This article was downloaded by: [Jennifer Botha-Brink]On: 09 November 2011, At: 01:06Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Vertebrate PaleontologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/ujvp20

Osteohistology of the Triassic archosauromorphsProlacerta, Proterosuchus, Euparkeria, andErythrosuchus from the Karoo Basin of South AfricaJennifer Botha-Brink a b & Roger M. H. Smith c da Karoo Palaeontology, National Museum, P. O. Box 266, Bloemfontein, 9300, South Africab Department of Zoology and Entomology, University of the Free State, Bloemfontein, 9300,South Africac Karoo Palaeontology, Iziko South African Museum, P. O. Box 61, Cape Town, 8000, SouthAfricad Department of Geology, University of Cape Town, South Africa

Available online: 08 Nov 2011

To cite this article: Jennifer Botha-Brink & Roger M. H. Smith (2011): Osteohistology of the Triassic archosauromorphsProlacerta, Proterosuchus, Euparkeria, and Erythrosuchus from the Karoo Basin of South Africa, Journal of VertebratePaleontology, 31:6, 1238-1254

To link to this article: http://dx.doi.org/10.1080/02724634.2011.621797

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

Journal of Vertebrate Paleontology 31(6):1238–1254, November 2011© 2011 by the Society of Vertebrate Paleontology

ARTICLE

OSTEOHISTOLOGY OF THE TRIASSIC ARCHOSAUROMORPHS PROLACERTA,PROTEROSUCHUS, EUPARKERIA, AND ERYTHROSUCHUS FROM THE KAROO BASIN

OF SOUTH AFRICA

JENNIFER BOTHA-BRINK*,1 and ROGER M. H. SMITH2

1Karoo Palaeontology, National Museum, P. O. Box 266, Bloemfontein 9300, South Africa, and Department of Zoology andEntomology, University of the Free State, Bloemfontein 9300, South Africa, jbotha@nasmus.co.za;

2Karoo Palaeontology, Iziko South African Museum, P. O. Box 61, Cape Town, 8000, South Africa, and Department of Geology,University of Cape Town, South Africa, rsmith@iziko.org.za

ABSTRACT—The South African non-archosauriform archosauromorph Prolacerta and the archosauriforms Proterosuchus,Erythrosuchus, and Euparkeria were important constituents of the Early to early Middle Triassic Karoo ecosystem followingthe end-Permian mass extinction. We present new data on the osteohistology of these stem archosaurs and provide insight intotheir paleobiology. Bone tissues of the Early Triassic Prolacerta contain a poorly defined fibro-lamellar complex, with parallel-fibered bone in some regions, whereas the contemporaneous Proterosuchus exhibits rapidly forming uninterrupted fibro-lamellar bone early in its ontogeny, which becomes slow forming lamellar-zonal bone with increasing age. The early MiddleTriassic Erythrosuchus deposited highly vascularized, uninterrupted fibro-lamellar bone throughout ontogeny, whereas thegrowth of the contemporaneous Euparkeria was relatively slow and cyclical. When our data are combined with those ofprevious studies, preliminary results reveal that Early and Middle Triassic non-crown group archosauromorphs generallyexhibit faster growth rates than many of those of the Late Triassic. Early rapid growth and rapid attainment of sexual maturityare consistent with life history expectations for taxa living in the unpredictable conditions following the end-Permian massextinction. Further research with larger sample sizes will be required to determine the nature of the environmental pressureson these basal archosaurs.

INTRODUCTION

Archosauria, which includes crocodilians, birds, and their ex-tinct relatives diversified during the Middle Triassic to becomethe dominant clade of terrestrial vertebrates for the remainderof the Mesozoic Era. The stem archosaurs, i.e., non-archosaurianArchosauromorpha (sensu Gauthier, 1986), first appeared duringthe Late Permian, but only became abundant during the Triassic(Gower and Sennikov, 2000). They include well-known taxa suchas Rhynchosauria, Prolacerta, Protorosaurus, Trilophosaurus,and non-archosaurian Archosauriformes (sensu Gauthier, 1986).The last group comprises proterosuchids, erythrosuchids, prote-rochampsids, euparkeriids, and several isolated genera.

There has been much debate regarding the phylogenetic rela-tionships of the basal Archosauromorpha (e.g., Gauthier, 1986;Benton and Clark, 1988; Sereno and Arcucci, 1990; Sereno, 1991;Parrish, 1992, 1993; Juul, 1994; Gower, 2003; Gower and Sen-nikov, 1996, 1997, 2000). Thus, research on this group has focusedon re-examining basal archosaur morphology in order to betterassess archosauromorph relationships (Gower, 2003). Relativelyfewer studies have directed their efforts towards understandingthe growth of these animals. As most basal archosauromorphtaxa lived during the Early and Middle Triassic, they formed animportant part of the recovery ecosystem immediately followingthe end-Permian mass extinction (Botha and Smith, 2006; Sahneyand Benton, 2007). The Early Triassic was a time of severe eco-logical crisis and instability in both the terrestrial (Roopnarineet al., 2007) and marine realms (Payne et al., 2004; Bottjer et al.,2008). The global climate during the Triassic is characterized asbeing particularly harsh and arid with highly seasonal, unreli-

*Corresponding author.

able quantities of rainfall (Smith, 1995; Smith and Ward, 2001).Animals living in unpredictable environments often respond bygrowing rapidly in order to reach sexual maturity as quickly aspossible, a strategy that has been observed in living taxa (e.g.,Cardinale and Modin, 1999; Curtin et al., 2009). Thus, exploringthe biology and ecology of basal archosauromorphs can provideinsight into the life histories of extinct terrestrial vertebrates liv-ing in an unpredictable, drought-prone environment. In this way,this study is a test of the application of life-history theory to ex-tinct communities.

Here we present new data on the osteohistology of the non-archosauriform archosauromorph Prolacerta broomi Parrington,1935, and the archosauriforms Proterosuchus fergusi Broom,1903, Erythrosuchus africanus Broom, 1905, and Euparkeriacapensis Broom, 1913, with some interpretation of the paleobi-ology of the basal archosaurs.

Taxa Examined in this Study

The Early Triassic non-archosauriform archosauromorph Pro-lacerta broomi (Fig. 1) was included in this study as an archosauri-form outgroup. It is a small (80 mm skull length) archosauro-morph, known from South Africa and Antarctica, that wasinitially thought to be the precursor of squamates (Robinson,1967, 1973). However, more complete descriptions of the animal(Gow, 1975; Modesto and Sues, 2004) have revealed it to be thesister taxon of archosauriforms (Dilkes, 1998; Modesto and Sues,2004).

The earliest archosauriform taxon known from South Africais the medium-sized (approximately 2 m in body length) Pro-terosuchus fergusi (Reig, 1970; Welman, 1998). The lifestyle ofthis animal has been speculated upon by several authors. Broom(1932), Broili and Schroder (1934), Reig (1970), and Cruickshank

1238

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1239

FIGURE 1. Phylogenetic relationships of the archosauromorph taxa ex-amined in this study (in bold). Phylogeny taken from Sues (2003), Dilkesand Sues (2009), and Nesbitt et al. (2009).

(1972) proposed a semi-aquatic lifestyle for the animal, basedon its body proportions and horizontal zygapophyses (which aresimilar to those of extant crocodilians), the supposedly wet cli-mate in which it lived, and its association with Lystrosaurus, adicynodont therapsid that was then presumed to be amphibi-ous. However, more recent sedimentological analyses interpretthe climate of this period as relatively arid, with highly sea-sonal unpredictable rainfall (Smith, 1995; Smith and Ward, 2001).The lifestyle of Lystrosaurus is also a contentious issue. Broom(1902, 1903) and Watson (1912, 1913) first proposed an aquaticlifestyle, which was recently supported by Ray et al. (2005). Incontrast, King (1991), King and Cluver (1991), and Botha andSmith (2006) proposed a terrestrial lifestyle for the animal. Fur-thermore, Cruickshank (1972) also noted that the position of theexternal nares in Proterosuchus, the vertically oriented occipi-tal region (which differs from typical aquatic crocodilians), andthe well-ossified limbs, carpus, and tarsus indicated a terrestriallifestyle for this taxon (Cruickshank, 1972). Similarly, Welman(1998) also observed the absence of dorsally oriented nares, afeature that Nesbitt et al. (2009) noted is typical of aquatic rep-tiles such as crocodilians, phytosaurs, mesosaurs, and plesiosaurs(Camp, 1930; Mazin, 2001; Modesto, 2006). Thus, the lifestyle ofProterosuchus remains unresolved, but the weight of evidence isfor a terrestrial rather than aquatic lifestyle.

Proterosuchus (Fig. 1) is important in Pangaean-scale recon-structions of Early Triassic paleobiogeography in that it is the firstnew terrestrial taxon to appear above the Permo-Triassic bound-ary in the Karoo-aged basins of South Africa and Russia (Bentonet al., 2004; Smith and Botha, 2005). It was the most prominentpredator in the main Karoo Basin during the earliest Triassic, butwas supplanted in the Middle Triassic by the more derived ar-chosauriform Erythrosuchus africanus.

Erythrosuchus (Fig. 1) represents the first of many radiationsof large carnivores during the Middle Triassic (Parrish, 1992). Itwas larger than Proterosuchus (up to 5 m in length) with a dis-proportionately large head (Hughes, 1963; Gower, 2003). Earlierresearchers proposed an aquatic mode of life to support the largehead (von Huene, 1911); however, later researchers consider it tobe completely terrestrial (e.g., Tatarinov, 1961; Parrish, 1992).

Another well-known archosauriform from the Middle Triassicof South Africa is the smaller (<1 m in length), but more derivedEuparkeria capensis (Fig. 1) (Sereno, 1991). Once includedwithin the Archosauria as a basal ‘ornithosuchian’ (Gauthier,1984), it is now considered to be an archosauriform (Parrish,1992; Juul, 1994; Gower and Wilkinson, 1996) and is widelyaccepted as being closely related to the common ancestor ofdinosaurs and crocodilians (Gower and Wilkinson, 1996).

Previous Studies

The morphology of the skulls and postcrania of Prolacerta,Proterosuchus, and Euparkeria have been described in detail(e.g., Broom, 1903, 1913; Ewer, 1965; Gow, 1975; Welman,1998; Gower, 2003; Modesto and Sues, 2004) and although thepostcranial skeleton of Erythrosuchus has yet to be adequatelydescribed (due to fragmentary material), the skull, ankle, and iso-lated postcranial elements have been described (e.g., von Huene,1911; Brink, 1955; Cruikshank, 1978; Parrish, 1992; Gower, 1996,2003). To date, very little is known about the growth patterns ofthese animals because, due to their rarity, complete ontogeneticseries are not available for any of these taxa.

