Historical Biology

An International Journal of Paleobiology

ISSN: 0891-2963 (Print) 1029-2381 (Online) Journal homepage: https://www.tandfonline.com/loi/ghbi20

Tribal allocation and biogeographical significance

of one of the largest sigmodontine rodent, the
extinct Galápagos Megaoryzomys (Cricetidae)

Christophe Ronez, Jorge Brito, Rainer Hutterer, Robert A. Martin & Ulyses F.
J. Pardiñas

To cite this article: Christophe Ronez, Jorge Brito, Rainer Hutterer, Robert A. Martin & Ulyses
F. J. Pardiñas (2020): Tribal allocation and biogeographical significance of one of the largest
sigmodontine rodent, the extinct Galápagos Megaoryzomys (Cricetidae), Historical Biology

To link to this article: https://doi.org/10.1080/08912963.2020.1752202

View supplementary material

Published online: 27 Apr 2020.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ghbi20
HISTORICAL BIOLOGY
https://doi.org/10.1080/08912963.2020.1752202

ARTICLE

Tribal allocation and biogeographical significance of one of the largest sigmodontine

rodent, the extinct Galápagos Megaoryzomys (Cricetidae)
a b
Christophe Ronez , Jorge Brito , Rainer Huttererc, Robert A. Martin d
and Ulyses F. J. Pardiñas a,b

a
Instituto de Diversidad y Evolución Austral (Ideaus-conicet), Puerto Madryn, Chubut, Argentina; bInstituto Nacional de Biodiversidad (INABIO), Quito,
Ecuador; cZoologisches Forschungsmuseum Alexander Koenig (ZFMK), Bonn, Deutschland; dDepartment of Biological Sciences, Murray State University,
Murray, KT, USA

ABSTRACT ARTICLE HISTORY

One of the largest members of Sigmodontinae, the extinct Megaoryzomys curioi from Santa Cruz Island in the Received 20 November 2019
Galápagos Archipelago (Ecuador), is traditionally treated as a representative of Thomasomyini. We contrasted Accepted 2 April 2020
this hypothesis based on a direct study of craniodental material of M. curioi, including a well-preserved skull KEYWORDS
assessed through CT-scan examination, within a broader sampling of brachydonts sigmodontines. M. curioi Thomasomyini; Oryzomyini;
lacks or presents a poorly expressed suspensory process of the squamosal bone, while in Thomasomyini this gigantism; insular evolution;
structure is well developed. In the same way, M. curioi does not exhibit the condition in the dorsal aperture of anatomy
the ectotympanic displayed by the Thomasomyini, and the morphology of the mandible also contrasts with
the widespread condition shown by Thomasomyini. Besides, the procingulum of the first lower molar in
M. curioi lacks the typical pattern of Thomasomyini composed of two defined conulids. Overall craniodental
morphology in M. curioi favours its allocation to the Oryzomyini. This alternative hypothesis resolves two
biogeographical issues connected with the target genus: (1) Galápagos Islands were colonised by members of
a single sigmodontine tribe (i.e., Oryzomyini) and (2) within the sigmodontine radiation; members of the
Oryzomyini were uniquely capable of reaching oceanic islands.

Introduction 1983; Weksler 2006; Dowler 2015; Prado and Percequillo 2018). If
Megaoryzomys is a thomasomyinin, it would imply that at least two
The Galápagos Islands’ Megaoryzomys curioi is considered one of the
tribes of sigmodontines colonised the archipelago. In addition,
largest sigmodontine rodent based on craniodental measurements
since living thomasomyinins are mostly Andean forms (Pacheco
(Steadman and Ray 1982; Turvey et al. 2010; Pardiñas et al. 2017). It
2003; Pacheco et al. 2015; Pardiñas et al. 2017), Megaoryzomys
was originally described under Megalomys, with the single species
considered as a Thomasomyini implies a major geographic exten-
M. curioi, based on materials recovered in Santa Cruz Island of the
sion for the tribe. Although both conditions are possible, a more
Galápagos Archipelago (Niethammer 1964; Hutterer and Oromí
parsimonious conclusion is that Megaoryzomys is an oryzomyinin.
1993). A revision including new specimens prompted Lenglet and
We examined this hypothesis based on direct inspection of
Coppois (1979) to describe the giant rat from Santa Cruz Island as
Megaoryzomys craniodental material, including a well-preserved
a new genus Megaoryzomys. At the same time, these authors treated it
skull. The results were then incorporated into a new biogeographi-
as a member of Oryzomyini. Steadman and Ray (1982) argued against
cal model limiting colonisation of the Galápagos to the Oryzomyini.
this tribal allocation and moved Megaoryzomys to the Thomasomyini,
a group coined by themselves. In almost four decades since
then, Megaoryzomys has received little attention and its suprageneric
affiliation freely varied between Oryzomyini, Thomasomyini or Material and methods
Sigmodontinae incertae sedis (Table 1). Although two species are Most of the available material belonging to Megaoryzomys is housed
referred to the genus, only one, M. curioi (or Curio’s giant rice rat), at the Smithsonian National Museum of Natural History (USNM;
was named (Steadman and Ray 1982). The other, treated as Washington D. C., USA) and was secured by D. Steadman during
Megaoryzomys sp., corresponds to populations from Isabela Island the excavation of several cave deposits (Steadman and Ray 1982;
(Steadman and Ray 1982; Steadman et al. 1991; Turvey 2009). The Steadman et al. 1991). We repeatedly requested access to these
genus is restricted to the Holocene and was apparently extirpated in specimens but failed because they were apparently involved in
historical times (Steadman et al. 1991; Turvey 2009). a long-term loan (M. Brett-Surman, pers. comm. to UFJP). We
Two biologically interesting issues are connected with the allo- also inquired about the unnumbered specimens studied by
cation of Megaoryzomys to the Thomasomyini. Other sigmodontine Lenglet and Coppois (1979), which belong to the collections of
rodents recorded from the Galápagos include Aegialomys galapa- Royal Belgian Institute of Natural Sciences (RBINS, Brussels,
goensis (Waterhouse, 1839), Nesoryzomys darwini Osgood, 1929 Belgium). In a study-visit made to the USNM collections, one of
(probably extinct), Nesoryzomys fernandinae Hutterer and Hirsch, the authors (CR) examined a limited sample of Megaoryzomys
1980, Nesoryzomys indefessus (Thomas, 1899) (probably extinct), curioi from Cueva de Kubler (excavation IIC), composed of isolated
Nesoryzomys narboroughi Heller, 1904, and Nesoryzomys swarthi molars and some fragmentary mandibles. We also examined a large
Orr, 1938; all six undisputed oryzomyinins (e.g., Patton and Hafner cranium fragment, almost complete but lacking the auditory

CONTACT Christophe Ronez [email protected]

Supplemental data for this article can be accessed here.
© 2020 Informa UK Limited, trading as Taylor & Francis Group

Published online 27 Apr 2020

2 C. RONEZ ET AL.

Table 1. Tribal allocation of Megaoryzomys since its original description.

