Neither a Rodent nor a Platypus: a Reexamination of Necrolestes patagonensis Ameghino

2007·9.6 MB·English

by  Robert J. Asher

PUBLISHED BY THE AMERICAN MUSEUM OF NATURAL HISTORY
CENTRAL PARK WEST AT 79TH STREET, NEW YORK, NY 10024
Number 3546, 40 pp., 21 figures, 1 table January 15, 2007
Neither a Rodent nor a Platypus: a Reexamination of
Necrolestes patagonensis Ameghino
ROBERT J. ASHER,1 INE´S HOROVITZ,2 THOMAS MARTIN,3
AND MARCELO R. SA´NCHEZ-VILLAGRA4
ABSTRACT
We present new descriptions, figures, and interpretations of well-preserved cranioskeletal
elements of the early Miocene, fossorial mammal Necrolestes patagonensis from Argentina. As
previously noted, this animal is highly apomorphic and its phylogenetic affinities are difficult to
interpret.Itsderivedfeaturesincludehypsodontuppercheekteeth,apartiallyfusedcervicalspine,
anunfusedatlasconsistingofseparaterightandleftmassae,andadistallyossifiedflexortendonof
the forearm. Characters that support its status as a therian mammal include a coiled cochlear
housingof theinnerear. Consistentwithitsstatusasa metatherian isthepresence offive upper
incisors, transverse canal foramina, and a broad proximal fibula. However, we cannot confirm
other characters claimed by previous researchers as evidence for affinity with marsupial or
nonplacentalmammals,suchasthepresenceofaninflectedmandibularangleandepipubicbones.
Furthermore,Necrolestesshowssimilaritiestoeutherianmammals,suchassmallincisiveforamina
and possibly three molars. On biogeographic and some anatomical grounds, identification of
Necrolestes as a metatherian remains a compelling option. However, pending a combined-data
phylogenetic analysis encompassing Theria and accounting for the anatomical diversity of
Necrolestes,possible membershipin Eutheria shouldnotbe ruledout.
1Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, United Kingdom
([email protected]).
2DepartmentofEcologyandEvolutionaryBiology,UniversityofCalifornia,621CharlesYoungDr.S.,LosAngeles,
CA90095,USA([email protected]).
3Institutfu¨rPala¨ontologie,Universita¨tBonn,Nussallee8,53115Bonn,Germany([email protected]).
4Palaeontologishes Institut and Museum, Karl Schmid-Strasse 4, CH-8006 Zu¨rich, Switzerland (m.sanchez@pim.
unich.ch).
CopyrightEAmericanMuseumofNaturalHistory2007 ISSN0003-0082
2 AMERICAN MUSEUMNOVITATES NO. 3546
INTRODUCTION Peterson specimens. We do not repeat all of
their descriptions, but point out issues of
Necrolestes patagonensis was first described disagreement between the two authors, our
by Florentino Ameghino in 1891 from the
view of who is correct, and describe several
early Miocene Santa Cruz beds of Patagonia.
anatomical regions not discussed by either
Ameghino, followed by Scott (1905), com-
author. In addition, we present the first
pared this animal favorably with African
detailed photographs of these fossils ever
golden moles (Chrysochloridae). It has sub-
published.
sequently been associated with marsupial
Like Plesiorycteropus (MacPhee, 1994),
moles (Leche, 1907), didelphoids (Winge,
Necrolestes has posed an enigma for multiple
1941;Patterson,1958),palaeanodonts(Saban,
generations of paleontologists. Despite the
1954), xenarthrans (McDowell, 1958), amer-
high quality of its osteological remains,
idelphian marsupials (Szalay, 1994: 346), and
making sense of its relationships within the
even as a middle Cenozoic holdover of the
framework of modern mammal diversity has
Gondwanatheria (Van Valen, 1988). Some
provendifficult.Intermsoftheattentionpaid
authors have doubted the marsupial affin-
to it by scientists over the past 100 years,
ities of Necrolestes, but have not offered an
Necrolestes shows an interesting resemblance
alternativeplacement(Turnbull,1971;Archer,
to the Malagasy subfossil Plesiorycteropus:
1984). The initial publication (Ameghino,
both were first named in the closing years of
1891)didnotincludefiguresorrefertoatype
the 19th century; both remained in obscurity
specimen, but based on Ameghino’s descrip-
for much of the 20th century; and both were
tive notes (p. 303), the material first described
the subjects of seemingly conclusive revisions
consisted of a relatively complete lower jaw
by Bryan Patterson (1958 for Necrolestes,
with an intact mandibular condyle. Shortly
1975 for Plesiorycteropus). Like MacPhee
after Ameghino’s initial work, more complete
(1994) did for Plesiorycteropus, we too take
material of Necrolestes was collected by J. B.