The study of bone histology (microstructure) is a well-established technique for analyzing the growth of extinctanimals. By comparing the bone tissues of extinct animals withthose of living taxa, the growth patterns, ontogenetic status,and, indirectly, aspects of the physiology of fossil taxa can beinferred (Amprino, 1947; Enlow and Brown, 1956, 1957; deRicqles, 1976). Gross (1934) and later de Ricqles (1976) brieflydescribed the bone microstructure of Erythrosuchus as rapidlyforming azonal fibro-lamellar bone, and their results wererecently confirmed by de Ricqles et al. (2008). These authorsassessed the bone histology of 12 Triassic archosaurs, includinga range of non-dinosaurian archosauriforms and dinosaurs, ina phylogenetic context. Euparkeria was also briefly mentionedin this study, but the description was based only on a possiblehumerus, osteoderms, and rib fragments (Ricqles et al., 2008).This sizeable contribution provided an important foundationfrom which further research could develop.

MATERIALS AND METHODS

The study material consists of 20 limb bones and ribs fromfour Early to Middle Triassic archosauromorph taxa from theKaroo Basin of South Africa (Table 1). These include one indi-vidual of Prolacerta broomi (NMQR 3763), three individuals ofProterosuchus fergusi (SAM-PK-K140, SAM-PK-11208, NMQR880), three individuals of Erythrosuchus africanus (SAM-PK-K10025, SAM-PK-1118, NMQR 3675), and three individuals ofEuparkeria capensis (SAM-PK-K10010, SAM-PK-13666, SAM-PK-K10548). Limb bones were preferentially selected becausethey contain the least secondary remodeling in the midshaft re-gions and hence exhibit the most complete growth record of theanimal (Chinsamy, 1990, 1991, 1995; Francillon-Vieillot et al.,1990; Horner et al., 1999). However, ribs and other fragmentswere also examined to provide further data. Histological analysisis a destructive sampling technique and limb bones of these taxaare relatively rare, which poses difficulties in obtaining enoughmaterial for this type of study. Despite this, we were fortunate inobtaining various associated limb bones from single individualsas well as homologous elements from multiple specimens, allow-ing us to consider inter-elemental and inter-individual variationwhen documenting the growth patterns of these taxa.

In South Africa, Prolacerta and Proterosuchus are known fromthe Lower Triassic Lystrosaurus Assemblage Zone and Erythro-suchus is restricted to the overlying Middle Triassic CynognathusAssemblage Zone (Cynognathus zone B; Hancox et al., 1995)(Table 1). The study material of all these taxa was positively iden-tified through the association of the elements with skull material.Euparkeria is known from a single locality on the outskirts of thetown Aliwal North from the Middle Triassic Cynognathus As-semblage Zone (also Cynognathus zone B). The material includesseveral similar-sized adult individuals (Ewer, 1965) preserved ona number of sandstone blocks intermingled with three specimensof the rhynchosaur Mesosuchus browni (Dilkes, 1998). The study

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1240 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

TABLE 1. Specimens thin-sectioned in this study.

Genus, Element Accession number LocalityAssemblage

zoneSkulllength

Totallength

Midshaftwidth

Proximalwidth

Distalwidth

%adult

Prolacerta broomiTibia NMQR 3763 Fairydale, Bethulie Lystrosaurus 76 4.8 12.2 95

Proterosuchus fergusiTibia SAM-PK-K140b Visgat, Conway Lystrosaurus 266 139.2 14.8 48.1 25.4 67Femur SAM-PK-K140c Visgat, Conway Lystrosaurus 266 167 19.7 55.2 53.8 67Fibula NMQR 880 Kruisvlei, Winburg Lystrosaurus 331.5 13.9 26.2 82Femur SAM-PK-11208a Barendskraal,

MiddelburgLystrosaurus 59.3 100

Fibula SAM-PK-11208b Barendskraal,Middelburg

Lystrosaurus 17.2 100

Erythrosuchus africanusRib SAM-PK-K1118a Lemoenfontein,

Aliwal NorthCynognathus 28.2

Fragment SAM-PK-K1118b Lemoenfontein,Aliwal North

Cynognathus 34.6

Tibia SAM-PK-K10025a Luiper Kop, Albert Cynognathus 34.7 68.5 56Rib SAM-PK-K10025b Luiper Kop, Albert Cynognathus 19.2 56Radius NMQR 3675a Slootkraal, Aliwal

NorthCynognathus 171.7 27.5 55.6 63

Rib NMQR 3675b Slootkraal, AliwalNorth

Cynognathus 15 63

Rib NMQR 3675c Slootkraal, AliwalNorth

Cynognathus 15.5 63

Rib NMQR 3675d Slootkraal, AliwalNorth

Cynognathus 13.1 63

Rib NMQR 3675e Slootkraal, AliwalNorth

Cynognathus 13.5 63

Euparkeria capensisHumerus SAM-PK-13666 Aliwal North Comm.,

Aliwal NorthCynognathus 42.4 4.4 17.7 11.4 88

Femur SAM-PK-K10010a Aliwal North Comm.,Aliwal North

Cynognathus 6.1 11.4 89

Tibia SAM-PK-K10010b Aliwal North Comm.,Aliwal North

Cynognathus 53.6 6.7 12 9.6 89

Fibula SAM-PK-K10010c Aliwal North Comm.,Aliwal North

Cynognathus 4.9 12.7 89

Femur SAM-PK-K10548 Aliwal North Comm.,Aliwal North

Cynognathus 68 7.2 16.5 17.1 96

The% adult size was estimated using the largest known specimens for each genus.

material taken from these blocks was positively identified as Eu-parkeria capensis by comparing the limb bone morphology to theassociated articulated skeletons with skulls. This allowed for con-clusive species identification, something that has not been donepreviously with Euparkeria capensis.

All elements were measured and photographed prior to thinsectioning. An estimate of ontogenetic age was also calculatedfor each element using skull lengths and associated limb bonemeasurements from the largest known specimens of each genus.As the mid-diaphysis experiences the least secondary remodelingand thus provides the most complete information about the lifehistory of the animal, elements were thin-sectioned in this regionwherever possible. The thin-sectioning process was conducted atthe National Museum, Bloemfontein, following procedures out-lined by Chinsamy and Raath (1992). The bone microstructurewas viewed and photographed using a Nikon Eclipse 50i Polariz-ing microscope and DS-Fi1 digital camera. Bone histology termi-nology follows that of Francillon-Vieillot et al. (1990) and Reid(1996).

The organization and density of the vascular canals in corticalbone is closely related to bone apposition rate (e.g., Amprino,1947; Margerie et al., 2002). Quantifying the relative area occu-pied by the vascular canals in a given section of bone provides aquantitative method for comparing relative vascular density be-tween different taxa. This process reflects the maximum possi-ble vascularization for each element, as the channels would haveincluded lymph and nerves as well as blood vessels (Starck and

Chinsamy, 2002). The relative vascular area or cortical porositywas quantified using the image analysis program NIH ImageJ. Itwas calculated by taking the total canal area and dividing it bythe total cortical area for each thin section and then multiply-ing by 100 to obtain a percentage. A completely solid, avascularbone would have a cortical porosity of 0%, whereas a completelyporous bone would have a cortical porosity of 100%.

The diaphyseal cortical thickness of the limb bones was alsocalculated where possible. Several factors, including phylogeny,biomechanics, ontogeny, body size, and environmental fluctua-tions, influence the microstructure of bones (e.g., Cubo et al.,2005; Kriloff et al., 2008). However, numerous studies haveshown that lifestyle preferences affect the bone microstruc-ture of an animal as well (Wall, 1983; Stein, 1989; Bou et al.,1990; Fish, 1993; Germain and Laurin, 2005; Kriloff et al., 2008;Canoville and Laurin, 2010). For example, a relatively thick cor-tex (>30% of the average diameter) may indicate an aquaticor fossorial lifestyle (Wall, 1983; de Buffrenil et al., 1990; Huaand de Buffrenil, 1996; Botha, 2002, 2003), because thick bonewalls counteract buoyancy while in the water and provide ex-tra strength and support while swimming or digging. In extremecases, aquatic animals exhibit particularly thick, compact cor-tices (e.g., the manatee, Dugong dugong) or completely infilledmedullary cavities with broad transition zones between the com-pacta and medullary cavity (e.g., the dolphin, Delphinus delphis)depending on the extent of the aquatic lifestyle (Canoville andLaurin, 2010).

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1241

In this study, both the thickness of the cortex and the overallcompactness of the bone were measured. The cortical thickness(or ‘k,’ according to Currey and Alexander, 1985) was recordedby measuring the ratio between the inner diameter and outer di-ameter of the bone. Cortical thickness is thinnest when k tends to-wards 0 (i.e., no bone walls) and thickest when k tends towards 1(i.e., solid bone) (Currey and Alexander, 1985). Secondly, a bonecompactness profile (defined as “the ratio between the surfaceoccupied by bone tissues and the total bone surface” [Germainand Laurin, 2005:337]) was obtained for each element (wherepossible) using the computer program Bone Profiler for Windows(Girondot and Laurin, 2003). This method involves converting animage of the bone cross-section to a simple black and white im-age, where the black areas represent bone and the white areas themedullary spaces and channels within the cortex. Bone Profilerthen calculates the ratio between the mineralized tissue surfaceand the total cross-sectional surface (Girondot and Laurin, 2003;Laurin et al., 2004; Germain and Laurin, 2005; Canoville and Lau-rin, 2010). Compactness is lower towards 0 and higher towards 1.The method of Currey and Alexander (1985) is easy to imple-ment, but it assumes that transverse bone sections have perfectlycircular profiles and a sharp distinction between the medullarycavity and cortex. This is not always the case, thus the com-pactness method of Girondot and Laurin (2003) was also usedbecause it is more biologically appropriate. However, the lattermethod also requires completely preserved transverse sections.Because some thin sections revealed broken or fragmentary cor-tices and because the cortical channels were not always clearthroughout the entire cross-section in all study elements, onlythe bones of Euparkeria, a tibia (SAM-PK-K140b) and fibula(NMQR 880) of Proterosuchus, and a section of an Erythro-suchus radius (NMQR 3675) could be assessed using this method.

Institutional Abbreviations—BP, Bernard Price Institute forPalaeontological Research, University of the Witwatersrand, Jo-hannesburg; NMQR, National Museum, Bloemfontein; SAM-PK, Iziko South African Museum, Cape Town.

RESULTS

Prolacerta broomi

NMQR 3763 is a tibia from a single individual. Prolacerta spec-imens are rare and thus it is not known how large this animalgrew. However, the neurocentral sutures of the cervical verte-brae of NMQR 3763 are fused and the skull length measures 76mm in length, which is close to the largest specimen known (80mm maximum skull length). Thus, it is likely that NMQR 3763represents an adult.

