Ontogenetic stages in the longbone histology of sauropod ... · Ontogenetic stages in the longbone histology of sauropod dinosaurs Nicole Klein and Martin Sander Abstract.—Long

� 2008 The Paleontological Society. All rights reserved. 0094-8373/08/3402-0006/$1.00

Paleobiology, 34(2), 2008, pp. 247–263

Ontogenetic stages in the long bone histology ofsauropod dinosaurs

Nicole Klein and Martin Sander

Abstract.—Long bones (femora, humeri) are the most abundant remains of sauropod dinosaurs.Their length is a good proxy for body length and body mass, and their histology is informativeabout ontogenetic age. Here we provide a comparative assessment of histologic changes in growthseries of several sauropod taxa, including diplodocids (Apatosaurus, Diplodocus, indeterminate Di-plodocinae from the Tendaguru Beds and from the Morrison Formation), basal macronarians (Ca-marasaurus, Brachiosaurus, Europasaurus), and titanosaurs (Phuwiangosaurus, Ampelosaurus). A totalof 167 long bones, mainly humeri and femora, and 18 limb girdle bones were sampled. Samplingwas performed by core drilling at prescribed locations at midshaft, and 13 histologic ontogeneticstages (HOS stages) were recognized. Because growth of all sauropod long bones is quite uniform,with laminar fibrolamellar bone being the dominant tissue, HOS stages could be recognized acrosstaxa, although with minor differences. Histologic ontogenetic stages generally correlate closelywith body size and thus provide a means to resolve important issue like the ontogenetic status ofquestionable specimens. We hypothesize that sexual maturity was attained at HOS-8, well beforemaximum size was attained, but we did not find sexually differentiated growth trajectories sub-sequent to HOS-8. On the basis of HOS stages, we detected two morphotypes in the Camarasaurussample, a small one (type 1) and a larger one (type 2), presumably representing different speciesor sexual dimorphism.

Nicole Klein and Martin Sander. Institut fur Palaontologie, Universitat Bonn, Nußallee 8, D-53115 Bonn,Germany. E-mail: [email protected], [email protected]

Accepted: 25 October 2007

Introduction

In extant wild vertebrates, age determina-tion is possible by mark-release-recapturestudies. Additionally morphology and size ofan individual can be indicators of age. Obvi-ously, the mark-release-recapture method isnot applicable to extinct animals, and alsomorphology and size as age indicators are dif-ficult to use in fossils. Often it is not clear, be-cause of the incompleteness of the fossil re-cord, whether differences in size and mor-phology reflect ontogenetic and allometric ortaxonomic differences. However, because ofontogenetic changes in growth rate and bonetissue type, long bone histology can also re-cord the ontogenetic history of an individual.

Since the last decade, histology of fossilbone has become a frequently used and pro-ductive method for studying the ontogenyand life history of extinct amniotes and espe-cially dinosaurs. Thus, long bone histologyprovides a unique approach to gathering in-formation about the biology and life history ofdinosaurs.

Early ontogenetic stages have been de-scribed on the basis of morphology and size

for other dinosaurs, but juveniles are ratherrare for sauropods (Gilmore 1925; Carpenterand McIntosh 1994; Martin 1994; Chiappe andCoria 2001; Lehmann and Coulson 2002; Wil-hite 2003; Bonnan 2004; Foster 2005; Ikejiri etal. 2005; Tidwell and Wilhite 2005; Sander etal. 2006; Schwarz et al. 2007).

Histologic determination of the ontogeneticstatus in sauropods has the potential to re-solve the question of whether a small sauro-pod specimen is a juvenile or a dwarf form.For example, the dwarf status of the diminu-tive basal macronarian Europasaurus holgeri(Sander et al. 2006) was proven histologically,as well as the juvenile status of a small diplo-docid specimen from the Morrison Formation(Schwarz et al. 2007). How difficult dwarfingis to prove without histology is illustrated bythe titanosaurid Magyarosaurus from the LateCretaceous of Romania; although this is theclassical example of insular dwarfing in di-nosaurs (Nopcsa 1914; Jianu and Weishampel1999), its dwarf status has recently been ques-tioned (Le Loeuff 2005b). Additionally, taxo-nomic diversity can be detected by bone his-tology in a morphologically homogeneous

248 NICOLE KLEIN AND MARTIN SANDER

sample (e.g., Camarasaurus, Ampelosaurus, Phu-wiangosaurus, this study).

Studies on ontogenetic stages using bonehistology have been done before in dinosaurs(Varricchio 1993; Chinsamy 1994; Curry 1999;Erickson and Tumanova 2000; Horner et al.2000; Sander 2000; Horner and Padian 2004;Chinsamy-Turan 2005; Erickson 2005). How-ever, because of biological and taphonomic fil-ters, juveniles and adults of a single speciesare rarely preserved together at the same lo-cality. Moreover, because juveniles rarelyshow morphological autapomorphies, formost species it is still difficult to assign thesespecimens to the corresponding adult speci-mens even when they are found together (e.g.,Chiappe et al. 2005; Schwarz et al. 2007). Ad-ditionally, most of the bone histological stud-ies in the past have dealt only with two on-togenetic stages, the late stage of activegrowth, indicated in most dinosaurs by lami-nar fibrolamellar bone, and the cessation ofgrowth (Varricchio 1993; Chinsamy 1994; Cur-ry 1999; Erickson and Tumanova 2000; Horneret al. 2000; Sander 2000; Horner et al. 2001;Horner and Padian 2004; Chinsamy-Turan2005; Erickson 2005). In fact, it is becomingcommon practice in the description of new di-nosaur taxa to use bone histology to ascertainthe ontogenetic stage of the type material(Sander et al. 2006; Erickson et al. 2006; Xu etal. 2007). The cessation of growth is indicatedby avascular lamellar-zonal bone in the out-ermost cortex, the so-called external funda-mental system (EFS) (Cormack 1987) or outercircumferential lamella (OCL) (Chinsamy-Tur-an 2005). Mainly because dinosaurs of earlyontogenetic stages are poorly represented inthe fossil record and difficult to identify, onlya few studies have dealt with earlier ontoge-netic stages or a range of ontogenetic stageson the basis of bone histology (Varricchio1993; Curry 1999; Erickson and Tumanova2000; Horner et al. 2000; Chinsamy-Turan2005; Erickson 2005).

One of the best and, concerning the ontog-eny, most complete studies is that of the had-rosaurid Maiasaura by Horner et al. (2000),who studied various long bones and otherbones. They described six gradational growthstages in the principally fibrolamellar primary

bone tissue: early and late nestling, early andlate juvenile, subadult, and adult. These his-tological growth stages are also well support-ed by the relative sizes of the bones (Horner etal. 2000: Table 1).

Erickson and Tumanova (2000) studied agrowth series from juveniles to adults of the cer-atopsian Psittacosaurus monogoliensis. They rec-ognized seven different growth stages (growthstages A–G), which were defined by the numberof lines of arrested growth and the femur size(Erickson and Tumanova 2000: Table 2).

Curry (1999) studied an ontogenetic series ofApatosaurus on the basis of the histology of radii,ulnae, and scapulae. She distinguished four dif-ferent osteogenic phases: a juvenile stage (�73%of adult size), which is subdivided into an early(�34% of adult size) and a late (�73% of adultsize) juvenile stage, a subadult stage (91% ofadult size), and an adult, fully grown stage.Sander (1999, 2000) independently recognizedessentially the same four stages in the sauropodtaxa from the Late Jurassic Tendaguru Beds, in-terpreting them as hatchling, juvenile, adult,and fully grown (‘‘senile’’) on the basis of thehypothesis that sexual maturity occurred wellbefore maximum size. Curry (1999), on the oth-er hand, had hypothesized that only the fullygrown individual in her sample was a sexuallymature adult.

It is the purpose of this paper to describe indetail ontogenetic changes in the histology ofsauropod long bones based on a large and tax-onomically diverse sample of mainly neosau-ropods, greatly expanding on the study bySander (2000). We erect 13 histologic ontoge-netic stages (HOS stages) that serve to com-pare life histories of sauropods. This approachis particularly useful in the virtual absence ofa growth mark record in sauropods, whichmakes quantitative life-history comparisonsimpossible. The terminology largely followsFrancillon-Vieillot et al. (1990).

