reinterpretation of a middle eocene record of tardigrada (pilosa, xenarthra, mammalia) from la...

PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY

CENTRAL PARK WEST A T 79TH STREET, NEW YOR K, NY 10024

Number 3689, 21 pp., 10 figures June 25, 2010

Reinterpretation of a Middle Eocene Record ofTardigrada (Pilosa, Xenarthra, Mammalia) from LaMeseta Formation, Seymour Island, West Antarctica

R. D. E. MACPHEE1 AND M. A. REGUERO2

ABSTRACT

An isolated and incomplete tooth, discovered in sediments of Middle Eocene La Meseta Fm onSeymour Island (northern Weddell Sea, West Antarctica), has previously been interpreted to bethat of a sloth. The specimen as preserved is composed of dentine, as in sloths and tooth-bearingxenarthrans generally. However, characters associated with the dentinal histology of definite slothsare either not represented on the Seymour tooth, or depart considerably from tardigradan andeven general xenarthran models according to new observations presented here. On the basis ofhistological criteria, the La Meseta tooth cannot be shown positively to be tardigradan; it may noteven be xenarthran. Further progress with establishing its relationships will depend on the recoveryof more (and better) specimens. For the moment, it is best attributed to Mammalia, incertae sedis.

INTRODUCTION

In 1995 Vizcaıno and Scillato-Yane report-ed the discovery of an unusual tooth (figs. 1,2), found during the previous year at DPVlocality 2/84, in what is now defined as theCucullaea I allomember of Middle Eocene LaMeseta Fm, Seymour Island (see Marenssi,2006; for abbreviations, see Materials andMethods). Although the tooth (MLP 94-III-

15-14), interpreted as an ‘‘upper caniniformlacking its base,’’ was incomplete, the authorsnoted that it was composed entirely of dentine,and that there was no evidence for the priorpresence of enamel (or, for that matter,cementum). Composition is significant; bar-ring a few primitive exceptions among arma-dillos (Martin, 1916; Simpson, 1932), xenar-thrans are the only South American landmammals whose teeth completely lack enamel.

Copyright E American Museum of Natural History 2010 ISSN 0003-0082

1 Division of Vertebrate Zoology/Mammalogy, American Museum of Natural History ([email protected]).2 Departamento Scientifico de Paleontologıa Vertebrados, Museo de La Plata, Paseo del Bosque s/n, 1900 La Plata,

Argentina ([email protected]).

Fig. 1. MLP 94-III-15-14, incomplete tooth (19.7 mm GL at time of discovery) from La MesetaFm, Seymour I., West Antarctica. A, apparent mesial view (assuming tooth is maxillary as stated by Vizcaınoand Scillato-Yane, 1995); B, lateral view; C, apparent distal view, stereopair; D, key to explanatory diagram:1, location of sample detached for SEM; 2, parts of pulp cavity; 3, intermediate ‘‘light layer’’ of Vizcaıno and

2 AMERICAN MUSEUM NOVITATES NO. 3689

For these and several other reasons discussedin following sections, Vizcaıno and Scillato-Yane (1995) concluded that the tooth mostprobably belonged to a member of Xenarthra,and, given its caniniform aspect, to a tardi-gradan rather than a cingulatan.

As recently relimited by McKenna and Bell(1997) and McKenna et al. (2006), Tardigrada(sloths) includes the most recent commonancestor of Bradypus and Choloepus plus allof its descendants. It is thus an avowedlycrown-group construction. Vizcaıno andScillato-Yane’s understanding of sloth diver-sity was similar, although expressed different-ly, and in any case no additional definition isrequired here. Pilosa + Cingulata with theirusual definitions and content compriseMcKenna and Bell’s (1997) Xenarthra, nowoften given superordinal status. Other taxagrouped within cohort Edentata by Gaudin(2004) shed no light on the affinities of the

specimen in question and accordingly play norole in this analysis.

Seymour Island is situated approximately90 km east of the northern end of the AntarcticPeninsula; not unexpectedly, the possibilitythat a sloth—indeed, that any xenarthran—existed during the Paleogene of what is nowWest Antarctica sparked much interest. First,at approximately 45 Ma, the Seymour speci-men immediately became not just one of theearliest-known xenarthrans, but the oldestapparent sloth in the entire fossil record,exceeding the age of the next oldest contenderby ca. 10 Ma (MacPhee, 2005). Its identifica-tion as a xenarthran strengthened the argumentthat the Seymour fauna was closely related toPaleogene faunas distinctive of southern SouthAmerica (cf. Case, 2006; Tejedor et al., 2009),adding additional weight to the notion that aterrestrial connection persisted betweenPatagonia and West Antarctica into the mid-Paleogene. Finally, the presence of a definitesloth in La Meseta sediments implied that, ca.45 Ma, West Antarctica possessed an environ-ment conducive to mammals preferring rea-sonably warm conditions (Carlini et al., 1990;but see Case, 2006).

MLP 94-III-15-14 is currently the onlyevidence of known whereabouts for theproposition that xenarthrans once lived inWest Antarctica. (A second possible xenar-thran from the La Meseta Fm of Seymour, adistal phalanx [MLP 88-I-1-95; DPV 6/84]possibly of vermilinguan or tardigradan affin-ity [Marenssi et al., 1994; Bargo and Reguero,1998], cannot now be located in the MLPcollections and therefore cannot be evaluatedin this paper.) In one regard, failure to recoveradditional specimens is unremarkable, asmammal fossils are notably rare in LaMeseta sediments and several nominal taxaare known only from one or two isolated teeth(Marenssi et al., 1994; MacPhee, 2008).Nonetheless, certain aspects of the originaldescription, discussed in the following sec-tions, put into question whether MLP 94-III-15-14 can be securely referred to Tardigradaor even to Xenarthra. To investigate thesedoubts, the present authors conducted a newinvestigation, concentrating on histologicalaspects of the tooth as revealed by scanningelectron microscopy (SEM). In addition to

Scillato-Yane (1995); 4, numerous darker layers,stacked as chevrons (also traceable on other side); 5,dark line between external thick layer and other layers;6, external thick layer of dentine; 7, central canal.

2010 MACPHEE AND REGUERO: REINTERPRETATION OF LA MESETA TARDIGRADAN 3

Fig. 2. MLP 94-III-15-14, incomplete tooth from La Meseta Fm, Seymour I., West Antarctica, samplecross section. A, entire section, with locations of areas illustrated in other views (B–E). Note indications ofgrowth layers. B, external margin; despite some chipping which obscures visualization, tubules are inevidence throughout (i.e., zone of secondarily atubular orthodentine is lacking). C, pulp–cavity margin;plane of section is almost parallel to plane of tubules. The substance(s) producing the black pigmentation isunknown, but it was presumably derived from surrounding sediments during fossilization. D, pulp–cavitymargin, close-up of apertures within dentine interpreted as channels for blood vessels; note that theirdistribution is very limited compared to the vascularized dentine of definite sloths. E, close-up of artifactsnoted in C. Note that black areas are largely confined to the walls of tubules, and that individual dentinaltubules traced for a reasonable interval might be black-walled in one area but not in another.


providing valuable data for assessing thestatus of the Seymour tooth, the SEM resultswere useful for re-evaluating previous inter-pretations of xenarthran dentinal histologybased on light-optical methods.