Transverse sections through the midshaft of the tibia reveal arelatively compact cortex (k = 0.54; compactness 0.69) and en-larged resorption cavities in the perimedullary region. The en-dosteal lamellae form a ring around the inner cortex and donot extend into the medullary cavity, leaving it relatively freeof bony trabeculae. The compacta contains small vascular canals(17 µm in diameter) arranged as longitudinally oriented pri-mary osteons (cortical porosity 2.2%), distributed haphazardlythroughout the cortex (Fig. 2A–C). The osteocyte lacunae areglobular and mostly arranged haphazardly throughout the cor-tex and around the primary osteons, suggesting a fibro-lamellarcomplex. However, there are patches, in the absence of primaryosteons, where the osteocyte lacunae form a more organized,parallel arrangement (Fig. 2D). Under polarized light, the fibersare arranged in a single direction, apart from those surround-ing the primary osteons, but there are regions showing a woven-fibered bone matrix. These combined features indicate the pres-ence of a weakly developed fibro-lamellar complex, with regionsof parallel-fibered bone in the innermost and outermost regionsof the bone. Several small secondary osteons are observed on theposterolateral side of the bone. The vascular canals extend to the

subperiosteal surface without decreasing in size or abundance.Canaliculi are not preserved and growth rings are absent.

Proterosuchus fergusi

The skull of the largest specimen known for this genus, SAM-PK-K10603, measures 400 mm in length and is regarded as hav-ing reached somatic maturity. This specimen was taken to rep-resent 100% adult size against which specimen SAM-PK-K140was measured and is estimated to be 67% of the maximum sizeknown for this genus (as defined by skull length and associatedlimb bones). A complete skull was available for NMQR 880 andis estimated to be 82% the size of SAM-PK-K10603. SAM-PK-11208 does not comprise a complete skull, but using compara-ble limb bone measurements it is estimated to be 100% adult(Table 1).

SAM-PK-K140 includes a tibia and femur from one individual.These elements, which both contain large, free medullary cavi-ties, are surrounded by relatively narrow cortices (tibia k = 0.6;compactness 0.573; femur k = 0.58). The osteocyte lacunae areglobular, abundant, and arranged haphazardly in a woven-fiberedbone matrix. Canaliculi are not preserved. The vascular canals inthe tibia (22 µm in diameter; cortical porosity 9%) and femur (39µm in diameter; cortical porosity 15%) are numerous and eitherradiate out from the medullary cavity or are arranged as longitu-dinally oriented primary osteons in radial rows (Fig. 3A, B). Thetibia also exhibits patches of randomly distributed longitudinallyoriented primary osteons. There are several resorption cavitieson one side of the bone in the tibia, but they are not extensiveand secondary osteons were not observed in either limb bone.There is a very thin layer of circumferential endosteal lamellaesurrounding the medullary cavities in both elements. The orien-tation of the osteocyte lacunae becomes more organized at thesubperiosteal surface of the tibia, suggesting a slight decrease ingrowth rate, but growth rings are absent in both elements.

The bone tissue of fibula NMQR 880 (k = 0.37; compactness0.778) consists of a moderately vascularized (cortical porosity4.1%), poorly developed fibro-lamellar complex in the innermostcortex, which changes into clearly defined lamellar-zonal bonetowards the periphery (Fig. 3C, D). Annuli (representing a tem-porary decrease in growth rate) are absent from the inner cortex,but appear in the peripheral lamellar-zonal region. The vascularcanals (27 µm in diameter) in the inner cortex comprise mostlysimple longitudinally oriented canals, but also poorly defined pri-mary osteons. The osteocyte lacunae in this region are abundantand globular, but become flattened in the peripheral lamellarregion. Canaliculi are preserved in patches and radiate out in alldirections from the osteocyte lacunae. A clear medullary cavity issurrounded by multiple, thick layers of endosteal circumferentiallamellae that contain flattened osteocyte lacunae. Enlargedresorption cavities surround the medullary cavity in the per-imedullary region and extend into the mid-cortex in the regionof a ridge that runs longitudinally down the side of the boneshaft. Sharpey’s fibers are observed on the medial side of thebone.

SAM-PK-11208 includes a femur and fibula from a singleindividual. The femur (Fig. 3E), which was sectioned belowthe midshaft region towards the distal end, comprises a large,free medullary cavity that is surrounded by a narrow cortex(cortical thickness could not be quantified due to incompletepreservation of the cortex). Secondary remodeling in the per-imedullary region is extensive, with large erosion cavities extend-ing periosteally (close to the periosteal surface in places), butno secondary osteons were observed. The primary bone tissuecomprises lamellar-zonal bone with highly organized osteocytelacunae arranged parallel to one another (Fig. 3E). Vasculariza-tion is poor (cortical porosity 2.3%) and changes from radiallyoriented canals (16 µm in diameter) in the innermost cortex, to a

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1242 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

FIGURE 2. Bone histology of a tibia of the Early Triassic Prolacerta broomi, NMQR 3673. A, B, uninterrupted poorly vascularized bone tissue,with fine layers of endosteal bone (arrow) and resorption cavities (rc) in the perimedullary region; C, high magnification showing a poorly definedfibro-lamellar complex with primary (black arrowheads) and secondary (white arrowhead) osteons; D, high magnification showing a more organizedparallel-fibered region (white arrowheads). Scale bars equal 100 µm.

mixture of these and longitudinally oriented simple vascularcanals in the outer cortex. There are fewer vascular canals atthe subperiosteal surface. Multiple annuli and lines of arrestedgrowth (LAGs; which represent a periodic, but complete ces-sation in growth) interrupt the bone tissue from the mid- toouter cortex. Sharpey’s fibers, for the attachment of muscles, arepresent in some regions.

The fibula of SAM-PK-11208 consists of a relatively narrowcortex (k = 0.73) that surrounds a large, clear medullary cav-ity. Secondary remodeling extends into the mid-cortex. Althoughlarge erosion cavities surround the medullary cavity, secondaryosteons are absent. The primary bone tissue consists of a parallel-fibered bone matrix with radiating vascular canals (23 µm in di-ameter) (Fig. 3F) interspersed with longitudinally oriented pri-mary osteons (cortical porosity 7%). The bone tissue becomespoorly vascularized lamellar-zonal bone with sparse longitudi-nally oriented simple canals and multiple annuli towards the pe-riphery. A small section of circumferential endosteal lamellaebordering the medullary cavity is present and Sharpey’s fiberswere observed in places.

Erythrosuchus africanus

Complete skulls are not available for any of the study spec-imens and, thus, identifications were based on partial skullmaterial. Charig et al. (1976) report 960 mm as the maximumskull length for this taxon. Specimen NMQR 3675 comprisesan almost complete articulated skeleton and two-thirds of theskull is preserved. Thus the length of the complete skull couldbe approximated and is estimated to be 63% of the maximumsize known for this genus. Using limb bone measurements,specimen SAM-PK-K10025 is estimated to be 56% maximumsize. SAM-PK-K1118 does not include any limb bones thatcould be compared with the other specimens and only skullfragments of this individual are preserved, thus the ontogeneticstatus of this individual could not be estimated using grossmeasurements.

The Erythrosuchus material includes a tibia and rib from oneindividual (SAM-PK-K10025), a rib and unidentified fragmentfrom another individual (SAM-PK-1118), and a radius and threerib fragments from a third individual (NMQR3675). All elements

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1243

FIGURE 3. Bone histology of the Early Triassic Proterosuchus fergusi. A, tibia SAM-PK-K140b showing uninterrupted fibro-lamellar bone with ra-dially oriented vascular canals; B, femur SAM-PK-K140c showing uninterrupted fibro-lamellar bone with a thin layer of circumferential endosteallamellae (arrow) surrounding the medullary cavity; C, fibula NMQR 880 showing longitudinally oriented vascular canals in the mid-cortex andlamellar-zonal bone towards the periphery (arrowhead). An annulus (arrow) is present in the lamellar-zonal region; D, high magnification of thelamellar-zonal region (arrowhead) of fibula NMQR 880; E, F, femur SAM-PK-11208a and fibula SAM-PK-11208b showing a change to lamellar-zonalbone towards the periphery (arrowheads). Scale bars equal 1000 µm in A, B, and C; 100 µm in D; 500 µm in E and F.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1244 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

FIGURE 4. Bone histology of the Middle Triassic Erythrosuchus africanus. A, B, radius NMQR 3675 and tibia SAM-PK-K10025 showing unin-terrupted fibro-lamellar bone; C, D, rib NMQR 3675c showing longitudinally oriented primary and secondary (arrows) osteons in a fibro-lamellarcomplex. Arrowhead indicates an annulus near the periphery. Scale bars equal 500 µm in A; 1000 µm in B and C; 100 µm in D.

contain numerous vascular canals in a woven-fibered bone ma-trix. The tibia of SAM-PK-K10025 (Fig. 4A) has undergone somediagenesis, particularly in the inner and mid-cortex, and is thusnot well preserved. Similarly, the outer cortex of the rib fromthe same individual has also been diagenetically altered. It isclear, however, that the tibia consists of a thick cortex (k = 0.35).A region of cancellous bone with large erosion cavities in theperimedullary region results in a small free region in the centerof the bone. The primary bone tissue consists of moderate tohighly vascularized fibro-lamellar bone in a plexiform network,similar to that seen in the femur/humerus in Gross (1934) and thefibula in de Ricqles et al. (2008). The canals are slightly smallerperiosteally, and longitudinally oriented primary osteons aremore common in the perimedullary and mid-cortical regions thanat the subperiosteal surface. The osteocyte lacunae are only pre-served in patches and appear to be globular in shape. Canaliculiare not preserved. Growth rings are absent. The rib of SAM-PK-K10025 has also been diagenetically altered, but is somewhatbetter preserved than the tibia. It contains a small medullarycavity that is filled with bony trabeculae. Resorption cavities areobserved in the perimedullary region. The fibro-lamellar bone of

the rib is highly vascularized, similar to the condition seen in therib of de Ricqles et al. (2008) and contains numerous large (67µm in diameter) longitudinally oriented primary and secondaryosteons randomly dispersed throughout the cortex. Anastomosesare absent and there is no decrease in vascular density towardsthe periphery. Osteocyte lacunae are only preserved in patchesand are globular in shape, and canaliculi are not preserved.