issue with some of the conclusions made by
Hatcher and O.A. Peterson between 1896
Patterson, in our case regarding Necrolestes.
and 1899 consisting of partial skeletons of
three individuals (Scott, 1905). Since then,
expeditions from the MACN–La Plata and INSTITUTIONAL ABBREVIATIONS
Duke University have recovered more BMNH The Natural History Museum
material (e.g., Goin et al., in press), but the London,United Kingdom
skeletons first described by Scott (1905) re- TM Transvaal Museum Pretoria, South
main the most complete yet discussed in the Africa
literature. YPM-PU Yale Peabody Museum Princeton
Universitycollection, USA
‘‘Insectivoran’’ (i.e., lipotyphlan and tenre-
ZIUT Zoologisches Institut Universita¨t
coid sensu Asher, 2005) and palaeanodont
Tu¨bingen, Germany
placental mammals are largely absent
ZMB ZoologischesMuseumBerlin,Germany
throughout the South American Cenozoic.
Among these taxa, only the single genus
ANATOMICAL DESCRIPTIONS
Cryptotis (Soricidae) has had a post-Pliocene
distribution in northern South America. Due CRANIUM
in part to the biogeographic enigma that
would be caused by the presence of either DENTITION: Scottgavetheformula4.1.3.3/
group in the South American Miocene, most 4.1.3.3; Patterson suggested 5.1.2.4/4.1.2.4.
recent authors have followedPatterson (1958) Regarding the incisor count, Patterson was
in identifying Necrolestes as a marsupial. correct. Posterior totheintactupperI1sthere
In this paper we reassess the status of are clearly four alveoli, including a partial
Necrolestes, and focus on the hypothesis that right I5 in YPM-PU 15065 (fig. 1). In
thisextinctSouthAmericanmammalisinfact addition, the left side of YPM-PU 15699
a metatherian. Both Scott (1905) and preserves the anterior four single-rooted in-
Patterson (1958) provided fairly detailed cisors, with a root fragment of left I5 just
descriptions of certain parts of the Hatcher- anterior to the canine; on the right side I5
2007 ASHER ETAL.: ARGENTINE FOSSORIALMAMMAL NECROLESTES 3
is entirely intact (fig. 2; Patterson, 1958: 6). and concave dorsally.’’ Ameghino (1891: 303)
Identification of incisors vs. more posterior alsodescribedtheangularprocesswithasmall
teethispossibleusingmorphologyandorigin, and internally directed inflection. However,
with incisors rooted at least partially in the we do not regard this kind of inflection to be
premaxilla, and canines, premolars, and mo- comparabletothatof,forexample,didelphids
lars in the maxilla. YPM-PU 15065 shows or borhyaenids. Rather, it shows some re-
breaks just anterior to the upper canines, semblance to that of Gypsonictops (Clemens,
within the fossa that receives each lower 1973: figs. 2, 3). In fact, and somewhat
canine, which we believe correspond to the ironically, the dentary of the placental
premaxilla-maxilla boundary (fig. 1). All of afrotherian Chrysochloris shows a much more
the aforementioned upper incisors occur prominent, medially directed angular process
withintheregionweidentifyasthepremaxilla. than that of Necrolestes (fig. 3), considerably
Lower incisors are more difficult to homolo- broader than that of the marsupial mole
gize. Of the anterior four lower teeth (i.e., the Notoryctes (fig. 3D). The morphology of the
putative incisors), the first is the largest, chrysochlorid angular process is no doubt
showing a root extending approximately to influenced by its articulation with the hyoid
the base of the jaw below the lower canine. apparatus (Bronner, 1991).
