JOURNAL OF MORPHOLOGY 271:729–737 (2010)

The Histological Structure of Glyptosaurine Osteoderms (Squamata: Anguidae), and the Problem of Osteoderm Development in Squamates Vivian de Buffre´nil,1* Jean-Yves Sire2, and Jean-Claude Rage1 1 2

De´partement Histoire de la Terre, Muse´um National d’Histoire Naturelle, UMR 7207 CNRS, Paris, France De´partment de Biologie, Universite´ Pierre et Marie Curie, UMR 7138 CNRS, Paris Cedex 05, France

ABSTRACT Glyptosaurinae, a fossil clade of anguid lizards, possess robust osteoderms, with granular ornamentation. In this study, the structural and histological features of these osteoderms were described in order to reconstruct their developmental pattern and further document the degree of homology that could exist between vertebrate integumentary skeletons. Glyptosaurine osteoderms have a diploe architecture and display an unusually complex structure that includes four tissue types: a core of woven-fibered bone intensely remodeled; a peripheral formation of the same tissue containing dense bundles of long Sharpey fibers; a thick basal layer of lamellar bone; and a superficial layer of a non-osseous material that belongs to the category of hypermineralized tissues such as ganoine, or enameloid and enamel tissues. The growth pattern of glyptosaurine osteoderms involved appositional processes due to osteoblast activity. In early growth stages, osseous metaplasia might have also been involved, but this possibility is not substantiated by histological observations. The superficial layer of the osteoderms must have resulted from epidermal contribution, a conclusion that would support previous hypotheses on the role of epidermal-dermal interactions in the formation of squamate osteoderms. J. Morphol. 271:729–737, 2010. Ó 2010 Wiley-Liss, Inc. KEY WORDS: integumantary skeleton; bone remodeling; metaplasia; bone tissue; hypermineralized tissues

INTRODUCTION The presence of dermal osseous plates, or osteoderms, is common in extant and extinct tetrapods [review in (Vickaryous and Sire, 2009); see also (Moss, 1969; Ruibal and Shoemaker, 1984)]. The taxonomic distribution of these structures is highly variable [cf. (Romer, 1997)]. Some distantly related taxa demonstrate osteoderms that are quite comparable in morphology and location (e.g., chroniosuchian anthracosaurs and crocodilians). Conversely, closely related taxa, such as various species of the genus Varanus, either have (Varanus (Megalania) priscus, Varanus komodoensis), or are deprived (most other Varanus species) of osteoderms (Auffenberg, 1981; Erickson et al., 2003). Available data for basal sarcopterygians suggest that the presence of osteoderms is plesiomorphic in tetrapods (Vickaryous and Sire, 2009). However, Ó 2010 WILEY-LISS, INC.

the analysis of osteoderm distribution in tetrapods indicates that they have been independently lost, and have then reappeared with diverse specialized designs, in various taxa [at least five times in Amniotes (Hill, 2005)]. Conversely, the ability of dermal tissues to undergo local mineralization and produce osseous integumental elements might represent a ‘‘deep homology’’ that would root in the most basal sarcopterygians (Main et al., 2005; Vickaryous and Hall, 2008). Moreover, the structural organization of osteoderms, as well as the nature of the tissues they are made of, displays a great diversity, even between taxa considered as close relatives (Ruibal and Shoemaker, 1984; Hill, 2006). This variability raises interesting questions about the value of the histological structure of osteoderms as a phylogenetically meaningful character (Hill, 2005). The formation of osteoderms is generally supposed to result, at least among extent diapsids (lepidosaurians and crocodilians), from a prominent mode of skeletogenesis: osseous metaplasia (Moss, 1969; Vickaryous and Sire, 2009). This process would consist in the mineralization of the dense and loose dermal strata, and their direct transformation into bone tissue in the absence of osteoblasts (Haines and Mohuiddin, 1968). Although a possible contribution of osteoblasts has been considered in some lepidosaurian taxa, it remains a controversial issue (Zylberberg and Castanet, 1985; Levrat-Calviac, 1986; Vickaryous and Hall, 2008; Vickaryous and Sire, 2009). Osteoderms are common in numerous lizard groups, especially Anguidae (Hoffstetter, 1962).

*Correspondence to: Vivian de Buffre´nil, Muse´um National d’Histoire Naturelle, De´partement Histoire de la Terre, UMR 7207 (CR 2P). Ba˜timent de Ge´ologie, CC 48, 57 rue Cuvier F-75005 Paris, France. E-mail: [email protected] Received 3 July 2009; Revised 19 October 2009; Accepted 21 October 2009 Published online 25 January 2010 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jmor.10829

730

V.

