Received: 23 March 2021

| Revised: 27 July 2021

| Accepted: 29 July 2021

DOI: 10.1111/zsc.12510

ORIGINAL ARTICLE

The Phyllotis xanthopygus complex (Rodentia, Cricetidae) in

central Andes, systematics and description of a new species

J. Pablo Jayat1 | Pablo Teta2 | Agustina A. Ojeda3 | Scott J. Steppan4 |

Jared M. Osland4 | Pablo E. Ortiz5 | Agustina Novillo6 | Cecilia Lanzone7 |
Ricardo A. Ojeda3

1
Unidad Ejecutora Lillo, CONICET-­
Fundación M. Lillo, Tucumán, Argentina
Abstract
2
CONICET, División Mastozoología, Phyllotis Waterhouse 1837 is one of the most studied genera of South American cri-
Museo Argentino de Ciencias Naturales cetid rodents. As currently understood, it includes 20 small to medium-­sized species
"Bernardino Rivadavia", Ciudad Autónoma
of predominantly rocky habitats. Among them, populations of the yellow-­rumped
de Buenos Aires, Argentina
3 leaf-­eared mouse, traditionally referred to P. xanthopygus (Waterhouse 1837), are
Grupo de Investigaciones de la
Biodiversidad-­IADIZA-­CCTMendoza-­ the most widely distributed, extending from central Peru to southern Chile and
CONICET, Mendoza, Argentina Argentina. Based mostly on molecular evidence, previous studies suggested that
4
Department of Biological Science, Florida P. xanthopygus constitutes a species complex, being characterized by geographically
State University, Tallahassee, FL, USA
5
structured and genetically divergent clades. In this work, we compare the molecu-
Instituto Superior de Correlación
Geológica, CONICET –­Universidad lar phylogenetic hypothesis for populations distributed on the eastern slopes of the
Nacional de Tucumán, San Miguel de central Andes with morphometric evidence using univariate and multivariate analy-
Tucumán, Argentina
6
ses. Quantitative morphological and molecular evidence suggests that at least four
Instituto de Biodiversidad Neotropical
(IBN), CCT-­CONICET, Tucumán,
nearly cryptic species of the P. xanthopygus complex occur from southern Bolivia
Argentina to west-­central Argentina. Three of these taxa have available names; one of them,
7
Laboratorio de Genética Evolutiva, IBS P. caprinus, is currently recognized to the species level; the other two, the clades of
(CONICET-­UNaM), Posadas, Argentina
P. x. posticalis-­P. x. rupestris and P. vaccarum, are both recognized as subspecies
Correspondence of P. xanthopygus. The remaining taxon represents a new species distributed in the
J. Pablo Jayat, Unidad Ejecutora Lillo west-­central Andes of Argentina. We discuss our morphological results in the light
(CONICET-­Fundación Miguel Lillo).
of other sources of evidence (e.g. qualitative and quantitative state characters, genetic
Miguel Lillo 251. CP 4000. San Miguel de
Tucumán, Tucumán, Argentina. and phylogenetic studies, and cytogenetic data) and name the new species as P. pe-
Email: eljayat@gmail.com huenche, honouring the original native people that historically inhabited west-­central
Funding information
Andes of Argentina.
National Council for Science
and Technology of Argentina, KEYWORDS
CONICET, Grant/Award Number: PIP cryptic species, integrative taxonomy, Myomorpha, new species, Phyllotini, sigmodontinae
11220150100258CO; Agencia Nacional de
Promoción Científica y Tecnológica, Grant/
Award Number: PICT 1636 and PICT
2012-­0050; National Science Foundation,
Grant/Award Number: DEB-­0108422,
DEB-­0841447 and DEB-­1754748

Zoologica Scripta. 2021;00:1–18. wileyonlinelibrary.com/journal/zsc © 2021 Royal Swedish Academy of Sciences. | 1

2
|    JAYAT et al.