Museum Abbreviations

BYU Museum of Earth Sciences, BrighamYoung University, Provo, Utah, USA

CM Carnegie Museum of Natural History,Pennsylvania, USA

DFMMh/FV Dinosaurier-FreilichtmuseumMunchehagen/Verein zur Forderung der

249SAUROPOD LONG BONE HISTOLOGY

TA

BL

E1.

On

tog

enet

icva

riat

ion

inlo

ng

bon

eh

isto

log

yis

sub

div

ided

into

seve

nb

one

tiss

ue

typ

esan

d13

his

tolo

gic

onto

gen

etic

stag

es(H

OS

stag

e).

Bon

eti

ssu

ety

pe

HO

Sst

age

Cor

tex

con

sist

sof

typ

eA

bon

eti

ssu

eH

OS-

1

Cor

tex

con

sist

sp

rim

aril

yof

typ

eA

bon

eti

ssu

ew

ith

typ

eB

bon

eti

ssu

ela

idd

own

inth

eou

ter

cort

exH

OS-

2

Cor

tex

con

sist

sp

rim

aril

yof

typ

eB

bon

eti

ssu

ew

hil

ein

the

inn

erco

rtex

rem

ain

sof

typ

eA

bon

eti

ssu

eca

nb

ep

rese

rved

HO

S-3

Cor

tex

con

sist

sp

rim

aril

yof

typ

eB

bon

eti

ssu

ew

ith

typ

eC

bon

eti

ssu

ela

idd

own

inth

eou

ter

cort

exH

OS-

4

Cor

tex

con

sist

sp

rim

aril

yof

typ

eC

bon

eti

ssu

ew

hil

ein

the

inn

erco

rtex

rem

ain

sof

typ

eB

bon

eti

ssu

eca

nb

ep

rese

rved

HO

S-5

Cor

tex

con

sist

sp

rim

aril

yof

typ

eC

bon

eti

ssu

ew

ith

typ

eD

bon

eti

ssu

ela

idd

own

inth

eou

ter

cort

exH

OS-

6

Cor

tex

con

sist

sp

rim

aril

yof

typ

eD

bon

eti

ssu

ew

hil

ein

the

inn

erco

rtex

rem

ain

sof

typ

eC

bon

eti

ssu

eca

nb

ep

rese

rved

HO

S-7

Cor

tex

con

sist

sp

rim

aril

yof

typ

eD

bon

eti

ssu

ew

ith

typ

eE

bon

eti

ssu

ela

idd

own

inth

eou

ter

cort

exH

OS-

8

Cor

tex

con

sist

sp

rim

aril

yof

typ

eE

bon

eti

ssu

ew

hil

ein

the

inn

erco

rtex

rem

ain

sof

typ

eD

bon

eti

ssu

eca

nb

ep

rese

rved

HO

S-9

Cor

tex

con

sist

sp

rim

aril

yof

typ

eE

bon

eti

ssu

ew

ith

typ

eF

bon

eti

ssu

ela

idd

own

inth

eou

ter

cort

exH

OS-

10

Cor

tex

con

sist

sp

rim

aril

yof

typ

eF

bon

eti

ssu

ew

hil

ein

the

inn

erco

rtex

rem

ain

sof

typ

eE

bon

eti

ssu

eca

nb

ep

rese

rved

HO

S-11

Cor

tex

con

sist

sp

rim

aril

yof

typ

eF

bon

eti

ssu

ew

hil

ein

the

oute

rco

rtex

anE

FS

isd

epos

ited

HO

S-12

Cor

tex

con

sist

sof

typ

eG

bon

eti

ssu

ew

hic

hm

ean

sit

isn

earl

yco

mp

lete

lyre

mod

eled

byse

con

dar

yos

teon

sH

OS-

13

Niedersachsischen Palaontologie (e.V.), Ger-many

IPB Institut fur Palaontologie, UniversitatBonn, Germany

LMC Local Museum Cruzy, Cruzy, FranceMDE Musee des Dinosaures, Esperaza,

FranceMfN Museum fur Naturkunde der Hum-

boldt-Universitat zu Berlin, Berlin, Germa-ny

OMNH Oklahoma Museum of Natural His-tory, Norman, Oklahoma, U.S.A.

PC.DMR Palaeontological collection, De-partment of Mineral Resources, Khon Kaen,Kalasin, Thailand

SMA Sauriermuseum Aathal, Aathal, Can-ton Zurich, Switzerland

Material

Specimens Studied. The histological database consists of 185 samples (167 long bonesamples, 18 girdle bone samples) derived fromover 12 well-known sauropod taxa (see Appen-dix online at http://dx.doi.org/10.1666/07017.s1). The material comes from different lo-calities in North America (Late Jurassic Morri-son Formation), East Africa (Late Jurassic Ten-daguru Beds), Europe (Late Jurassic of the Low-er Saxony Basin, northern Germany; Late Cre-taceous of the Marnes Rouges InferieuresFormation and Marnes de la Maurine member,department of Aude, France), and northernThailand (Late Triassic Nam Phong Formation;Early Cretaceous Sao Khua Formation). It wasobtained during several trips to different collec-tions over the last decade. Identification of thematerial is based on the collections labels and,in some cases, on more current published infor-mation (see Appendix). Species assignmentbased on long bones is difficult in sauropods,and many isolated long bones were identifiedonly to genus or subfamily level. This is why wediscuss the material largely at this level. Figure1 gives an overview of the phylogeny of thestudied sauropods. All samples were obtainedand studied first hand, except for the smallApatosaurus tibia OMNH 1285, which was sam-pled and processed by Andrew Lee (Universityof California, Berkeley), who kindly provided ahigh-resolution photomicrograph for this study.


FIGURE 1. Sauropod relationships after Wilson (2002).The genera and the subfamily printed in bold were sam-pled for this study.

Methods

Sampling by Coring and Cross-Sectioning.Because of their relatively simple appositionalgrowth and for a better comparison, we studiedlong bones predominately. The material wasusually sampled by core drilling with a dia-mond drill bit at a standardized sampling lo-cation in the midshaft region of the bones (Sand-er 1999, 2000; Klein and Sander 2007). Addition-ally, cross-sections of the entire midshaft weretaken in Isanosaurus, Europasaurus, and Ampelo-saurus. The midshaft region of the long bonesequals the neutral growth region, which con-tains the most complete growth record. If at allpossible, the cores were drilled on the anteriorside of the middle of the femur shafts and on theposterior side of the middle of the humerusshafts. It is crucial that core samples are takenfrom these standardized locations because oth-erwise comparability is compromised. Howev-er, in a few specimens, the bone surface wasdamaged at the prescribed location or the bone

was reconstructed in plaster, necessitating sam-pling at a different location in the midshaft re-gion. We drilled completely through some limbgirdle bones, thus sampling the medial and thelateral cortex.

The cores as well as the cross-sections wereprocessed into polished thin sections by stan-dard petrographic methods. For more detailssee Sander (1999, 2000) and Klein and Sander(2007).

Methods of Histological and Skeletochronologi-cal Study. Thin sections were photographedwith a digital camera (Nikon D1) using a bel-lows, resulting in a field of view of about 3 cm.Images of sections that were larger than thiswere created from two or three exposureswith Photoshop computer software. To ob-serve the histology in greater detail, all thinsections were examined by standard light mi-croscopic techniques (normal transmittedlight, polarized light) with a Leica DMLPcompound microscope (16� to 400� magni-fication). Polished sections were studied in in-cident light in dark-field and bright-field il-lumination also with the Leica DMLP com-pound microscope and with a binocular mi-croscope (Sander 2000).

Ontogeny and Changes in Bone Histology

General Ontogenetic Changes in Long Bone His-tology. All sauropod long bones that westudied grew fairly uniformly, laying downlaminar fibrolamellar bone tissue (FBL) intheir cortex. Differences in primary bone tis-sue types mainly concern the organization ofthe vascular system, the degree of vasculari-zation, and the presence and degree of devel-opment of primary osteons (i.e., the amount oflamellar bone deposited centripetally in thevascular canals). In cortical bone, vasculardensity is closely linked with bone appositionrate; i.e., less vascularization means a lowerapposition rate (Amprino 1947; Chinsamy1993; Curry 1999; Erickson and Tumanova2000; Horner et al. 2000; Erickson 2005). Thiscorrelation can be applied to qualitative stud-ies of growth rate where midshaft bone ap-position rate is used as a proxy for the growthrate of the entire animal. Vascularization, andtherefore growth rate, decreases in generalwith increasing age. This was noted in many


previous bone histological studies (Chinsamy1993; Curry 1999; Erickson and Tumanova2000; Horner et al. 2000; Chinsamy-Turan2005; summarized in Erickson 2005). This ob-servation is also made in the current sauropodsample, but the extensive database from verysmall to very large individuals allows a re-finement of this general observation.

