Toronto Studies, Geological Series 37: 1--45.

Pinna, G. 1979. Osteologia dello schelerro di Kritosaurus notabilis (Lambe, 1914 ) del Mu seo Civico di Storia N atural e di Milano. Mem orie della Societa Italiana di Scienze N aturali Museo Civ ico Milan o 22: 33-56.

Prieto-Ma rquez, A. 2000. On the postcrania of Bracb ylopbosaurus good­w ini (Dinosaur ia: Ornithopoda ): Implications for hadrosaur mor­phol ogy. Journal of Vertebrate Paleont ology 20(3, Suppl. ): 63A.

Rasmussen, M. E. 199 8. Notes on the morp hology and the orientation of the fore limb of O uranosaurus . Oryctos 1: 127-130.

Romer, A. S. 1927. The pelvic musculature of orni thischian dinosaurs. Acta Z oologica 8: 225-275.

___ . 195 6. O steology of the Reptiles. Chicago: University of Chicago Press.

Rowe, T. 1986. Homology and evolution of the deep dorsal thigh muscles in birds and other Reptili a. Journal of Morph ology 189: 327- 346.

Rozhd estvensky, A. K. 1957. Duck-bill dinosaur-Saur olophus from the Upper Cretaceo us of Mongolia. Verteb rata PalA siati ca 1: 129- 149 .

Sereno, P. C. 1999. The evolution of dinosaurs: Supplementa ry material. Sci­ence. Available at http://www.sciencemag.org/feature/data/1041760. sh!.

Sternberg, C. M. 1935. Hooded hadro saur s of the Belly River Series of the Upper Cretaceous. Bulletin of the N ational Mu seum of Canada 77 : 1-37.

Taq uet, P. 1976. Geologie et palaeontologie du gisement de Gado ufao ua (Aptian du N iger). Cash iers de Paleontologie du Centre National Recherche Scien tifique, Paris, pp. 1-1 91.

Wagner, J. R. 2001. Th e hadr osaurian dinosaurs (Orn ithischia: Hadro­saur ia ) of Big Bend National Park , Brewster Co unty, Texas, with implications for Late Cretaceous paleozoogeography. M.S. thesis. Texas Tech University.

Weishamp el, D. B., and J. R. Horn er. 1990. Hadrosaurid ae. In D. B. Weishampel, P. Dodson, and H. Osm61ska (eds.), The Dinosauria , pp. 534-561. Berkeley: University of California Press.

Weishampel, D. B., D. Grigorescu, and D. B. Norman. 1993. Tel­matosaurus transsyluan icus from the Late Cretaceo us of Romania: Th e most basal hadrosaur id dino saur. Palaeontology 36 : 361-385 .

Wiman, C. 1929. Die Kriede-D inosaurier aus Shantung. Palaeontologia Sini ca n.s.c., 6: 1-67.

Xu Z ., Zh ao X.-J ., Lu jc., Huang W.-B., Li Z .-Y., and Dong Z.-M. 2000. A new Iguan odontian from Sangping Form ation of Neixiang,

166 • Mich ael K. Brett-Surman and Jonathan R. Wagner

Henan ~.:

38 : 179- : ~

Young, C. C. : c Palaeon:

Appendix 8.1. - ­

ACF-acrornior; i ACR-acromio- ­ACT-acetabul AST- astraga lus CAL-ca1cane CRF-coracoid i CRL-carpals CRR-coraco id , DE-dentary DLF-dcltoid foss DLP-deltopecro::j DLR-deltoid ri FEM-femur FIB-fibula FMH-femoral l-, ­GL-glenoid GTR-greater H-height HH-humeral hea HUFL-hindlimL HND-sterna l ha HUM-humerus ILP-iliac ped uncle ISFH-ischial f, ISFT- ischial fQO.:­ISP-ischial ped JU-jugal L-Iength LA-lacrimal LTR-lesser rroca MC3-metacarp:. MC31H-metaca: ratio MT3-metata rsa.; MT3/F-metata-- ­NA-nasal OBF-obtura tor i, OBP-obturator r

See Figure

Humerus

chodon. Memoirs of the American Museum of Natural History 1. 33-54.

Ostrom. J. H. 1961. Parasaurolophus cyrtocristatus, a crested hadro- saurian dinosaur from New Mexico, l-ieldiana Geologx 14: 143- 168.