Author Tribe Related taxa/Remarks
Niethammer (1964) Oryzomyini Megalomys
Lenglet and Coppois (1979) Oryzomyini Nesoryzomys, Oryzomys
Steadman and Ray (1982) Thomasomyini Thomasomys, Rhipidomys
Eisenberg (1989) Oryzomyini -
Hutterer and Oromí (1993) Incertae sedis Highly differentiated genus
Leo and Gardner (1993) Thomasomyini Thomasomys
Tribe (1996) “plesiomorphic Neotropical muroids”a -
McKenna and Bell (1997) Thomasomyini -
Pacheco (2003) Oryzomyini Noronhomys
Musser and Carleton (2005) Incertae sedis -
Pardiñas et al. (2017) Oryzomyini or Thomasomyini -
a
After Voss (1993:25)

capsules and the jugals, and a right dentary of M. curioi from Cueva We focused exclusively on craniodental traits potentially indicating
del Cascajo, both housed at the collection of the Zoologisches tribal affiliation; several of them are here described for the first time. In
Forschungsmuseum Alexander Koenig (ZFMK; Bonn, Germany), addition, in an attempt to advance a morphology-based phylogeny
originally described and figured by Hutterer and Oromí (1993; the including Megaoryzomys, we incorporated the genus into the extensive
remaining specimens studied by this team have been either data matrix provided by Weksler (2006; with the taxonomy updated
returned to a Galapagoan Institution or deposited in the Canary after Weksler et al. 2006). We performed the phylogenetic analysis
Island). A large sample of sigmodontine belonging to the tribes using maximum parsimony with the TNT software (Goloboff et al.
Wiedomyini, Oryzomyini, Thomasomyini and several sigmodon- 2008), conducting a traditional search with tree bisection reconnection
tines of uncertain tribal affiliation (e.g., Abrawayaomys, Delomys) (TBR) using 100 random addition sequences, saving 10 trees per round
was used for comparison, based on material from the under equal weight.
following collections of mammals: Colección de Mamíferos del
Centro Nacional Patagónico (CNP; Puerto Madryn, Argentina);
Colección Nacional de Mastozoología, Museo de Ciencias Results
Naturales ‘Bernardino Rivadavia’ (MACN; Buenos Aires,
Cranium
Argentina); Colección de Mamíferos del Instituto Nacional de
Biodiversidad (MECN; Quito, Ecuador); Colección de Mamíferos Megaoryzomys has a markedly short palate, with the anterior border of
del Instituto de Ciencias Biológicas de la Escuela Politécnica the mesopterygoid fossa reaching the plane defined by the anterior
Nacional (MEPN; Quito, Ecuador); Museo Nacional de Historia faces of the M3s (Figures 1 and 2(a)). The length of the palate is
Natural de Montevideo (MNHM; Montevideo, Uruguay); Museu a variable trait in Oryzomyini. Although Voss and Carleton (1993,
de Zoologia, Departamento de Biologia Animal, Universidade p. 31) stated ‘long palate with prominent posterolateral pits’ as
Federal de Viçosa (MZUFV; Viçosa, Minas Gerais, Brazil); Museo a diagnostic feature for Oryzomyini and Weksler (2006, p. 34) indicated
Pontificia Universidad Católica del Ecuador (QCAZ; Quito, that long palates were the widespread condition in Oryzomyini,
Ecuador); Departamento de Ecologia e Zoologia, Universidade Percequillo et al. (2011) showed that short palates are also present in
Federal de Santa Catarina (UFSC; Florianópolis, Brasil). A list of this tribe (e.g., Drymoreomys; see also in Aegialomys, Prado and
this comparative material is presented in Table S1. Percequillo 2018). According to Pacheco (2003, p. 58) ‘. . . no tribe [of
We worked by direct comparison and examination of photographs sigmodontine] . . . can be characterized by a uniform palatal morphol-
and CT-scan models obtained by one of the authors (JB). The speci- ogy.’ Nevertheless, Wiedomyini and unique lines (Figure 2(b–h)) show
men ZFMK 2016_0981 (Figure 1; ‘Cráneo B’ of Hutterer and Oromí long palates, while living Thomasomyini have typically short palates
1993; the mandible lacks individual museum number) was scanned (except for Chilomys and Rhagomys; see Pacheco 2003, p. 58; Figure 2
with a micro-CT (Bruker Skyscan 1173) using a 70 kV voltage (i–l)). In addition, Megaoryzomys tends to display narrow palates,
source, with a resulting resolution of 29 μm. The raw data were although the typical condition is partially clouded by the dispropor-
reconstructed with NRecon (version 1.7.1.6, Bruker microCT), and tionate palatal expansion caused by the lingual roots of M1. Most
the 3D models were built with the CTvox software, version 3.0.0 r1114, Thomasomyini have broad palates (except Thomasomys aureus and
Bruker microCT (https://www.bruker.com). In addition, we examined Thomasomys macrotis; Pacheco 2003, p. 59), and the same is usually
a complete cranium with preserved left auditory capsule and associated true for Oryzomyini, but Weksler (2006, p. 111) discarded the trait as
mandible, the anterior part of a cranium with complete left zygomatic a tribal character, arguing continuous variation. Wiedomyinins are
arch, and a fragment of maxillary (RBINS 41840, ‘Specimen A’; RBINS typically narrow-palate forms (Pacheco 2003, p. 59), but Gonçalves
41841, ‘Specimen B’; and RBINS 41842, ‘Specimen C’ of Lenglet and et al. (2018) identified variation in this trait.
Coppois 1979, respectively). All three specimens are illustrated in Megaoryzomys has a massive squamosal bone; its posterior border
Figure S1. Cranial terminology employed in this contribution was varies between total absence (ZFMK 2016_0981) and an inconspic-
based mostly on Carleton and Musser (1989) while palate definitions uous expression (RBINS 41840, a very old individual) of the posterior
follow Hershkovitz (1962). Dental nomenclature follows Reig (1977) suspensory process to overlap the tegmen tympany (Figuress 1 and
and the ICAMER system recently proposed by Barbière et al. (2019), S1). Living Thomasomyini have well-developed suspensory processes
when necessary. Upper and lower molars are designated as M/m, (Figure 3(e–h)). The same is true for other brachydont sigmodon-
respectively. Figure S2 provides labelled diagrams of the M1/m1 of tines such as Wiedomyini (i.e., Juliomys, Phaenomys, Wiedomys and
M. curioi with both nomenclatures. The anatomy of Megaoryzomys has Wilfredomys; see Pardiñas et al. 2014; Gonçalves et al. 2018; Figure 3
been exhaustively described in three contributions (Lenglet and (i)), and unique lineages (Abrawayaomys, Delomys; Figure 3(d,j)).
Coppois 1979; Steadman and Ray 1982; Hutterer and Oromí 1993) One of the diagnostic traits indicated for Oryzomyini is ‘no posterior
and, for the sake of economy, many observations are not repeated here. suspensory process of the squamosal attached to the tegmen
 HISTORICAL BIOLOGY 3

Figure 1. Scanned specimen of Megaoryzomys curioi (ZFMK 2016_0981). Scale bar is 5 mm for both skull and jaw, 1 mm for the augmented tooth row.

tympany’ (Voss and Carleton 1993, p. 31; see also Pacheco 2003, the posterior ascending process, where the suture between alisphenoid and
p. 50; Weksler 2006, p. 40; Figure 3(a–c)). squamosal is straight (Figure 3(a–c)). A posterior ascending process is present
Megaoryzomys lacks an alisphenoid strut separating the bucci- in many other sigmodontines, where this suture is typically curved. The latter
nator-masticatory and accessory oval foramina, a condition pattern is observed in Thomasomyini, Wiedomyini Abrawayaomys and
reported as diagnostic of Oryzomyini (see Voss and Carleton Delomys (Figure 3(d–j)).
1993, p. 31). On the contrary, Thomasomyini typically have an According to the specimen RBINS 41840, Megaoryzomys has
alisphenoid strut, although moderate inter- and intraspecific varia- a ‘closed’ condition of the ectotympanic ring (Pacheco 2003), charac-
bility has been recorded (Pacheco 2003, p. 46). terised by the close proximity between the anterior part of the ectotym-
The posterior ascending process of the alisphenoid (sensu Myers et al. panic and the mastoid tubercle (Figure 4). Overall, the otic capsule of
1990, p. 20) is absent in Megaoryzomys (Figure 1). Oryzomyini similarly lack Megaoryzomys is minimally developed and flattened, but the auditory
4 C. RONEZ ET AL.