DE

´ NIL ET AL. BUFFRE

Within this group, the fossil Glyptosaurinae [Palaeocene to Oligocene in North America, Eocene only in Eurasia (Estes, 1983)] is distinctive for the presence of robust integumentary skeleton. The current study documents the histology of Glyptosaurinae osteoderms, and provides comments on the development and homology of the integumentary skeleton. It is aimed at documenting how glyptosaurine osteoderms formed and grew, and assessing the degree of homology that could exist between vertebrate integumentary skeletons. MATERIAL AND METHODS The fossil material comprises 26 well-preserved osteoderms (several are broken or incomplete) attributed to Glyptosaurinae. Since their discovery, the Glyptosaurinae have always been included in the Anguidae (Conrad and Norell, 2008), although Hill (2005) proposed to consider them as the sister group of a clade including Scincoidea (Scincidae, Cordylidae, Gerrhosauridae) and Anguidae. The specimens used for this study were collected in the 1970s and 1980s in two sites of the Phosphorites du Quercy (Bonis et al., 1973; Crochet et al., 1981; Rage, 2006). The Phosphorites du Quercy consist of more than 100 fossiliferous sites located in a restricted area in south-western France. They range from the early Eocene to the early Miocene (Rage, 2006). The studied osteoderms come from two of these localities. Sixteen of them are from the Bartonian (standard level MP 16; middle Eocene) of Lavergne site, and are registered under the references LAV 1258 to LAV 1273 in the paleontological collections of Montpellier 2 University. The other 10 osteoderms are from the Priabonian (standard level MP 18; late Eocene) of Sainte Ne´boule site. They are recorded in the collections of the Muse´um National d’Histoire Naturelle (Paris, France) under the references SNB 1022–SNB 1031. Morphologically, these osteoderms (see Fig. 1) are characterized by a large size (up to 8.2 3 5 mm), an important thickness (up to 2 mm, and even more), a hexagonal or rectangular shape, a variably pronounced dorsal keel, and a typical granular ornamentation (Hoffstetter, 1962). This osteoderm morphology is distinctive of the Glyptosaurinae [the osteoderms of another anguid taxon, the Gerrhonotinae, are morphologically close but they are relatively thinner; cf. (Auge´, 2005)]. The generic identification of our osteoderms is problematic. According to Sullivan and Auge´ (2006), osteoderms do not allow identification within Glyptosaurinae. Because all the osteoderms used in this study were not found in association with taxonomically significant remains, they are consequently considered belonging to the Glyptosaurinae, with no further specification. The two sets are nevertheless distinguished by their geographic origin: Lavergne or Sainte Ne´boule. Ten of the sixteen osteoderms from Lavergne, and seven of those from Sainte Ne´boule, have a roughly rectangular shape, with an area up to 37 mm2, and a mean thickness 0.9 to 1.25 mm (Fig. 1A–D). The other osteoderms (six from Lavergne, three from Sainte Ne´boule) are irregular hexagons (Fig. 1E–G) measuring 10–23 mm2 for a mean thickness 1.97–2.4 mm. According to previous descriptions (Hoffstetter, 1962; Auge´, 2005; Sullivan and Auge´, 2006) the hexagonal osteoderms are from the cephalic region (they were not fused to skull bones), and the rectangular ones are postcranial (mainly dorsal). After photography and measurement of the 26 osteoderms, twelve of them, forming a sub-sample in which both sites and the two morphologies quoted above were represented, were embedded in a polyester resin for making 80-lm thick (610 lm) ground sections. The latter were observed using a compound microscope in natural or polarized transmitted light. All measurements of osteoderm sections were from line drawings made with a camera lucida (403 to 503 magnifications). Only intact

Journal of Morphology

osteoderms were used for this purpose. Overall osteoderm compactness, or parameter C, represents the area actually occupied by biological mineralized tissues, expressed as a percentage of total sectional area. This parameter was measured with the software ImageJ (Abramoff et al., 2004) on digitalized images of the line drawings. Cavities less than 25 lm in diameter were not considered. Comparisons between the two sub-samples (Lavergne and Sainte Ne´boule) were only qualitative because the number of complete, intact osteoderms was too small to perform statistical tests. For the purpose of orientation, the side of the osteoderms bearing the granular ornamentation will be designated by the term superficial. Conversely, the other side (with non-ornamented surface) will be named deep, or basal (see Fig. 1).