MATERIALS AND METHODS

A small piece of the root end of MLP 94-III-15-14 was detached and prepared for SEM(figs. 1, 2). The sample, a transverse slicerunning from pulp cavity to external surface,was embedded in epoxy resin and polished withincreasingly finer grades of abrasives, endingwith successive suspension slurries of 5.0 and1.0 mm Al2O3. The preparation was examinedwithout coating on a Zeiss EVO 60 EP-SEM(‘‘environmental SEM’’) at 15 kV and photo-graphed at various magnifications. Specimensfrom a selected array of xenarthrans, living andextinct, were prepared in the same manner andcompared to the Seymour tooth (figs. 4–10;appendix 1). Previous workers interested insloth dental histology have tended to concen-trate on mylodontans, megatheriid megather-ians, and the extant genus Bradypus. Here weemphasize imaging of megalonychids (includingextant Choloepus), partly because members ofthis family possess the most ‘‘caniniform’’caniniforms among living xenarthrans (cf.Pujos et al., 2007). Cingulatans were not anoriginal focus of this study but became increas-ingly relevant as work progressed. Publishedinformation on dental histology exists for anumber of dasypodines and glyptodontids, butis sparse or nonexistent for the majority ofother cingulatan taxa. We selected for studyEuphractus and Chaetophratcus, two euphrac-tines that appear to be particularly apt choicesfor comparison with the Seymour tooth becausethey similarly lack dentinal complexity.

Unfortunately, all the fossil specimensproved to be brittle during preparation, andmost images are marred by hairline fractures.However, on the whole actual disruption wasminimal and histological interpretation wasnot seriously affected.

ABBREVIATIONS

AMNHM American Museum of NaturalHistory, Department of Mammalogy

AMNHVP American Museum of NaturalHistory, Department of VertebratePaleontology

bv blood vesselce cementumCU cuticle (dense connective tissue en-

wrapping tooth)DPV Departamento Scientifico de

Paleontologıa Vertebrados (used withlocality identifier)

Fm formationGL greatest lengthMLP Museo de La Plata, Departamento

Scientifico de PaleontologıaVertebrados (used with specimennumbers)

OL outermost layer (of orthodentine)SEM scanning electron microscopy

REDESCRIPTION OF MLP 94-III-15-14

Vizcaıno and Scillato-Yane (1995: 407)drew particular attention to four features ofMLP 94-III-15-14 that, they argued, indicatedin combination that the specimen could beattributed to a sloth: (1) tooth composed ofdentine only, no enamel; (2) caniniform shape,with associated wear facet; (3) pulp cavitywidely open, an indication of hypselodonty;and (4) three layers of dentine (‘‘a hardexternal layer… separated from a softercentral part by a thinner and clearer layer’’).On the whole, they concluded that thestrongest resemblances were to non-‘‘oropho-dontid’’ tardigradans, ‘‘in which a wide core ofsoft dentine is surrounded by a thick layer ofcompact dentine.’’

Figures 1 and 2 illustrate the tooth in itspresent condition. The specimen appearsuniformly black in ordinary light, but withSEM dark areas (i.e., areas that appear darkin electron-optical contrast) are unevenlydistributed, especially internally (fig. 2B–E).Although the condition of the tooth is suchthat one cannot infer how much has been lostby abrasion, SEM inspection failed to revealany kind of enamel, prismatic or nonpris-matic, on the sectioned specimen. Absent alsois any indication of cementum. By contrast,histological indications of dentine (tubules inamorphous ground substance) were conclusivefor that tissue type. The external surface islightly dimpled, smooth rather than rough,and without evidence of damage due to


abrasion (as opposed to breakage). The speci-men is gently rounded, with no crests apparent.As noted, the pulp cavity is apically widelyopen, which in adult mammals is stronglycorrelated with hypselodonty (MacPhee andFlemming, 2003).

While the specimen clearly tapered to apoint when intact, half of the crown andprobably much of the root is absent; thus, itsprecise conformation and original length areuncertain (fig. 1). Whether a major wear facetwas ever present is impossible to say, becausethe tooth’s apparent distal surface was alreadymissing at the time of discovery. The few spallmarks on the tip noted by Vizcaıno andScillato-Yane (1995) are actually on theapparent mesial surface, and could be entirelythe result of damage rather than wear. Itsresemblance to caniniforms of any knownsloth (including those with non-oblique wearfacets, such as certain mylodontids) is thusrather vague. Certainly, there is no evidencethat the specimen would have exhibited thebroad, vertically oriented chisel surfaces seenon caniniforms of late Cenozoic megalony-chids, which are especially trenchant in thisregard (Gaudin, 2004; fig. 3).

As stated in the original description of MLP94-III-15-14, lamellations or growth–layergroups (cf. Hohn, 2009) bounded by deposi-tional arrest lines can be identified in ordinarylight. Arrest lines are not frequently encoun-tered in sloths (but see fig. 10), most of whomlived in warm climates where seasonality doesnot leave a strong imprint on tooth ontogeny(Klevezal, 1996). As is evident in figure 1C–D,there are many more lamellations than thethree identified by Vizcaıno and Scillato-Yane(1995). These continue as discrete layers to theroot end as a series of stacked chevrons,although this is difficult to ascertain in thephotograph because of the specimen’s condi-tion.

If it is assumed that the arrest lines (such asthe very prominent ‘‘dark line between exter-nal thick layer and other layers,’’ feature 5 infig. 1D) on the tooth mark the position of thedentinopulpal junction at specific times duringthe Seymour animal’s life, then early in thetooth’s development the pulp cavity must haveextended to the tip of the crown. Thisinference is corroborated by the position of

the central canal (feature 7 in fig. 1D), whichmarks the ontogenetically primary position ofthe pulp cavity. Such a canal can usually bedetected in teeth of adult xenarthrans, but thefeature also occurs in other mammals in whichsignificant dentine deposition continues longinto postnatal life (e.g., aardvarks, odontocetewhales, dugongs; e.g., Owen, 1840–45).

Fig. 3. Megalonychid maxillary caniniformsfrom a late Cenozoic locality in Grenada, WestIndies. A, medial and B, adoral aspects (AMNHVP132715); C, medial and D, adoral aspects(AMNHVP 132716). Note large chisellike wearfacets. It is improbable that the Seymour specimenexpressed a similar morphology. After MacPhee etal. (2000).