Initial deposition of FBL is so fast that thevascular canals are large and only later arecentripetally infilled by lamellar bone (Chin-samy-Turan 2005) forming primary osteons.We used the degree of this infilling for our on-togenetic division of bone tissue types. How-ever, it should be noted that differences in de-velopment of the primary osteons among dif-ferent bone tissue types do not represent com-pleteness of infilling as the bone is depositedby the periosteum but rather varying degreesof final development, after periosteal bone de-position has ceased. In fact, many specimensshow growing laminar fibrolamellar bone inthe outermost three to five laminae where thegradual infilling of the primary osteons can beobserved in addition to the ontogenetic se-quence of development of the primary osteonsdeeper in the cortex.

Remodeling of bone means the replacementof primary bone by bone resorption and sub-sequent deposition of secondary bone. In con-trast to some other dinosaurs, sauropods oftenshow well-developed dense secondary oste-ons (Sander 2000; Chinsamy-Turan 2005: p.84), even several generations of secondary os-teons can be deposited in a single sample(� Haversian bone). In our study we found aclose correlation between the histologic onto-genetic stage of the primary bone tissue andthe number of secondary osteons. With in-creasing size and therefore also age, the num-ber and density of secondary osteons increas-es, too. Thus, we included the degree of re-modeling of the cortical bone in the character-ization of the HOS stages.

The primary cortex is usually not stratified bygrowth marks in sauropods. Only very fewspecimens of various taxa develop growthmarks in their middle and outer cortex, mainlylines of arrested growth (Sander 2000; Sanderand Tuckmantel 2003). The only known excep-tion is the dwarf sauropod Europasaurus (Sander

et al. 2006), which combines fibrolamellar bonewith regularly spaced lines of arrested growth.However, the development of growth marks oran EFS is also an important indicator of onto-genetic age, as was documented in several pre-vious studies (Chinsamy 1993; Varricchio 1993;Curry 1999; Erickson and Tumanova 2000; Hor-ner et al. 2000; Chinsamy-Turan 2005; Erickson2005).

Using changes in laminar fibrolamellar bonetissue and in intensity of remodeling by sec-ondary osteons (number and density), we wereable to differentiate seven bone tissue typesthat correspond to 13 HOS stages. The numberof HOS stages is greater than that of the de-scribed bone tissue types because the transi-tions between the ontogenetic bone tissuetypes are always gradual, resulting in transi-tional stages where one or more bone tissuetypes are preserved in a single specimen. TheHOS stages include these gradual transitionsand thus allow a more precise description ofthe ontogenetic status of an individual than doontogenetic bone tissue types alone.

Ontogenetic Bone Tissue Types in Long Bones.Horner et al. (2000: Fig. 2A) and Horner et al.(2001) described and figured embryonic bonetissue of embryos of Maiasaura and of some re-cent archosaurs. The embryonic stage is notpreserved in the current sauropod sample butit is likely that sauropod embryonic bonelooked similar to that in Maiasaura. From thisand because of the non-laminar organizationof the subsequent bone tissue type (type B) itis quite certain that the vascularization of thebone tissue laid down earliest in sauropods,designated here as type A, did not have a lam-inar organization. Type B bone tissue is fibro-lamellar bone that is dominated by woven orfibrous bone. The vascularization is not lami-nar but mainly longitudinal, and the densityof the vascular canals is very high. The vas-cular canals are large and essentially circularin cross-section (Fig. 2A,B), but with an irreg-ular margin, similar to small erosion cavitiesof the remodeling zone. However, the smallerosion cavities are easily distinguished fromprimary vascular canals of type B bone tissueby their resorption margins (Fig. 2O). Type Bbone tissue normally has no true primary os-teons developed, and only a thin sheath of la-



←

FIGURE 2. A–P, Ontogenetic bone tissue types in the long bone cortex of sauropod dinosaurs. All photomicrographswere taken in normal transmitted light unless noted otherwise. A, Type B bone tissue in Phuwiangosaurus PC.DMRPW4-6 (femur fragment of unknown bone length). B, Detail of type B bone tissue in Apatosaurus OMNH 1279 (femur,34.0 cm). Note the irregular arrangement of the vascular canals and their high density. C, Type C bone tissue inApatosaurus OMNH 1278 (humerus, 25.8 cm) with longitudinal and circular vascular canals. Note the increasinglaminar organization of the tissue type but the still wide-open vascular canals. D, Radial to reticular vascular or-ganization in type C bone tissue in Phuwiangosaurus PC.DMR K16-33 (left femur, 39.0 cm). E, F, Detail of vascularcanals in type C bone tissue in a diplodocid from the Tendaguru Beds (MfN NW4, femur 135.0 cm). The formationof the primary osteons has started, as can be seen in polarized light. E, Normal transmitted light, F, Polarized light.G, Type D bone tissue in Apatosaurus (CM 21715, humerus 79.2 cm). Note the laminar organization of this tissue.Polarized light. H, I, Detail of vascular canals in type D bone tissue in a diplodocid from the Tendaguru Beds (MfNNW4, femur 135.0 cm). The formation of the primary osteons has continued but the vascular canals are still fairlylarge. H, Normal transmitted light, I, Polarized light. J, Type E bone tissue in a diplodocid from the Howe Ranchlocality (SMA 647-87-1, femur 120.0 cm). Note the narrow vascular canals and the strongly laminar organization.K, Type E bone tissue in a diplodocid from the Howe Ranch locality (SMA 647-87-1, femur 120.0 cm) under polar-ized light. L, Detail of a primary osteon in type E bone tissue in Apatosaurus (BYU 601-17328, femur 158.0 cm).Polarized light. M, Primary osteon in type F bone tissue in Apatosaurus (BYU 601-17328, femur 158.0 cm). Note theincreasing amount of lamellar bone within the fibrolamellar complex. Polarized light. N, An EFS deposited in theouter cortex of Apatosaurus (SMA M4/10-1, femur 144.0 cm). Note the closely spaced growth lines and the nearlyavascular bone between the outermost growth lines. Polarized light. O, Young and large secondary osteons in theremodeling zone of a diplodocid from the Tendaguru Beds (MfN XI a7, humerus 80.5 cm). Polarized light. P, Densesecondary osteons in the cortex of Ampelosaurus (LMC3 602 humerus fragment of unknown bone length). Polarizedlight. Q, Tissue reminiscent of type B or type C bone tissue in a small Apatosaurus scapula (OMNH 1300, 35.5 cm).R, Opposite bone side with much more mature bone tissue in the same Apatosaurus scapula (OMNH 1300, 35.5 cm).Note the irregular vascularization.

mellar bone lines the vascular canals, indicat-ing that primary osteon formation has started.No secondary osteons or growth marks aredeveloped in type B bone tissue. If preservedat all, type B bone tissue occurs in our sampleonly in interstitial areas between the erosioncavities of the remodeling zone. It grades in allspecimens into laminar type C bone tissue.Type B bone tissue is shown in Figure 3A.

Type C bone tissue (Fig. 3A–C) consists of aprimarily laminar fibrolamellar bone with a stillvery high vascular canal density. Type C bonetissue usually also starts with longitudinal vas-cular canals that later grade into vascular canalswith a more circumferential organization. How-ever, longitudinal vascular canals in type Cbone tissue are easy to distinguish from thosein type B by a more regularly round margincompared to the irregular longitudinal vascularcanals in type B bone tissue (Fig. 2C). With in-creasing age, the vascularization gradually ac-quires a completely laminar organization, andsimultaneously the formation of true primaryosteons starts, indicated by a more distinct lin-ing of the vascular canals by lamellar bone (Fig.2E,F). A few specimens (e.g., Apatosaurus tibiaOMNH 1285; Phuwiangosaurus femur PW 5 A-1)have type C bone tissue with radial vasculari-zation (Fig. 2D), at least locally. No secondary

osteons or growth marks are developed in typeC bone tissue.