Parks, W. A. 1919. Preliminary description oi a new species of trachodont dinosaur or the genus Kritosaurus, Kritosaurus tncurvinuinus. Transactions of the Royal Society of Canada, scr. 3, 13: 51-59. 1920. The osteology of the trachodont dinosaur Kritosaurus in-

airvimanus. University of Toronto Studies, Geological Series 11: 1- 74. 1935. New species of trachodont dinosaurs from the Cretaceous

formations of Alberta with notes on other species. University of Toronto Studies, Geological Series 37: 1—45.

Pinna, (i. 1979. Osteologia dcllo scheletro di Kritosaurus notahilis (Lambe. 1914) del .Viuseo Civico di Storia Naturale di Milano. Memorie delta Soaela llaliana di Scienze Naturali Musco Cwico Mitmo 22: 33-56.

Pneto-Marquez. A. 2000. On the postcrama ol Brachylophasaurtis good- wmi (Dinosauria: Ornithopodai: Implications for hadrosaur mor- phology. Journal of Vertebrate Paleontology 20(3. Suppl.): 63A.

Uasmusscn, M. E. 1998. Notes on the morphology and the orientation ol the lorelimb of Oiiranosaurus. Oryctos 1: 127-1 30.

Romer, A. S. 1927 The pelvic musculature of nrr.ithtschi.in dinosaurs. ,L w Zuo/cpca 8: 225-27). 1956. Osteology of the Reptiles. Chicago: University of Chicago

Press. Rnwc, T. 1986. Homology and evolution of the deep dorsal thigh muscles

in birds and other Repulia. Journal of Moiplmlogy 189: 327-346 Rozhdcsrvensky, A. K. 1957. Uuck bill dinosaur—Saurolophus from I he

Upper Cretaceous of Mongolia. Vertebrata PalAsiatica 1: 129-149. Seteno, P. C. 1999. The evolution of dinosaurs: Supplementary material. Sci-

ence. Available at http://www.sciencemaR.org/fedtiirc/data/1041760 Ud.

Sternbcrg, C. M. 1935. Hooded hadrosaurs of the Belly River Series of the Upper Cretaceous. Bulletin o/ the National Museum of Canada 77: 1-37.

Taquct, P. 1976. Geologic er palaeonrologie du gisemeni de Gadoufaoua (Aptian du Niger). Cashiers de I'aleontologie du Centre National Recherche Scientifique, Paris, pp. 1-191.

Wagner, |. R. 2001. The hadrosaurian dinosaurs (Ornithischia: Hadro- saunai of Big Bend National Park, Brewster County, Texas, with implications for I .ate Cretaceous paieozoogcographj. M.S. thesis. Texas Tech University.

Wewhampd, D. B.. mod |. R. Hornet 1990. Hadroauridac. In D. B Weishampel. P. Dodson, and II. Osmolska leds.l. The Dinosauria, pp. 534-561. Berkeley: University of California Press.

Wmdumpcl. D. B_ D. GhpKwcu, and D. B. Noroun. 1993. TM- matosaurus transsylvanuus from the Late Cretaceous of Romania: The most basal hadrosaund dinosaur. Palaeontology 36: 361-385.

Wiman, C. 1929. Die Kriede-Dinosaurier aus Shantung. I'alaeontologia Simca n.s.c, 6: 1-67.

Xu Z_ Zhao X.J. Im J-C. Huang W. &. U Z.Y., and Don* Z.-M. 2000. A new Iguanodonnan from Sangping Formation of Nci\:iang,

166 • Michael K. Breit-Surmar. and Jonathan R. Wagner

-..tural History 1:

.1 crested hadro­Geology 14: 143­

-cies of trachodont •1$ /neurVlmanus.

t, ~ . 13: 51-59. .::- Kritosaurus in­igica! Series 11: 1­

rn the Cretaceous .cies , University of

.oss urus notabilis . ra le di Milano. ..li Museo Civico

ophosaurus good­:- hadrosaur mor­

3. Supp!.): 63A. ~ the orientation of 3 J. ischian dinosaurs.

ersiry of Chicago

rsal thigh muscles c-. ' 189: 327-346.

ophus from the siztica 1: 129-149.

•._,.n y materia!' Sci­..re/data/l 041760.