Fig. 4. Euphractus sexcinctus AMNHM 100075, upper molariform in cross section. A, tooth is formedcompletely from typical (avascular) orthodentine, including the central region where some evidence of bloodvessels might be expected. Key to numbered, morphologically distinguishable tissues/areas: 2a, outermostorthodentine layer or OL (extent indicated by double-headed arrow) in which tubules absent or not readilyapparent, no blood vessels; 2b, inner orthodentine with abundant, increasingly recumbent tubules, no bloodvessels. The narrow strip of tissue (CU) separated from the rest of the tooth at the top of thephotomicrograph is described as ‘‘cuticle’’ in Da Silva Sasso and Della Serra’s (1965) histochemical study ofEuphractus teeth; they stated that cementum as such is absent, although Ferigolo (1985: 72) claimed that a‘‘thin and acellular’’ cementum was present on his specimens. Actually, the histochemical tests used by DaSilva Sasso and Della Serra to determine presence of glycoproteins and muccopolysaccharides are fairlynonspecific; they only establish that the cuticle is composed of some form of dense connective tissue which isnot differentiated in the direction of bone, as would be expected if fully developed cementum were present.SEM observations are otherwise consistent with their interpretation. In this specimen the cuticle separatedfrom the dentine during preparation, indicating it was only lightly attached. B, close-up of a small regionalong the tooth’s periphery; recumbent tubules (bright features) decline sharply in abundance in theoutermost layer of the orthodentine (double-headed arrow). On close inspection, ghostly outlines of filled-intubules can be detected.


Using SEM, no gross difference could bedetected between the dentines forming theexterior margin of the tooth, the dentineadjacent to the pulp cavity, and anything inbetween, unlike most of the taxa in thecomparative set (figs. 4–10). Such uniformityimplies that growth layers are compositionallyextremely homogeneous. The only departurefrom gross structural uniformity in the avail-able section of MLP 94-III-15-14 occurs in thezone adjacent to the pulp cavity, where a fewtortuous holes indicate the probable presenceof blood vessels in life (fig. 2D).

Other features seen in the cross section arebest explained as artifactual. Under highmagnification, ‘‘black’’ areas showing strongelectron-optical contrast are almost entirelyconfined to the walls of the dentinal tubules(fig. 2E); intertubular areas are usually clear.Uptake, however, is highly variable, withsome areas being completely bereft of darkmatter. This suggests that the ‘‘thinner andclearer’’ areas may themselves be artifactsrather than a consequence of some sort ofsclerotizing process during life (cf. Klevezal,1996; cf. similar artifacts in a specimen ofAcratocnus, fig. 9). It is also fairly apparent, at

least on the basis of the one section available,that tubules do not undergo significantchanges in size or direction. In mammals andperhaps many other vertebrates, dentinaltubules often curve and branch markedly;indeed, Boyde (1971: 87) thought that branch-ing features were sufficiently diverse andpossibly taxon-specific enough for them tobe utilized in systematics (cf. Hildebolt et al.,1986). Unfortunately, this idea does not seemto have been followed up by subsequentresearchers, at least for xenarthrans, butpromising new techniques that will makeinvestigation simpler are on the horizon(Kalthoff, 2008). In any case, with only asingle preparation we cannot establish branch-ing pattern or related variables for theSeymour tooth.

COMPARISONS AND ANALYSES

Our analysis in this section is based oncomparisons of six features, treated in detail infollowing paragraphs, which we consider to begenerally characteristic of xenarthran teeth.The first four have been examined sufficientlywell in the literature for us to be reasonably

Fig. 5. Chaetophractus vellerosus AMNHM 246457, molariform in cross section. Within the dense arrayof tubules typical of orthodentine, there are occasional larger defects that appear to be for small bloodvessels (bv). However, apertures are tiny and their frequency per unit area is much lower than in typicalvasodentine (see fig. 6).


certain that they are widespread within thegroup, if not precisely universal. The last twocharacters should be considered on test; at thehistological level, tooth-bearing xenarthransdiffer dentally much more than one mightglean from available studies. Individuals,developmental stages, and even tooth loci insome species may differ in substantial waysthat have rarely been taken into account.

To anticipate our chief conclusion, there areno features of the Seymour tooth thatunambiguously tie it to known Xenarthra tothe exclusion of other possibilities.

1. Enamel normally absentIt is of course parsimonious to assume that

xenarthrans derive from eutherians in whichenamel was present, as prismatic enameloccurs on tooth crowns of the early Paleo-gene dasypodoid Utaetus buccatus (Simpson,1932), and enamel matrix has also been dem-onstrated conclusively in perinatal Dasypusnovemcinctus (Martin, 1916). The relevant pointis that in these cases the enamel layer amounts tono more than a thin scale, of little apparentsignificance for tooth function.

Wherever the correct affinities of theSeymour tooth may lie, there is no positiveevidence for the presence of enamel in anyform, while there is some negative evidence forits complete absence. For example, all externalsurfaces on the tooth are relatively smoothand shiny, which implies that no other hardtissues enwrapped it during life. If they had,then one would expect to see some histologicalevidence of a dentinoenamel or dentinocemen-tum junction, as occurs in the teeth of typicalmammals. Such junctions, when seen insection at high magnification (cf. Fawcett,1986), are marked by tiny, irregular asperities,pits, and spindle-shaped processes of dentinalmatrix (enamel spindles) that invade theenamel or cementum for a short distance.Functionally, these features presumably serveto limit the amount of torque developed alongthe junction interface during mastication, that,if left unconstrained, might cause the devel-opment of cracks between different tissuetypes. These microstructures are also respon-sible for the dull, rather rough appearance ofdentinal surfaces exposed on fossil teeth withmissing enamel or cementum—in markedcontrast to the gently scalloped, reflective

external surface of MLP 94-III-15-14. Insum, although not all relevant surfaces arepreserved on the Seymour tooth, it is unlikelythat an exterior coating of enamel could havebeen present at any stage.

2. Cementum normally presentAn external wrapping of cementum, some-

times highly vascularized, has been counted asa universal feature of all adequately investigat-ed xenarthrans (Gaudin, 2004), although itscomplete absence in some extant dasypodoidshas also been claimed (see fig. 4). In any case,ample observational evidence reveals thatcementum thickness varies in tooth-bearingxenarthrans from scarcely noticeable, as insome armadillos, to very significant, as inMegatherium (Ferigolo, 1985; Gaudin, 2004).In the latter taxon there is good evidence that athick casing of cementum is mechanicallysignificant in occlusal performance (Vizcaıno,2009). Ferigolo (1985) noted a positive corre-lation in xenarthran taxa between the degree ofvascularization and the amount of cementumpresent; those with the most vascularized teeth(such as Megalocnus and Neocnus in this study)also exhibited the greatest thickness of cemen-tum. By contrast, in xenarthrans in which thefunctional role of cementum is more limited,this tissue tends to be thin and possibly onlylightly anchored to the underlying dentine (cf.fig. 10). Thus, while we cannot absolutelydiscount the presence of cementum on MLP94-III-14-15 during life, as in the case of enamel(see above) existing external surfaces provideno conclusive evidence for its presence.

3. At least two kinds of structurally distinctdentine present (orthodentine and vasoden-tine)Since the time of Owen’s investigations

(1840–1845) it has been widely recognizedthat at least two kinds of descriptivelydifferent dentine comprise the teeth of almostall dentulous xenarthrans. However, this pointrequires some elaboration, as there are excep-tions among armadillos (see below), and thenature of so-called ‘‘hard’’ vs. ‘‘soft’’ dentinehas been inadequately explored.