The transition from type C to type D bonetissue also is gradual and not abrupt. Thechange is again indicated by the increase in la-mellar bone in the primary osteons (Fig.2G–I), and the vascular canals in type D bonetissue have a thick lining of lamellar bone.However, vascularization is still high in typeD: the vascular canals are large, but smallerthan in type C bone tissue. Vascularization isprimarily laminar, but a few areas have ver-miform or more reticular organization. Intype D bone tissue, the formation of secondaryosteons starts. Thus, incompletely filled(‘‘young’’) large secondary osteons (Currey2002) develop mainly between larger erosioncavities of the remodeling zone and are closelyassociated with the medullary cavity (Fig.2O). Growth marks are rather rare in this tis-sue type. Type D bone tissue type is figuredin Figure 3B,C.

Type E bone tissue (Fig. 3C,D) represents astill relatively fast growing tissue. The transi-tion from type D to type E bone tissue againis gradual. Both types, in fact, differ only inorganization and degree of vascular density.The vascular spaces in type E bone tissue arestill present but smaller than in type D be-


FIGURE 3. A growth series of Apatosaurus specimens reaching from stage HOS 4 to HOS 13. A, Type B and type Cbone tissue in humerus OMNH 1278 (25.8 cm). B, Type C and type D bone tissues in humerus CM 21715 (79.2 cm).C, Type C, D, E bone tissues in humerus BYU 681-4749 (88.0 cm). Note absence of any growth marks. D, Type Ebone tissue in femur SMA 0014 (164.0 cm). Note the faint growth cycles and the appearance of secondary osteons.E, Type F bone tissue with LAGs, finally resulting in an EFS in femur BYU 601-17328 (158.0 cm). F, Completelyremodeled cortex in femur OMNH 01991 (180.0 cm).


cause the thickness of the layer of lamellarbone lining the vascular canals increases, re-sulting in very distinctive primary osteons(Fig. 2J–L). The secondary osteons betweenthe erosion cavities in the inner cortex aremore densely spaced (Fig. 3D) in type E bonetissue. Additionally, the spread of secondaryosteons into the primary cortex has started bynow, resulting in scattered secondary osteonsthat may extend in some specimens up to themiddle of the primary cortex. Growth marksmay occur, but they remain rare and are nottypical for type E bone tissue.

Type F bone tissue (Fig. 3E) is characterizedby a clear decrease in vascularization, finally re-sulting in a near complete infilling of the pri-mary vascular canals by lamellar bone. Lines ofarrested growth (LAGs) may occur in this bonetissue type. In some specimens a change fromfibrolamellar to lamellar-zonal bone and the de-position of an EFS (Fig. 2N) are initiated. TheEFS indicates that a growth plateau has beenreached. The vascular canals of primary osteonsare more or less completely filled by lamellarbone tissue (Fig. 2M). Remodeling by secondaryosteons has increased significantly, and densesecondary osteons are deposited at least up tothe middle to inner two-thirds of the primarycortex (Fig. 3E). In type F bone tissue, growthmarks are usually apparent, including the close-ly spaced LAGs of the EFS. However, somespecimens do not show growth marks or an EFSin type F bone tissue.

Type G bone tissue (Fig. 3F) is characterizedby an almost complete remodeling of the pri-mary cortex by secondary osteons.

Histologic Ontogenetic Stages in Long Bones.The thirteen HOS stages recognized in sauro-pod long bones are listed in Table 1. As notedabove, there are more HOS stages than onto-genetic bone tissue types (types A–G) becausein any one growth series there are specimensthat preserve more than one bone tissue type insequence. Of course, the preservation of succes-sive bone tissue types depends on remodelingand resorption activity because strong resorp-tion will result in a relatively thin cortex andstrong remodeling activity will obliterate theprimary growth record. This is why the HOSstages that are based on the transition from onebone tissue type to the next are defined such

that this transition occurs in the outer cortex,which is least affected by variability in resorp-tion and remodeling.

Ontogenetic Changes in the Histology of LimbGirdle Bones. Our large sample size (see Ap-pendix) also enabled us to study histologicalfeatures during ontogeny in limb girdle bones.However, most of these bones grew differentlythan long bones, mainly with much lower ap-position rates, owing to their smaller size orthinner cortex (sauropod scapulae are as largeas long bones just thinner). We will describethese histologic changes during ontogeny usingthe scapula as an example. Scapulae are flatbones, and their sampling location differs fromthat in long bones. The scapulae were sampledin the anteroposterior middle of the proximalmidshaft region (Klein and Sander 2007; seealso Curry 1999: Fig. 1). The scapula (Apatosau-rus; OMNH 1300, 35.5 cm) shows some kind oftype B and type C tissue in the lateral bone cor-tex (Fig. 2L). The inner cortex consists of type Bbone tissue with loosely arranged, very largeround and elliptical vascular canals reminiscentof the longitudinal to radial vascular organiza-tion of fibrolamellar bone. Toward the outer cor-tex, laminar organization increases. However,the whole tissue is very cavernous or primarilycancellous, consisting more of vascular canalsthan of primary bone. The medial side ofOMNH 1300 shows a completely different bonetissue. First, the remodeling zone reaches muchfarther into the primary cortex, and it has al-ready relatively dense secondary osteons. Theremaining primary bone tissue is only a fewmillimeters thick. The primary bone tissue typeis longitudinal fibrolamellar bone, but differentfrom that of long bones or from that of the op-posite side. Scapula SMA 0011 (diplodocid;138.0 cm) is almost completely remodeled onthe medial side of the bone, whereas the lateralside shows only a few scattered secondary os-teons and consists otherwise of longitudinallyvascularized primary bone. BYU 725-16614(Apatosaurus; 127.0 cm) has only scattered sec-ondary osteons in the outer cortex of the bone’slateral side. In the exterior cortex and betweenscattered secondary osteons primary bone isvisible, consisting of fibrolamellar bone with alarge amount of parallel-fibered bone. The bonetissue in the exterior cortex is lamellar-zonal.


TABLE 2. Histologic ontogenetic stages (HOS stage), the bone tissue types (A–G), bone dimensions, and body sizeand mass of Apatosaurus specimens. Percentage adult length of Apatosaurus was estimated from femur/body lengthratio (Seebacher 2001; Mazzetta et al. 2004), which is roughly 8%; the maximum femur length of Apatosaurus in oursample is 180.0 cm. The body mass was calculated after Seebacher (2001). The corresponding body mass for thehumeri was based on their corresponding femur length. Specimens were arranged by femur size. cfl � correspond-ing femur length. Btt � bone tissue type.

Specimen Bone length and cfl HOS Btt% Adultlength Body mass

Tibia OMNH 1285 20.05 cm (� 31.8 cm cfl) HOS-4 type B 17.6 147 kgFemur OMNH 01279 34.0 cm HOS-4 type B 19 179 kgHumerus OMNH 1278 25.8 cm (� 39.7 cm cfl) HOS-4 type C 22 291 kgFemur BYU 601-17103 83.0 cm HOS-7 type D 46 2599 kgFemur BYU 725-17014 97.0 cm HOS-8 type E 54 4168 kgHumerus CM 21715 79.2 cm (� 121.8 cm cfl) HOS-8 type E 68 8066 kgFemur BYU 681-11940 133.0 cm HOS-10 type F 74 10,755 kgHumerus BYU 6812-4749 88.0 cm (� 135.4 cm cfl) HOS-8 type E 75 10,979 kgFemur SMA M4/10-1 144.0 cm HOS-12 type F 80 13,668 kgHumerus CM 3378 98.0 cm (� 150.7 cm cfl) HOS-11 type F 84 15,461 kgFemur BYU 601-17328 158.0 cm HOS-12 type F 88 17,926 kgFemur SMA Jacques 164.0 cm HOS-10 type F 91 20,166 kgFemur OMNH 01991 176.0 cm HOS-12 to HOS-13 type G 98 24,872 kgFemur OMNH 4020 180.0 cm HOS-13 type G 100 26,664 kg

Four LAGs are deposited in this tissue. Only theincompletely drilled medial side of the bone is,as far as preserved, completely remodeled bysecondary osteons. Most of the other scapulae(Apatosaurus; BYU 681-16795, 116.0 cm; BYU681-16782, 125.0 cm; OMNH 01375, 145.5 cm)were almost completely remodeled by densesecondary osteons in both cortices.