RESULTS Morphological Features of the Osteoderms The superficial ornamentation of all osteoderms consists of numerous tubercles that often fuse with each other at the top of the osteoderm keel (Fig. 1A,C,E,H,I). These tubercles create a rough surface, typical of the granular ornamentation. In a single osteoderm the diameter at the base of the tubercles may vary from 0.3 to 0.5 mm. The tubercles display several differences according to the fossiliferous site they come from. In Lavergne specimens (Fig. 1A,H), they often have a sharp (but frequently broken) apex, whereas they are always blunt in Sainte Ne´boule osteoderms (Fig. 1C). Moreover, the spatial density of the tubercles in the latter is more elevated: ca 4.9/mm2 versus 3.4 for Lavergne specimens. The tubercles have a whitish, vitreous coloration that clearly contrasts with the brown color acquired by the rest of the osteoderms during fossilization (Fig. 1A,H). The anterior, ‘‘gliding surface’’ of each rectangular osteoderm (Fig. 1A,C) is smooth [this part is covered by the preceding osteoderm; (Hoffstetter, 1962)]. The superficial side of the osteoderms also displays vascular pits, located between the tubercles (Fig. 1A,C,E). The deep surfaces of all osteoderms from Lavergne (Fig. 1B,F) are flat and smooth, whereas they are folded in some specimens from Sainte Ne´boule (Fig. 1G). In all osteoderms, the deep surface bears 6–8 vascular pits the diameter of which ranges from 120 to 240 lm in the specimens from Sainte Ne´boule, and from 60 to 90 lm in those from Lavergne (Fig. 1B,D,F,G). The peripheral (equatorial) margin of the cephalic osteoderms, and the lateral sides of postcranial ones, display deep folds and indentations indicative of suture surfaces (Fig. 1B,D,G,I). Vascular pits also perforate these surfaces. Microanatomical Organization of the Osteoderms At a microanatomical level, there is no noticeable difference between Lavergne and Sainte Ne´boule specimens; therefore, the following descriptions apply to both locations. All cephalic

STRUCTURE OF GLYPTOSAURINE OSTEODERMS

731

Fig. 1. General morphology of glyptosaurine osteoderms from two sites of the Phosphorites du Quercy. The arrows point to some vascular pits. A: View of the superficial side of a rectangular, postcranial osteoderm from Lavergne. B: View of the deep side of the same specimen as in (A). C: View of the superficial side of a postcranial osteoderm from Sainte Ne´boule. D: View of the deep side of the same specimen as in (C). E: View of the superficial side of a cephalic osteoderm from Lavergne. F: View of the deep side of the same specimen as in (E). G: Folded, deep surface of an osteoderm from Sainte Ne´boule. H: Close view of the tubercles of a specimen from Lavergne. Remark the sharp tip of most tubercles. I: Sutural surface of a cephalic osteoderm from Lavergne. Scale bars 5 2 mm (A–D); 1 mm (E–G, I); 0.5 mm (H). ant, anterior side; sup, superficial side; sut, sutural surface; deep, deep, or basal, side.

osteoderms share a basic diploe architecture, with a central cancellous core bordered by two compact cortices (Figs. 2A, 3A). However, the relative thickness of the cortices, and the relative area of the central cancellous core, vary among osteoderms. Postcranial osteoderms are more compact (C 5 93–97% versus 80– 88% in cephalic osteoderms) and display only small rounded cavities (Fig. 2B). In all cases, the cortices house vascular canals that are generally more numerous in cranial than in postcranial osteoderms. Vascular canals consist in primary or secondary osteons that communicate with the large inner cavities and open on the deep and superficial surfaces of the osteoderms by the pits mentioned above. Primary, simple vascular canals occur in some specimens, but are generally sparse.

Histological Structure Lavergne and Sainte Ne´boule osteoderms are histologically comparable, though the former are more degraded by diagenetic processes than the latter. Cephalic osteoderms (the thickest ones) display four, clearly distinct types of osseous tissue (Figs. 3 and 4) that are also found, but less clearly developed, in postcranial osteoderms. The core of each osteoderm is composed of woven-fibered bone (Figs. 3B,C and 4) that forms a laterally extensive, but relatively thin, layer: its thickness is less than one third of total osteoderm thickness. This tissue is typically defined by i) its monorefringence in polarized light and ii) a high density of spherical osteocyte lacunae, randomly distributed within the bone matrix (Fig. 3B). VasJournal of Morphology

732

V.