The elements of SAM-PK-1118 are better preserved. Both therib (cortical porosity 13%) and fragment consist of highly vascu-larized fibro-lamellar bone where vascular density remains con-stant up to the subperiosteal surface. The vascular canals aremostly primary osteons with short anastomoses and form a retic-ular network in places. They are smaller in diameter (average33 µm) than those of rib SAM-PK-K10025. The rib contains alarge medullary cavity that is filled with broken fragments oftrabeculae. Some secondary remodeling is apparent in the per-imedullary region in the form of enlarged resorption cavities.Small secondary osteons are common throughout the cortex. Theosteocyte lacunae are globular and abundant, but canaliculi arenot preserved. Two faint annuli were observed in each elementnear the subperiosteal surfaces.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1245

The bone tissue of radius NMQR 3675 (Fig. 4B) is similar totibia SAM-PK-K10025, but is a little better preserved. A rela-tively thick cortex (k = 0.2; compactness 0.856) surrounds a largemedullary cavity. A few broken fragments of trabecular bonelie within the medullary cavity but two-thirds of the region pre-serves a thin layer of circumferential endosteal lamellae and ero-sion cavities are largely absent, suggesting that the medullary cav-ity was probably relatively free of bony trabeculae during life.The cortex consists of uninterrupted highly vascularized (corti-cal porosity 9%) fibro-lamellar bone. The vascular canals are ar-ranged either in a plexiform network or they radiate out towardsthe subperiosteal surface. There is no decrease in channel densitytoward the periphery. The vascular canals are relatively narrow(27 µm in diameter). Numerous canaliculi are preserved and radi-ate out from the abundant, globular osteocyte lacunae in all direc-tions. Small secondary osteons are observed in the perimedullaryregion.

Endosteal trabeculae extend into the medullary cavities of allthree ribs of NMQR 3675. The cortex is relatively compact andcomprises a highly vascularized woven-fibered bone matrix. Amixture of large (43 µm in diameter) longitudinally oriented pri-mary and secondary osteons are arranged haphazardly through-out the cortex. A few isolated enlarged resorption cavities are ob-served in the mid-cortex. Canaliculi are preserved in patches andradiate out from the abundant, globular osteocyte lacunae, whichare distributed randomly throughout the cortex. One of the ribscontains an annulus near the subperiosteal surface (Fig. 4C, D).New bone containing large primary osteons was deposited afterthis annulus.

Euparkeria capensis

Ewer (1965) described all of the Euparkeria individuals asadults, although there is some variability with complete skullsranging from 69 to 100 mm in length. This latter measurementwas used along with comparable limb bone measurements to es-timate the ontogenetic status of the Euparkeria material in thisstudy. All study material has been assigned an adult status be-cause estimates range from 88% to 96% of the maximum knownsize of this genus (Table 1).

SAM-PK-13666 is a humerus that comprises a relatively thickcortex (k = 0.44; compactness 0.771). The small medullary cav-ity is devoid of trabecular infilling. Secondary remodeling is al-most non-existent; one enlarged resorption cavity is present onthe posteromedial side of the bone. The primary bone tissue con-sists of highly organized (the osteocyte lacunae are arranged par-allel to one another in places) parallel-fibered bone. The vascu-lar canals (cortical porosity 5.4%) are arranged as longitudinallyoriented primary osteons and some regions also have short anas-tomoses forming a subreticular pattern (Fig. 5A). The abundantglobular osteocyte lacunae radiate short canaliculi. Two indistinctwide annuli interrupt the bone tissue and a layer of circumferen-tial endosteal lamellae surrounds the medullary cavity. Calcifiedcartilage was observed in the proximal epiphyseal region.

SAM-PK-K10010 includes a femur, tibia, and fibula from a sin-gle individual. As femur SAM-PK-K10010a was not complete,thin sections had to be taken slightly proximal to the midshaft re-gion. In all three elements, a narrow cortex (compactness: femur0.448, tibia 0.738, fibula 0.692) surrounds a large free medullarycavity (k: femur 0.73, tibia 0.48, fibula 0.51). The primary bonetissue is poor to moderately vascularized (depending on the el-ement), highly organized parallel-fibered bone (Fig. 5B, C, D).The osteocyte lacunae are globular in shape and abundant, butcanaliculi are not preserved in any element. Cortical porosity is3.3% in the femur, 3.5% in the tibia, and 5% in the fibula. Thecanals mostly consist of longitudinally oriented simple canals,with a few randomly distributed primary osteons. The tibia con-tains more canals with short anastomoses forming a subreticu-

lar pattern in places (Fig. 5C). Cortical porosity remains constantthroughout the cortex. Two LAGs interrupt the bone tissues ofboth the femur and fibula. Multiple LAGs and annuli are presentin the tibia. Secondary remodeling is more extensive in the femur,on the posterolateral side of the bone (Fig. 5B), probably becauseit was thin-sectioned closer to the metaphysis than the other el-ements. In the tibia and fibula, enlarged resorption cavities arepresent only in a small region on the anterior and posterolateralsides of the bones, respectively. A fairly substantial layer of cir-cumferential endosteal lamellae surrounds the medullary cavityin the fibula (Fig. 5D). Calcified cartilage was observed in the dis-tal epiphyseal region of the tibia.

The femur SAM-PK-K10548 (Fig. 6) consists of a relativelynarrow cortex (k = 0.53; compactness 0.661) that surrounds alarge free medullary cavity. The moderately vascularized (4.4%),highly organized parallel-fibered bone contains mostly longitu-dinally oriented simple canals with a few primary osteons andshort anastomoses in places (Fig. 6A, B). The osteocyte lacu-nae vary between globular and flattened, and canaliculi are notpreserved. The vascular density remains constant throughout thecortex (Fig. 6A). A few enlarged resorption cavities are presenton the anteromedial side of the bone, but they are not exten-sive. A thin layer of inner circumferential endosteal lamellae sur-rounds most of the medullary cavity. Calcified cartilage was ob-served in the distal epiphyseal region (Fig. 6C).

DISCUSSION

Growth Patterns of the Studied Taxa

Table 2 provides a summary of the quantifiable characters andbone tissue patterns of the taxa examined in this study. The bonetissue pattern of Prolacerta is characterized by parallel-fiberedbone with longitudinal primary osteons and localized woven bonematrix. Some regions contain primary osteons and haphazardlyarranged globular osteocyte lacunae, suggesting relatively rapidgrowth, and others contain fewer primary osteons and more or-ganized bone tissue, with flattened osteocyte lacunae arrangedparallel to one another, suggesting relatively slow growth. Thebone matrix of Prolacerta can thus best be described as interme-diate between parallel fibered (Fig. 7A) and woven (Fig. 7B).Transitions between these categories are known in other taxa(e.g., some ‘pelycosaurs’ and dinosaurs [Enlow and Brown, 1957;Redelstorff and Sander, 2009; Huttenlocker and Rega, in press],and some extant squamates and proboscidians [de Ricqles et al.,1991]) and can even appear within the same section (Francillon-Vieillot et al., 1990). The individual studied is close in size to thelargest Prolacerta specimen known and the neurocentral suturesof the cervical vertebrae are fused, suggesting that this animalhad reached adult status at time of death. However, there is noindication of a decrease in growth rate towards the periphery andgrowth rings are absent. The closure of the cervical neurocentralsutures has been found to be a reliable indicator of maturity in ex-tant crocodilians (Brochu, 1996) and squamates (Maisano, 2002),but does not necessarily indicate that maximum size has beenreached. Although it is possible that Prolacerta grew considerablylarger than is currently known, it is clear from both gross mor-phology and the bone microstructure that specimen NMQR 3673was not a juvenile. Experiments on extant reptiles have shownthat growth rings, which represent a temporary slowing downor cessation in growth, are deposited annually, during the unfa-vorable growing season (Hutton, 1986). Because the climate ofthe Early Triassic Karoo is characterized by highly seasonal rain-fall and possibly drought conditions, it is noteworthy that eitherthe growth of Prolacerta (at least during early to mid-ontogeny)did not temporarily slow down or cease during the unfavourablegrowing season or adult status was attained within one year.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1246 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

FIGURE 5. Bone histology of the Middle Triassic Euparkeria capensis. A, humerus SAM-PK-13666 showing parallel-fibered bone (black arrow-head) and an annulus (white arrowhead) in the mid-cortex; B, femur SAM-PK-K10010a showing poorly vascularized parallel-fibered bone with LAGs(arrowheads). Arrows show enlarged resorption cavities in the perimedullary region; C, tibia SAM-PK-K10010b showing a subreticular vascularpattern and LAGs (arrowheads), which become multiple in places; D, fibula SAM-PK-K10010c showing a thick region of parallel-fibered bone inter-spersed with LAGs (arrowheads), but continued growth at the subperiosteal surface, and a thick region of circumferential endosteal lamellae (arrow)surrounding the medullary cavity (mc). Scale bars equal 100 µm in A; 500 µm in B, C, and D.

The presence of abundant radiating vascular canals in a fibro-lamellar complex and absence of growth rings in ProterosuchusSAM-PK-K140 (67% maximum size) suggests that this individualwas growing relatively rapidly at time of death (Fig. 7C). In con-trast, the bone tissue patterns of the ontogenetically older spec-imens, NMQR 880 and SAM-PK-11208 (82% and 100% maxi-mum size, resepctively), reveal a dramatic change in growth fromrapidly forming fibro-lamellar bone to slowly forming lamellar-zonal bone tissue (Fig. 7D). This change in tissue type indicatesthat sexual maturity had probably already been reached. Sex-ual maturity has been shown to be associated with a dramaticchange to a slowly forming bone tissue type (parallel-fibered orlamellar bone matrix) in living taxa (e.g., amphibians and rep-tiles [Castanet and Smirina, 1990; Castanet and Baez, 1991] andwalrus Odobenus rosmarus [Klevezal, 1996]) and the associa-tion has also been proposed for several fossil taxa (non-avian di-nosaurs [Reid, 1996; Sander, 2000] and non-mammaliaform cyn-odont therapsids [Botha and Chinsamy, 2005]). Lee and Werning

(2008) demonstrated that sexual maturity in dinosaurs may haveoccurred with a decrease in bone deposition rate, but prior to theonset of a different bone tissue type. The growth patterns of Pro-terosuchus are interpreted as relatively rapid continuous growthuntil fairly late in ontogeny (at least until 67% adult size), with atransformation to slow, interrupted growth once sexual maturitywas reached. The appearance of growth rings in the peripherallamellar-zonal region indicates that growth became intermittentlate in ontogeny and either decreased or ceased temporarily dur-ing the unfavourable growing season.