_ iver Series of the m of Canada 77:

r - de Gadoufaoua Centre National

~-hisch i a : Hadro­. ry, Texas, with

_::- hy. M.S. rhesis.

saur idae. In D. B. . The Dinosauria, _::-$5.

an . 1993. Tel­.eous of Romania:

",., 36: 361-385. "'g_ Palaeontologia

Dong Z.-M. 2000. -=1);] of Neixiang,

Henan and its stratigraphical implications. Vertebrata PalAsiatiea 38: 179-191.

Young, C. C. 1958. The dinosaurian remains of Laiyang, Shantung. Palaeontologiea Siniea, n.s., C, 16: 53-138.

Appendix 8.1. List of abbreviations.

ACF-acromion fossa ACR-acromion ridge ACT-acetabulum AST-astragalus CAL-ealcaneum CRF-coracoid fossa CRL-carpals CRR-coracoid ridge DE-dentary DLF-deltoid fossa DLP-deltopectoral crest DLR-deltoid ridge FEM-femur FIB-fibula FMH-femoral head GL-glenoid GTR-greater trochanter H-height HH-humeral head HLIFL-hindlimblforelimb ratio HND-sternal handle HUM-humerus ILP-iliac peduncle ISFH-ischial foot-heel ISFT-ischial foot-toe ISP-ischial peduncle ]U-jugal L-length LA-lacrimal LTR-lesser trochanter MC3-metacarpal III MC31H-metacarpal III/humerus ratio MT3-metatarsal III MT3IF-metatarsal III/femur ratio NA-nasal OBF-obturator fossa OBP-obturator process

OLN-olecranon notch OLP-olecranon process PBH-pubic blade height PBL-pubic blade length PD-predentary PDL-sternal paddle PF-prefrontal PMl-premaxilla-l PM2-premaxilla-2 PO-postorbital POEP-postorbital eye pocket POP-postacetabuJar process PP-postpubis PRP-preacetabular process PUB-pubic blade PUN-pubic neck PUP-pubic peduncle QJ-quadratojugal QU-quadrate RIH-radius/humerus ratio RAD-radius RPL-reflected premaxillary lips SCAP-scapula SCBL-scapular blade SHF-shaft SQ-squamosal STB-sternal body (paddle) STH-sternal handle SUIC-supraiJiac crest TIF-tibialfemur ratio TIB-tibia TR3-third trochanter TR4-fourth trochanter TRS-tarsals UIH-ulna/humerus ratio ULN-ulna UNG-ungual W-width

Appendix 8.2. System of standardized measurements.

See Figure 8.8 for location of measurements.

Humerus

Length from 'a' to 'b' Length of deltopectoral process from 'e' to 'f' Width of deltopectoral process from 'c' to 'd'

Appendicular Anatomy in Campanian and Maastrichtian North American Hadrosaurids • 167

Henan and us gtrarjgraphical implications. Vertebrate falAsiatica 36: 179-191.

Young, C. C. 1956. The dinosaurisn remains of Laiyang. Shantung. PaknwifokgMra inwrw. n.*.. C 16: 53-138.

Appendix 8.1. List of abbreviations.

ACF—acromion fossa ACR—acromion ridge ACT—acetabulum AST—astragalus (IAL—calcaneuoi CRF—coracoid lossa CRl.—carpal; C.RR—coracoid ndgc DE—dentary DI.F—deltoid fossa Dl P—deltopectoral crest DLR—deltoid ridge FFM—femur FIB—fibula I Ml I- femoral head GL—glenoid CiTR—greatei trochantcr H—height HH—humeral head HUH.—liindlimb/foiclimb ratio FIND—sternal handle HUM—huinerus 1LI'—iliac peduncle ISFH—ischial foot—heel ISFT—ischial foot—toe ISP—ischial peduncle JU-iuRal L—length LA—lacnmal I.TR—lesser rrochanter MCI—inctac.irp.il III MC'.VFI—iiictjc.irpal lll/humertis ratio MT3—mctatarsal III MT.VF—metatarsa! Ill/femur ratio NA—nasal O111-—obturator fossa OBP—obturator process