DE

´ NIL ET AL. BUFFRE

Fig. 2. Inner architecture of glyptosaurine osteoderms. A: Section through a thick, cephalic osteoderm. Superficial side above. Vascular cavities in black. B: Section along the small axis of a postcranial osteoderm. Postcranial osteoderms are more compact and less vascularized than cephalic ones. Scale bar 5 2 mm. The dotted line shows the limit between the superficial, ornamented layer and the underlying bone.

cular canals are mainly located in the core of the osteoderm, a region that was submitted to intense remodeling (see below: Bone remodeling). The basal region of the osteoderms (Fig. 3A,D) is composed of a lamellar tissue clearly characterized, in polarized light, by the alternation of light and dark thin lamellae (from one lamella to the following one, collagen fibers are perpendicularly oriented). This tissue forms a thick layer, especially in cephalic osteoderms, in which it can be 1 mm thick and can represent half the total thickness of the osteoderm (Fig. 3A). In transmitted light, this tissue displays typical, flat osteocyte lacunae that are all oriented parallel to the outer surface of the cortex (Fig. 3E). It also contains numerous Sharpey’s fibers that create two systems with different locations and orientations: 1) a system of short (length 30–40 lm), robust (diameter up to 2.5 lm) and vertically-oriented fibers located deep into the lamellar cortex (Fig. 3D); 2), long bundles of Sharpey’s fibers oriented obliquely and located at the extremities of the lamellar bone layer. Periodic growth lines occur throughout the basal lamellar layer (Fig. 3E). The basal layer laterally merges with a third type of bone tissue, dominated by Sharpey’s fibers. These fibers are long, parallel to each other, but their direction makes an angle of some 30–408 with the fibers of the basal region (Fig. 3E). The spatial density of these fibers is so high that they tend to blur the type of bone tissue in which they are inserted. This type of bone tissue is intermediate between the woven-fibered bone, with rounded osteocyte lacunae oriented at random, and the parallel-fibered bone with irregular birefringence in polarized light. It is located at the lateral margins of the osteoderms, just under the sutural surfaces, where it forms an equatorial ring. Toward the center of the osteoderm, this tissue comes in contact with the core of woven-fibered bone, and is also subjected to extensive remodeling (Fig. 3A,E). Journal of Morphology

The superficial layer forming the granular ornamentation is made of a distinct tissue type that can be up to 400 lm thick at the tubercle apices. It is continuous in Sainte Ne´boule osteoderms (Figs. 1C,4A), while it is interrupted and limited to the tubercles in Lavergne specimens (Fig. 4B,E,F). Frequently, especially in Lavergne osteoderms, two tubercles can be superimposed (Fig. 4B–G) in the thickness of this layer. The histological characteristics of this layer completely differ from those of the underlying bone tissues (Figs. 3B and 4). It is indeed composed of a vitreous, translucent, avascular and acellular tissue. The latter is monorefringent and contains no differentiated structures, with exception for poorly-defined periodical growth marks that can be seen only in polarized light. The limit between this tissue and the underlying woven-fibered bone is always sharp, and sometimes underlined by a reversion line that indicates that the woven-fibered bone has been submitted to partial resorption before the deposition of the superficial layer (Fig. 3C). It is noteworthy that the superficial layer has the same histological aspect in all osteoderms studied, and that it displays no sign of diagenetic alteration, even when the underlying bone tissue is heavily degraded (Fig. 4B,E,F). Remodeling All osteoderms examined histologically, showed pattern of intense perivascular remodeling, i.e., the formation of secondary osteons and endosteal deposits of lamellar bone on the walls of broad resorption bays created by osteoclast activity. Extensive remodeling is characteristic of the woven-fibered bone of the core, the adjacent parts of the basal lamellar layer (Fig. 3D), and the fiberrich tissue edging sutural contacts (Fig. 3E). The superficial-most acellular layer lacks Haversian systems but there is evidence of superficial erosion of the tubercles in some osteoderms. This erosion was either followed by the formation of a new tubercle or by a deposition of woven-fibered bone that was generally followed by the formation of a new tubercle (Fig. 4C–G). Some poorly developed tubercles are covered by woven-fibered bone and do not protrude at the osteoderm surface (Fig. 4F,G). DISCUSSION Morphological Differences Between the Two Sets of Osteoderms Osteoderms from Lavergne and Sainte Ne´boule apparently differ by the shape and spatial density of their tubercles; however, our sample is too small for statistical tests. Such differences can be due to various causes like topographic location on body, individual age, and developmental stage, as well as sex that are undocumented parameters in our sample. These differences could also have a taxo-