The growth of Erythrosuchus is characterized by rapid, unin-terrupted growth, as revealed by relatively highly vascularizedfibro-lamellar bone (Fig. 7E), and supports previous observationsof earlier studies (Gross, 1934; de Ricqles, 1976; de Ricqles etal., 2008). Growth rings have not been reported previously forErythrosuchus, nor were they observed in the limb bones in thisstudy. However, annuli were observed in a fragment and two ribs,indicating that Erythrosuchus was capable of varying its growth

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1247

FIGURE 6. Bone histology of the Middle Triassic Euparkeria capen-sis, femur SAM-PK-K10548. A, B, showing parallel-fibered bone withlongitudinally oriented simple canals, interrupted by annuli and LAGs(arrowheads). A thin layer of circumferential endosteal lamellae (arrow)surrounds the medullary cavity (mc); C, high magnification of the epiphy-seal region showing calcified cartilage (arrow). Scale bars equal 1000 µmin A; 100 µm in B and C.

rate. The general absence of growth rings, however, until at least63% adult size (Table 1), indicates early rapid growth to at leastthe subadult stage. De Ricqles et al. (2008) noted a decrease ingrowth rate and a poorly defined external fundamental system orEFS (as defined by Cormack, 1987) in a rib (BP/1/4680), whichindicates the attainment of maximum size and thus a virtual ces-sation in growth (although an increase in mass may still occur).Interestingly, the authors noted that growth rings were absent inall study elements, including the rib containing the EFS, contraryto our findings. It is possible that the change in bone tissue to-wards the periphery in their rib represented a temporary decreasein growth rate, but not necessarily an EFS, or this difference maybe the result of individual variation (caused by either genetic orenvironmental differences).

All the Euparkeria elements used in this analysis were consid-ered to be adult a priori and the bone microstructure of all thelimb bones supports this assumption. The presence of parallel-fibered bone and prominence of annuli and LAGs indicates acomparatively slow growth rate for this genus (Fig. 7F). These ob-servations differ from what was noted by de Ricqles et al. (2008).They examined an unidentified limb bone (a possible humerusor femur), osteoderms, and rib fragments of Euparkeria. The au-thors noted that growth rings were absent and the bone tissueconsisted of a poorly defined fibro-lamellar complex. It is pos-sible that the limb bone in their study belonged to a juvenileindividual and growth rings were absent early in ontogeny (al-though Ewer [1965] noted that the Euparkeria material consistedof an aggregation of adults based on gross morphology), or thatthe element may belong to Mesosuchus and not Euparkeria. Fos-sils of these taxa are closely associated on the sandstone blocksin which they are preserved. The growth of Euparkeria differsmarkedly from the other taxa studied, in the presence of poorlyvascularized parallel-fibered bone throughout the entire cortexand prominence of annuli and LAGs. It was a relatively small an-imal (approximately 1 m body length) and considerably smallerthan Proterosuchus (2 m body length) or Erythrosuchus (5 mbody length). Within a given clade, large taxa often grow abso-lutely faster than smaller ones do (Case, 1978; Padian et al., 2004),and the slower growth in Euparkeria may simply be size related.

The compactness values of all the studied taxa range from0.448 in a Euparkeria femur SAM-PK-K10010a to 0.856 in theErythrosuchus radius NMQR 3675. Femur SAM-PK-K10010awas sectioned slightly below the mid-diaphysis towards the meta-physeal region (and thus accounts for the lower compactness be-cause metaphyses are more spongy than the diaphysis), but therest of the Euparkeria values range from 0.661 in femur SAM-PK-K10548 to 0.771 in humerus SAM-PK-13666. The Prolacertaand Proterosuchus compactness values are similar to those of Eu-parkeria. The k values indicate moderately thick cortices in Pro-lacerta (0.54), Proterosuchus (average 0.57), and Euparkeria (av-erage 0.54). These results agree with Cubo et al.’s (2005) findings;these authors described the last common ancestor of archosaursas having moderately thick cortices (0.476). In contrast, Erythro-suchus has a relatively thicker cortex (average 0.28), similar tothat of living crocodilians (0.22; Cubo et al., 2005).

None of the taxa in this study exhibit pachyostosis (thickeningof the cortex with no medullary cavity) as seen in aquatic animalssuch as plesiosaurs, nor do they have particularly thick compactcortices with tiny medullary cavities (to counteract buoyancy) asseen in shallow water aquatic animals such as Dugong dugong(manatee) or Ornithorhynchus anatinus (duckbilled playpus)(Germain and Laurin, 2005). Completely infilled medullarycavities with broad transition zones between the compacta andmedullary cavity as is typical of aquatic diving animals (suchas the dolphin Delphinus delphis and the southern elephantseal Mirounga leonina: Germain and Laurin, 2005; Kriloff etal., 2008) are also absent (Fig. 8). Prolacerta, Euparkeria, and

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1248 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

TABLE 2. Quantifiable characters and bone tissue patterns for the specimens examined in this study.

Genus, Element Accession number kCompactness

valueCortical

porosity (%)Vascular canaldiameter (µm)

Bone tissuetype

Growthrings

Prolacerta broomiTibia NMQR 3763 0.54 0.69 2.2 17 FLB/PFB Absent

Proterosuchus fergusiTibia SAM-PK-K140b 0.6 0.573 9 22 FLB AbsentFemur SAM-PK-K140c 0.58 15 39 FLB AbsentFibula NMQR 880 0.37 0.778 4.1 27 FLB/LZB PresentFemur SAM-PK-11208a — — 2.3 16 LZB MultipleFibula SAM-PK-11208b 0.73 — 7 23 PFB/LZB Multiple

Erythrosuchus africanusRib SAM-PK-K1118a — 13 33 FLB PresentTibia SAM-PK-K10025a 0.35 — — 66 FLB AbsentRib SAM-PK-K10025b — — — 67 FLB AbsentRadius NMQR 3675a 0.2 0.856 9 27 FLB AbsentRib NMQR 3675c — — — 43 FLB Present

Euparkeria capensisHumerus SAM-PK-13666 0.44 0.771 5.4 13 PFB PresentFemur SAM-PK-K10010a 0.73 0.448 3.3 13 PFB PresentTibia SAM-PK-K10010b 0.48 0.738 3.5 17 PFB PresentFibula SAM-PK-K10010c 0.51 0.692 5 18 PFB PresentFemur SAM-PK-K10548 0.53 0.661 4.4 15 PFB Present

Abbreviations: FLB, fibro-lamellar bone; LZB, lamellar-zonal bone; PFB, parallel-fibered bone.

Erythrosuchus are considered to be terrestrial; Proterosuchusis the only taxon in this study whose lifestyle is still activelydebated. The cortical thickness values of Proterosuchus aresimilar to (Prolacerta, Euparkeria) or less than (Erythrosuchus)those of the other taxa. Moreover, the femoral cortical thicknessof Proterosuchus is 0.58, whereas the amphibious Nile crocodile(Crocodylus niloticus) has a femoral cortical thickness of 0.22(Cubo et al., 2005). Similarly, the tibial compactness valueof Proterosuchus (0.573) is also lower than that of Alligatormississipiensis, which exhibits a tibial compactness value of0.78 (Kriloff et al., 2008). Given the difference in these valuesand the observation that Proterosuchus does not exhibit any ofthe aquatic histological features listed above, we conclude thatthe bone microstructure of Proterosuchus does not support anaquatic lifestyle. Interestingly, the cortical thickness of Erythro-suchus (radius k = 0.2; tibia k = 0.35) is similar to that of thesemi-aquatic Alligator mississipiensis (0.22; Cubo et al., 2005).However, Currey and Alexander (1985) found that althoughshallow water aquatic animals (i.e., non-divers) such as the alliga-tor and manatee revealed low k values (i.e., thick cortices), theyalso found that larger animals also have low k values (e.g., thesloth bear Melursus ursinus, k = 0.26). Because Erythrosuchusexhibits no morphological features suggesting that it was aquaticor even semi-aquatic, we attribute the low k values of this animalto its large size, because thick cortices are stronger under local-ized impact than thinner ones (Currey and Alexander, 1985).

Archosauromorph Growth Patterns in a Phylogenetic Context

Although early studies (e.g., Seitz, 1907; Gross, 1934; Enlowand Brown, 1957; de Ricqles, 1976) set a solid foundation onwhich further research could develop, several recent studies onarchosauromorph bone histology have began to provide morecomprehensive information about the growth patterns of thisgroup (e.g., de Ricqles et al., 2008; Nesbitt et al., 2009; Wern-ing and Irmis, 2010). These new data have shed light on the evo-lutionary relationships between archosauromorph taxa and withthe archosaur crown group in a histological context (Ricqles etal., 2008; Werning and Irmis, 2010).

Werning and Irmis (2010) recently revealed a slow over-all growth pattern for the Late Triassic basal archosauro-morph Trilophosaurus buettneri. They described the bonemicrostructure of this taxon as low to moderately vascularized,lamellar-zonal bone. This type of bone tissue, in association withsometimes multiple LAGs, was evident throughout ontogenyand indicates an overall slow and cyclical growth rate (Wern-ing and Irmis, 2010). Similarly, Nesbitt et al. (2009) recentlydescribed the adult femoral bone histology of the Late Triassicarchosauriform Vancleavea campi as slow, noting a poorlyvascularized lamellar-zonal bone tissue with distinct LAGsthroughout the cortex. These authors also noted a particularlythick cortex and infilled medullary cavity and suggested, inassociation with gross morphological features, that Vancleaveawas semi-aquatic (Nesbitt et al., 2009). De Ricqles et al. (2008)also examined several Middle–Late Triassic indeterminaterhynchosaur archosauromorph limb bones and found slowlyforming lamellar-zonal bone tissue with distinct growth cycles,typical of extant reptiles. They also studied the Middle Triassicsemi-aquatic proterochampsid archosauriform Chanaresuchusand noted an early rapid growth rate (based on an inner thick re-gion of fibro-lamellar bone), which decreased dramatically laterin ontogeny (based on the presence of peripheral lamellar-zonalbone: de Ricqles et al., 2008).

When the data obtained in these studies is added to this studyand examined in a phylogenetic context, a mixture of bone tissuepatterns emerges. The basal archosauromorph Trilophosaurusand the rhynchosaur archosauromorphs (including Scaphonyx;see Enlow and Brown, 1957) display slow, cyclical growth simi-lar to living reptiles (de Ricqles et al., 2008; Werning and Irmis,2010). Prolacerta, a slightly more derived archosauromorph, re-veals relatively faster, uninterrupted growth. If we assume thatthe cortex of Prolacerta was deposited in a single year (basedon the absence of growth rings), we estimate the bone deposi-tion rate to have been approximately 2.1 µm/day (using a year of380 days), which is similar to extant crocodilians (2.48 µm/day;Montes et al., 2010), and faster than extant varanid lizards (1.18µm/day; Montes et al., 2010). The more derived archosauri-forms reveal progressively increasing growth rates, with Protero-suchus (11.3 µm/day) and Chanaresuchus (de Ricqles et al., 2008)

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1249

FIGURE 7. Transverse sections in polarized light. A, Prolacerta NMQR 3673, tibia showing parallel-fibered bone; B, Prolacerta NMQR 3673, tibiashowing woven-fibered bone; C, Proterosuchus SAM-PK-K140b tibia showing woven-fibered bone; D, Proterosuchus NMQR 880 fibula showinglamellar-zonal bone towards the periphery; E, Erythrosuchus NMQR 3675 radius showing woven-fibered bone; F, Euparkeria SAM-PK-K10548femur showing parallel-fibered bone. Scale bars equal 100 µm in A, B, C, and F; 500 µm in D and E.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1250 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

FIGURE 8. Mid-diaphyseal cross-sections of the study taxa showing therelative cortical thickness and cortical compactness (apart from D). A,Prolacerta tibia; B, Euparkeria tibia; C, Proterosuchus tibia; D, Erythro-suchus tibia; and E, Erythrosuchus radius. Bone shaded in black. The vas-cular canals in D are not shown due to patchy preservation, but E givesan indication of the size of the vascular canals in Eythrosuchus. Imagesare not to scale. Abbreviation: mc, medullary cavity.