OI.N—olccranon notch OI.I'—olccranon process PBF1—pubic Made height PBL—pubic blade length I'D—predentary PDF— sternal paddle PF—pretrontal I'M I—pre maxilla—1 PM2—premaxill?.—2 PO—post orbital I'OF.P—postorbital eye pocket POP—postacetabular process I'!'—postpubis PRP—prcacetabular process PUB—pubic blade PUN—pubic neck PUP—pubic peduncle Q_| — t|iiadratojugal QU—quadrate R/H—radius/humcrus ratio RAD—radiw RPL—reflected premaxillary lips SCAP—scapula SCBL—scapular blade SHF-*hak SQ—squatnosal STB—sternal body (paddle) STH—sternal handle SUIC— suprailiac crest T/F—tibia/femur ratio TIIJ—ribi.i TR3—third trochantcr TR4 — fourth trochantcr TRS—taraah U/H—iilna/humcrus ratio ULN—ulna UNG—wngual W—width

Appendix 8.2. System of standardized measurements.

Sec Figure 8.8 for location of measurements.

Huinerus

Length from "a* to 'b' Length ol deltopectoral process Irom 'e' to "!' Width ol deltopectoral process from 'c' to 'd'

Appi-ndicular Anatomy in Lam pa man and Maasrriclin.in North American Hadrosaunds • 167

c

E

~

B

E

A

c

A Figure 8.8. Location for 8 standardized measurements (see Appendix 8.l) . (A) Humerus; (B) pubis; (C) ilium.

Pubis

Length from 'a' to 'b' Length of blad e from 'a ' to If' Width of blad e from 'c' to ' d' Width of neck from 'h' to 'i '

llium

Length from 'a ' to 'b' Length of postacerabular pr ocess from 'a' to 'h' Width of postacetabular proc ess from "e' to If' Width of iliac bod y from 'c' to 'd ' The placement of points for measurements are as follows: On the

humerus, line "ab' indicates the maximum distances where 'a' fall s on the midpoint of the hea d and 'b' is on the asymptote of the distal trochlea. Line 'ef' defines the deltopectoral crest, and each point falls on the asymp­tote. Line 'cd' is perp endicular to 'ef.' Other measurements proved to be taxonomically useless.

O n the pubis: Line "ab' is the length wh ere line 'af' bisects the blade into do rsal and ventra l segments. Point ' b' is on the asymptote of the ace­

168 • Michael K. Brett-Surman and jonathan R. Wagner

c

tabulum. Line 'cO:' pendicular to 'at" imum width of , to •

height. On the ilium: :

lars. Line 'cd' is :: acetabulum. Poin: antitrochanter. P, ­cleo Line 'hg' is 2:; now defines the '; and is perpen dicu ­sects the preaceta good measure of ~

diagnostic elernenzs Measu rements

twe en perpendicu point except in G: the fourth rr ochaz;

of the scapula is " the asympto te oi 8.1A) . A line half<' scapular blade 0 :';'" becau se of the ex .':=­is th erefore rneaser now th e diam eter

Caut ionary ~ . the measurements viou s publications graphs varied fr- ­from the same crea sing size of ct~ ~

on ph otographs,

Appen •

hgiire S.S. Location for standardized measurements (sec

pubis; ICl ilium

Pubis

Length from 'a' to 'b' Length of blade from 'a' to T Width ol blade Irom "c" to'd' Width of neck from 'h' to Y

Ilium

Length Iron: *a' to 'b' Length of posracetabular process from a' to *h' Width ol postacetabular process from 'e' to T Width of iliac body from c' to 'd' The placement of points for measurements are as follows: On the

humerus, line *ab' indicates the maximum distances where "a" falls on the midpoint ol the head and 'b' is on the asymptote of the distal trochlea. Line "ef defines the dcltopcctoral crest, and each point falls on the asymp- tote, line cd' is perpendicular to "cf.' Other measurements proved to be taxonoitiitally uselew.

On the pubis: Line 'ab' is the length where line 'at" bisects the blade into dorsal and ventral segments. Point b' is on the asymptote ol the acc-

168 • Michael K. Rreti-Surman and Jonathan It. Wagner

8

tabulum. Line 'cd' is perpendicular to 'af' and bisects it. Line 'eg' is per ­pendi cular to 'af' where poi nt 'e' is on the asymp tote. Line ' ih' is the min­imum width of the neck. Line 'af' is th e blade length , and line 'cd' is the height.