STRUCTURE OF GLYPTOSAURINE OSTEODERMS

733

Fig. 3. Structure of glyptosaurine osteoderms (ground sections). A: General view (polarized light) of a cephalic osteoderm in cross section. The overall architecture is that of a diploe, with large resorption bays. B: Superficial region of a cephalic osteoderm (transverse section; transmitted light). The superficial layer is made of an avascular, acellular tissue laying over a well vascularized formation of woven-fibered bone. Most vascular canals are primary osteons. C: Superficial region of a cephalic osteoderm (transverse section; polarized light). The tubercles of the superficial layer are monorefringent, as is also the narrow subjacent layer of woven-fibered bone that was not subjected to remodeling. There is a clear-cut limit between the superficial layer and the underlying woven-fibered bone. The rest of the woven-fibered bone region was extensively remodeled and replaced by secondary lamellar or parallel-fibered bone (asterisk). The arrow points to a resorption bay. D: Lamellar bone composing the basal region of the osteoderms (polarized light). Toward the central region this layer is extensively resorbed and reconstructed through deposition of secondary endosteal bone. The enlarged region on the left (transmitted light) shows numerous short, anchorage fibers that are located within this layer. E: Structure of the bone underlying the sutural surfaces (transmitted light). This layer shares characteristics with woven-fibered and parallel-fibered bone tissue types. It encloses numerous, extended bundles of Sharpey’s fibers that are also present (but with different orientations) in the lateral regions of the basal layer of lamellar bone. The dotted line shows the limit between the two bone layers. Periodic growth marks are located in the basal layer but vanish in the Sharpey-fibered bone underlying the sutures. The deep regions of this bone are submitted to resorption. Scale bars 5 500 lm (A); 200 lm (B, D, E); 300 lm (C). deep, deep side of the osteoderm; gm, periodic growth mark; lam. b, lamellar bone; sup, superficial side of the osteoderm; rec, reconstructed bone; rem. wfb, remodeled woven-fibered bone; res, resorption process.

Journal of Morphology

734

V.

DE

´ NIL ET AL. BUFFRE

Fig. 4. Superficial layer of the osteoderms. A: Cephalic osteoderm (transverse section in transmitted light). General view of the superficial layer and the underlying, vascularized woven-fibered bone. Vascular canals (arrows) open between the tubercles. B: Postcranial osteoderm (transverse section in transmitted light). Several superimposed tubercles occur in the superficial layer. The bone tissue was damaged by diagenesis, whereas the tissue composing the tubercles was not. C: Close view at two successive generations (1, 2) of tubercles occurring at the same location in a cephalic osteoderm (transmitted light). The top of tubercle 1 displays signs of an erosion process that occurred before the formation of tubercle 2. D: Same section as in (C), polarized light. The tubercles and the subjacent woven-fibered bone are monorefringent. E: Two superimposed generations of tubercles (1, 2; transmitted light). F: Two superimposed generations of tubercles (1, 2), and a tubercle that was re-covered by bone after its formation (arrow; transmitted light). During fossilization, the bone tissue became degraded by diagenesis, whereas the tissue composing the tubercles remained intact. G: Same section as in (F), polarized light. Scale bars 5 300 lm (A,B); 150 lm (C–G). rec. b, reconstructive, endosteal bone deposit; res, resorption process; vc, vascular canals; wfb, woven-fibered bone.

nomic significance. This possibility was not considered by Sullivan and Auge´ (2006), but no study dealing with the characteristics of osteoderm ornamentation in Glyptosaurinae has ever been published. The glyptosaurine fauna of the Phosphorites du Quercy includes at least two genera (Auge´, 2005; Auge´ and Sullivan, 2006; Sullivan and Auge´, 2006). One of them, Paraplacosauriops, is known to occur in two sites, Lavergne and Perrie`re. Therefore, the osteoderms from Lavergne could possibly belong to this genus. Those from Sainte Journal of Morphology

Ne´boule would then represent the other genus, Placosaurus, but this remains hypothetical. Structural Complexity of the Osteoderms The main result of this study is to show that glyptosaurine osteoderms have a more complex structure than those of Anguis fragilis, the only anguid species precisely studied in this respect (Zylberberg and Castanet, 1985; Levrat-Calviac, 1986). The osteoderms of A. fragilis have no vascu-