1 | IN TRO D U C T ION occur in central Argentina, within temperate grassland en-

vironments (Ojeda et al., 2021; Steppan & Ramírez, 2015;
The leaf-­eared mice of the genus Phyllotis Waterhouse 1837 Teta et al., 2018).
include 20 species of small to medium-­sized sigmodontine The complex taxonomic history of the nominal forms
rodents found mainly in rocky habitats (Jayat et al., 2016; related to P. xanthopygus has been extensively documented
Rengifo & Pacheco, 2015; Steppan & Ramírez, 2015). The in several publications (e.g. Hershkovitz, 1962; Ojeda
geographic distribution of this genus extends widely in et al., 2021; Pearson, 1958; Steppan & Ramírez, 2015;
South America, from the highlands of Ecuador, through- Teta et al., 2018); here, we will provide a brief resume, but
out the Andes and Andean foothills, to the southern tip of readers interested in going further into this topic should
the continent (Steppan & Ramírez, 2015). Most species of consult basic literature. Pearson (1958) and Hershkovitz
Phyllotis inhabit open, arid to semiarid, mostly rocky and (1962), in his revision of the genus, considered P. dar-
brushy habitats, from sea level up to over 6,700 m (Steppan wini as a widely distributed and polytypic species, in-
& Ramírez, 2015; Storz et al., 2020), but a few species are as- cluding the nominal forms xanthopygus, chilensis Mann
sociated with more humid and forested areas of Yungas envi- 1945, posticalis Thomas, 1912, ricardulus Thomas, 1919,
ronments, on eastern sub-­Andean slopes (Ferro et al., 2010; rupestris P. Gervais 1841, limatus Thomas, 1912 and vac-
Jayat et al., 2007; Jayat et al., 2016). Furthermore, isolated carum Thomas, 1912 as subspecies (later, Crespo [1964]
populations of these mice are distributed in montane grass- added the nominal form bonariensis). Based on karyotypic
land habitats of the Sierras Pampeanas and the Sierra de la and crossbreeding evidence, Walker et al. (1984) proved
Ventana, in central and east-­central Argentina, respectively that true P. darwini is geographically restricted to central
(Crespo, 1964; Pearson, 1972; Steppan & Ramírez, 2015; Chile, leaving all the other nominal taxa under P. xantho-
Teta et al., 2018). pygus. Since then, the status of P. xanthopygus underwent
Several systematic studies, based both on morphological additional removal of the nominal forms bonariensis and
and molecular data, have contributed to clarifying most of the limatus, which were both considered as valid species (e.g.
phylogenetic relationships, species boundaries and taxon- Reig, 1978; Steppan, 1998; Steppan & Ramírez, 2015).
omy among the species of this genus (e.g. Ferro et al., 2010; Finally, an important milestone should be mentioned; re-
Jayat et al., 2007; Jayat et al., 2016; Steppan, 1995, 1998; cent molecular studies have challenged the traditional
Steppan et al., 2007; Pacheco et al., 2014; Rengifo & classification of xanthopygus into six subspecies, show-
Pacheco, 2015, 2017; Ojeda et al., 2021). All these stud- ing that, as it is currently understood, this taxon is a com-
ies have substantially expanded our understanding of the plex of two or more cryptic species (Albright, 2004; Jayat
species-­level taxonomy of the genus, but several uncertain- et al., 2016; Ojeda et al., 2021; Riverón, 2011). The taxon-
ties still remain, especially for populations representing the omy of most of the nominal forms related to the P. xantho-
nominal forms described for central and southern Andean pygus complex was initially addressed using morphology
areas (Ojeda et al., 2021). (e.g. Hershkovitz, 1962; Pearson, 1958). However, incon-
Based on morphological and molecular evidence, three sistencies between these taxonomic delimitations and more
species groups are currently recognized within Phyllotis: recent molecular analyses have become evident (see Teta
the andium-­amicus, the osilae and the darwini groups et al., 2018 and Ojeda et al., 2021). In this context, the
(Rengifo & Pacheco, 2017; Steppan, 1993, 1995; Steppan integration of existing molecular results with additional
et al., 2007). The P. darwini species group, which is the most sources of data (e.g. morphological, cytogenetic) seems
speciose, includes P. darwini (Waterhouse 1837), P. bonar- to be the next step in delimiting species and resolving the
iensis Crespo, 1964, P. caprinus Pearson, 1958, P. limatus taxonomic status of some nominal forms (e.g. ricardulus
Thomas, 1912, P. magister Thomas, 1912, P. osgoodi Mann Thomas, 1919, oreigenus Cabrera, 1926 and darwinii vac-
1945 and P. xanthopygus (Waterhouse 1837). Among these, carum Thomas, 1912).
the yellow-­r umped leaf-­eared mouse, P. xanthopygus, is the The aim of our study was to test the taxonomic status of
most widely distributed species (having the largest altitudi- the main clades obtained in the most recent molecular phy-
nal range of any vertebrate), occurring from Junín depart- logenetic hypothesis for the P. xanthopygus complex (Ojeda
ment, in central Peru to the provinces of Magallanes and et al., 2021) primarily using morphological evidence but
Santa Cruz in southern Chile and Argentina, respectively. supplemented with new DNA sequence data. The molecular
Across this 6,000 km range, populations assigned to P. xan- analyses of Ojeda et al. (2021) recovered eight well supported
thopygus are often the most common rodent species among monophyletic groups, with two major clades in the species
small mammal assemblages (Steppan & Ramírez, 2015). complex. One included P. xanthopygus s.s., P. bonariensis,
Although this rodent shows an apparently continuous geo- P. caprinus, P. limatus, and P. vaccarum, and two additional
graphic distribution across the arid and semiarid Andean clades named P. sp. 1, and P. sp. 2. The other was a diverse
ecosystems, some geographically isolated populations clade of specimens from different populations and localities
JAYAT et al.   
| 3