Figure 2. Ventral view of palates of a selected sample of Sigmodontinae. (a) Megaoryzomys curioi (ZFMK 2016_0981); (b) Oryzomys palustris (CNP 6491); (c) Sigmodontomys
alfari (MECN 6021); (d) Neacomys sp. (MEPN 12081); (e) Tanyuromys thomasleei (MECN 3407); (f) Delomys sublineatus (UFSC 711); (g) Abrawayaomys chebezi (CNP 3631); (h)
Wilfredomys oenax (MHNM 8112); (i) Rhagomys rufescens (MZUFV 3706); (j) Rhipidomys albujai (MECN 3791); (k) Thomasomys cinerus (QCAZ 16173); (l) Chilomys instans
(MECN 5854). Scaled to the same size.

meatus is huge. Pacheco (2003, p. 55; Pacheco et al. 2015) regarded the Dentary
‘open’ condition of the ectotympanic ring, where the dorsal aperture
The mandible of Megaoryzomys exhibits a comparatively straight
largely exposes the petrosal, as characteristic of the ‘Andean thomaso-
and rugged but small coronoid process that does not reach the
myine clade’ (including Aepeomys, Chilomys, Rhipidomys and
height of the condyle (Figure 1). Pacheco (2003, p. 250) stated
Thomasomys).
that most Thomasomyini have a long and falciform coronoid
 HISTORICAL BIOLOGY 5

Figure 3. Lateral view of the posterior region of the cranium of (a) Oryzomys palustris (CNP 6491); (b) Sigmodontomys alfari (MECN 6021); (c) Neacomys sp. (MEPN 12081); (d)
Delomys sublineatus (UFSC 711); (e) Rhagomys rufescens (MZUFV 3706); (f) Rhipidomys albujai (MECN 3791); (g) Thomasomys cinerus (QCAZ 16,173); (h) Chilomys instans
(MECN 5854); (i) Wilfredomys oenax (MHNM 8112); (j) Abrawayaomys chebezi (CNP 3631). Dashed line highlights the suture between squamosal and alisphenoid bones,
arrows point the suspensory process of the squamosal. Scaled to the same size.

process, whereas the presence of a shorter coronoid process may Megaoryzomys does not possess the ‘flag’ morphology in its cor-
vary among other clades. In concordance with most Oryzomyini, onoid process characterising many Thomasomyini (Figure 5(d,e)).
6 C. RONEZ ET AL.

Figure 4. Lateral view of the left auditory region of (a) Megaoryzomys curioi (RBINS 81840); (b) Sigmodontomys alfari (MECN 6021); (c) Thomasomys cinereus (QCAZ 16173).
Abbreviations: ect, ectotympanic; h, hamular; mt, mastoid tubercle; pgf, postglenoid foramen; psp, posterior suspensory process of squamosal; tt, tegmen tympani.
Arrows point the ectotympanic condition. Scaled to the same size.

Considerable variation is displayed by the coronoid process among change between unworn and worn M1s is the depth of penetration
studied forms, including a process with bifid ends (e.g. of the hypoflexus, overlapping both paraflexus and metaflexus. In
Wilfredomys; Figure 5(g)) or small projections (Abrawayaomys, addition, these labial flexi not totally surround the main labial cusps.
Figure 5(c)). In Delomys the coronoid process is higher than the The juvenile m1 also shows a complex occlusal pattern (Figure 6
condyle (Figure 5(a)). (k)). The procingulum appears as an entire structure composed of
The mandibular foramen of Megaoryzomys is large, elongated, two conulids (anterolabial and anterolingual) and two stylids (ante-
slightly hidden by the crista ramus mandibulae (sensu Răduleţ romedian stylid and protostylid). The respective arms of the con-
2007), and placed under the condylar process (Figure 1). In ulids and the extension of the area of each conulid and stylid shape
Thomasomyini and Wiedomyini this foramen is also elongated a rounded procingulum delimiting a deep central fossetid. The
but can be seen in lateral view, and never reaches the anterior connection with the postcingulum is central, but the structures
portion of the condyle (Figure 5(d–g)). Oryzomyinins have an involved other than the anterior arm of the protoconid are not
elongated mandibular foramen that is partially hidden by the crista obvious. There is no anterolophid. The mesolophid is isolated and
ramus mandibulae and is located at least under the anterior portion reaches the labial border. In advanced stages of wear, it may fuse
of the condylar foramen (Figure 5(b)). Abrawayaomys exhibits with the entoconid (Steadman and Ray 1982). The posterior arm of
a small and elongated mandibular foramen, whereas in Delomys the protoconid is separated from the anterior arm of the hypoconid
this foramen is wide and rounded (in both cases it does not reach by a proto-posterolophulid and a contribution of the entoconid
the level of the anterior part of the condylar process, Figure 5(a,c)). cuspal area. Lingual flexids are provergent but never surround the
Megaoryzomys has a developed chin (mental) process that is obser- lingual cusps. The hypoflexid is horizontal and its interior enamel
vable on both lateral and internal views (Figure 1); this condition is also border is opposite the base of the mesolophid.
recorded in extant Oryzomyini (Figure 5(i)). On the contrary, most of The basic M1 morphology of Megaoryzomys is also seen in many
the remaining studied taxa (Thomasomyini except for Rhagomys; brachydont and pentalophodont sigmodontines, including the
Wiedomyini; Abrawayaomys and Delomys) exhibit inconspicuous Thomasomyini (Figure 6(h–j)), Oryzomyini (Figure 6(b–d)),
chin processes when examined from internal view (Figure 5(i–l)). Wiedomyini (Figure 6(g)) and several incertae sedis genera (e.g.,
Delomys, Abrawayaomys; Figure 5(e,f)). Typically, among the mem-
bers of these taxa, the procingulum of M1 is divided into two conules
Dentition and a third, anterolingual conule, may also be present (e.g., Delomys).
Megaoryzomys has brachydont and pentalophodont molars, with the The enamel platform observed in Megaoryzomys is not seen in the
main cusps arranged in opposite (M1) or alternate (m1) pairs. On M1 Thomasomyini, but many Oryzomyini display a platform.
(Figures 1 and 6(a)), unworn individuals have a procingulum with In general appearance the upper molars of Megaoryzomys strongly
two well-defined conules, the labial smaller than the lingual, and an resemble those of the Thomasomyini, particularly when worn; this
anterior enamel platform; the latter structure lacks expression in likely explains why Steadman and Ray (1982) allied Megaoryzomys
worn animals but the separation between both conules persists, with Rhipidomys and Thomasomys. However, inspection of unworn
forming a classic anteromedian flexus. The M1 also has a wide and specimens reveals that connections point to a closer relationship with
long anteroloph and mesoloph, the latter reaching the labial tooth Oryzomyini and Delomys than to the Thomasomyini. In the M1 of
border and maintained free of fusion with other structures. As wear Megaoryzomys, the procingulum is connected to the protocone exclu-
advances the anteroloph merges with the labial conule, increasing the sively by the lingual conule. The same situation is observed in
occlusal surface of the latter and producing two conules similar in Oryzomyini and Delomys (Figure 6(b–e)). On the contrary, the proto-
size. A small posteroloph is present but quickly fuses with the meta- cone cuspal area on M1 is connected to the procingulum by both
cone with wear. The lingual conule connects with the proto- lingual and labial conules in Thomasomyini and Wiedomyini (Figure
anterolophule (ICAMER system; Figure S2). The protocone has 6(g–j)). Another difference of the M1 procingulum is the differential
a small, free posterior arm in juveniles (Figure 6(a)) and a paraloph expression of both conules. Although in adult Megaoryzomys these
that extends medially, ending anterior to the free, short posterior arm structures acquire approximately equal sizes and are still separated by
of the protocone. With advancing wear the paraloph connects an anteriomedian flexus (Figure 1), in unworn specimens the lingual
directly with the protocone, erasing all the small structures observed conule is distinctly larger, opposite to the condition recorded in
in unworn specimens. The hypocone is anteriorly extended by Thomasomyini (Pacheco 2003). Moreover, the anteroloph and
a hypo-anterolophule connecting to the protocone. A noticeable a mesoloph are slender in Thomasomyini and Wiedomyini, whereas
 HISTORICAL BIOLOGY 7