STRUCTURE OF GLYPTOSAURINE OSTEODERMS

lar canals and are exclusively made of bone tissue. The latter displays two types, histologically and topographically distinct: a basal layer of lamellar bone, and a superficial (dorsal) layer of slightly hypermineralized woven-fibered bone. Considering the structural differences that exist between glyptosaurine osteoderms and those of A. fragilis, the assessment of homologies between the tissues they are made of may look uncertain. Our interpretation is as follows: 1. The layer of woven-fibered bone in the core of glyptosaurine osteoderms would correspond to the layer made of the same tissue in A. fragilis (superficial region). The correspondence would thus concern both the tissue type and its position, i.e., dorsal to the basal lamellar layer. The same correspondence could also apply to the osteoderms of other squamates such as Eumeces inexpectatus or Heloderma sp. (Moss, 1969). 2. The basal layer of lamellar bone in glyptosaurine osteoderms obviously parallels that described in A. fragilis, although this layer is markedly less developed in A. fragilis. 3. The tissue adjacent to sutural contacts that displays a high density of Sharpey’s fibers has not been mentioned by Zylberberg and Castanet (1985) in A. fragilis, the osteoderms of which lack sutures. Conversely, in the peripheral region of the osteoderm of a scincid, Corucia zebrata, Moss (1969, p. 515) observed and illustrated a ‘‘sclerification . . . intermediate between calcified tendon and bone’’ that could be comparable to that tissue in aspect and position. The same tissue would also occur in Tarentola mauritanica, a gekkonid species, in which Levrat-Calviac (1986) and Levrat-Calviac and Zylberberg (1986) described long, parallel and densely-packed Sharpey’s fibers that deeply enter the peripheral (equatorial) regions of the osteoderms and make a strong bond between neighboring elements. In the armadillo (Dasypus novemcinctus, a xenarthran mammal), the bone occupying the peripheral regions of the osteoderms, just under the sutures, looks similar to this tissue because of its high fibrillar content. This bone is called ‘‘Sharpey-fibered bone’’ by Vickaryous and Hall (2006); see also Vickarious and Sire (2009). 4. The vitreous, acellular tissue composing the superficial layer of glyptosaurine osteoderms neither exists in A. fragilis nor in Ophisaurus s. l. The nature of this tissue cannot be settled with certainty, as there are no histochemical, developmental or ultrastructural data available for a similar tissue in a living squamate species. Histologically, this tissue displays none of the typical characteristics of bone: it lacks altogether vascular canals, osteocyte lacunae, canaliculi or any kind of intra-osseous tubules that could

735

have housed cytoplasmic extensions of osteoblasts. Moreover, its completely monorefringent aspect in polarized light suggests the absence (or poor development) of a fibrillar, collagenous meshwork. The histology of this tissue is inconsistent with bone. In the osteoderms of other, nonglyptosaurine squamates, the existence of a similar, non-osseous layer has been considered by Moss (1969, 1972) in Heloderma. Vickaryous and Sire (2009), also concluded to the existence of such a tissue on the osteoderms of Tarentola annularis and T. mauritanica, although LevratCalviac and Zylberberg (1986) had discarded this possibility. Apart from lizards, such a tissue is unknown in tetrapod osteoderms, be they from lissamphibians (Ruibal and Shoemaker, 1984), pareiasaurs (Scheyer and Sander, 2009), chelonians (Ewert, 1985; Scheyer and SanchezVillagra, 2007), crocodilians (Buffre´nil, 1982; Vickaryous and Hall, 2008), dinosaurians (Buffre´nil et al., 1986; Ricqle`s et al., 2001) or mammals (Hill, 2006; Vickaryous and Hall, 2006). At a broader comparative level, the tissue forming the superficial layer of glyptosaurine osteoderms looks similar to various hypermineralized tissues covering the integumentary skeleton of nontetrapod vertebrates [review in (Sire et al., 2009)]. These tissues include enamel and enameloid (odontodes and dermal denticles), ganoine (scales of the Polypteridae and Lepisosteidae), hyaloine (scutes of armored Siluriformes) and limiting layer (elasmoid scales). Various published pictures showing superposition or superficial resorption of odontodes on the scales of the Polypteridae (Meunier, 1980; Meunier and Gayet, 1992) are strikingly similar to the observations presented in this study. Therefore, as for other lizards (Tarentola spp., Heloderma), the superficial layer of glyptosaurine osteoderms is structurally consistent with the hypermineralized tissues in nontetrapods. These tissues result from epidermal-dermal interactions, with an ultimate contribution of the epidermal basal layer cells in the deposition of material at the scale surface (Sire et al., 2009). Such a contribution would occur also during the deposition of the superficial tissue in squamate osteoderms. This conclusion would substantiate the concept of a deep homology (unevenly expressed among tetrapod taxa) concerning the ability of epidermal cells to contribute to the deposition of such hypermineralized tissues on elements of the integumental skeleton. Role and Limits of Metaplasia Previous studies pointed to the prominent role of osseous metaplasia, as defined above in the Introduction, in the formation of osteoderms in at least 10 genera (12 species) of lizards [reviews in (Moss, Journal of Morphology

736

V.