To make headway here there is also a needto reduce the abundant—and notably redun-dant—nomenclatural terminology used todescribe qualitatively different dentines found


in vertebrates (e.g., ‘‘orthodentine,’’ ‘‘duro-dentine,’’ ‘‘normodentine,’’ ‘‘modified ortho-dentine,’’ ‘‘orthovasodentine,’’ ‘‘vasodentine,’’‘‘secondary dentine,’’ ‘‘intermediary tissue,’’‘‘osteodentine’’). For present purposes, wefind the most useful contrast to be betweenthe categories orthodentine and vasodentine.Avoiding unnecessary complication, vasoden-tine may be defined as ‘‘an incrementallydeposited circumpulpar tissue permeated bypassages for capillaries, but [generally] lackingtubules for extended odontoblastic processes’’(Lund et al., 1992: 61), while orthodentine,which classically lacks blood vessels, is char-

acterized by ‘‘numerous fine tubules fromodontoblastic processes coursing parallel toeach other’’ through the ground substance(Lund et al., 1992: 58).

As defined here, orthodentine occurs inalmost every gnathostome class; vasodentinehas a more restricted but still significantdistribution (Arsuffi, 1938; Bradford, 1967).In xenarthrans the two types exhibit differentmechanical properties (Keil and Venema,1963; Vizcaıno, 2009; see below). Functionalblood vessels incarcerated within dentinaltissues are rare in adult vertebrates generally,and decidedly so in mammals (for a list of taxa

Fig. 6. Megalocnus rodens AMNHVP 143454, molariform in cross section. A, overview, with locationsof image montages B and C indicated. Key to numbered, morphologically distinguishable tissues/areas: 1,cementum, with numerous blood vessels; at least some of the rounded apertures appear to be lacunae forcementocytes; 2a, outermost orthodentine layer (OL), tubules absent or not readily apparent, very few bloodvessels; 2b, inner orthodentine with abundant, increasingly recumbent tubules and randomly distributedblood vessels; 3, vasodentine, with patterned distribution of many small blood vessels, some anastomosingwith larger feeder vessels. D and E, from another part of the tooth, are noncontiguous close-ups of thetransitional zones between outermost orthodentine and cementum (D) and inner orthodentine (E),respectively. Note numerous recumbent, infilled tubules, much more evident at this magnification than in B.


allegedly possessing this feature, see Arsuffi,1938). Various kinds of intermediate statesbetween typical orthodentine and vasodentinehave been identified (e.g., Arsuffi, 1938;Ferigolo, 1985), but, as with all classificationsof this type, it is vital to clearly separate endmembers, against which transitional variants

can be arrayed and compared. As in verte-brate dentines generally, in xenarthrans odon-toblast cell bodies are absent within thesetissues.

The outermost dentine of a tooth is alwayslaid down in ontogeny as orthodentine, even ifblood vessels also occur within it (cf. Neocnus;

Fig. 6. Continued.


Fig. 7. Neocnus gliriformis AMNHVP 143453, molariform in cross section. Key to numbered,morphologically distinguishable tissues/areas: 1, cementum; 2a, outermost layer (OL) of orthodentine; 2binner orthodentine; 3, vasodentine. Difference between 2a and 2b is obvious in ordinary light as well, due tovariable uptake of artifactual pigments in tubular vs. atubular orthodentine. A, overview of highly vascularcementum and orthodentine. B, close-up of cementum/orthodentine (OL) interface, marked by numerousblood vessels (bv) along borders of OL and within cementum. These appear to comport with similarconcentrations of distinctively U-shaped blood vessels seen along this interface by Ferigolo (1985) in certainother sloths. C, close-up of entire section of orthodentine; long irregular features (st) appear to be theoutlines of infilled vascular (bv) channels, although some may be infilled tubules. Because of uncertainty onthis point they are simply identified as ‘‘striae.’’ D, fresh fracture surface prior to grinding and polishing.


fig. 7A–C). This process continues as thedentinopulpal junction retreats rootward, asa consequence of later matrix being under-plated onto pre-existing dentine. Zones ofdentinal vascularization have been found in allsloths and most adult armadillos that havebeen properly investigated, which could beevidence that presence of vasodentine isprimitive for these groups (cf. Gaudin, 2004;but see Ferigolo, 1985). There is, however,substantial variation in the size and number ofblood vessels, as well as other features (figs. 4–10). For example, Ferigolo (1985) noted thatsome taxa display an ‘‘intermediary tissue,’’possessing both tubules and blood vessels(resembling Arsuffi’s [1938] ‘‘orthovasoden-tine’’), while in Bradypus tridactylus ‘‘in somecases the central dentine was almost amor-phous [i.e., atubular] with only a few vascularcanals or none’’ (Ferigolo, 1985: 72). Thesame author noted that, in Euphractus andBradypus, vascularization is actually lostduring the course of development (cf. figs. 4,5). (Ontogenetic loss of vascularization alsooccurs in the dentine core of the incisors ofsome rodents according to Moss-Salentijn andMoss [1975] and is well attested in variousfishes [Ørvig, 1951].)

In another deviation from the generalxenarthran condition, the central figure inglyptodont teeth is comprised of a differentstructurally defined dentine (osteodentine)

rather than typical vasodentine. However, largeapertures occur within glyptodont osteoden-tine, which is presumably the reason that Owen(1840–1845: 325) referred to the ‘‘vascularosseous texture’’ of the latter (see also com-ments by Lund et al., 1992: 60). Ferigolo (1985)believed that osteodentine occurs more widelyamong xenarthrans, but did not illustrate anynew examples. A tissue with some characteris-tics of bone (e.g., lacunae) lines the pulp cavityof the specimen of Glossotherium illustrated infig. 10, but its preservation is too poor fordefinitive identification. Microanatomical con-ditions in the teeth of Pseudoglyptodon, regard-ed by some authorities as a close relative oftardigradans (McKenna et al., 2006), areunknown; there are certainly no macroana-tomical resemblances to the Seymour tooth.

Although the dentines of unquestionablexenarthrans do show some amount of mor-phological variation, the Seymour specimenstill stands apart. As noted previously, onlyone type of dentine (with tubules, thusorthodentine by default) seems to composethe latter, with the dubious exception of arestricted vascularized zone near the pulpcavity. Nor do the numerous growth layergroups correlate in any obvious way withexpected gross differences between ‘‘hard’’and ‘‘soft’’ dentines as seen in definite sloths.These points bear on the evaluation of thenext two characters as well.

Fig. 8. Choloepus hoffmanni AMNHM 173551, caniniform in cross section. This tooth is composed oforthodentine (seen here) and a small core of vasodentine (not shown). The tubules immediately beneath thecementum (ce) become gradually more indistinct externally, but the true atubular outermost zone (OL) isvery thin.


Fig. 9. Acratocnus antillensis AMNHVP 143452, caniniform in cross section. The tooth was broken atmidshaft in order to undertake SEM; images are of the unpolished fracture surface (cf. fig. 7D). As is thecase with other fossil sloth teeth depicted in this paper, image intelligibility is affected by noisy artifactualpigmentation. In A it can be seen that the tubules become more recumbent toward the exterior, but after ashort transition cannot be traced further. In B, a close-up of the transitional area, the line of tubuletermination (black arrows) is very distinct. At least in this area, no tubules can be traced into the amorphousoutermost layer to the left. Cementum may not have been present on this specimen; in any case, there isnone now.