Histologic Ontogenetic Stage andBody Size

As first conceptualized by Sander (2000) forthe Tendaguru sauropods, the HOS stages insauropod humeri and femora can be used toconstruct a growth curve by plotting HOSstages against femur length (as a proxy forbody size) (Fig. 4). Bones of very small ani-mals show type B and type C bone tissue(HOS-3 to HOS-6), which represent very fastgrowth, whereas medium-sized animals showtype D and type E bone tissue (HOS-7 to HOS-9), in which a successive slow-down in growthrate starts but still a substantial gain in bodysize was achieved. Large animals usually havetype E bone tissue (HOS-10 to HOS-13) wheregrowth significantly decreased or ceased en-tirely (HOS-10 to HOS-11) and finally pla-teaued (HOS-12). HOS-13 records the pro-longed life of the individual after growth hadceased. This stage corresponds to the ‘‘senile’’stage of Sander (2000) (Fig. 3, Table 1). At the

moment, no time scale for the different onto-genetic stages (HOS stages) is available.

An exemplary interpretation of the Apato-saurus sample is offered in Table 2. In this tax-on, HOS-3 occurs in individuals of up to 20%of adult body length as represented by femurlength. HOS-5 occurs in individuals of up to40% of adult body length. HOS-7 seems to oc-cur only between 40% and 50% of adult bodylength. HOS-9 occurs in animals of up to 75%of adult size. HOS-10 to HOS-12, representingtype F bone tissue, is seen in the size class be-tween 75% and 95% of adult body length, andHOS-13 only occurs in the largest individualsof Apatosaurus.

Plots of body size versus HOS stage indicatethat these stages generally correspond well tobody size in sauropods (Fig. 4). This is in con-trast to what was observed in the basal sau-ropodomorph Plateosaurus, which shows de-velopmental plasticity in its bone histology(Sander and Klein 2005; Klein and Sander2007) as indicated by the lack of correlation be-tween individual age and body size. In Figure4, HOS stage (� age) is plotted against the fe-mur length (� body size). For all bones otherthan the femora, the value of correspondingfemur length was used. Apatosaurus, Brachio-saurus, and Diplodocus (Fig. 4A,B,G) show arelatively good relation between HOS stageand femur length. The diplodocine samples


FIGURE 4. Histologic ontogenetic stage (HOS stage) versus femur length in various taxa. A, Apatosaurus. B, Diplod-ocus. C, Diplodocinae indet. from Howe Ranch locality. D, Diplodocinae indet. from the Tendaguru Beds. E, Ca-marasaurus. F, Europasaurus. G, Brachiosaurus. H, Phuwiangosaurus. Black rectangles represent femur samples, whitediamonds humerus samples, and triangles tibia samples. Humeri and tibiae were calculated to their correspondingfemur length. The corresponding femur length for the humeri is based on ratios found in the literature (Janensch1961; McIntosh 1990; Remes 2006). Although some taxa (Apatosaurus, Brachiosaurus, and Diplodocus) show a verygood agreement between body size and HOS stage, others appear more inhomogeneous. See text for discussion.


from Tendaguru and the Howe Ranch, as wellas the Camarasaurus, Europasaurus, and Phu-wiangosaurus samples (Fig. 4C–F,H) are lesshomogeneous for reasons that will be dis-cussed below.

In spite of the very large sample size, con-structing a diagram for Ampelosaurus is at themoment not possible because of the fracturedor flattened condition of the sampled bones,which makes calculation of the original bonesize difficult. From the different diagrams inFigure 4 we conclude that there are some dif-ferences among life-history strategies in thesauropods we studied. For example, in theApatosaurus sample (Fig. 4A) the curve fromjuveniles (small) to adults (large) is slightlyconcave, contrary to the Diplodocus and Bra-chiosaurus sample (Fig. 4B,G), which show amore convex curve. This difference couldmean that growth in Apatosaurus juveniles wasnot as fast as in Diplodocus or Brachiosaurus ju-veniles. On the other hand, growth in Apato-saurus adults was faster than in Diplodocusadults.

Discussion

Quality of the Growth Record in Humeri, Fem-ora and Other Bones. One important result ofthis study is that the long bones of the sauro-pods studied (specifically humeri and femora)show a clear correlation between bone size andhistologic ontogenetic stage (Table 2, Fig. 4).Nevertheless, these graphs (Fig. 4) also showthat there is a small difference between femoraand humeri because bone tissue appositionrates are higher in the femora than in the hu-meri. This was to be expected, at least for taxawith a size discrepancy between humeri andfemora such as diplodocid sauropods. For somesauropod taxa, there are no known humeri orfemora, raising the question whether otherbones might be useful for detecting the onto-genetic status of an individual. Our survey isadmittedly incomplete in this regard, havingonly covered some lower limb bones (ulna, tibia,fibula) and pubis and scapula (the scapula wasshown by Curry [1999] to reflect ontogeneticstatus). Although different ontogenetic stages inbone histology in other bones could be identi-fied, the bone tissue types and the bone appo-sition rates are, at least in the flat bones (pubis

and scapula), different from those observed inthe long bones. The comparison of lower limbbones (tibia, fibula, ulna) with humeri and fem-ora is still problematic because of the small sam-ple size for these bones.

Histological Ontogenetic Stages in the DifferentSauropod Taxa. Not all sauropod taxa exhibitall HOS stages in the same manner; i.e. thereare taxon-specific differences in bone histolo-gy and thus growth trajectory, as already not-ed by Sander (2000). For example, the youngof Ampelosaurus and Phuwiangosaurus alreadyshow very well-developed primary osteons intype B and type C bone tissue. These primaryosteons in Ampelosaurus and Phuwiangosaurusare much more mature than in other sauro-pods with similar bone tissue types. However,the very young status of these bone tissuetypes is indicated by other bone histologicalfeatures (e.g., size, number, and organizationof vascular canals) and by the number andform of osteocytes and amount of fibrous andparallel-fibered bone in the fibrolamellar tis-sue. Ampelosaurus and Phuwiangosaurus showstronger remodeling by dense secondary os-teons than do the diplodocids and more basaltitanosauromorphs at the same HOS stage.Currently, we cannot explain these differencesbut they are the focus of more detailed re-search on titanosaur bone histology (Kleinand Sander 2006).

Another difference is the occurrence ofgrowth marks such as lines of arrested growth(LAGs). In Diplodocidae, Brachiosaurus, andCamarasaurus, growth marks are more fre-quent than in Ampelosaurus and Phuwiangosau-rus. In Europasaurus growth marks occur reg-ularly (Sander et al. 2006). In the other taxa,sample size is too small to evaluate this fea-ture. Although in sauropods it is generally un-clear what events late in the life history of theanimal triggered the deposition of LAGs,within several taxa some presumable adultsshow growth marks but not others. Because ofthe large sample size and the size spectrum ofthe studied specimens (Appendix), a sam-pling artifact seems unlikely, and a biologicalcause (endogenous or exogenous factors) ishypothesized.

HOS-13 (complete remodeling by secondaryosteons) is seen only in the femora and humeri


FIGURE 5. HOS stages versus femur length in Camara-saurus are interpreted in this graph as reflecting twomorphotypes. Individual SMA 0002 (humerus 70.5 cm,femur 93.5 cm) thus would represent morphotype 1 andindividual CM 36664 (humerus 117.2 cm, femur 145.2cm), morphotype 2.

of Apatosaurus, Ampelosaurus, and some speci-mens of Camarasaurus (type 1). This may be be-cause the largest individuals of the other taxaare poorly represented in the fossil record.Thus, the recorded growth history as repre-sented by this histologic stage is simply not pre-served in the other taxa. There is no evidencethat the pattern of secondary osteon develop-ment is dependent on body mass because Bra-chiosaurus as the heaviest taxon sampled showsless remodeling than Apatosaurus or the medi-um-sized Ampelosaurus. As noted earlier, the is-sue of the controls on secondary remodeling isbeyond the scope of this paper and will not bediscussed further. Another difference betweenApatosaurus and some other taxa is that adultbone tissue type I is not as well developed in thistaxon, suggesting that HOS-9 in Apatosaurus,seen only in a few specimens, did not last verylong. All these differences in growth trajectoriesare of particular interest when viewed in anevolutionary context.