Figure 5. Internal view of the ascending ramus (a to g) and the diastema (h to l) of (a, h) Delomys sublineatus (UFSC 711); (b, i) Sigmodontomys alfari (MECN 6021); (c, j)
Abrawayaomys chebezi (CNP 3631); (d, k) Chilomys instans (MECN 5854); (e, l) Thomasomys aureus (QCAZ 16173); (f) Rhagomys rufescens (MZUFV 3706); (g) Wilfredomys
oenax (MHNM 8117). Scaled to the same size.

they are wider in Oryzomyini and Delomys (Figure 6(a–j)). The shape (Figure 6(k–n)). In the Thomasomyini, two conulids are typi-
of the anteroloph and mesoloph of Megaoryzomys more closely resem- cally well defined (less defined in Aepeomys; a single conulid in
bles that of the Oryzomyini rather than the Thomasomyini. Rhagomys), separated by an anteromedian flexid (Figure 6(r)).
The m1 pattern of Oryzomyini (Figure 6(l–n)), In Wiedomyini, the procingulum of m1 is composed of two or
Thomasomyini (Figure 6(r–t)), Wiedomyini (Figure 6(q)) and three aligned structures (as in Delomys). The connection
Delomys (Figure 6(o)) is similar, but some differences are between the procingulum and the postcingulum is central in
observed. The morphology observed in most Oryzomyini is Oryzomyini and Delomys, whereas it is shifted either lingually
more similar to Megaoryzomys, with the lack of an anterolophid (Wiedomyini, Rhipidomys, Thomasomys) or labially (Rhagomys)
and the presence of a rounded procingulum, with no clearly in the remaining taxa. Finally, the flexids of Megaoryzomys are
differentiated individual conulids enclosing a central fossetid not organised as in Thomasomyini or Wiedomyini, where the
8 C. RONEZ ET AL.

Figure 6. First upper (a to j) and lower (k to t) molars of (a, k) Megaoryzomys curoi (USNM 284210); (b, l) Oryzomys palustris (CNP 6491); (c, m) Cerradomys langguthi (CNP
337); (d, n) Lundomys molitor (MNHM 735, MNHM 780); (e, o) Delomys sublinateus (UFSC 711); (f, p) Abrawayaomys chebezi (CNP 3631); (g, q) Wilfredomys oenax (CNP 2378);
(h, r) Rhipidomys austrinus (CNP 6512); (i, s) Thomasomys baeops (MACN 3–20); (j, t) Rhagomys rufescens (MZUFV 3706). All teeth are oriented to be left (upper) or right
(lower), and scaled to same size.

mesoflexid and entoflexid are well-pronounced anteriorly and of the Neotominae, Tylomyinae, Thomasomyini and incertae sedis
may surround their respective cusps (metaconid and entoconid) sigmodontines, always appears as a unified clade, as well as mono-
in the latter tribes. phyly of the genera Euryoryzomys, Holochilus, Neacomys, Nectomys,
Nephelomys, Nesoryzomys, Oecomys, Oligoryzomys, Oryzomys, and
Zygodontomys. Some of the clades recorded by Weksler et al. (2006)
Phylogeny were recovered in our analysis, but the position of several taxa varied
from their study (e.g., Amphinectomys is not always in their group D;
We obtained 11 most parsimonious trees with 521 steps, and we Figures S3 and S4). Megaoryzomys was retrieved as sister to
present the strict consensus tree in Figure 7. The outgroup, composed Mindomys, a poorly known monotypic genus endemic from mid-
 HISTORICAL BIOLOGY 9