from the Altiplano of northern Argentina, northern Chile, 2.2 | Phylogenetic framework of the study
Bolivia and southern Peru, named the P. posticalis-­P. rup-
estris clade. Using a unilocus species delimitation method, We structured all qualitative and quantitative morphologic
the Bayesian implementation of the Poisson tree processes analyses on the basis of the phylogenetic hypothesis devel-
(bPTP), these authors corroborate most of these clades as oped by Ojeda et al. (2021). Additionally, we added 47 new
species-­level lineages, but four species were delimited within samples not included in Ojeda et al. (2021) that greatly ex-
the clade of P. posticalis –­P. rupestris. The mean genetic panded the geographic coverage for DNA data and analysed
distances between clades ranged from 3.0% between P. vac- them in a more restricted and detailed genealogic analysis,
carum and P. limatus, to 10.6% between P. sp. 1 and P. sp. 2 testing the monophyly of the new species (see Supplementary
and between P. posticalis-­P. rupestris and P. sp. 2. Material S1).
In this work, we use morphology to test the status of those
clades distributed on the eastern Andean slopes of the west-­
central Andes of Argentina, which include, from north to 2.3 | Taxonomic assignment of the samples
south, P. x. posticalis -­P. x. rupestris, P. caprinus, P. vac-
carum and P. sp. 2. We focus our analysis on populations We examined skull and dental morphology of the Argentinean
of the central Andes in western Argentina because is a geo- specimens of Phyllotis used in the phylogenetic analysis de-
graphic area highly underrepresented in previous studies and veloped by Ojeda et al. (2021) in search of qualitative state
for which we have gathered a great quantity of information characters associated to each one of the recognized mitochon-
in recent years. We characterize at least four clades that we drial clades. Then, we looked for these same characters in the
hypothesize correspond to distinct species. Three of these specimens examined in systematic collections not included in
clades could be associated with previously available names the phylogenetic analysis of Ojeda et al. (2021). We follow
but we recognize one as representing a new species. We original morphological descriptions of the nominal forms (e.g.
discuss our results in the light of other sources of evidence Cabrera, 1926; Crespo, 1964; Thomas, 1912, 1919) and revi-
(including genetic, phylogenetic and cytogenetic published sionary studies of the genus Phyllotis (e.g. Hershkovitz, 1962;
data) and describe the newly recognized species, which was Pearson, 1958), to relate the characters’ state observed in our
named honouring the original native people that historically samples with specimens of the several nominal forms de-
inhabited west-­central Andes of Argentina and nearby areas scribed for the genus. When qualitative state characters were
of Chile. not useful (e.g. damaged or broken skulls and obliterated mo-
lars patterns), we followed a criterion of geographic proximity
among localities (respect to the provenance localities of the
2 | M ATE R IA L S A N D ME T HODS sequenced specimens used in the phylogenetic analyses con-