4. Outermost orthodentine is moderately tocompletely atubularA little-noted peculiarity of many, if not all,

xenarthrans is that the dentine formed first inontogeny mostly lacks patent tubules in theadult. In the montages of SEM images of theMegalocnus rodens molariform in figure 6, itcan be seen that tubules, very prominent in theorthodentine forming the middle portion ofthe tooth, become suddenly much sparser andtend to disappear altogether in the outermostlayer (OL), except for occasional ghostlyoutlines. The effect is not necessarily constantall the way around the tooth; the degree oftubule obliteration may vary from area toarea, but it is distinctly noticeable when deeperand more superficial layers are compared.

The fact that tubule remnants can beidentified at all indicates that the OL issecondarily atubular. In M. rodens vascularchannels also occur in the OL, but are infre-quent. In the related taxon Neocnus gliriformis(fig. 7C) both structures occur in the OL, but

outlines (striae) of infilled vascular channelsseem to be particularly abundant.

Da Silva Saso and Della Serra (1965: 160,fig. 3) appear to have been the first to detectthe existence of an external layer of relativelyatubular dentine in the tooth of a xenarthran(Euphractus; see fig. 4). However, they inter-preted the lack of tubules in a non-ontogeneticmanner, suggesting that the majority ofdentinal tubules simply terminated beforereaching the outer margin of the tooth. Thisdescribes the adult condition, but does notaccount for its development. In all probability,the earliest deposition of dentinal matrix inEuphractus occurs around long cellular pro-cesses of active odontoblasts in the lining ofthe primordial pulp cavity, as in the standardvertebrate pattern. Later, however, mosttubules (and vascular channels) traversingthe OL become infilled, yielding the definitivecondition of truncated tubules described here.

How this happens would have to beestablished experimentally and is therefore

Fig. 10. Glossotherium sp. AMNHVP 143455, molariform (root end) in cross section. Unfortunately thisspecimen, from a tar seep in southwestern Trinidad, suffered substantial fracturing during preparation.However, tissue types are well differentiated. Key to numbered, morphologically distinguishable tissues/areas: 2a, outermost layer of orthodentine (OL); 2b, inner orthodentine; 3, vasodentine; 4, ?osteodentine.(Cementum is present as a thin layer on other parts of this tooth, but had spalled off on the root end and isthus not represented.) The outermost layer is very thin, as in the specimen of Choloepus (fig. 8), but as wouldbe expected in light of conditions in other sloths examined here, the layer lacks patent tubules. Arrest lines,perhaps equivalent to growth layer groups, are moderately prominent in the lower part of the image withinthe inner orthodentine.


well beyond the scope of this study. However,it bears mentioning that, in the teeth of manynonmammals, orthodentine is covered byenameloid, a hypermineralized tissue. Thus,in the wrasse Labrus, tubules of orthodentinegenerally terminate at the enameloid border,but ‘‘some cross the boundary and run forsome distance into the enameloid’’ (Bradford,1967: 11), which is primary evidence thatodontoblasts are involved in the formation ofthis tissue as well. Gillis and Donoghue (2007:41), describing the histology and developmentof single-crystallite enameloid in variouschondricthyans, noted that this tissue consistsof ‘‘a monolayer of hypermineralized cappingtissue… [that among other features] containstraces of odontoblast tubules… and lackinglines of incremental growth.’’ In human dentalanatomy, tubule obliteration is typically view-ed as a pathological process, which weakensthe dentine (e.g., production of ‘‘translucent’’dentine by secondary hypermineralization,which alters the refractive index of the tubulesin such a way that they become indistinguish-able from ground substance; Warwick andWilliams, 1973; Arnold et al., 2007).

True enameloid has not been recognized inany mammal, and it seems fairly clear that thexenarthran solution is different from that offishes, in which a greater degree of tissuedifferentiation occurs. Nonetheless, the func-tional convergence is interesting, because in atleast some xenarthrans the OL is actually thehardest part of the tooth, as Keil and Venema(1963: 181, fig. 8) demonstrated quantitativelyin their study of microhardness (but withoutclarifying histological structure, which makesinterpretation of their results somewhat diffi-cult). In Euphractus, they showed that theexterior surface yielded the highest hardnessmeasurement, but in this taxon cementum iseither much reduced or replaced by a thinlayer of cuticle only a few micra thick (seefig. 4), which implies that Keil and Venema(1963) had actually measured the outermostlayer of orthodentine instead. The sameinterpretation applies to their results forCabassous, which also exhibits a negligibleamount of cementum (Ferigolo, 1985). Thegraph for the molariform of Dasypus indicatesthat its hardest part lies more internally. This,however, is still consistent with the observa-

tion that the outer orthodentine is hardest,because in this taxon the teeth are externallywrapped in a thick layer of cementum (whichis the tissue that they measured first). In anycase, close inspection of molariforms ofseveral different xenarthran taxa indicates thatthe true boundary between the ‘‘hard’’ and‘‘soft’’ dentines of gross morphology mostprobably lies not between the orthodentineand vasodentine (or between cementum anddentine), as in most traditional descriptions,but within the orthodentine as we have definedit here.

In smaller xenarthrans (e.g., Euphractus,fig. 4) the absolute width of the OL is, ofcourse, narrower than in a comparatively largetaxon like Megalocnus (fig. 6). However,notice needs to be taken of variable conditionsin megalonychids, in which the molariforms ofsome taxa (e.g., Neocnus; fig. 7A, D) possess arelatively well-developed OL, but the canini-forms of others (e.g., Choloepus; fig. 8) do not.The difference may underlie functional dis-tinctions between the two tooth morphs: thecaniniforms have a morphology consistentwith their use for shearing, while the occlusalsurfaces of molariforms function in crushing/milling. Whether or not it is correct, thisobservation underlines our earlier point thatsignificant morphological variation occursamong sloth tooth loci.

This observation is also pertinent for theevaluation of the Seymour specimen.Figure 2B is a highly magnified view of thefriable outer margin of the tooth; the edge wasslightly chipped during preparation, but it isotherwise intact. Even though the quality ofimagery is compromised by artifactual staining,no secondarily atubular zone is apparent withSEM, either here or elsewhere, and tubulescontinue to the margin of the tooth’s externalsurface. In this regard the Seymour specimendiffers from all examples utilized in this study,including Euphractus and Choloepus, both ofwhich possess extremely narrow atubular zonesin the outer orthodentine.