Discussion
Tribal affiliation
Musser and Carleton (2005, p. 1125) summarised the unstable
position of Megaoryzomys indicating ‘While not an oryzomyine
per se, as demonstrated by Steadman and Ray (1982), the relation-
ships and tribal affiliation of the unfortunately christened
Megaoryzomys deserve reconsideration within a broader sampling
of New World cricetids and from a cladistic perspective.’ All the
selected characters discussed above allow a better-supported
hypothesis for the tribal allocation of the giant Galápagos rat.
Megaoryzomys is characterised by the combination of a short
palate, the lack of a posterior ascending process of the alisphenoid
and the absence or inconspicuous expression of the suspensory
process of the squamosal. The jaw expresses a comparatively small
coronoid process that does not reach the height of the condyle,
a developed chin process (in internal view), and a large and elongated
mandibular foramen slightly hidden by the crista ramus mandibulae
(located under the condylar process). Molars are brachydont and
pentalophodont with two well-defined conules on M1 and
a procingulum composed of two conulids (with associated stylids)
on m1 encircling a central fossetid. Anteroloph and mesoloph are
present on M1 and the m1 displays a short mesolophid. Certain flexi/
ids are relatively shallow on M1/m1 because cusp lophids tend to be
horizontal, rather than anterior, in position, and the cusps lack
double connections with surrounding structures.
The Oryzomyini can be diagnosed by seven putative synapomor-
phies (Voss and Carleton 1993; Weksler 2006, 2015, p. 291–292): long
palate with prominent posterolateral pits; absence of an alisphenoid
strut; absence of a posterior suspensory process of the squamosal
attached to tegmen tympani; absence of a gall bladder; 12 thoracic
vertebrae; absence of a haemal arch on the first caudal vertebra; and
fewer than 36 caudal vertebrae. With these elements at hand, we pose
the hypothesis that Megaoryzomys is not a Thomasomyini but rather
a unique Oryzomyini. The proposed tribal allocation is not free of
potential conflict since palatal morphology has been a central element
in the conundrum of distinction among Oryzomyini, Thomasomyini
and several pentalophodont Sigmodontinae (Thomas 1906). The
importance of palatal morphology in the shaping of current ideas
about tribal configurations was reviewed by Voss (1993, p. 21–25).
According to this author, ‘a salient aspect of all the published literature
on the relationships of Thomasomys to other pentalophodont genera is
the paucity of characters mentioned: mammary counts, zygomatic
plate development, interorbital shape, and palatal morphology virtually
exhaust the list of cited attributes’ (Voss 1993, p. 23). Although clearly
he overlooked the specific contribution of Steadman and Ray (1982),
where the Thomasomyini was named and diagnosed (including
Megaoryzomys, Rhipidomys, and Thomasomys), this does not change
the equation much. Differentiation of the Thomasomyini was based on
a combination of a short palate and pentalophodont molars of medium
to large size. After a dense taxonomic sampling, Pacheco (2003, 2015)
provided the first set of anatomical traits to distinguish an ‘Andean
thomasomyine clade.’ According to Pacheco (2003; Pacheco et al. 2015,
p. 572), this group is characterised by several craniodental traits includ-
ing shallow zygomatic notch (depth less than half width), an ‘open’
Figure 7. Strict consensus of 11 minimum-length trees resulting from cladistics condition of the ectotympanic ring (well-exposed tegmen tympani),
parsimony of 99 morphological characters (from Weksler 2006). Tree
length = 521, consistency index = 0.27, retention index = 0.63. and a procingulum on M1 composed by a distinctly reduced antero-
lingual conule relative to the anterolabial conule. Certainly,
Megaoryzomys has shallow zygomatic notches, a trait amplified by
the shortening of the rostrum, and partially compensated by a broad
elevation montane cloud forest in the Andes of northwestern Ecuador anteriorly well-developed zygomatic plate. However, shallow zygo-
(Weksler et al. 2006). The clade Megaoryzomys plus Mindomys occu- matic notches are unusual for Thomasomyini, which are typically
pies a basal position in 9 of the 11 trees recovered (Figures S3 and S4). characterised by narrow and high zygomatic plates as found in giant
10 C. RONEZ ET AL.

Table 2. Extant and extinct (†) sigmodontines represented in islands; Tierra del T. apeco (4.3 mm) and M. curioi (8.1 mm; see Gardner and Romo
Fuego and several minor Atlantic islands are not included, as well as taxa at generic 1993:Table 1; Leo and Gardner 1993:Table 2). Even the Oryzomyini
or suprageneric level mentioned from Caribbean and Galapagoan islands (taxon-
omy after Pardiñas et al. 2017). Data from several sources; SCI = southern Chilean Antillomys, uncritically treated as ‘large or larger than any extant
islands; † combined with an island denotes fossil record. sigmodontine’ (Brace et al. 2015: supplementary information pp. 2),
Tribe Taxa Islands has a narrower zygomatic plate (6.5 mm) compared with the
Abrotrichini Abrothrix hirta Chiloé
Galápagos giant rat. Regarding the condition of the auditory region,
Abrothrix lanosa Madre de Dios, Estados Megaoryzomys shares with other several sigmodontines the ‘closed’
Abrothrix manni Chiloé condition of the ectotympanic ring. Finally, the anterolingual conule
Abrothrix olivacea Chiloé, Estados, SCI appears larger than the anterolabial conule in Megaoryzomys, possibly
Abrothrix sanborni Chiloé ruling out an association with the Thomasomyini. In addition, accord-
Abrothrix xanthorhina Chiloé, SCI
Geoxus lefkenche Guafo ing to Leo and Gardner (1993, p. 425) ‘dental characteristics (e.g.,
Geoxus valdivianus Mocha, Chiloé paraloph joins paracone to median mure . . .) place M. curioi well
Akodontini Necromys urichi Trinidad, Tobago apart from other known species of larger thomasomyines.’ However,
Euneomyini Irenomys tarsalis Chiloé in other ways, the dentition of Megaoryzomys is too symplesiomorphic
Euneomys chinchilloides Cabo de Hornos
Oryzomyini †Agathaeromys Bonaire
to allow definitive tribal allocation.
donovani An alternative hypothesis is that Megaoryzomys is so divergent
†Agathaeromys Bonaire in morphology that it constitutes a separate and unnamed tribe.
praeuniversitatis The combination of a short palate and the virtual absence of
†Antillomys rayi Antigua, Barbuda, Guadeloupe, suspensory process of the squamosal give some support to this
M. Galante
†Dushimys larsi Curaçao proposition. However, several of the most ‘bizarre’ craniodental
†Megalomys audreyae Antigua and Barbuda traits expressed by this rat must be interpreted in the context of
†Megalomys curazensis Curaçao a tendency to gigantism in island isolation. Megaoryzomys is also
†Megalomys desmarestii Martinique characterised by powerful and displaced backward zygomatic
†Megalomys georginae Barbados
†Megalomys luciae Santa Lucia
arches to a degree where the zygomatic root of the squamosal is
†Megalomys sp. Granada displaced to the rear of the cranium. The entire calvarium is shor-
†Megaoryzomys curioi Galápagos tened anteriorly-posteriorly and broadened transversally. More
†Megaoryzomys sp. Galápagos impressive, the height of the skull along the braincase is equal or
†Noronhomys vespuccii Fernando de Noronha less than the same dimension at the zygomatic plate level, and the
†Oligoryzomys victus S. Vincent, Grenadines
†Oryzomys antillarum Jamaica rear profile is markedly flat. In addition, probably associated with
†Pennatomys nivalis Nevis mechanical advantage and stress, the zygomatic plate is particularly
Aegialomys Galápagos broad and a peculiar denticulate border is displayed by the dentary
galapagoensis bone in its contact with the incisor (Figure 1).
Hylaemys megacephalus Trinidad
Nectomys palmipes Trinidad
Hutterer and Oromí (1993) considered Megaoryzomys as
Nesoryzomys darwini Galápagos a product of a long history of island evolution and highlighted
Nesoryzomys Galápagos a potential association of this giant rat with a volcanic tube envir-
fernandinae onment. The morphological divergence of island rodents, including
Nesoryzomys indefessus Galápagos body size variations, is a well-known biological process apparently
Nesoryzomys Galápagos
narboroughi connected to the interplay of multiple environmental factors (e.g.,
Nesoryzomys swarthi Galápagos Foster 1964; Lomolino 2005; Moncunill-Solé et al. 2014; Van der
Oecomys speciosus Trinidad Geer 2018). Palaeontological paradigmatic examples are repre-
Oecomys trinitatis Trinidad, Tobago sented, among many other, by the Atlantic and Mediterranean
Oligoryzomys delicatus Trinidad, Tobago
Oligoryzomys Chiloé, SCI, Estados
murids Canariomys (see Crusafont-Pairo and Petter 1964; López-
longicaudatus Martínez and López Jurado 1987), the recently described
Oligoryzomys yatesi Capitán Aracena, Harrison Apocricetus darderi, the well-known selenodont forms Myotragus
Oryzomys gorgasi Curaçao and Tragomys (see Torres-Roig et al. 2019 and the references cited
Oryzomys nelsoni María Madre therein), and the cricetids Hattomys (see Freudenthal 1985;
Zygodontomys Coiba, Cébaco, Pearl, Margarita,
brevicauda Trinidad, Tobago Savorelli 2013) and Mystemys (see Savorelli and Masini 2016).
Zygodontomys sp. Granada Prima facie we can trace several traits that may link
Phyllotini Calomys hummelincki Aruba, Curaçao, †Tobago Megaoryzomys to Oryzomyini. However, the potential extraction
Loxodontomys micropus Chiloé of ancient DNA from Megaoryzomys remains seems to be a better
Reithrodontini Reithrodon auritus Malvinas/Falklands
Thomasomyini Rhipidomys couesi Margarita, Trinidad
way to determine the phylogenetic relationships of this rodent in
Rhipidomys nitela Little Tobago the near future, as was recently addressed for extinct Caribbean
Sigmodontini Sigmodon hispidus †Aruba sigmodontines (Brace et al. 2015). Mindomys, the oryzomyinin
Sigmodon alstoni †Tobago genus recorded as the sister to Megaoryzomys in our phylogeny,
Sigmodon peruanus Puná also lacks published molecular data.
Incertae sedis †Cordimus debuisonjei Curaçao
†Cordimus hooijeri Bonaire
†Cordimus raton Curaçao
Insular colonisation in sigmodontine rodents
The question of the arrival of Megaoryzomys on the Galápagos
forms such as Thomasomys aureus, Thomasomys apeco, and Islands raises the vast topic of insular colonisation and evolution.
Thomasomys macrotis (Gardner and Romo 1993; Leo and Gardner Currently, more than 50 extant or extinct sigmodontines species
1993). For example, there is a sharp contrast in the breadth of the from at least 8 tribes have been reported in an insular context (Table
zygomatic plate among T. aureus (3.1 mm), T. macrotis (3.9 mm), 2), representing a noticeable diversity. However, when the sea level
 HISTORICAL BIOLOGY 11