FIGURE 9. Stratocladogram of archosauromorph bone histology stud-ied to date. White shading indicates lamellar-zonal bone, black shadingindicates fibro-lamellar bone, and grey shading indicates parallel-fiberedbone tissue. Phylogeny taken from Sues (2003), Dilkes and Sues (2009),and Nesbitt et al. (2009). Histological information taken from de Ricqleset al. (2008), Nesbitt et al. (2009), Werning and Irmis (2010), and thisstudy. Time scale taken from Walker and Geissman (2009). Numbersindicate million years ago. Abbreviations: Chsn, Changhsingian; Olkn,Olenekian; PTB, Permo-Triassic boundary.

exhibiting rapid early growth, but slow, cyclical late growth, andErythrosuchus, uninterrupted rapid growth (21 µm/day) through-out ontogeny. Montes et al. (2010) found that the last commonancestor of archosaurs had a bone deposition rate of approxi-mately 13.5 µm/day (Montes et al., 2010). Non-avian dinosaursranged from 3 to 24 µm/day (Sander and Tuckmantel, 2003;Cooper et al., 2008). Thus, the growth rates of Proterosuchus andErythrosuchus are relatively high, with Erythrosuchus exhibitinga growth rate similar to the fastest-growing dinosaurs. It shouldbe noted, however, that these rates are mere estimates and sev-eral assumptions were made in order to calculate these rates; forexample, the number of days in a Triassic year (380 days) andthe active growth period of these animals in one year (i.e., an-nual growth may have been restricted to a 6-month period dur-ing the more favorable growing months). Substantial variationexists both among different elements of a single individual andwithin a single bone type. For example, Starck and Chinsamy(2002) found that the depositional rates of fibro-lamellar bonein Japanese quail varied from 10 to 50 µm/day.

In general, however, there appears to be an increase in the pre-dominance of rapidly forming fibro-lamellar bone within the ar-chosauriforms, as previously noted by de Ricqles et al. (2008).However, Euparkeria (estimated bone deposition rate of 1.4µm/day) and Vancleavea deviate from this pattern. There may beseveral possible reasons for their relatively slower growth strate-gies, such as differences in body size or nutrient availability. On-togenetic variation is unlikely to be the cause of these differencesas all these specimens are considered to be adults.

The bone tissue patterns of these basal archosaurs differ fromthose of many non-archosaurian reptiles, which predominantlycontain slowly forming lamellar bone that is usually cyclical in na-ture. The Permian basal eureptiles Captorhinus, Moradiasaurus(de Ricqles, 1976), and Claudiosaurus (de Buffrenil and Mazin,

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1251

1989), as well as the basal diapsid eureptile Youngina (pers. ob-serv., 2010), all reveal a slowly forming lamellar bone matrix,which, in Claudiosaurus, is cyclical in nature. Triassic sauroptery-gian eureptiles such as Pachypleurosaurus, Neusticosaurus, andNothosaurus also contain slowly forming lamellar-zonal bone(Enlow and Brown, 1957; Sander, 1990; Hua and Buffrenil, 1996;Scheyer et al., 2010). In contrast, the bone tissues of the Tri-assic aquatic ichthyosaurs Omphalosaurus, Stenopterygius, andIchthyosaurus reveal rapidly forming fibro-lamellar bone, whichhas been attributed to their active predatory, aquatic lifestyles(de Buffrenil and Mazin, 1989). Chelonia appear to exhibitlamellar-zonal bone as a group (e.g., Enlow and Brown, 1957;de Ricqles, 1976; Castanet and Cheylan, 1979). Few pararep-tiles, which forms the sister group of Eureptilia, have beenstudied; however, de Ricqles (1976) and Scheyer and Sander(2009) found slowly forming lamellar-zonal bone in Permianpareiasaurs. More research on parareptiles is needed, particularlyon Triassic parareptiles, in order to elucidate the bone tissue pat-terns of this group and to compare them with archosaurs.

De Ricqles et al. (2008:72, 74) suggested that the ability to“reach and maintain” rapid growth rates during at least early andmid-ontogeny was plesiomorphic for archosauriforms and thatthe Triassic was a “time of experimentation in growth strategies.”Although archosauromorph growth patterns do appear to re-flect a phylogenetic signal, in that rapidly forming fibro-lamellarbone becomes increasingly more prominent, derived taxa suchas Vancleavea and Euparkeria, which are considered to be rel-atively closely related to crown group archosaurs, deviate fromthis pattern and suggest that a relatively weak phylogenetic sig-nal was present and that factors other than phylogeny must beconsidered when interpreting the observed growth patterns ofarchosauriforms; something which de Ricqles et al. (2008) alsoemphasized.

Archosauromorph Growth Patterns in an Ecological Context

Stem archosaurs that lived during the Early Triassic (e.g., Pro-lacerta, Proterosuchus) and Middle Triassic (e.g., Erythrosuchus,Chanaresuchus) exhibit relatively rapid growth rates that werefaster than many of those that lived during the Late Triassic(e.g., Trilophosaurus, Scaphonyx, Vancleavea). It is interestingto note that although the Early Triassic Prolacerta is similar inbody size to the Middle Triassic Euparkeria, it grew relativelymore quickly. The climate of the Early Triassic Karoo Basin ischaracterized by unreliable monsoonal-type torrential rainfall in-terspersed with extreme drought (Smith and Botha, 2005). Therelatively rapid continuous growth seen in some of these non-dinosaurian archosaurs may have allowed these taxa to reachsexual maturity relatively quickly (Fig. 9). Early rapid, uninter-rupted growth strategies have been observed in other taxa suchas the dicynodont Lystrosaurus (Botha-Brink and Angielczyk,2010) and the therocephalian Moschorhinus (Huttenlocker et al.,2010), both of which survived the end-Permian mass extinctionand formed an important part of the Early Triassic fauna. At-taining sexual maturity relatively quickly is advantageous in un-predictable environments and has been observed in living taxa(e.g., Cardinale and Modin, 1999; Curtin et al., 2009). Fast earlygrowth and rapid attainment of sexual maturity are consideredto be evolutionary adaptations against low juvenile survivorshipand a significantly shorter life span in unpredictable, stressful en-vironmental conditions (e.g., Tinkle, 1969; Gasser et al., 2000;Curtin et al., 2009). Rapid early growth in Early Triassic verte-brates is consistent with life history theory and we therefore sug-gest that early attainment of sexual maturity was advantageousin the unpredictable rainfall regime during this time period whenjuvenile (as well as adult) mortality was high. As the climate laterameliorated, some animals (such as Vancleavea and Euparkeria)returned to slow early growth (and possibly delayed sexual ma-

turity). However, rapid early growth (i.e., rapidly forming fibro-lamellar bone) was retained by the early crown group archosaursand the later, more derived ornithodirans throughout the LateTriassic into the Jurassic (de Ricqles et al., 2008).

As our sample size is small, this hypothesis must remainpreliminary. However, future research may be able to testthe environmental stress theory by increasing the Triassicarchosauromorph sample size, including Permian eureptiles andTriassic parareptiles, and analyzing the bone microstructure ofvarious vertebrate clades (i.e., other than reptiles) across thePermo-Triassic boundary to determine if rapid early growth wasan important factor in the Early Triassic recovery.

ACKNOWLEDGMENTS

We thank A. Crean of the Iziko South African Museum ofCape Town and J. Nyaphuli of the National Museum, Bloem-fontein for the mechanical preparation of the study material.K. Angielczyk, S. Werning, A. Lee, and K. Curry-Rogers arekindly acknowledged for their comments on an earlier draft ofthe manuscript. This project was funded by research grants fromthe National Research Foundation of South Africa to J.B.B.(UID65244) and R.M.H.S. (UID68944).

LITERATURE CITED

Amprino, R. 1947. La structure du tissue osseux envisage comme expres-sion de differences dans la vitesse de l’accroissement. Archives deBiologie 58:315–330.

Benton, M. J., and J. M. Clark. 1988. Archosaur phylogeny and the re-lationships of the Crocodilia; pp. 295–338 in M. J. Benton (ed.),The Phylogeny and Classification of the Tetrapods. 1. Amphibians,Reptiles, Birds. Systematic Association Special Volume. ClarendonPress, Oxford, U.K.

Benton, M. J., V. P. Tverdokhlebov, and M. V. Surkov. 2004. Ecosystemremodelling among vertebrates at the Permian-Triassic boundary inRussia. Nature 432:97–100.

Botha, J. 2002. The palaeobiology of the non-mammalian cynodonts de-duced from bone microstructure and stable isotopes. Ph.D. disserta-tion, University of Cape Town, Cape Town, South Africa, 218 pp.

Botha, J. 2003. Biological aspects of the Permian dicynodont Ouden-odon (Therapsida, Dicynodontia), deduced from bone histology andcross-sectional geometry. Palaeontologia Africana 39:37–44.

Botha, J., and A. Chinsamy. 2005. Growth patterns of Thrinaxodon, anon-mammalian cynodont from the Early Triassic of South Africa.Palaeontology 48:385–394.

Botha, J., and R. Smith. 2006. Rapid vertebrate recuperation in the KarooBasin of South Africa following the end-Permian extinction. Journalof African Earth Sciences 45:502–514.

Botha-Brink, J., and K. D. Angielczyk. 2010. Do extraordinarily highgrowth rates in Permo-Triassic dicynodonts (Therapsida, Anomod-ontia) explain their success before and after the end-Permian extinc-tion? Zoological Journal of the Linnean Society 160:341–365.

Bottjer, D. J., M. E. Clapham, M. L. Fraiser, and C. M. Powers. 2008. Un-derstanding mechanisms for the end-Permian mass extinction andthe protracted Early Triassic aftermath and recovery. GSA Today18:4–10.

Bou, J., M. J. Castiella, J. Ocana, and A. Casinos. 1990. Multivariate anal-ysis and locomotor morphology in insectivores and rodents. Zoolo-gischer Anzeiger 225:287–294.