5. Orthodentine restricted compared to vaso-dentine in molariformsVizcaıno and Scillato-Yane (1995) assumed

that the rather thick outer layer of dentineseen on the broken face of the Seymour tooth


(fig. 1C) was equivalent to the ‘‘hard’’ dentineof xenarthrans, while the layers internal to itwere compositionally ‘‘soft.’’ This interpreta-tion seemed to fit well with gross conditionsseen in the majority of xenarthrans, in whichorthodentine forms a comparatively narrowcollar around a larger central core of vaso-dentine (cf. Gaudin, 2004). However, as notedabove, SEM reveals that the Seymour tooth ishistologically homogeneous except for a re-stricted zone next to the pulp cavity, in whicha small number of vascular channels occur inaddition to tubules. If, for the sake ofargument, the latter is regarded as a form ofvasodentine (or Ferigolo’s [1985] ‘‘intermedi-ary tissue’’), and the rest of the tooth asorthodentine, even though there is no percep-tible structural difference between the two,then the resulting ratio of the two dentines (asmeasured from the pulp cavity to the externalsurface) is on the order of 1:9—essentially thereverse of the condition regarded as thedominant one in tardigradans, in whichvasodentine forms the bulk of the tooth (e.g.,fig. 6A).

However, once again conditions in definitexenarthrans display more complexity thantends to appear in systematic treatments. Inat least some euphractine dasypodids, adultteeth lack vascularization completely, oralmost so (figs. 5, 6). In Euphractus itself,vasculature is said to be present in youngstages but later disappears (Ferigolo, 1985),although how channels are remodeled orinfilled remains unexplored. In any case,although not all euphractine species have beenstudied, lack of dentinal vascularization maybe a general feature of adult stages of thissubfamily. Thus, in the central part of thetooth of the adult specimen of Chaetophractusvellerosus utilized in the present study (fig. 5),there are some holes, slightly larger than theapertures of the surrounding tubules, whichvery likely carried blood vessels. To thislimited degree, the Seymour tooth may besaid to resemble a morphological feature alsoencountered in a definite xenarthran.

6. Flabellate arrangement of dentinal tubulesA final interesting aspect of xenarthran

dentinal histology that has not been widelycommented upon by recent authors is adistinctive pattern of primary curvature in

dentinal tubules. Tubules very rarely follow astraight course; in humans, in which tubulepatterning has been most closely investigated,primary, secondary, and even tertiary curva-tures have been described (Warwick andWilliams, 1973). In dentulous xenarthrans,primary tubule orientation, viewed fromcentral toward more peripheral regions in agiven tooth, follows an arc that is initiallysemivertical but that becomes increasinglyrecumbent toward the tooth’s external surface.This is best seen in longitudinal section inordinary light, where tubule content anddirection can be detected across the entiretooth. Cross sections are less reliable in thisregard unless one has many sections in whichtubule orientation can be evaluated.

This classic fan-shaped or ‘‘flabellate’’arrangement of tubules was first described byOwen (1840–45: 324 and associated plates) inan armadillo: the ‘‘calcigerous tubes’’ (i.e.,dentinal tubules), at first vertical, ‘‘rapidlydiverge like the outer streams of a fountain’’;they ‘‘bend outwards, and direct their courseat right angles to the axis of the tooth, butwith a slight convexity directed towards thegrinding surface.’’ In other cases the flabellatearrangement is more muted: Owen (1840–45:336) noted for Mylodon that the ‘‘vasculardentine is permeated… by long, nearly straightand parallel vascular canals, proceeding, forthe most part, outwards, and with a slightinclination to the grinding surface; this incli-nation is least at the base, and greatest at thesummit of the tooth, where the vascular canalsare parallel with the axis of the tooth.’’ Similarconditions obtain in Bradypus and Dasypus(cf. Ferigolo, 1985, pls. 1–4). Using polarizedlight, Keil and Venema (1963) and Da SilvaSasso and Della Serra (1965) also documentedassociated transitions in collagen-fiber orien-tation in dentine ground substance.

Although variation certainly exists amongextant xenarthrans for this character, andmore comparative and quantitative data areneeded to evaluate it properly, it seemsreasonable to conclude that marked curvatureis a dominant feature of dentinal organizationin members of the superorder. All that can besaid regarding the Seymour specimen is thatevidence of similar curvature is lacking in theavailable section: tubules are essentially recti-


linear, if somewhat undulating, and highlyrecumbent (trending horizontally with respectto the tooth’s long axis) in all areas, includingthe areas closest as well as farthest from thepulp cavity. The broken distal surface gives noindication of tubule direction, but increasingverticality of tubules toward the mid-part ofthe crown seems inconsistent with the stacked-chevron appearance of the different growthlayers.

CONCLUSIONS

Microscopic study of MLP 94-III-15-14reveals that this specimen—the sole specimenattributed to the ‘‘Seymour Island sloth’’—departs markedly from the morphology ex-pected in definite tardigradans. The fewmicroanatomical resemblances (mainly toarmadillos) that can be cited are not compel-ling. Because of the incompleteness and poorcondition of the specimen, it cannot beconclusively established that enamel andcementum were absent from the tooth duringlife, although it seems extremely unlikely thatthey were present. However, there are otherinformative features that can be assessed, andthese provide a useful basis for making aninformed decision about the specimen’s taxo-nomic allocation:

1. Prior to damage, the Seymour tooth wasprobably conical, but there is no compel-ling morphological reason to identify it asa caniniform (as opposed to any otherlocus in a tooth row) in the absence ofknowledge concerning the appearance ofother teeth in the jaw from which it came.

2. The Seymour tooth appears to be com-posed entirely of one kind of dentine—orthodentine, with the usual array oftightly packed tubules in an amorphousmatrix. Laminations (growth–layergroups) can be identified under ordinarylight, but with SEM they do not displayhistological or compositional differencesthat would provide a basis for identifying‘‘hard’’ vs. ‘‘soft’’ dentine, contra Vizcaınoand Scillato-Yane (1995). Nevertheless, afew small channels, consistent with thepresence of blood vessels, occur close tothe pulp cavity. This observation mightbe grounds for recognizing a small zone

of vasodentine; if so, as in other respects,the Seymour tooth stands at the far endof accepted xenarthran variation.

3. In the orthodentine of most of thedefinite xenarthrans examined here, amarked boundary exists between an innerzone in which tubules (and, in some taxa,vascular channels) are patent and anoutermost zone (OL) in which tubulesand/or vascular conduits are not evident.An identifiable OL is absent in theavailable section of the Seymour tooth.

4. At least in the available cross section,tubule orientation (curvature) does notappear to change in the Seymour tooth.Thus a flabellate arrangement of tubulesis probably absent, although to demon-strate this conclusively a longitudinalprofile would be required.

5. Considered together, these facts raisesignificant doubts concerning the previoussystematic allocation of MLP 94-III-15-14.If the taxon it represents is to be retainedwithin Xenarthra on the ground that noother allocation is likely, it cannot beplaced with certainty in either of the majortooth-bearing groups within the superor-der. To the limited extent that the Seymourspecimen can be said to resemble teeth ofany definite tardigradans or cingulatans, itis actually closer histologically to somearmadillos than to any known sloths. Eventhen the resemblances are of a largelynegative kind. A cogent example is thedentition of the adult euphractines inves-tigated in this paper: the teeth are mereplugs, lacking cementum, vasodentine, andany sort of significant external layer ofatubular orthodentine. The Seymour toothdisplays a similar lack of complexity, butthis would seem much more likely to be theresult of convergence than common inher-itance. In this connection it is relevant tonote that scutes are almost always in highabundance whenever armadillos are pres-ent in a paleontological locality or strati-graphic section (Vizcaıno et al., 1998).After a quarter century of intense pros-pecting in La Meseta sediments by manydifferent teams, it is surely meaningful thatunambiguously cingulatan osteodermshave never shown up.