Figure 8. Map showing the distribution of sigmodontine tribes with taxa occupying central and South American islands. Grey represents current continental limits, dashed
line connotes the 200 m isobaths level. Scale bars are in km.

of the Last Glacial Maximum is considered, several of the insular New Jersey, USA (for Oryzomys palustris, the northernmost
masses could have been colonised over land bridge connections (at Sigmodontinae), to Cabo del Hornos Islands, Chile (for Oligoryzomys
least for 26 insular species; Figure 8). Members of the Oryzomyini longicaudatus) (Pardiñas et al. 2017). The capacity of dispersal is
and Phyllotini are the only clades that reached islands across marine magnified among Oryzomyini because of their large size and variety
barriers. The only Phyllotini member is represented by Calomys of adaptations. The Oryzomyini demonstrate remarkable examples of
hummelincki from Curaçao (Husson 1960); the island is 40 km off forms adapted to riparian and semiaquatic conditions, although not
the South American continent. Among the Sigmodontinae, only rivalling the aquatic specialisations achieved by other living cricetids
Nesoryzomys, Aegialomys, Megaoryzomys and Noronhomys have (e.g., Ondatrini; Carleton and Musser 1984). Natatory abilities and
been able to reach distant islands. Indeed, even with a 200 m sea associated morphological traits (e.g., natatory fringes, interdigital web-
level regression, both Galápagos and Fernando de Noronha Island, bing) have been recorded in the oryzomyinin genera Amphinectomys,
the home of Noronhomys, are around 800 and 200 km away from Holochilus, Lundomys, Nectomys, and Oryzomys palustris (Weksler
South America, respectively (Patton and Hafner 1983; Carleton and 2006; Pardiñas et al. 2017). Although nothing is known specifically
Olson 1999). Other extinct and extant Oryzomyni from the about their capacities to cross marine barriers, swimming proficiency
Caribbean are found on islands within 50 km of the mainland of several of these rats has been tested experimentally (e.g., Esher et al.
(Turvey et al. 2010). 1978; Santori et al. 2008, 2014; Torres et al. 2020). Thus, the anatomy,
Hutterer and Oromí (1993) considered that the record of dispersal potential, and biology of Oryzomyini suggest further indirect
Megaoryzomys suggested three independent colonisations of these evidence that Megaoryzomys was likely a large, distinct Oryzomyini.
islands by sigmodontine rodents, Nesoryzomys and Aegialomys com- The available evidence appears to connect and restrict the
prising the other two. This hypothesis has also been recently tested by Oryzomyini as oceanic island colonisers within the sigmodontine
Castañeda-Rico et al. (2019) who interpreted that both Nesoryzomys radiation (Carleton and Olson 1999; Turvey et al. 2010). However,
and Aegialomys reached the Galápagos during the Pliocene and it is premature to decide if this role was driven by anatomical
Middle Pleistocene, respectively. Several episodes of colonisation design, behavioural tendencies, or geographical and ecological dis-
are also allowed due to the age of the archipelago, calculated to 4 tribution. Within the latter, temporal segregation in a restricted
My for the extant islands (Geist et al. 2014). If our affiliation of geographical context (e.g., coastal aquatic or semiaquatic commu-
Megaoryzomys with the Oryzomyini is accepted, it would resolve nities) could have favoured Oryzomyini as island invaders. Since we
the issue about the supposed presence of two sigmodontines tribes have no idea when Megaoryzomys, Nesoryzomys, or their ancestors
in the Galápagos Archipelago, since the remaining sigmodontine taxa reached the Galápagos, this discussion is necessarily speculative.
involved in these colonisations is also Oryzomyini. Nevertheless, the semi-aquatic adaptations of some Oryzomyini
Oryzomyini exhibit the largest geographic distribution in the sub- invite contemplation of an ancient episode conducive to island
family Sigmodontinae. Representatives of the tribe are found from colonisation. Perhaps Oryzomyini was the dominant sigmodontine
12 C. RONEZ ET AL.