In addition to the variation seen among dif-ferent taxa, it is obvious from the graphs in Fig-ure 4 that there is also variation within taxa,more properly within the operational taxonomicunits (OTUs) used in this study. The HoweRanch Diplodocinae sample, the Camarasaurussample, and the Phuwiangosaurus sample did notshow a good correlation between femur lengthand HOS stage (Fig. 4D,E,H). There are severalexplanations for this variability that are not mu-tually exclusive: (1) Errors in histologic stage as-signment; (2) true individual variation as seenin any biological species; (3) sexual dimorphismin growth trajectories; and (4) cryptic biologicalspecies that differ in histology and growth tra-jectory but cannot be separated morphological-ly, at least in the known skeletal parts. Whereasexplanations 1 and 2 certainly play a role but aredifficult to assess, explanations 3 and 4 deservefurther comment.

The current study did not find support forexplanation 3, sexually differentiated growthtrajectories, in any of the OTUs. AlthoughSander (2000) interpreted the data for the Ten-daguru diplodocine under the premise thatthis material represents a single biologicalspecies, the work of Remes (2006) indicatesthat at least two diplodocine species are rep-resented by the material listed as Barosaurus

africanus by Janensch (1961). It now appearslikely that Sander’s histotypes A and B of‘‘Barosaurus africanus’’ represent two differentspecies: type A is Tornieria africana (Remes2006) and type B could be Australodocus, an-other diplodocine from the Tendaguru Beds(Remes 2007).

The material ascribed to the genus Camara-saurus may represent an example of explana-tion 4 as suggested by Figure 5. The bone his-tology shows clearly that some individuals inthis sample are older at a generally smallerbody size. Although explanation 3 might beworth considering as well, this would mean asize difference of nearly 50% between the sex-es. Phuwiangosaurus (Fig. 4H) and Ampelosau-rus (not shown) also exhibit an inhomoge-neous pattern of size (femur length) plottedversus HOS stage samples despite the largesample size for these OTUs. However, the ma-terial derives in both cases from several local-ities spread over a relatively extensive geo-graphic area. In addition, the calculation ofbone length is problematic for Ampelosaurusbecause some bones are badly compressed.

Comparison with Other Studies on HistologicOntogenetic Stages. Erickson and Tumanova(2000) erected a similar system for a sample ofthe ceratopsian Psittacosaurus monogoliensis, as-signing the number of LAGs (� estimated agein years) in a given femur size to a growth stagesystem, running from growth stage A to growth


stage G. However, although they did not discussbiological meaning of each of these stages, theydid find that these stages represented growthseries from juvenile through adult (Ericksonand Tumanova (2000).

The HOS stages described by Horner et al.(2000) for the hadrosaur Maiasaura are compa-rable with our observations in their general pat-tern. These authors describe rather fast growthin the young animals (nestling, juvenile, sub-adult). Growth slows down in adults, and theanimals finally reach a growth plateau. The dif-ferences between their study and ours in detailsof bone histology and in the interpretation andnomenclature of the described bone tissue types(A–G) are related to the different taxa (hadro-saur versus sauropod) and life histories (altricialyoung in Maiasaura versus presumably precocialyoung in sauropods [Carpenter 1999; Sander etal. in press]).

The study of an ontogenetic series of Apato-saurus bones by Curry (1999) is difficult tocompare with our HOS stages because Currydid not study the same long bones (humerusand femur). However, the bone tissue types inthe radii and ulnae that she studied (Curry1999: Fig. 2) are similar to our type C and typeD bone tissue in the humeri and femora of sau-ropods (HOS-5 to HOS-7), although the radiiand ulnae (Curry 1999: Fig. 2) already show anumber of scattered secondary osteons. Thusin comparison to our results, her age estimatesof ‘‘early juvenile’’ for these lower limb bonesof Apatosaurus are too young and, on the otherhand, her percentages of adult body sizes forApatosaurus early ontogenetic stages are toohigh; e.g., the early juveniles are 61% of adultsize (Curry 1999: Table 1). The scapula (117.4cm long) from the Cactus Park locality thatshe sampled is the same bone as our sampledspecimen BYU 681-16795 (116.0 cm). Curry’sevaluation of bone tissue type (late juvenile)(1999: Fig. 5A) is similar to our evaluation(type D bone tissue � subadult to type E bonetissue � adult I), keeping in mind that wesampled both the median and lateral sides ofthe bone, which results in some additional in-formation to Curry’s sample. However, we donot agree with her extrapolated adult bodysize of this bone (Curry 1999: Table 1; 56% ofadult size), because of a different calculation

method. According to our extrapolation meth-od this scapula has a corresponding size of�70% of maximal adult body size.

Sander (2000) and Sander and Tuckmantel(2003) studied the transition from type D bonetissue (subadult) to type E bone tissue (adultbone) in sauropods in some detail. Note, how-ever, that they did not distinguish a type D bonetissue but called both the current type C bonetissue and type D bone tissue ‘‘juvenile bone.’’Sander (2000) was able to quantify the drop invascularity from our type D bone tissue to typeE bone tissue, and Sander and Tuckmantel(2003) showed that the decrease in vasculariza-tion was not because of a decrease in initial vas-cular canal size but to an increase in infilling ofthe vascular canals, because lamina thicknesswas rather uniform in both type D and type Ebone tissue, and also uniform across differentspecimens and taxa of sauropods.

Biologic Ontogenetic Stages. Terms such asembryo, hatchling, juvenile, subadult, andadult, which originate from the study of life his-tory in recent amniotes, are also in common usefor extinct taxa. However, these terms are usu-ally used in a morphological and size-relatedcontext. Their definition and their ontogeneticmeaning for living animals vary from group togroup, mainly depending on the specific life his-tories. The use of these terms for extinct speciesis problematic and only hypothetical. However,the bone tissue types described here (types A–G) are hypothesized to correspond to biologicontogenetic stages: Bone tissue type A may cor-respond to the bone tissue of an embryo; type Bto that of a hatchling; type C to that of a juvenile;type D to that of a subadult; type E to that of ayoung, still growing adult; type F to that of anadult, where growth had stopped (developmentof an EFS); and finally type G, the completelyremodeled bone tissue of a senescent individ-ual.

Determination of Sexual Maturity from the On-togenetic Histologic Stages. Determining sex-ual maturity in extinct animals is rather con-troversial (Curry 1999; Sander 2000; Horner etal. 2000; Chinsamy-Turan 2005; Erickson2005; Erickson et al. 2007; Klein and Sander2007) and remains hypothetical until clear ev-idence, such as medullary bone, is discoveredfor a taxon. Medullary bone is known from re-


cent ovulating birds (Chinsamy-Turan 2005)and also from the theropod Tyrannosaurus(Schweitzer et al. 2005), in which it clearly in-dicates a sexually mature female. However, inour large sauropodomorph sample (Appen-dix; see also Klein and Sander 2007), no med-ullary bone is preserved.

Sander (2000) hypothesized that sexual ma-turity in sauropods is recorded in HOS stages,but this has not met with universal agreement(e.g., Erickson 2005; Chinsamy-Turan 2005).Although sexual maturity has not been iden-tified with certainty in the bone histology ofextinct vertebrates, one can argue the pro andcons for several possible moments in an indi-vidual’s histological growth record as indica-tors of the onset of sexual maturity. Sander’s(2000) reasoning was that the attainment ofsexual maturity resulted in a decrease ingrowth rate due to the shift in resources fromgrowth to reproduction as a sauropod becamesexually mature. This is the pattern seen invirtually all living tetrapods and presumablyapplied to sauropods as well. As the currentstudy shows, bone histology in sauropodsdocuments several decreases in linear growthrate (Fig. 3), and the question arises which ofthem might reflect sexual maturity.