The succeeding three teeth are small and non- Scott(1905:369)suggestedthatthedentary
hypsodont,andinYPM15699appeartoshow of Necrolestes is ‘‘exceedingly’’ like that of
enamel restricted to the crown, in contrast to Chrysochloris. Because the coronoid process
Necrolestes’hypsodontcheekteeth.Wefollow of the YPM-PU 15384 Necrolestes dentary
both Scott and Patterson in identifying the (fig. 3B) is broken off close to its base, it
pair of hypertrophied, trenchant lower ante- shows a somewhat chrysochlorid-like form.
molars as canines based on their form. However, not only is the coronoid process
The best criterion to distinguish molars prominent in an intact specimen (YPM-PU
frompremolarsistoothreplacement,afeature 15699), but it also shows an additional,
for which we unfortunately do not have any posteriorly directed process close to its apex
direct evidence in the YPM-PU specimens. (fig. 3A). Furthermore, compared with the
However, new material described by Goin et jaw of Chrysochloris (fig. 3C), Necrolestes has
al. (in press) shows an erupting premolar a more gracile mandibular angle, a prominent
anterior to three more posterior cheek teeth, masseteric fossa, a more inferiorly situated
suggesting that Scott (1905) was correct in mandibular foramen, and a shorter, dorso-
identifying three molars. We would therefore ventrally deeper mandibular symphysis.
argue that the correct dental formula for Hence, we disagree with Scott’s assessment
Necrolestes is 5.1.3.3/4.1.3.3. of similarity in jaw structure between the two
A break in the maxilla posterolateral to the taxa.
lacrimal foramen on the right side of YPM- CUSP HOMOLOGIES: Patterson (1958:4) re-
PU 15065 exposes the dorsal extremes of two ferredtothemaincuspoftheuppermolarsof
posterior cheek teeth (M1 and M2 following Necrolestes as ‘‘presumably paracones,’’ an
Goin et al., in press). These teeth are interpretation that at first glance might be
hypsodont, without roots, and retain enamel sharedbymanycontemporarypaleomammal-
on their lingual aspect dorsally into the ogists.However,wenotethatthemaincuspin
maxilla (fig. 1D). When viewed from an molars of zalambdodont marsupials such as
anterior perspective, the cheek teeth of Notoryctes has been interpreted to be the
Necrolestes are curved from dorsal apex to metacone, not the paracone (Archer et al.,
occlusal surface, showing a lateral concavity. 2000; Long et al., 2002: 67; Asher and
From a lateral perspective the cheek teeth are Sa´nchez-Villagra, 2005). Similarly, Goin and
also curved, showing a posterior concavity Candela(2004)recentlydescribedtheputative
(fig. 1C). metatherian genus Kiruwamaq based on a
JAW: Patterson (1958: 8) suggested that zalambdodont-like upper cheek tooth, show-
although the angular process was small, it ingalargemetaconeanddiminutiveparacone.
‘‘bears an internal ledge and this is inflected Occlusal relations can elucidate some of these
4 AMERICAN MUSEUMNOVITATES NO. 3546
2007 ASHER ETAL.: ARGENTINE FOSSORIALMAMMAL NECROLESTES 5
Fig. 1. Continued.
Fig.1. SkullofNecrolestesYPM-PU15065instereo(A)dorsal,(B)ventral,(C)lateralviewswith(D)
closeup of dorsal maxillary region showing hypsodont, rootless cheek teeth. Arrows in C and D indicate
locationofinternal temporal space andcourse ofinfraorbital canal, respectively. Scalebars5 5mm.
6 AMERICAN MUSEUMNOVITATES NO. 3546
Fig. 2. Necrolestes YPM-PU 15699 in stereo (A) dorsal, (B) ventral, and (C) lateral views, with inset
showinganterior fragment of nasalbone in dorsalview.Scale bars55 mm.
2007 ASHER ETAL.: ARGENTINE FOSSORIALMAMMAL NECROLESTES 7
Fig. 2. Continued.
cusp homologies: for example, the paracone molar ofthesame tooth locus.Weregard this
occludesinoradjacenttothelowerectoflexid, occlusal pattern as suggestive that the main
lateral to the talonid basin; the protocone upper molar cusp in Necrolestes is the
occludes within the talonid basin; and com- metacone, not the paracone.