DE

´ NIL ET AL. BUFFRE

1969; Vickaryous and Sire, 2009); see also (Zylberberg and Castanet, 1985; Levrat-Calviac and Zylberberg, 1986)], one crocodilian (Vickaryous and Hall, 2008), and perhaps also an unidentified ankylosaurian dinosaur studied by Ricqle`s et al. (2001). It remains uncertain if glyptosaurine osteoderms developed via osseous metaplasia. However, our observations indicate that metaplasia (if any) would have been restricted to the initial stage of osteoderm formation. It indeed ceases to be an acceptable explanation for subsequent osteoderm growth because the involvement of osteoblasts is then unquestionable for the reasons exposed below. Intense bone remodeling is an outstanding feature of glyptosaurine osteoderms [an uncommon situation in squamates; (Moss, 1969, 1972; Erickson et al., 2003)]. Bone remodeling involves successive resorption and reconstruction phases that are due to the activity of osteoclasts and endosteal osteoblasts, respectively. The origin of endosteal osteoblasts is basically local: they derive from the mesenchymal cells that externally surround the growing bone. These cells initially reach the sites where reconstruction must occur by progressing within the bone along the perivascular spaces (Karaplis, 2008; Krstic, 1988). Therefore, the occurrence of a reconstruction phase implies the presence of osteoblasts, forming the functional equivalent of a periosteal cambium, in contact with the outer surfaces of the osteoderm. It is likely that the growth of the basal layer of lamellar bone on the deep side of the osteoderms, and possibly also the growth of the fiber-rich woven-fibered bone bordering the sutures, were the result of the activity of these cells. Conversely, the presence of active osteoblasts on the superficial, ornamented, side of the osteoderm is unlikely (or the activity of these cells must, at least, have presented a complex pattern) because most of the mineralized tissue formed at this level was not bone. In conclusion, the results of the present study show that glyptosaurine osteoderms have a complex histological structure, and suggest different origins for the tissues composing the osteoderms: 1) the basal layer of lamellar bone, and possibly also the fiber-rich bone underlying the sutures, must have resulted from sub-periosteal accretion. 2) The superficial, hypermineralized-like tissue covering the osteoderms resulted from epidermal contribution. 3) The woven-fibered bone forming the core of the osteoderms is of dubious origin, and its mode of ossification, either osteoblastic osteogenesis or osseous metaplasia, remains equivocal. This interpretation consistently explains the structural complexity of glyptosaurine osteoderms. Unfortunately, the paleontological record for anguid osteoderms is too scanty to assess whether this complexity is a derived condition, as compared to the more simple structure displayed by Anguis Journal of Morphology

osteoderms, or conversely, if the simple structure of Anguis is an evolutionary simplification (de-differentiation) of the elaborated structure observed in the Glyptosaurinae. ACKNOWLEDGMENTS The authors are grateful to the members of the crews of the Montpellier 2 and Paris 6 Universities who collected the osteoderms. They also thank M. Michel Lemoine (CNRS, France) for his high quality technical assistance. LITERATURE CITED Abramoff MD, Magelhaes PJ, Ram SJ. 2004. Image processing with ImageJ. Biophoton Int 11:36–42. Auffenberg W. 1981. The behavioral ecology of the Komodo monitor. Gainesville: University of Florida Press. Auge´ ML. 2005. Evolution des le´zards du Pale´oge`ne en Europe. Mem Mus Nat Hist Nat 192:3–369. Auge´ ML, Sullivan RM. 2006. A new genus. Paraplacosauriops (Squamata, Anguidae, Glyptosaurinae), from the Eocene of France. J Vertebr Palentol 26:133–137. Conrad JL, Norell MA. 2008. The braincases of two glyptosaurines (Anguidae. Squamata) and anguid phylogeny. Amer Mus Novit 3613:1–24. Crochet JY, Hartenberger JL, Rage JC, Re´my JA, Sige´ B, Sudre J, Vianey-Liaud M. 1981. Les nouvelles faunes de verte´bre´s ante´rieures a` la ‘‘Grande Coupure’’ de´couvertes dans les phosphorites du Quercy. Bull Mus Nat Hist Nat C 3:245–266. de Bonis L, Crochet JY, Rage JC, Sige´ B, Sudre J, Vianey-Liaud M. 1973. Nouvelles faunes de verte´bre´s oligoce`nes des phosphorites du Quercy. Bull Mus Nat Hist Nat, Sciences de la Terre 28:105–113. de Buffre´nil V. 1982. Morphogenesis of bone ornamentation in extant and extinct crocodilians. Zoomorphology 99:155–166. de Buffre´nil V, Farlow JO, Ricqle`s A de. 1986. Growth and function of Stegosaurus plates: Evidence from bone histology. Paleobiology 12:459–473. de Ricqle`s A, Pereda Suberbiola X, Gasparini Z, Olivero E. 2001. Histology of dermal ossifications in an ankylosaurian dinosaur from the Late Cretaceous of Antarctica. Asoc Paleontol Argentina. Publication Especial 7:171–174. Erickson GM, Ricqle`s A de, Buffre´nil V de, Molnar RE, Bayless MK. 2003. Vermiform bones and the evolution of gigantism in Megalania—How a reptilian fox became a lion. J Vertebr Palentol 23:966–970. Estes R. 1983. Sauria terrestrial, amphisbaenia. In: Wellnhofer P, editor. Handbuch der Pala¨oherptologie. Part 10A. Stuttgart: Gustav Fisher.xxii 1 249 p. Ewert MA. 1985. Embryology of turtles. In: Gans C, Billett F, Maderson PFA, editors. Biology of the Reptilia, Vol. 14, Development A. New York: John Wiley and Son. pp 75–268. Haines RW, Mohuiddin A. 1968. Metaplastic bone. J Anat 103: 527–538. Hill RV. 2005. Integration of morphological data sets for phylogenetic analyses of Amniota: The importance of integumentary characters and increased taxonomic sampling. Syst Biol 54:530–547. Hill RW. 2006. Comparative anatomy and histology of xenarthran osteoderms. J Morphol 267:1441–1460. Hoffstetter R. 1962. Observations sur les oste´odermes et la classification des anguide´s actuels et fossiles (Reptiles. Sauriens). Bull Mus Nat Hist Nat, 2e`me se´rie 34:149–157. Karaplis AC. 2008. Embryonic development of bone and regulation of intramembranous and endochondral bone formation. In: Belezikian JP, Raisz G, Martin TJ, editors. Principle of Bone Biology, Vol. 1. Amsterdam: Academic Press. pp 53–84.