Brink, A. S. 1955. Notes on some thecodonts. Navorsinge van die Na-sionale Museum Bloemfontein 1:41–148.

Brochu, C. A. 1996. Closure of neurocentral sutures during crocodilianontogeny: implications for maturity assessment in fossil archosaurs.Journal of Vertebrate Paleontology 16:49–62.

Broili, K., and J. Schroder. 1934. Beobachungen an Wirbeltieren derKarrooformation. V. Uber Chasmatosaurus van hoepeni Haughton.Sitzungsberichte der Bayerischen Akademie der Wissenschaften,Mathematisch-naturwieenschaftliche Abteiliung 3:225–264.

Broom, R. 1902. On the leg and toe bones of Ptychosiagum. Transactionsof the South African Philosophical Society 11:233–235.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1252 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

Broom, R. 1903. On a new reptile (Proterosuchus fergusi) from the Kar-roo beds of Tarkastad, South Africa. Annals of the South AfricanMuseum 4:159–164.

Broom, R. 1905. Notice of some new fossil reptiles from the Karroo bedsof South Africa. Records of the Albany Museum 1:331–337.

Broom, R. 1913. On the South African pseudosuchian Euparkeria andallied genera. Proceedings of the Zoological Society of London83:619–633.

Broom, R. 1932. On some South African pseudosuchians. Annals of theNatal Museum 7:55–59.

Buffrenil, V. de, and J.-M. Mazin. 1989. Bone histology of Claudiosaurusgermaini (Reptilia, Claudiosauridae) and the problem of pachyosto-sis in aquatic tetrapods. Historial Biology 2:311–322.

Buffrenil, V. de, A. de Ricqles, C. E. Ray, and D. P. Domning.1990. Bone histology of the ribs of the archaeocetes (Mam-malia: Cetacea). Journal of Vertebrate Paleontology 10:455–466.

Camp, C. L. 1930. A study of the phytosaurs with description of new ma-terial from western North America. Memoirs of the University ofCalifornia 10:1–174.

Canoville, A., and M. Laurin. 2010. Evolution of humeral microanatomyand lifestyle in amniotes, and some comments on palaeobiologicalinferences. Biological Journal of the Linnean Society 100:384–406.

Cardinale, M., and J. Modin. 1999. Changes in size-at-maturity of Balticcod (Gadus morhua) during a period of large variations in stock sizeand environmental conditions. Fisheries Research 41:285–295.

Case, T. J. 1978. Speculations on the growth rate and reproduction ofsome dinosaurs. Paleobiology 4:320–328.

Castanet, J., and M. Baez. 1991. Adaptation and evolution in Gallotializards from the Canary Islands: age, growth, maturity and longevity.Amphibia-Reptilia 12:81–102.

Castanet, J., and M. Cheylan. 1979. Les marques de croissance des oset des ecailles comme indicateur de l’age chez Testudo hermanni etTestudo graeca (Reptilia, Chelonia, Testudinidae). Canadian Jour-nal of Zoology 57:1649–1665.

Castanet, J., and E. Smirina. 1990. Introduction to the skeletochronolog-ical method in amphibians and reptiles. Annales des Sciences Na-turelles, Zoologie 11:191–196.

Charig, A. J., B. Krebs, H.-D. Sues, and F. Westphal. 1976. Thecodontia;pp. 1–137 in O. Kuhn (ed.), Handbuch der Palaoherpetologie. Part13. Thecodontia. Gustav Fischer Verlag, Stuttgart.

Chinsamy, A. 1990. Physiological implications of the bone histologyof Syntarsus rhodesiensis (Saurischia: Theropoda). PalaeontologiaAfricana 27:77–82.

Chinsamy, A. 1991. The osteohistology of femoral growth within aclade: a comparison of the crocodile Crocodylus, the dinosaurs Mas-sospondylus and Syntarsus, and the birds Struthio and Sagittarius.Ph.D. dissertation, University of the Witwatersrand, Johannesburg,200 pp.

Chinsamy, A. 1995. Ontogenetic changes in the bone histology of the LateJurassic ornithopod Dryosaurus lettowvorbecki. Journal of Verte-brate Paleontology 15:96–104.

Chinsamy, A., and M. A. Raath. 1992. Preparation of fossil bone for his-tological examination. Palaeontologia Africana 29:3–44.

Cooper, L. N., A. H. Lee, M. L. Taper, and J. R. Horner. 2008. Relativegrowth rates of predator and prey dinosaurs reflect effects of preda-tion. Proceedings of the Royal Society B 275:2609–2615.

Cormack, D. 1987. Ham’s Histology. Lippincott, New York, 732 pp.Cruikshank, A. R. I. 1972. The proterosuchian thecodonts; pp. 89–119 in

K. A. Joysey and T. S. Kemp (eds.), Studies in Vertebrate Evolution.Oliver and Boyd, Edinburgh.

Cruikshank, A. R. I. 1978. The pes of Erythrosuchus africanus Broom.Zoological Journal of the Linnean Society 62:161–177.

Cubo, J., F. Pontin, M. Laurin, E. de Margerie, and J. Castanet. 2005.Phylogenetic signal in bone microstructure of sauropsids. System-atic Biology 54:562–574.

Currey, J. D., and R. M. Alexander. 1985. The thickness of the walls oftubular bones. Journal of Zoology 206:453–468.

Curtin, A. J., G. R. Zug, and J. R. Spotila. 2009. Longevity and growthstrategies of the desert tortoise (Gopherus agassizii) in two Ameri-can deserts. Journal of Arid Environments 73:463–471.

Dilkes, D. W. 1998. The Early Triassic rhynchosaur Mesosuchusbrowni and the interrelationships of basal archosauromorph rep-tiles. Philosophical Transactions of the Royal Society of London B353:501–541.

Dilkes, D. W., and H.-D. Sues. 2009. Redescription and phylogenetic re-lationships of Doswellia kaltenbachi (Diapsida: Archosauriformes)from the Upper Triassic of Virginia. Journal of Vertebrate Paleon-tology 29:58–79.

Enlow, D. H., and S. O. Brown. 1956. A comparative histological study offossil and recent bone tissues. Part I. The Texas Journal of Science8:405–443.

Enlow, D. H., and S. O. Brown. 1957. A comparative histological study offossil and recent bone tissues. Part II. The Texas Journal of Science9:136–214.

Ewer, R. F. 1965. The anatomy of the thecodont reptile Euparkeriacapensis Broom. Philosophical Transactions of the Royal Society ofLondon B 248:379–435.

Fish, F. E. 1993. Comparison of swimming kinematics between terres-trial and semiaquatic opossums. Journal of Mammalogy 74:275–284.

Francillon-Vieillot, H., V. de Buffrenil, J. Castanet, J. Geraudie, F.J. Meunier, J. Y. Sire, L. Zylberberg, and A. de Ricqles. 1990.Microstructure and mineralization of vertebrate skeletal tissues;pp. 471–548 in J. G. Carter (ed.), Skeletal Biomineralization: Pat-terns, Processes and Evolutionary Trends. Van Nestrand Reinhold,New York.

Gasser, M., M. Kaiser, D. Berrigan, and S. C. Stearns. 2000. Life-historycorrelates of evolution under high and low adult mortality. Evolu-tion 54:1260–1272.

Gauthier, J. A. 1984. A cladistic analysis of the higher systematic cat-egories of Diapsida. Ph.D. dissertation, University of California,Berkeley, California, 564 pp.

Gauthier, J. 1986. Saurischian monophyly and the origin of birds; pp. 1–58in K. Padian (ed.), The Origin of Birds and the Evolution of Flight.Memoirs of the California Academy of Sciences 11.

Germain, D., and M. Laurin. 2005. Microanatomy of the radius andlifestyle in amniotes (Vertebrata, Tetrapoda). Zoologica Scripta34:335–350.

Girondot, M., and M. Laurin. 2003. Bone Profiler: a tool to quantify,model, and statistically compare bone-section compactness profiles.Journal of Vertebrate Paleontology 23:458–461.

Gow, C. E. 1975. The morphology and relationships of Youngina capensisBroom and Prolacerta broomi Parrington. Palaeontologia Africana18:89–131.

Gower, D. J. 1996. The tarsus of erythrosuchid archosaurs, and implica-tions for early diapsid phylogeny. Zoological Journal of the LinneanSociety 116:347–375.

Gower, D. J. 2003. Osteology of the early archosaurian reptile Ery-throsuchus africanus Broom. Annals of the South African Museum110:1–88.

Gower, D. J., and A. G. Sennikov. 1996. Morphology and phyloge-netic informativeness of early archosaur braincases. Palaeontology39:883–906.

Gower, D. J., and A. G. Sennikov. 1997. Sarmatosuchus and the earlyhistory of the Archosauria. Journal of Vertebrate Paleontology17:60–73.

Gower, D. J., and A. G. Sennikov. 2000. Early archosaurs from Russia;pp. 140–159 in M. J. Benton, E. N. Kurochkin, M. A. Shishkin, andD. M. Unwin (eds.), The Age of Dinosaurs in Russia and Mongolia.Cambridge University Press, London.

Gower, D. J., and M. Wilkinson. 1996. Is there any consensus onbasal archosaur phylogeny? Proceedings of the Royal Society B263:1399–1406.

Gross, W. 1934. Die Typen des mikroskopischen Knochenbaues beifoslen Stegocephalen und Reptilien. Zeitschrift fur Anatomie103:731–764.

Hancox, P. J., M. A. Shishkin, B. S. Rubidge, and J. W. Kitching. 1995. Athreefold subdivision of the Cynognathus Assemblage Zone (Beau-fort Group, South Africa) and its palaeogeographical implications.South African Journal of Science 91:143–144.

Horner, J. R., A. de Ricqles, and K. Padian. 1999. Variation in dinosaurskeletochronology indicators: implications for age assessment andphysiology. Paleobiology 25:295–304.

Hua, S., and V. de Buffrenil. 1996. Bone histology as a clue in the inter-pretation of functional adaptations in the Thalattosuchia (Reptilia,Crocodylia). Journal of Vertebrate Paleontology 16:703–717.

Huene, F. von. 1911. Uber Erythrosuchus, Vertreter der neuen Reptil-Ordnung Pelycosimia. Geologische und Palaontologische Abhand-lungen, N. F. 10:67–122.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

BOTHA-BRINK AND SMITH—ARCHOSAUROMORPH OSTEOHISTOLOGY 1253

Hughes, B. 1963. The earliest archosaurian reptiles. South African Jour-nal of Science 59:221–241.

Huttenlocker, A., and E. Rega. In press. The paleobiology and bone mi-crostructure of pelycosaurian-grade synapsids; in A. Chinsamy (ed.),The Forerunners of Mammals: Radiation, Histology and Biology.Indiana University Press, Bloomington and Indianapolis, Indiana.