pared to the paracone, the metacone occludes ENAMEL MICROSTRUCTURE: An isolated
closer to the paracristid of the more posterior rightmolarfromYPM-PU15384wasembed-
lower molar (e.g., M1 metacone-m2 paracris- ded in resin and sectioned longitudinally and
tid). With the caveat that occlusal relations transversely to obtain information about its
across tooth loci are obviously contingent enamel microstructure. Enamel terminology
upon jaw mobility, and that noninterlocking follows Koenigswald and Sander (1997a) and
upper and lower teeth are not as canalized as Martin(1999a,b);forpreparationtechniques,
those that interlock (Polly et al., 2005), we see Martin (2004). The schmelzmuster of
tentatively note that the main upper cusp of Necrolestes (fig. 4) was compared to that of
Necrolestes could be the metacone. As dis- various Mesozoic taxa, as well as to data on
cussed by Asher and Sa´nchez-Villagra (2005: small placentals and marsupials drawn from
fig. 2), the main upper cusp of each molar in a large body of literature (see Koenigswald
YPM-PU 15699, which preserves uppers and and Sander [1997b] and literature cited there-
lowers inocclusion,showscloser proximityto in). In addition, we compared the pattern of
the preparacristid of the next most-posterior Necrolestes with that of sectioned right upper
lowermolarthantoanystructureonthelower molars of Chrysochloris (ZMB 76872),
8 AMERICAN MUSEUMNOVITATES NO. 3546
Notoryctes (ZIUT SZ10068), and Potamogale
(ZMB 71592).
The enamel distribution on the Necrolestes
upper molar is very asymmetrical. Enamel is
thickest (140 mm) on the lingual side and very
thin or even missing on the labial side. In the
longitudinal section (fig. 4A) two zones are
evident. The inner zone consists of radial
enamel with the prisms inclined about 40u
apically. After three fifths of the enamel
thickness, prisms turn simultaneously in hor-
izontal direction (inclination 0u), forming an
outer layer of tangential enamel. Prisms and
interprismaticmatrix(IPM)runparallelinthe
outer zone and therefore are difficult to
distinguish. Prisms and IPM seem to be at
least partially confluent, forming a prismless
external layer (PLEX). In the cross section
(fig. 4B), prisms are cut transversely in the
inner zone and longitudinally in the outer
zone. In this section the distinction between
prismsandIPMintheouterzoneissomewhat
moreclear.Fromthecrosssectionitisevident
that the PLEX apparently is restricted to the
outer 10–20 mm of the enamel layer. The
radialenamelofNecrolesteswithasingleturn
of prisms is very similar to that of the other
small placentals and marsupials that have
been studied.
In longitudinal section (fig. 4C),the marsu-
pialmoleNotoryctesexhibitsaschmelzmuster
with radial enamel and steeply apically in-
clined prisms in the inner zone and a simulta-
neous antapical turn of prisms in the outer
zone.Inthecrosssection(fig. 4D),theprisms
arecutobliquelytransverselyintheinnerzone
and almost longitudinally in the outer zone;
the thick IPM is well exposed. The golden
mole Chrysochloris has a very similar
schmelzmusterwith radial enamel intheinner
zone and tangential enamel in the outer zone;
prisms and IPM tend to be confluent. In the
longitudinal section of the protocone, the
radial prisms of the inner zone are cut
longitudinally (fig. 4E), and in the cross
section they are cut transversely (fig. 4F).
r
Fig.3. Mandiblesof(A,B)Necrolestespatago-
nensis (A shows coronoid process of YPM-PU Notoryctes typhlops BMNH 39.4210 in external
15699; B shows jaw of YPM-PU 15384), (C) (top), internal (middle) and occlusal (bottom)
Chlorotalpa leucorhina ZMB 31505, and (D) views. Scalebars5 5mm.
2007 ASHER ETAL.: ARGENTINE FOSSORIALMAMMAL NECROLESTES 9
The rounded-prism cross sections are sur- ectotympanic. They nevertheless clearly lack
rounded by thick IPM. In the outer zone, alisphenoid, basisphenoid, and petrosal con-
prisms and IPM are confluent and hardly tributions to an ossified bulla. Terrestrial
distinguishable. The otter shrew Potamogale mammals with prominent, ossified ecto- or
has a generally similar schmelzmuster with entotympanic bullae tend to have strong
apically inclined prisms in the inner zone and coossifications and/or articulation scars, such
a simultaneous antapical turn in the outer that the contribution of tympanic elements to
zone(figs. 4G,4H).PrismsandIPMaremore a bulla can often be inferred. Since the
clearly distinct in the outer zone than in the basicranial remains of Necrolestes show no
other taxa studied. sign of an ecto- or entotympanic bulla, we
The schmelzmuster of Necrolestes and the agree with Patterson (1958) that it lacked an
other taxa studied here represent the plesio- ossified bulla.