STRUCTURE OF GLYPTOSAURINE OSTEODERMS Krstic RV. 1988. Atlas d’histologie ge´ne´rale. Paris: Masson. Levrat-Calviac V. 1986. Etude compare´e des oste´odermes de Tarentola mauritanica et de T. neglecta (Gekkonidae. Squamata). Arch Anat Microscop Morphol Exp 75:29–43. Levrat-Calviac V, Zylberberg L. 1986. The structure of the osteoderms in the Gekko: Tarentola mauritanica. Am J Anat 176:437–446. Main RP, Ricqle`s A de, Horner JR, Padian K. 2005. The evolution and function of thyreophoran dinosaur scutes: Implications for plate function in stegosaurs. Paleobiology 31:291–314. Meunier FJ. 1980. Recherches histologiques sur le squelette dermique des Polypteridae. Arch Zool Exp Ge´n 121:279–295. Meunier FJ, Gayet M. 1992. Remaniement de la ganoı¨ne chez un Semiodontidae nouveau du Cre´tace´ Supe´rieur de Bolivie: Inte´reˆt pale´obiologique. Geobios 25:767–774. Moss ML. 1969. Comparative histology of dermal sclerifications in reptiles. Acta Anat 73:510–533. Moss ML. 1972. The vertebrate dermis and the integumental skeleton. Amer Zool 12:27–34. Rage JC. 2006. The lower vertebrates from the Eocene and Oligocene of the Phosphorites du Quercy (France): An overview. In: Pe´lissie´ T, Sige´ B, editors. 30 Millions d’Anne´es de Biodiversite´ Dynamique dans le Pale´okarst du Quercy, ‘‘Journe´es Bernard Ge`ze’’. Strata 13:161–173. Romer AS. 1997. Osteology of the Reptiles, 2nd ed. Malabar (USA): Krieger Publishing Company. Ruibal R, Shoemaker V. 1984. Osteoderms in anurans. J Herpetol 18:313–328.

737

Scheyer TM, Sa´nchez-Villagra MR. 2007. Carapace bone histology in the giant pleurodiran turtle Stupendemys geographicus: Phylogeny and function. Acta Palaeontol Pol 52: 137–154. Scheyer TM, Sander M. 2009. Bone microstructure and mode of skeletogenesis in osteoderms of three pareiasaur taxa from the Permian of South Africa. J Evol Biol 22:1153–1162. Sire J-Y, Donoghue CJ, Vickaryous MK. 2009. Origin and evolution of the integumentary skeleton in non-tetrapod vertebrates. J Anat 214:409–440. Sullivan RM, Auge´ M. 2006. Redescription of the holotype of Placosaurus rugosus Gervais 1848-1852 (Squamata, Anguidae, Glyptosaurinae), from the Eocene of France and a revision of the genus. J Vertebr Paleontol 26:127–132. Vickaryous MK, Hall BK. 2006. Osteoderm morphology and development in the nine-banded armadillo, Dasypus novemcinctus (Mammalia, Xenarthra, Cingulata). J Morphol 267:1273– 1283. Vickaryous MK, Hall BK. 2008. Development of the dermal skeleton in Alligator mississippiensis (Archosauria. Crocodylia) with comments on the homology of the osteoderms. J Morphol 269:398–422. Vickaryous MK, Sire J-Y. 2009. The integumentary skeleton of tetrapods: origin, evolution, and development. J Anat 214: 441–464. Zylberberg L, Castanet J. 1985. New data on the structure and the growth of the osteoderms in the reptile Anguis fragilis L (Anguidae. Squamata). J Morphol 186:327–342.

Journal of Morphology