6. Interestingly, the Seymour tooth couldalso be said to be morphologically asclose to certain toothed cetaceans as toany xenarthran. Thus the first incremen-tal line seen on the broken face of thetooth, bounding the layer that Vizcaınoand Scillato-Yane (1995) regarded as‘‘hard’’ dentine, recalls the neonatal linethat separates, in many extant odonto-cetes, prenatally deposited dentine fromdentine deposited later (Hohn, 2009).Also, dentine tubules in modern toothedwhales tend to be radially organized witha very slight primary curvature, and inyoung stages very little cement is present.Enamel occurs as a thin cap that may beworn away with use (Maas, 2009; R.MacPhee, personal obs.), although actualabsence of enamel in unworn cetaceanteeth seems never to have been noted inthe neontological or paleontological lit-erature (cf. O’Leary, 1998). Cetaceanremains have been discovered in LaMeseta deposits (e.g., Mitchell, 1989;Fordyce, 1989; Fostowicz-Frelik, 2003),although none so far described is anobvious source for the Seymour tooth.Nevertheless, Cetacea may be the logicalplace to look should more examples ofthis kind of tooth be found in future.

7. In light of the numerous qualificationsthat must now be applied to the accep-tance of the xenarthran status of MLP 94-III-15-14, we conclude that it is betterrelegated to Mammalia, incertae sedis,until its status can be more meaningfullyresolved. (The more restricted allocation‘‘Eutheria, incertae sedis’’ would also beappropriate, but as it offers no actualgain in understanding it is not adoptedhere.) Regrettably, this action removesthe only available evidence for xenarthranpresence in the Paleogene of Antarctica.Although there are some compellingcompositional correlations between theSeymour mammal assemblage and vari-ous Patagonian faunas of Riochican andVacan age (Reguero et al., 2002; Case,2006; Tejedor et al., 2009), the precise fitof the former within the sequence ofSouth American land mammal ages is notsettled. Of great interest is the recent

argument by Tejedor et al. (2009) that theLa Meseta mammal assemblage corre-lates well with that of Paso del Sapo andcoeval localities in western Patagonia,whose age may lie in the unconstrained‘‘grey’’ area between Riochican andVacan time (thus late Early Eocene). Atthese localities xenarthrans are represent-ed in the form of several astegotheredasypodoids, including a new species ofRiostegotherium, a genus otherwiseknown only from Itaboraı (long regardedas late Paleocene, but possibly consider-ably younger according to Gelfo et al.[2009]). However, the strength of the Pasodel Sapo/La Meseta correlation will haveto be tested with taxa other than xenar-thrans: the Paso del Sapo fauna and itspossible correlates lack recognizablesloths, while La Meseta Fm so far lacksany evidence of armadillos (and now,sloths as well).

ACKNOWLEDGMENTS

The senior author thanks his AMNHcolleagues James Webster (facilities and ma-terials for grinding and polishing sections),Becky Rudolph (SEM), and Lorraine Meeker(illustrations). We are grateful to Tim Gaudin(University of Tennessee, Chatanooga) andanother reviewer for their helpful comments,Clare Flemming for editing the final manu-script, and Mary Knight for copyediting andoverseeing the production of the publishedpaper. Research related to this investigationwas conducted under grant NSF OPP ANT0636639 to the senior author.

REFERENCES

Arnold, W.H., V. Bietau, P.O. Renner, and P.Gaengler. 2007. Micromorphological and mi-croanalytical characterization of stagnatingand progressing root caries lesions. Archivesof Oral Biology 52: 591–597.

Arsuffi, E. 1938. Beitrage zur Kenntnis desVasodentins. Zeitschrift fur Anatomie undEntwicklungsgeschichte 108: 749–760.

Bargo, M.S., and M.A. Reguero. 1998. Annotatedcatalog of the fossil vertebrates from Antarcticahoused in the Museo de La Plata, Argentina. I.Birds and land mammals from La Meseta


Formation (Eocene–?Early Oligocene). In S.Casadıo (editor), Paleogeno de America del Sury de la Peninsula Antartica, Asociacion Paleon-tologica Argentina, Publicacion Especial 5.Buenos Aires, 211–221.

Bradford, E.W. 1967. Microanatomy and histo-chemistry of dentine. In A.E.W. Miles (editor),Structural and chemical organization of teeth,vol. 2. New York: Academic Press, 3–34.

Boyde, A. 1971. Comparative histology of mam-malian teeth. In A.A. Dahlberg (editor), Dentalmorphology and evolution. Chicago: ChicagoUniversity Press, 81–94.

Carlini, F., et al. (1990). The first Palaeogene landplacental mammal from Antarctica: its paleo-climatic and paleobiogeographical bearings.Abstracts, Fourth International Congress ofSystematic and Evolutionary Biology, Univer-sity of Maryland, College Park, 1–7 July, 1990:325.

Case, J.A. 2006. The late Middle Eocene terrestrialvertebrate fauna from Seymour Island: the tailsof the Eocene Patagonian size distribution.Special Publication of the Geological Society ofLondon 258: 177–186.

Da Silva Sasso, W., and O. Della Serra. 1965.Observacoes sobre as estruturas de dentes dexenartros pertencentes aos generos ‘‘Dasypus’’,‘‘Euphractus’’ e ‘‘Bradypus’’ (Edentata, Mam-malia). Revista Brasileira de Biologia 25:157–164.

Fawcett, D. 1986. A textbook of histology. 11th ed.Philadelphia: Saunders.

Ferigolo, J. 1985. Evolutionary trends of thehistological pattern in the teeth of Edentata(Xenarthra). Archives of Oral Biology 30:71–82.

Fordyce, R.E. 1989. Origins and evolution ofAntarctic marine mammals. In J.A. Crame(editor), Origins and evolution of theAntarctic biota. Special Publication of theGeological Society of London 47: 269–281.

Fostowicz-Frelik, L. 2003. An enigmatic whaletooth from the Upper Eocene of SeymourIsland, Antarctica. Polish Polar Research 24:13–28.

Gaudin, T.J. 2004. Phylogenetic relationshipsamong sloths (Mammalia, Xenarthra, Tardi-grada): the craniodental evidence. ZoologicalJournal of the Linnaean Society 140: 255–305.

Gelfo, J.N., Goin, F.J., M.O. Woodburne, and C.de Muizon. 2009. Biochronological relation-ships of the earliest South American Paleogenemammalian faunas. Palaeontology 52: 251–-269.

Gillis, J.A., and P.C.J. Donoghue. 2007. Thehomology and phylogeny of chondrichthyan

tooth enameloid. Journal of Morphology 268:33–49.

Hildebolt, C.F., G. Bate, J.K. McKee, and G.C.Conroy. 1986. The microstructure of dentine intaxonomic and phylogenetic studies. AmericanJournal of Physical Anthropology 70: 39–46.

Hohn, A.A. 2009. Age estimation. In W.F. Perrin,B.G. Wursig and J.G.M. Thewissen (editors),Encyclopedia of marine mammals. Amsterdam:Elsevier/Academic Press, 6–13.