morphic therian schmelzmuster type that is CTscansoftheinnerearofYPM-PU15699
typical for small marsupials (Peradectia and and15384indicatethatthecochleawascoiled,
Didelphimorphia) and placental insectivorans makingnearlyafullturn.Whilethisdegreeof
(Koenigswald,1988,1997a,b;Koenigswaldet coiling is not as tight as in some placental
al., 1987; Koenigswald and Goin, 2000). A mammals, it is considerably more than that
very similar schmelzmuster has also been seen in monotremes. Interestingly, hedgehogs,
detected in the molars of various Mesozoic sea cows, vombats, and the marsupial mole
non-therian taxa (Wood and Stern, 1997; Notoryctesareamongtheonlyextanttherians
Wood et al., 1999) as well as in the incisors that also show reduced cochlear coiling
of zalambdalestids, pseudictopids, and early (Gray, 1907; Sa´nchez-Villagra and Schmelzle,
gliriforms such as Eurymylus and Eomylus in press).
(Martin, 1999a, b, 2004). Evidently derived Adult monotremes, didelphids, dasyurids,
characters of gondwanatheres (Koenigswald and other metatherians such as Pucadelphys
et al., 1999), such as an IPM that runs at and Deltatheridium, but not Dromiciops,
a distinct angle to the prisms in the radial Mayulestes, or borhyaenids, possess a prootic
enamel of themolars and prisms increasing in canal (Wible and Hopson, 1995; Wible et al.,
thicknesstowardstheexteriorofthetooth,are 2001). In didelphids, this structure transmits
notevidentinourobservationsofNecrolestes. venousbloodfromtheprooticsinus(Sa´nchez-
The schmelzmuster of Necrolestes represents Villagra and Wible, 2002), passes through an
the plesiomorphic therian condition that osseous canal connecting the ventrum of the
characterizes many small placentals and mar- pars canalicularis of the petrosal and a small
supials. foramen immediately lateral to the secondary
BASICRANIUM: Scott(1905:368)referredto facialforamen,andleavesthetympanicregion
an ‘‘ossified and moderately inflated’’ audito- posteroventrally adjacent to the facial nerve.
ry bulla. However, we agree with Patterson Except for petrosals assigned to the eutherian
(1958) that Scott was actually referring to the taxon Prokennalestes (Wible et al., 2001) and
ventrum of the pars cochlearis itself. None of specimens assigned to the taxonomically
the Necrolestes material shows any sign of ambiguous ‘‘zhelestids’’ (Ekdale et al., 2004),
ossifications enclosing the middle ear (figs. 1, the prootic canal has not been explicitly
2).Infairness,theYPM-PUspecimensdonot documented in any eutherian mammal. The
have completely intact, articulated basicrania. anteroventral parscanalicularisof Necrolestes
The most complete is on 15699, which has is smaller than that of didelphids and shows
a petrosal still articulated with the skull, no sign of a prootic canal (fig. 5).
bounded laterally by the jaw joint and Wibleetal.(2001:character153)notedthat
anteromedially by the basisphenoid; however, the state of the internal acoustic meatus
regions posterior and medial to the petrosal (‘‘shallow, with thin prefacial commissure’’)
are missing (fig. 2). YPM-PU 15384 preserves comprises a synapomorphy forEutheriain an
a fragment consisting of petrosal, squamosal, analysis of selected crown and stem Theria,
and a part of the still-articulated mandibular focusingonextincttaxa(Rougieretal.,1998).
condyle(fig. 5).Neitherspecimenpreservesan However, they noted in addition that the
10 AMERICAN MUSEUMNOVITATES NO. 3546
Fig.4. Scanningelectronmicroscope(SEM)photosofsectionedandetchedrightM1sshowingenamel
microstructure of(A, B)Necrolestes(YPM-PU 15384;occlusalsurfaceto theleftinA),(C, D)Notoryctes
(ZIUT SZ10068; tip of protocone to the right in C), (E, F) Chrysochloris (ZMB 76872), and (G, H)
Potamogale (ZMB 71592; occlusal surface to the top in G). Longitudinal sections are listed in the left