clade during Late Miocene–Early Pliocene in a limited portion of Gardner AL, Romo RM. 1993. A new Thomasomys (Mammalia: Rodentia) from
northern South American coastal lowlands. This geographic dis- the Peruvian Andes. Proc Biol Soc Wash. 106:762–764.
Geist DJ, Snell H, Snell H, Goddard C, Kurz MD. 2014. A paleogeographic
tribution would have facilitated dispersal to the Galápagos and
model of the Galápagos Islands and biogeographical and evolutionary impli-
Caribbean archipelagos. Despite the current absence of confirma- cations. In: Harpp KS, Mittelstaedt E, d’Ozouville N, Graham DW, editors.
tory evidence, the degree of morphological divergence exemplified The Galápagos: A natural laboratory for earth sciences. Hobokens: John
by Megaoryzomys points to a unique experiment in insular evolu- Wiley & Sons; p. 145–167.
tion among sigmodontines. Goloboff PA, Farris JS, Nixon KC. 2008. TNT, a free program for phylogenetic
analysis. Cladistics. 24:774–786. doi:10.1111/j.1096-0031.2008.00217.x.
Gonçalves PR, Christoff AU, Machado LF, Bonvicino CR, Peters FB,
Percequillo AR. 2018. Unraveling deep branches of the Sigmodontinae tree
Acknowledgments (Rodentia: cricetidae) in Eastern South America. J Mamm Evol. 27:1–22.
We thank A. Milhouse for allowing access to specimens housed in the USNM as Hershkovitz P. 1962. Evolution of Neotropical cricetine rodents (Muridae) with
well as A. Folie, O. Pauwels and S. Sweydan for providing pictures of the sample special reference to the Phyllotine group. Fieldiana Zool. 46:1–524.
housed in the RBINS. JB thanks the “Germany-Brazil-Ecuador Trilateral Husson AM. 1960. Die zoogdieren van de Nederlandse Antillen [Mammals of
Cooperation Program,” funded by the international cooperation GIZ for facil- the Netherlands Antilles]. Curaçao: Natuurwetenschappelijke Werkgroep
itating the trip to the Alexander Koenig Museum for review and scanning of Nederlandse Antillen. German.
specimens. We are also indebted to the curators D. Flores (MACN), E. González Hutterer R, Oromí P. 1993. La rata gigante de la Isla Santa Cruz, Galápagos:
(MNHM), G. Lessa (MZUFV), J. Cherem (UFSC), who kindly loaned us valu- algunos datos y problemas [The giant rat from the Santa Cruz Island,
able comparative material. The help of C. Cañon with the phylogenetic analyses Galapagos: some data and problems]. Resultados científicos del proyecto
is deeply appreciated. UP is grateful to A. Percequillo, who generously shared his Galápagos: Patrimonio de la Humanidad. Museo de Ciencias Naturales,
expertise of Oryzomyini anatomy. P. Hadler and an anonymous reviewer greatly Tenerife. 4:63–76. Spanish.
improved this contribution. PICT 2014-1039 (to UJFP) partially funded this Lenglet G, Coppois G. 1979. Description du crâne et de quelques ossements d’un
research. genre nouveau éteint de Cricetidae (Mammalia, Rodentia) géant des
Galapagos: Megaoryzomys (gen. nov.) [Description of the skull and bony
elements of a new giant extinct genus of Cricetidae (Mammalia, Rodentia)
Disclosure statement from the Galapagos: Megaoryzomys (gen. nov.)]. Bull Acad Roy Belgique
Classe des Sci. 5:633–648. French.
No potential conflict of interest was reported by the authors. Leo LM, Gardner AL. 1993. A new species of a giant Thomasomys (Mammalia,
Muridae, Sigmodontinae) from the Andes of north-central Peru. Proc Biol
Soc Wash. 106:417–428.
ORCID Lomolino M. 2005. Body size evolution in insular vertebrates: generality of the
island rule. J Biogeogr. 32:1683–1699. doi:10.1111/j.1365-2699.2005.01314.x.
Christophe Ronez http://orcid.org/0000-0001-5163-7931 López-Martínez N, López Jurado LF. 1987. Un nuevo Múrido gigante del
Jorge Brito http://orcid.org/0000-0002-3410-6669 Cuaternario de Gran Canaria: canariomys tamarani n. sp. (Rod. Mamm.)
Robert A. Martin http://orcid.org/0000-0001-8678-0316 Interpretación filogenética y biogeográfica. [A new giant murid from the
Ulyses F. J. Pardiñas http://orcid.org/0000-0001-9496-5433 quaternary of Gran Canaria: canariomys tamarani n. sp. (Rod. Mamm.).
Phyllogenetic and biogeographic interpretations]. Doñana Acta Vertebr.
2:1–60. Spanish.
References McKenna MC, Bell SK. 1997. Classification of mammals above the species level.
New York: Columbia University Press.
Barbière F, Ronez C, Ortiz PE, Martin RA, Pardiñas UFJ. 2019. A new nomen- Moncunill-Solé B, Jordana X, Marín-Moratalla N, Moyà-Solà S, Köhler M. 2014.
clatural system for the study of sigmodontine rodent molars: first step How large are the extinct giant insular rodents? New body mass estimations
towards an integrative phylogeny of fossil and living cricetids. Bio J Linn from teeth and bones. Integr Zool. 9:197–212. doi:10.1111/1749-4877.12063.
Soc. 127:224–244. doi:10.1093/biolinnean/blz021. Musser GG, Carleton MD. 2005. Superfamily Muroidea. In: Wilson DE,
Brace S, Turvey ST, Weksler M, Hoogland MLP, Barnes I. 2015. Unexpected Reeder DM, editors. Mammal species of the world: a taxonomic and geo-
evolutionary diversity in a recently extinct Caribbean mammal radiation. graphic reference. Baltimore: Johns Hopkins University; p. 894–1531.
Proc R Soc Lond B. 282:20142371. doi:10.1098/rspb.2014.2371. Myers P, Patton JL, Smith MF. 1990. A review of the boliviensis group of Akodon
Carleton MD, Musser GG. 1984. Muroid rodents. In: Anderson S, Jones JK, (Muridae: sigmodontinae), with emphasis on Peru and Bolivia. Miscellaneous
editors. Orders and families of recent mammals of the world. New York: John Publications Museum of Zoology, University of Michigan. 177:1–104.
Wiley & Sons; p. 289–379. Niethammer J. 1964. Contribution à la connaissance des Mammifères terrestres de
Carleton MD, Musser GG. 1989. Systematic studies of oryzomyine rodents l’ile Indéfatigable (=Santa Cruz), Galápagos [Contribution to the knowledge of
(Muridae, Sigmodontinae): a synopsis of Microryzomys. B Am Mus Nat terrestrial mammals of the Indefatigable Island (=Santa Cruz), Galapagos].
Hist. 191:1–83. Mammalia. 28:593–606. French. doi:10.1515/mamm.1964.28.4.593.
Carleton MD, Olson SL. 1999. Amerigo Vespucci and the rat of Fernando de Pacheco VR 2003. Phylogenetic analyses of the Thomasomyini (Muroidea:
Noronha: a new genus and species of Rodentia (Muridae: sigmodontinae) Sigmodontinae) based on morphological data [dissertation]. New York:
from a volcanic island off Brazil’s continental shelf. Am Mus Novit. The University of New York.
3256:1–59. Pacheco VR, Patton JL, D’Elía G. 2015. Tribe Thomasomyini Steadman and Ray,
Castañeda-Rico S, Johnson SA, Clement SA, Dowler RC, Maldonado JE, 1982. In: Patton JL, Pardiñas UFJ, D’Elia G, editors. Mammals of South
Edwards CW. 2019. Insights into the evolutionary and demographic history America (Vol. 2). Chicago: The University of Chicago Press; p. 571–682.
of the extant endemic rodents of the Galápagos Islands. Therya. 10:213–228. Pardiñas UFJ, Myers P, León-Paniagua L, Ordoñez Garza N, Cook J,
doi:10.12933/therya-19-873. Krystufek B, Haslauer B, Bradley R, Shenbrot G, Patton JL. 2017. Family
Crusafont-Pairo M, Petter F. 1964. Un Muriné géant fossile des iles Canaries Cricetidae. In: Wilson DE, Lacher TE, Mittermeier RA, editors. Handbook of
Canariomys bravoi gen. nov., sp. nov [A giant fossil murine from Canary mammals of the world, volume 7, rodents 2. Barcelona: Lynx Edition; p.
Islands Canariomys bravoi gen. nov., sp. nov]. Mammalia. 28:607–612. 204–535.
French. doi:10.1515/mamm.1964.28.4.607. Pardiñas UFJ, D’Elia G, Lessa G, Passamani M, Teta P. 2014. Nuevos datos
Dowler RC. 2015. Genus Nesoryzomys Heller, 1904. In: Patton JL, Pardiñas UFJ, morfológicos y una hipótesis filogenética para Phaenomys (Rodentia,
D’Elia G, editors. Mammals of South America (Vol. 2). Chicago: The Cricetidae) [New mophological data and a phylogenetic hypothesis for
University of Chicago Press; p. 390–393. Phaenomys (Rodentia, Cricetidae)]. Mast Neotrop. 21:251–261. Spanish.
Eisenberg JF. 1989. Mammals of the neotropics: the Northern neotropics. Patton JL, Hafner MS. 1983. Biosystematics of the native rodents of the
Chicago: University of Chicago Press. Galápagos Archipelago, Ecuador. In: Bowman RI, Benson M, Leviton AE,
Esher RJ, Wolfe JL, Layne JN. 1978. Swimming behavior of rice rats (Oryzomys editors. Patterns of evolution in Galapagos organisms. San Francisco:
palustris) and cotton rats (Sigmodon hispidus). J Mammal. 59:551–558. American Association for the Advancement of Science, Pacific Division; p.
doi:10.2307/1380231. 539–568.
Foster JB. 1964. Evolution of mammals on islands. Nature. 202:234–235. Percequillo AR, Weksler M, Costa LP. 2011. A new genus and species of rodent
doi:10.1038/202234a0. from the Brazilian Atlantic Forest (Rodentia: Cricetidae: Sigmodontinae:
Freudenthal M. 1985. Cricetidae (Rodentia) from the Neogene of Gargano Oryzomyini), with comments on oryzomyine biogeography. Zool J Linn
(Prov. of Foggia, Italy). Scripta Geol. 77:29–76. Soc. 161:357–390. doi:10.1111/j.1096-3642.2010.00643.x.
 HISTORICAL BIOLOGY 13