Kalthoff, D. 2008. Tooth microstructure in fossiland recent sloth (Mammalia, Folivora).Journal of Vertebrate Paleontology 28 (3,suppl.): 98A.

Keil, A., and B. Venema. 1963. Struktur- undMikroharteuntersuchungen an Zahnen vonGurteltieren (Xenarthra). Zoologische Beitrage9: 173–195.

Klevezal, G.A. 1996. Recording structures ofmammals: determination of age and recon-struction of life history. Rotterdam: Balkema.

Lund, R., P.R. Bartholomew, and A. Kemp. 1992.The composition of the dental hard tissues offishes. In P. Smith (editor), Structure, functionand evolution of teeth. Tel Aviv: Freund,35–71.

Maas, M.C. 2009. Bones and teeth, histology of. InW.F. Perrin, B.G. Wursig and J.G.M.Thewissen (editors), Encyclopedia of marinemammals. Amsterdam: Elsevier/AcademicPress, 116–122.

MacPhee, R.D.E. 2005. ‘‘First’’ appearances in theCenozoic land-mammal record of the GreaterAntilles: significance and comparison withSouth American and Antarctic records.Journal of Biogeography 32: 551–564.

MacPhee, R.D.E. 2008. Austral portals and thepaleobiogeography of Antarctic land mam-mals. Abstracts of Papers, Annual Meeting ofthe Geological Society of America 40 (6): 333.

MacPhee, R.D.E., and C. Flemming. 2003. Apossible heptaxodontine and other caviidanrodents from the Late Quaternary of Jamaica.American Museum Novitates 3422: 1–42.

MacPhee, R.D.E., R. Singer, and M. Diamond.2000. Late Cenozoic mammals from Grenada,Lesser Antilles Island-Arc. American MuseumNovitates 3303: 1–20.

Marenssi, S.A. 2006. Eustatically controlled sedi-mentation recorded by Eocene strata of theJames Ross Basin, Antarctica. SpecialPublication of the Geological Society ofLondon 258: 125–133.

Marenssi, S.A., M.A. Reguero, S.N. Santillana,and S.F. Vizcaıno. 1994. Eocene land mammalsfrom Seymour Island, Antarctica: paleobiogeo-graphical implications. Antarctic Science 6:3–15.


Martin, B. 1916. Tooth development in Dasypusnovemcinctus. Journal of Morphology 27:647–691.

McKenna, M.C., and S.K. Bell. 1997. Classificationof mammals above the species level. New York:Columbia University Press.

McKenna, M.C., A.R. Wyss, and J.J. Flynn. 2006.Paleogene pseudoglyptodont xenarthrans fromcentral Chile and Argentine Patagonia.American Museum Novitates 3536: 1–18.

Mitchell, E.D. 1989. A new cetacean from the LateEocene La Meseta Formation, Seymour Island,Antarctic Peninsula. Canadian Journal ofFisheries and Aquatic Sciences 46: 2219–2235.

Moss-Salentijn, L., and M.L. Moss. 1975. Studieson dentin. 2. Transient vasodentin in the incisorteeth of a rodent (Perognathus longimembris).Acta Anatomica 91: 386–404.

O’Leary, M.A. 1998. Phylogenetic and morpho-metric reassessment of the dental evidence for amesonychian and cetacean clade. In J.G.M.Thewissen (editor), The emergence of whales,evolutionary patterns in the origin of Cetacea.New York: Plenum Press, 133–162.

Ørvig, T. 1951. Histologic studies of placodermsand fossil elasmobranches. 1: The endoskeletonwith remarks on the hard tissues of lowervertebrates in general. Arkiv for Zoologi 2:321–454.

Owen, R. 1840–1845. Odontography, or, a treatiseon the comparative anatomy of the teeth; theirphysiological relations, mode of development,and microscopic structure, in the vertebrateanimals. London: Balliere.

Pujos, F., G. De Iuliis, C. Argot, and L. Werdelin.2007. A peculiar climbing Megalonychidaefrom the Pleistocene of Peru and its implicationfor sloth history. Zoological Journal of theLinnean Society 149: 179–235.

Reguero, M.A., S.A. Marenssi, and S.N.Santillana. 2002. Antarctic Peninsula andSouth America (Patagonia) Paleogene terres-trial faunas and environments: biogeographicrelationships. Palaeogeography, Palaeoclima-tology, Palaeoecology 179: 189–210.

Simpson, G.G. 1932. Enamel on the teeth of anEocene edentate. American Museum Novitates567: 1–4.

Tejedor, M.F., et al. (2009). New Early Eocenemammalian fauna from western Patagonia,Argentina. American Museum Novitates3638: 1–43.

Vizcaıno, S.F. 2009. The teeth of the ‘‘toothless’’:novelties and key innovations in the evolutionof xenarthrans (Mammalia, Xenarthra). Paleo-biology 35: 343–366.

Vizcaıno, S.F., and G.J. Scillato-Yane. 1995. AnEocene tardigrade (Mammalia, Xenarthra)from Seymour Island, West Antarctica.Antarctic Science 7: 407–408.

Vizcaıno, S.F., R. Pascual, M.A. Reguero, and F.J.Goin. 1998. Antarctica as a background formammalian evolution. In S. Casadıo (editor),Paleogeno de America del Sur y de la PeninsulaAntartica, Asociacion Paleontologica Argen-tina, Publicacion Especial 5. Buenos Aires,199–209.

Warwick, R., and P.L. Williams. 1973. Gray’sanatomy. 35th British ed. Philadelphia: Saunders.

APPENDIX 1COMPARATIVE SET

Cingulata, Dasypodidae, EuphractinaeChaetophractus vellerosus AMNHM 246457

(Holocene); Bolivia; lower molariform cross sectionEuphractus sexcinctus AMNHM 100075 (Holo-

cene), zoo specimen; upper molariform crossand longitudinal sections

Tardigrada, Megatheria, MegalonychidaeCholoepus hoffmanni AMNHM 173551 (Holocene);

zoo specimen; upper caniniform cross section

Megalocnus rodens AMNHVP FM 143454 (L.Pleistocene/Holocene); Cuba; lower molariformcross section

Neocnus gliriformis AMNHVP 143453 (L. Pleistocene/Holocene); Cuba; molariform cross section

Acratocnus antillensis AMNHVP 143452 (L.Pleistocene/Holocene); Cuba; upper caniniformcross section

Tardigrada, Mylodonta, MylodontidaeGlossotherium sp. AMNHVP 143455 (L.

Pleistocene); Trinidad; molariform (root end)cross section


This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

Complete lists of all issues of the Novitates and the Bulletin are available at World Wide

Web site http://library.amnh.org/pubs. Inquire about ordering printed copies via e-mail from

[email protected] or via standard mail from: American Museum of Natural History,

LibraryÐScientific Publications, Central Park West at 79th St., New York, NY 10024. TEL:

(212) 769-5545. FAX: (212) 769-5009.

reinterpretation of a middle eocene record of tardigrada (pilosa, xenarthra, mammalia) from la...

Documents