Huttenlocker, A., J. Botha-Brink, and C. Sidor. 2010. Bone histol-ogy and growth in the Permo-Triassic therocephalian Moschorhi-nus kitchingi (Synapsida: Therapsida) from the Karoo Basinof South Africa. Journal of Vertebrate Paleontology 30(3,Supplement):108A–109A.

Hutton, J. M. 1986. Age determination of living Nile crocodiles from thecortical stratification of bone. Copeia 2:332–341.

Juul, L. 1994. The phylogeny of basal archosaurs. PalaeontologiaAfricana 31:1–38.

King, G. M. 1991. The aquatic Lystrosaurus: a palaeontological myth.Historical Biology 4:285–322.

King, G. M., and M. A. Cluver. 1991. The aquatic Lystrosaurus: an alter-native lifestyle. Historical Biology 4:323–342.

Klevezal, G. A. 1996. Recording Structures of Mammals. Determinationof Age and Reconstruction of Life History. A. A. Balkema, Rotter-dam, 274 pp.

Kriloff, A., D. Germain, A. Canoville, P. Vincent, M. Sache, and M. Lau-rin. 2008. Evolution of bone microanatomy of the tetrapod tibia andits use in palaeobiological inference. Journal of Evolutionary Biol-ogy 21:807–826.

Laurin, M., M. Girondot, and M.-M. Loth. 2004. The evolution of longbone microanatomy and lifestyle in lissamphibians. Paleobiology30:589–613.

Lee, A. H., and S. Werning. 2008. Sexual maturity in growing dinosaursdoes not fit reptilian growth models. Proceedings of the NationalAcademy of Sciences of the United States of America 105:582–587.

Maisano, J. A. 2002. Terminal fusions of skeletal elements as indica-tors of maturity in squamates. Journal of Vertebrate Paleontology22:268–275.

Margerie, E. de, J. Cubo, and J. Castanet. 2002. Bone typology andgrowth rate: testing and quantifying ‘Amprino’s rule’ in the mallard(Anas platyrhynchos). Les Comptes Rendus de l’Academie des sci-ences, Serie Biologies 325:221–230.

Mazin, J.-M. 2001. Mesozoic marine reptiles: an overview; pp. 95–117 inJ.-M. Mazin and V. de Buffrenil (eds.), Secondary Adaptation ofTetrapods to Life in Water. Verlag Dr. Friedrich Pfeil, Munich.

Modesto, S. P. 2006. The cranial skeleton of the Early Permianaquatic reptile Mesosaurus tenuidens: implications for relation-ships and palaeobiology. Zoological Journal of the Linnean Society146:345–368.

Modesto, S. P., and H.-D. Sues. 2004. The skull of the Early Triassic ar-chosauromorph reptile Prolacerta broomi and its phylogenetic sig-nificance. Zoological Journal of the Linnean Society 140:335–351.

Montes, L., J. Castanet, and J. Cubo. 2010. Relationship between bonegrowth rate and bone tissue organization in amniotes: first testof Amprino’s rule in a phylogenetic context. Animal Biology60:25–41.

Nesbitt, S. J., M. R. Stocker, B. J. Small, and A. Downs. 2009. The oste-ology and relationships of Vancleavea campi (Reptilia: Archosauri-formes). Zoological Journal of the Linnean Society 157:814–864.

Padian, K., J. R. Horner, and A. de Ricqles. 2004. Growth in small di-nosaurs and pterosaurs: the evolution of archosaurian growth strate-gies. Journal of Vertebrate Paleontology 24:555–571.

Parrington, F. R. 1935. On Prolacerta broomi, gen. et sp. n. and the originof lizards. Annals and Magazine of Natural History 16:197–205.

Parrish, J. M. 1992. Phylogeny of the Erythrosuchidae (Reptilia: Ar-chosauriformes). Journal of Vertebrate Paleontology 12:93–102.

Parrish, J. M. 1993. Phylogeny of the Crocodylotarsi with reference to ar-chosaurian and crurotarsan monophyly. Journal of Vertebrate Pale-ontology 13:287–308.

Payne, J. L., D. L. Lehrmann, J. Wei, M. J. Orchard, D. P. Schrag, andA. H. Knoll. 2004. Large perturbations of the carbon cycle duringrecovery from the end-Permian extinction. Science 305:506–509.

Ray, S., A. Chinsamy, and S. Bandyopadhyay. 2005. Lystrosaurus murrayi(Therapsida, Dicynodontia): bone histology, growth and lifestyleadaptations. Palaeontology 48:1169–1185.

Redelstorff, R., and P. M. Sander. 2009. Long and girdle bone histologyof Stegosaurus: implications for growth and life history. Journal ofVertebrate Paleontology 29:1087–1099.

Reid, R. E. H. 1996. Bone histology of the Cleveland-Lloyd dinosaursand of dinosaurs in general, Part I: Introduction: introduction tobone tissues. Brigham Young University Geology Studies 41:25–71.

Reig, O. A. 1970. The Proterosuchia and the early evolution of the Ar-chosaurs; an essay about the origin of a major taxon. Bulletin of theMuseum of Comparative Zoology 139:229–292.

Ricqles, A. de. 1976. On bone histology of fossil and living reptiles,with comments on its functional and evolutionary significance; pp.123–150 in A. d’A. Bellairs and C. B. Cox (eds.), Morphology andBiology of Reptiles. Academic Press, London.

Ricqles, A. de, F. J. Meunier, J. Castanet, and H. Francillon-Vieillot.1991. Comparative microstructure of bone; pp. 1–77 in B. K. Hall(ed.), Bone Matrix and Bone Specific Products. CRC Press, BocaRaton, Florida.

Ricqles, A. de, K. Padian, F. Knoll, and J. R. Horner. 2008. On the originof high growth rates in archosaurs and their ancient relatives: com-plementary histological studies on Triassic archosauriforms and theproblems of a “phylogenetic signal” in bone histology. Annales dePaleontologie 94:57–76.

Robinson, P. L. 1967. The evolution of the Lacertilia. ProblemesActuels de Paleontologie (Evolution des Vertebres), ColloquesInternationaux du Centre. National de la Recherche Scientifique163:395–407.

Robinson, P. L. 1973. A problematic reptile from the British Upper Trias.Journal of the Geological Society of London 129:457–479.

Roopnarine, P. D., K. D. Angielczyk, S. C. Wang, and R. Hertog.2007. Food web models explain instability of Early Triassic terres-trial communities. Proceedings of the Royal Society B 274:2077–2086.

Sahney, S., and M. J. Benton. 2007. Recovery from the most profoundmass extinction of all time. Proceedings of the Royal Society B275:759–765.

Sander, M. A. 1990. Skeletochronology in the small Triassic reptileNeusticosaurus. Annales des Sciences Naturelles Zoologie Paris11:213–217.

Sander, M. A. 2000. Longbone histology of the Tendaguru sauropods:implications for growth and biology. Paleobiology 26:466–488.

Sander, M. A., and C. Tuckmantel. 2003. Bone lamina thickness, boneapposition rates, and age estimates in sauropod humeri and femoraPalaontologische Zeitschrift 76:161–172.

Scheyer, T. M., and P. M. Sander. 2009. Bone microstructures andmode of skeletogenesis in osteoderms of three pareiasaur taxa fromthe Permian of South Africa. Journal of Evolutionary Biology22:1153–1162.

Scheyer, T. M., N. Klein, and P. M. Sander. 2010. Developmentalpalaeontology of Reptilia as revealed by histological studies. Semi-nars in Cell and Developmental Biology 21:462–470.

Seitz, A. L. 1907. Vergleichende Studien uber den mikroskopischenKnochenbau Fossiler und rezenter Reptilien. Nova Acta Leopold-ina Abhandlungen der Deutschen Akademie der Naturforscher37:230–370.

Sereno, P. C. 1991. Basal archosaurs: phylogenetic relationships and func-tional implications. Society of Vertebrate Paleontology Memoir2:1–53.

Sereno, P. C., and A. B. Arcucci. 1990. The monophyly of crurotarsal ar-chosaurs and the origin of bird and crocodile ankle joints. NeuesJahrbuch fur Geologie und Palaontologie, Abhandlung 180:21–52.

Smith, R. M. H. 1995. Changing fluvial environments across the Permian-Triassic boundary in the Karoo Basin, South Africa and possiblecauses of tetrapod extinctions. Palaeogeography, Palaeoclimatol-ogy, Palaeoecology 117:81–104.

Smith, R., and J. Botha. 2005. The recovery of terrestrial vertebrate di-versity in the South African Karoo Basin after the End-Permian ex-tinction. Comptes Rendus Palevol 4:555–568.

Smith, R. M. H., and P. D. Ward. 2001. Pattern of vertebrate extinctionsacross an event bed at the Permian-Triassic boundary in the KarooBasin of South Africa. Geology 29:1147–1150.

Starck, J. M., and A. Chinsamy. 2002. Bone microstructure and develop-mental plasticity in birds and other dinosaurs. Journal of Morphol-ogy 254:232–246.

Stein, B. R. 1989. Bone density and adaptation in semi-aquatic mammals.Journal of Mammalogy 70:467–476.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011

1254 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

Sues, H.-D. 2003. An unusual new archosauromorph reptile from the Up-per Triassic Wolfville Formation of Nova Scotia. Canadian Journalof Earth Sciences 40:635–649.

Tatarinov, L. P. 1961. [The pseudosuchian material of the USSR]. Pale-ontologicheskii Zhurnal 1961:117–132. [Russian]

Tinkle, D. W. 1969. The concept of reproductive effort and its relation tothe evolution of life histories of lizards. The American Naturalist103:501–516.

Walker, J. D., and J. W. Geissman. 2009. Geologic Time Scale. GeologicalSociety of America, Boulder, Colorado.

Wall, W. P. 1983. The correlation between limb-bone density and aquatichabits in recent mammals. Journal of Vertebrate Paleontology57:197–207.

Watson, D. M. S. 1912. The skeleton of Lystrosaurus. Records of the Al-bany Museum 2:287–299.

Watson, D. M. S. 1913. The limbs of Lystrosaurus. Geological Magazine10:256–258.

Welman, J. 1998. The taxonomy of the South African proterosuchids(Reptilia, Archosauromorpha). Journal of Vertebrate Paleontology18:340–347.

Werning, S., and R. Irmis. 2010. Reconstructing the ontogeny of the Tri-assic basal archosauromorph Trilophosaurus using bone histologyand limb bone morphometrics. Journal of Vertebrate Paleontology30(3, Supplement):185A–186A.

Submitted December 3, 2010; revisions received August 16, 2011;accepted August 22, 2011.Handling editor: Patrick O’Connor.

Dow

nloa

ded

by [

Jenn

ifer

Bot

ha-B

rink

] at

01:

06 0

9 N

ovem

ber

2011