On the ilium: Line 'a b' is the maximum length between perpendicu­lars. Line 'cd' is perpendicular to 'ab' where'd' is on the asympto te of th e acetabulum. Poin t 'c ' mayor may not fall on the anterior bor der of the antitrochanter. Point 'g' lies on the posterior border of th e ischial pedun­cle. Line 'hg' is drawn perpendicular to line 'ab' at this point. Line 'a h' now defines th e pos tace tabular process length. Line 'e f bisects this line and is perpend icular to it, thus defining the height. Line 'cb' usua lly bi­sects the preacetabular process, thus form ing angle L DCB, which is a good measure of the preacetabular process deflect ion. Th ese are the most diagnostic elements.

Measur ements for other elements are simply the maximum length be­tween perpend iculars and the min imum diameter of the sha ft at its mid­point except in the case of the femur where it is taken halfway betwee n the fourth trochan ter and the ex treme distal end of the femur. The length of the scap ula is tak en along the dorsal border from the coracoid ridge to the asympt ote of th e dorsal bor der at the end of the scap ular blade (Fig. 8.1A ). A line halfway along this line perpendicular to the trend of the scapular blad e defines the height . Thi s is not possible in camptosaur ids because of th e exaggera ted development of the acromial ridg e. Th e length is therefore measured between perpendiculars (Fig. 8.1B). Th e height is now the diameter of the blade midway between these two points.

Cautiona ry N ote. We have found dur ing these studies th at many of the measurements in Lull and Wright (1942) are tak en directly from pre ­vious publicat ions. Many measurements that were made from photo­graphs varied from actua l measurement s tha t one of us (M. B.-S.) made from the same bone. They seem to vary in direct prop ortion to the in­creasing size of the bone. Th e reade r is cautio ned against relying tot ally on photographs.

w s: On the -re 'a' falls on th e distal trochlea. ~ , s on the asyrnp­

rsproved to be

App endicular Anatomy in Campanian and Maastricht ian North Amer ican Hadrosaurids • 169

I--'--' vhweprnm :.':'->;-,-'- l=iMf'ih'A;hfmm: - - "-'- '"" — ^ '-- -' '"'" ''"'pWvh,dmllmw\i = i'-

"'"'-ii ' ' '--")tm4mlmHhlwm:^"i"i '--

*FHh-' ' .^HWyrnHMl' lUHftW"'-'

d& UMf 'h@= k WMWH pwppmWm in lm= yh' 4f ih;» mum: IHHF ^h mwfWmmihfpiW":''.'", --''-,"- '=- ' '::,.!':nwlim

(RhlhfH"-'-'"'", ''-"'IliimiW-:'' "'W:Wh!t'- 'mWWfFUffl ' ,- =1 "" ".. II.. :„„*|

dw#H8MiteWwii*:

Mr*m pefpemkwkf* 4ml ihtmmimmi' '"-'-' '"- --' -"til f (m ihe H# m# ihf fpmm whop ii M wkF' ';ii'= "< 1^1^4:1: '^' .1:'. -- ' ic; 4HW fm= HHffmf ""-' - -' -'"'- ' ' ht= kmiwh '-'- 1.....1 .1..:... .:.. .1 1... 1 1...1... ..

*:IA'= A !im= iHtlf 4* fi"?!g ilm Ime ppri-- '= -'- in ihp mmd fi ;hp , - ='-' ' '-''khm!': " "-HiU'whkm^m^iMHfitk

' ' -'ll*MAW!Pf4lHlA:^!t'l '''-^'nm'*illtl@&Th(rl Alhm^' ' ' "'i'ripPHWl^'""':,. M:IMl: fWnHiNhiA mwvihf Wn 1 '- " ' ' -. '^*m«iWiwup'.,...

- 4#m#wn'\,«k: ^h4*pmHmW4wM" I- ' ' uu! ikmwmfmt:-";-' -" -""' * = e'"' 'a=Ui ni^ wkm tJifhin imm pm: -1 -'" "'-it Mam mw -.'* dwi we# '-». ...,-

....:.,., ...;..! I...; ^lm^*MIM;HlhllMlWK"li4lM '- "- -4tk: ' - - ' - "- - - , ^m

ApjVHtliiHkf Aiwimm m WmpHHHM 4Hd MwHwhwii NwHh Am-= - ''..:..-. J* * Im*