One possibility, favored by Curry (1999), isthat sexual maturity coincided with attain-ment of final size, as reflected by the deposi-tion of an EFS or a similar bone tissue type(HOS-12 in the current study). This possibilitycan be evaluated by using the limited quanti-tative life-history data available for sauropods.Using growth cycles, Sander (2000) estimatedthat an individual of the basal titanosaur Ja-nenschia required at least 26 years to reach fi-nal size. However, an individual of Apatosau-rus of 91% maximum size (SMA 0014; Table 2)shows 25 growth cycles in the outer cortex(Sander and Tuckmantel 2003), suggestingthat it was at least 30 years old. In this casesexual maturity would not be attained untilthe end of the third decade of life, which wasconsidered highly unlikely by Dunham et al.(1989) for any dinosaur. HOS-12 thus canprobably be excluded from consideration forrecording sexual maturity.

Earlier possible stages are HOS-10 and HOS-8. Arguing against HOS-10 is again the rather

large size and thus old age of the animals in-volved. This is why we favor HOS-8 to HOS-9.As Erickson et al. (2007) and Lee and Werning(2008) recently showed, in non-avian theropodand ornithischian dinosaurs sexual maturityalso occurred well before full adult size wasreached. Our Apatosaurus sample (Table 2) sug-gests a body size of �50% maximum adult sizeat HOS-8. This would correspond to a femurlength of 90.0 cm or 11.25 m body length. At-tainment of sexual maturity at about one-halfmaximum body size is consistent with the lifehistories of modern large reptiles such as croc-odiles and turtles but not with those of largeherbivorous mammals.

Conclusions

This study shows that well-constrained coresamples of the large long bones of sauropod di-nosaurs can be used to identify seven definedbone tissue types (A–G) that correspond to theontogenetic status of a given sauropod individ-ual with great accuracy, assigning it to one ofthe 13 HOS stages designated herein. We wereable to designate so many HOS stages becauseof our large sample size and the uniformgrowth of sauropod long bones. Although var-iations of HOS stages do occur among differentneosauropod taxa, these are mainly related tothe relative importance of the stages in their on-togeny. We thus feel that these stages can be ap-plied to other neosauropods, and their utilitycan be tested by other workers. In fact, we hopethat it will become accepted practice in thestudy of sauropods and other dinosaurs thatany description of a new taxon will be accom-panied by determination of its ontogenetic sta-tus using bone histology. This would help avoidmuch controversy surrounding specimens thatare rather similar morphologically but differgreatly in body size.

In all sauropod taxa studied, there is a closecorrelation between body size and ontogenet-ic stage, suggesting that sauropods had lostdevelopmental plasticity and grew along a ge-netically predetermined growth trajectory.However, although sauropods appear to showa more reptilian growth curve in probablyhaving had an early onset of sexual maturity,as was also shown recently for some other di-nosaurs (Erickson et al. 2007; Lee and Werning


2008), their growth rates must have been ashigh as in birds or mammals.

In the virtual absence of cyclical growthmarks in sauropod long bones, we believeHOS stages are a considerable improvementover a purely qualitative analysis of thegrowth record. Nevertheless, we would like toobtain fully quantitative sauropod growthcurves in which size and body mass gain areplotted as a function of time. This could beachieved if specific growth rates could be de-rived for the different bone tissue types (i.e.,juvenile, subadult, and adult bone) by mea-suring them in recent equivalents of these tis-sues or deriving them from other fossil spec-imens that show these tissues to be interrupt-ed by growth marks such as bones of the sau-ropod skeleton growing with negativeallometry.

Comparison of growth trajectories of sistergroups reveals evolutionary change by heter-ochrony. Because our work produces growthtrajectories of a wide variety of sauropod taxa,it provides the foundation for a much refinedunderstanding of body-size evolution in sau-ropods. We have shown this already for theevolution of dwarfing in the basal macronar-ian Europasaurus (Sander et al. 2006) and theevolution of the first giant sauropods frommuch smaller prosauropod ancestors (Sanderet al. 2004).

Acknowledgments

In a study based on destructive methods,the open minds and help of curators are of ut-most importance. It therefore is our particularpleasure to acknowledge the following indi-viduals for giving permission for and aid insampling specimens in their care: D. Berman(CM), E. Buffetaut (LMC), R. Cifelli and K. Da-vies (OMNH), W.-D. Heinrich (MfN), N. Knot-schke (DFMMh/FV), J. Le Loeuff (MDE), H.-P. Schultze (MfN), K. Stadtman (BYU), H. H.J. Siber (SMA), and V. Suteethorn (PC.DMR).R. Wilhite (Louisiana State University, BatonRouge) is acknowledged for providing iden-tifications and measurements of sauropodlong bones in the BYU collections. We alsowould like to thank O. Dulfer and G. Oles-chinski (both Institute of Paleontology, Uni-versity of Bonn), who did a great job in cutting

the thin sections and photographing them forstudy. The manuscript has greatly benefitedfrom reviews by G. Erickson and an anony-mous reviewer. Our research was funded bythe Deutsche Forschungsgemeinschaft (DFG)through grants SA 469/7 and SA 469/16. Thisis contribution number 40 of the DFG Re-search Unit 533, ‘‘Biology of the Sauropod Di-nosaurs.’’

Literature CitedAmprino, R. 1947. La structure du tissue osseux envisagee com-

me expression de differences dans la vitesse del’accroissement. Archives de Biologie 58:315–330.

Ayer, J. 2000. The Howe Ranch dinosaurs. Sauriermuseum Aath-al, Zurich.

Bonnan, M. F. 2004. Morphometric analysis of humerus and fe-mur shape in Morrison sauropods: implications for functionalmorphology and paleobiology. Paleobiology 30:444–470.

Carpenter, K. 1999. Eggs, nests, and baby dinosaurs: a look at di-nosaur reproduction. Indiana University Press, Bloomington.

Carpenter, K., and J. McIntosh. 1994. Upper Jurassic sauropodbabies from the Morrison Formation. Pp. 372 in K. Carpenter,K. Hirsch, and J. Horner, eds. Dinosaur eggs and babies. Cam-bridge University Press, Cambridge.

Chiappe, L. M., and S. L. Coria. 2001. Embryonic skulls of ti-tanosaur sauropod dinosaurs. Science 293:2444–2446.

Chiappe, L. M., F. Jackson., R. A. Coria, and L. Dingus. 2005.Nesting titanosaurs from Auca Mahuevo and adjacent sites:understanding sauropod reproductive behaviour and embry-onic development. Pp. 285–302 in K. Curry Rogers and J. A.Wilson, eds. The sauropods: evolution and paleobiology. Uni-versity of California Press, Berkeley.

Chinsamy, A. 1993. Bone histology and growth trajectory of theprosauropod dinosaur Massospondylus carinatus (Owen). Mod-ern Geology 18:319–329.

———. 1994. Dinosaur bone histology: implications and inferences.Pp. 213–227 in G. D. Rosenberg and D. L. Wolberg, eds. DinoFest.Paleontological Society Special Publication 7:213–227.

Chinsamy-Turan, A. 2005. The microstructure of dinosaur bone.Johns Hopkins University Press, Baltimore.

Cormack, D. 1987. Ham’s histology. Lippincott, New York.Currey, J. D. 2002. Bones: structure and mechanics. Princeton

University Press, Princeton, N.J.Curry, K. 1999. Ontogenetic histology of Apatosaurus (Dinosau-

ria: Sauropoda): new insights on growth rates and longevity.Journal of Vertebrate Paleontology 19:654–665.

Dunham, A. E., K. L. Overall, W. P. Porter, and C. A. Forster.1989. Implications of ecological energetics and biophysicaland developmental constraints for life-history variation in di-nosaurs. In J. O. Farlow, ed. Paleobiology of the dinosaurs.Geological Society of America Special Paper 238:1–21.

Erickson, G. M. 2005. Assessing dinosaur growth patterns: a mi-croscopic revolution. Trends in Ecology and Evolution 20:677–684.

Erickson, G. M., and T. A. Tumanova. 2000. Growth curve ofPsittacosaurus mongoliensis Osborn (Ceratopsia: Psittacosaur-idae) inferred from long bone histology. Zoological Journal ofthe Linnean Society 130:551–566.

Erickson, G. M., P. J. Currie, B. D. Inouye, and A. A. Winn. 2006.Tyrannosaur life tables: an example of nonavian dinosaurpopulation biology. Science 313:213–217.