Prado JR, Percequillo AR. 2018. Systematic studies of the genus Aegialomys Weksler, (Brants, 1827) (Cricetidae, Sigmodontinae). Mammalia. In press.
Percequillo and Voss, 2006 (Rodentia: Cricetidae: Sigmodontinae): annotated doi:10.1515/mammalia-2019-0023.
catalogue of the types of the species-group taxa. Zootaxa. 4144:447–498. Torres-Roig E, Agustí J, Bover P, Alcover JA. 2019. A new giant cricetine from
Răduleţ N. 2007. Morphology of the inner side of the mandible in micromam- the basal Pliocene of Mallorca (Balearic Islands, western Mediterranean):
mals (Mammalia: Insectivora, Chiroptera, Rodentia) of Romania. Trav Mus biostratigraphic nexus with continental mammal zones. Hist Biol.
Natl Hist Nat Grigore Antipa. 50:371–393. 31:559–573. doi:10.1080/08912963.2017.1377194.
Reig OA. 1977. A proposed unified nomenclature for the enameled components Tribe CJ 1996. The Neotropical rodent genus Rhipidomys (Cricetidae,
of the molar teeth of the Cricetidae (Rodentia). J Zool. 181:227–241. Sigmodontinae) – a taxonomic revision [dissertation]. London: University College.
doi:10.1111/j.1469-7998.1977.tb03238.x. Turvey ST. 2009. Holocene mammal extinction. In: Turvey ST, editor. Holocene
Santori RT, Delciellos AC, Vieira MV, Gobbi N, Loguercio MFC, Rocha–Barbosa O. extinction. New York: Oxford University Press; p. 41–61.
2014. Swimming performance in semiaquatic and terrestrial oryzomyine Turvey ST, Weksler M, Morris EL, Nokkert M. 2010. Taxonomy, phylogeny, and
rodents. Mammal Biol. 79:189–194. doi:10.1016/j.mambio.2013.12.004. diversity of the extinct Lesser Antillean rice rats (Sigmodontinae:
Santori RT, Vieira MV, Rocha-Barbosa O, Magnan-Neto JA, Gobbi N. 2008. Oryzomyini), with description of a new genus and species. Zool J Linn Soc-
Water absorption of the fur and swimming behavior of semiaquatic and Lond. 160:748–772. doi:10.1111/j.1096-3642.2009.00628.x.
terrestrial Oryzomine rodents. J Mammal. 89:1152–1161. doi:10.1644/07- Van der Geer AAE. 2018. Changing invaders: trends of gigantism in insular
MAMM-A-327.1. introduced rats. Environ Conserv. 45:203–211. doi:10.1017/S0376892918000085.
Savorelli A. 2013. New data on the Cricetidae from the Miocene “Terre Rosse” of Voss RS. 1993. A revision of the Brazilian muroid rodent genus Delomys with
Gargano (Apulia, Italy). Geobios. 46:77–88. doi:10.1016/j.geobios.2012.10.002. remarks on “thomasomyine” characters. Am Mus Novit. 3073:1–44.
Savorelli A, Masini F. 2016. Mystemys giganteus n. gen. et sp.: an enigmatic and Voss RS, Carleton MD. 1993. A new genus for Hesperomys molitor Winge and
rare cricetid from the Terre Rosse M013 fissure filling (Gargano, Southeastern Holochilus magnus Hershkovitz (Mammalia, Muridae): with an analysis of its
Italy). Palaeontogr Abt A. 306:1–23. doi:10.1127/pala/306/2016/1. phylogenetic relationships. Am Mus Novit. 3085:1–39.
Steadman DW, Ray CE. 1982. Relationships of Megaoryzomys curioi, an Extinct Weksler M. 2006. Phylogenetic relationships of oryzomine rodents (Muroidea:
Cricetine Rodent (Muroidea: muridae) from the Galápagos Islands, Ecuador. Sigmodontinae): separate and combined analyses of morphological and
Sm C Paleob. 51:1–23. molecular data. B Am Mus Nat Hist. 296:1–149. doi:10.1206/0003-0090-
Steadman DW, Stafford TW, Donahue DJ, Jull AJ. 1991. Chronology of (2006)296[0001:PROORM]2.0.CO;2.
Holocene vertebrate extinction in the Galápagos Islands. Quat Res. Weksler M. 2015. Tribe Oryzomyini Vorontsov, 1959. In: Patton JL, Pardiñas
36:126–133. doi:10.1016/0033-5894(91)90021-V. UFJ, D’Elia G, editors. Mammals of South America (Vol. 2). Chicago: The
Thomas O. 1906. Notes on South-American rodents. Ann Mag Nat Hist Ser 7. University of Chicago Press; p. 291–293.
18:442–448. doi:10.1080/00222930608562646. Weksler M, Percequillo AR, Voss RS. 2006. Ten new genera of Oryzomyine
Torres J, Santori RT, Rocha-Barbosa O, Candela AM, Pardiñas UFJ. 2020. rodents (Cricetidae: Sigmodontinae). Am Mus Novit. 3537:1–29.
Swimming behavior and performance of the marsh rat Holochilus vulpinus doi:10.1206/0003-0082(2006)3537[1:TNGOOR]2.0.CO;2.