Erickson, G. M., K. Curry-Rogers, D. J. Varricchio, M. A. Norell,and X. Xu. 2007. Growth pattern in brooding dinosaurs re-


veals the timing of sexual maturity in non-avian dinosaursand genesis of the avian condition. Biology Letters 3:558–561.

Foster, J. R. 2005. New juvenile sauropod material from WesternColorado, and the record of juvenile sauropods from the Up-per Jurassic Morrison Formation. Pp. 141–153 in Tidwell andCarpenter 2005.

Francillon-Vieillot, H., V. de Buffrenil, J. Castanet, J. Geraudie,F. J. Meunier, J. Y. Sire, L. Zylenberberg, and A. de Ricqles.1990. Microstructure and mineralization of vertebrate skeletaltissues. Pp. 471–530 in J. E. Carter, ed. Skeletal biominerali-zation: patterns, processes and evolutionary trends. Van Nos-trand Reinhold, New York.

Gilmore, C. W. 1925. A nearly complete articulated skeleton ofCamarasaurus, a saurischian from the Dinosaur National Mon-ument, Utah. Memoirs of the Carnegie Museum 10:347–384.

Horner, J. R., and K. Padian. 2004. Age and growth dynamics ofTyrannosaurus rex . Proceedings of the Royal Society of LondonB 27:1875–1880.

Horner, J. R., A. de Ricqles, and K. Padian. 2000. Long bone his-tology of the hadrosaurid dinosaur Maiasaura peeblesorum:growth dynamics and physiology based on an ontogenetic se-ries of skeletal elements. Journal of Vertebrate Paleontology20:115–129.

Horner, J. R., K. Padian, and A. de Ricqles. 2001. Comparativeosteohistology of some embryonic and perinatal archosaurs:developmental and behavioral implications for dinosaurs. Pa-leobiology 27:39–58.

Ikejiri, T., V. Tidwell, and D. Trexler. 2005. New adult specimensof Camarasaurus lentus highlight ontogenetic variation withinthe species. Pp. 154–179 in Tidwell and Carpenter 2005.

Janensch, W. 1961. Die Gliedmaßen und Gliedmaßengurtel derSauropoden der Tendaguru-Schichten. Palaeontographica3(Suppl. 7):177–235.

Jianu, C. M., and D. B. Weishampel. 1999. The smallest of thelargest: a new look at possible dwarfing in sauropod dino-saurs. Geologie en Mijnbouw 78:335–343.

Klein, N., and P. M. Sander. 2006. An unusual bone histologyand growth pattern in Ampelosaurus atacis, a titanosaurid sau-ropod from South France. Journal of Vertebrate Paleontology26(Suppl. to No. 3):85A.

———. 2007. Bone histology and growth of the prosauropod di-nosaur Plateosaurus engelhardti (von Meyer 1837) from the No-rian bonebeds of Trossingen (Germany) and Frick (Switzer-land). Special Papers in Palaeontology 77:169–206.

Lee, A., and S. Werning. 2008. Sexual maturity in growing di-nodaurs does not fit reptilian growth models. Proceedings ofthe National Academy of Sciences USA 105:582–587.

Lehmann, T. M., and A. B. Coulson. 2002. A juvenile specimenof the sauropod dinosaur Alamosaurus sanjuanensis from theUpper Cretaceous of Big Bend National Park, Texas. Journalof Paleontology 76:156–172.

Le Loeuff, J. 2005a. Osteology of Ampelosaurus atacis (Titanosau-ria) from Southern France. Pp.115–137 in Tidwell and Car-penter 2005.

———. 2005b. Romanian Late Cretaceous dinosaurs: big dwarfsor small giants? Historical Biology 17:15–17.

Martin, V. 1994. Baby sauropods from the Sao Khua Formation(Lower Cretaceous) in Northeastern Thailand. Gaia 10:147–153.

Mazzetta, G. V., P. Christiansen, and R. A. Farin. 2004. Giantsand bizarres: body size of some southern South AmericanCretaceous dinosaurs. Historical Biology 2004:1–13.

McIntosh, J. 1990. Species determination in sauropod dinosaurswith tentative suggestions for their classification. Pp. 53–69 inK. Carpenter and P. Currie, eds. Dinosaur systematics: per-

spectives and approaches. Cambridge University Press, Cam-bridge.

Reisz, R. R., D. Scott, H.-D. Sues, D. C. Evans, and M. A. Raath.2005. Embryos of an Early Jurassic prosauropod dinosaur andtheir evolutionary significance. Science 309:761–764.

Remes, K. 2006. Revision of the Tendaguru sauropod dinosaurTornieria africana Fraas and its relevance for sauropod paleo-biogeography. Journal of Vertebrate Paleontology 26:651–669.

———. 2007. A second Gondwana diplodocid dinosaur from theUpper Jurassic Tendaguru Beds of Tanzania, East Africa. Pa-laeontology 50:653–667.

Sander, P. M. 1999. Life history of Tendaguru sauropods as in-ferred from long bone histology. Mitteilungen Museum furNaturkunde Berlin, Geowissenschaftliche Reihe 2:103–112.

———. 2000. Longbone histology of the Tendaguru sauropods: im-plications for growth and biology. Paleobiology 26:466–488.

Sander, P. M., and N. Klein. 2005. Unexpected developmentalplasticity in the life history of an early dinosaur. Science 310:1800–1802.

Sander, P. M., and C. Tuckmantel. 2003. Bone lamina thickness,bone apposition rates, and age estimates in sauropod humeriand femora. Palaontologische Zeitschrift 77:161–172.

Sander, P. M., N. Klein, E. Buffetaut, G. Cuny, V. Suteethorn, J.Le Loeuff. 2004. Adaptive radiation in sauropod dinosaurs:bone histology indicates rapid evolution of giant body sizethrough acceleration. Organisms, Diversity, and Evolution 4:165–173.

Sander, P. M., O. Mateus, T. Laven, and N. Knotschke. 2006. Bonehistology indicates insular dwarfism in a new Late Jurassicsauropod dinosaur. Nature 441:739–741.

Sander, P. M., F. Jackson, and L. M. Chiappe. In press. UpperCretaceous titanosaur nesting sites and their implications forsauropod dinosaur reproductive biology. Palaeontographica,Abteilung A.

Schwarz, D., T. Ikejiri, B. H. Breithaupt, P. M. Sander, and N.Klein. 2007. A nearly complete skeleton of an early juvenilediplodocid (Dinosauria: Sauropoda) from the Lower Morri-son Formation (Late Jurassic) of North Central Wyoming andits implications for early ontogeny and pneumaticity in sau-ropods. Historical Biology 19:225–253.

Schweitzer, M. H., J. L. Wittmeyer, and J. R. Horner. 2005. Gen-der-specific reproductive tissue in ratites and Tyrannosaurusrex. Nature 308:1456–1460.

Seebacher, F. 2001. A new method to calculate allometric length-mass relationships of dinosaurs. Journal of Vertebrate Pale-ontology 21:51–60.

Tidwell, V., and K. Carpenter, eds. 2005. Thunder-lizards. In-diana University Press, Bloomington.

Tidwell, V., and R. Wilhite. 2005. Ontogenetic variation and iso-metric growth in the forelimb of the Early Cretaceous sau-ropod Venenosaurus. Pp. 187–196 in Tidwell and Carpenter2005.

Varricchio, D. J. 1993. Bone microstructure of the Upper Creta-ceous theropod dinosaur Troodon formosus . Journal of Verte-brate Paleontology 13:99–104.

Wilhite, R. 2003. Biomechanical reconstruction of the appendic-ular skeleton in three North American Jurassic sauropods.Ph.D. dissertation. Louisiana State University, Baton Rouge.

Wilson, J. A. 2002. Sauropod dinosaur phylogeny: critique andcladistic analysis. Zoological Journal of the Linnean Societyof London 136:217–276.

Xu, X., Q. Tan, J. Wang, X. Zhao and L. Tan. 2007. A giganticbird-like dinosaur from the Late Cretaceous of China. Nature447:884–847.

Documents

Ontogenetic stages in the longbone histology of sauropod ... · Ontogenetic stages in the longbone histology of sauropod dinosaurs Nicole Klein and Martin Sander Abstract.—Long