ducted by Ojeda et al., 2021 and to the collecting localities of
2.1 | Specimens examined confidently identified specimens). In using this last criterion,
we took into account the presence of major geographic barri-
Specimens of Phyllotis used in this study (Figure 1 and ers between populations, environmental continuity and the ab-
Supplementary Material S1) are housed in the following sence of obvious discontinuities in size and shape of the skull
Argentinean natural history collections: Centro Regional (and in the external appearance of the skins) within groups (see
de Investigaciones Científicas y Transferencia Tecnológica Teta et al., 2018 and references there for a similar procedure).
de La Rioja (CRILAR), La Rioja; Instituto Argentino
de Investigaciones de Zonas Áridas (CMI), Mendoza;
Fundación-­Instituto Miguel Lillo (CML), San Miguel 2.4 | DNA Sequencing
de Tucumán; Museo Argentino de Ciencias Naturales
‘Bernardino Rivadavia’ (MACN-­ Ma), Ciudad Autónoma We sequenced the partial or complete cytochrome b gene
de Buenos Aires; Museo Municipal de Ciencias Naturales (cytb; 419 to 1,144 bp) for 47 individuals from localities of
‘Lorenzo Scaglia’ (MMPMa), Mar del Plata; Centro Nacional Mendoza and Neuquén provinces (Supplementary Material
Patagónico (CNP), Puerto Madryn. We examined 654 speci- S1). These specimens were collected from an overlapping but
mens of this genus, including types and topotypes of several more extensive range to that of P. sp. 2 in Ojeda et al. (2021).
nominal forms allied to the P. xanthopygus species com- Amplifications and sequencing followed Steppan et al. (2007)
plex, including P. bonariensis Crespo, 1964 (holotype and using primers P484 and P485 for PCR with 35–­40 cycles
paratypes); P. caprinus Pearson, 1958 (topotypes); P. dar- of 94°C (30–­45 s), 55°C (45 s), and 72°C (90 s) between
wini vaccarum Thomas, 1912 (topotypes); P. ricardulus an initial denaturation at 94°C (2 min) and a terminal 72°C
Thomas, 1919 (topotypes); and P. oreigenus Cabrera, 1926 elongation (6–­7 min). Sequences were aligned and individu-
(topotypes). ally checked for quality in Sequencher (GeneCodes Corp.)
4
|    JAYAT et al.

F I G U R E 1 Maps showing type localities of the nominal forms related to the P. xanthopygus complex (stars) and the sequenced specimens
included in Ojeda et al. (2021) (left), and the collecting localities of the specimens studied in the morphometric analyses (right). Phyllotis
bonariensis (turquoise circles); P. caprinus (orange circles); clade of P. x. posticalis-­P. x. rupestris (light blue circles); clade of P. sp. 1 (green
circles); clade of P. sp. 2 (red circles); clade of P. vaccarum (yellow circles); clade of P. xanthopygus s.s. (pink circles)

and Mesquite v3.08 (Maddison & Maddison, 2017) before 2.6 | Morphometric analyses
further analysis. Sequences have been deposited on Genbank
with accession numbers MZ298848-­MZ298894. Standard external measurements were recorded from speci-
men labels or museum catalogs: total body length (includ-
ing body plus tail length), TBL; tail length, T; hind foot
2.5 | Phylogenetic analysis length (including claw), HF; ear length, E; and weight, W.
The following 24 skull measurements were recorded with
The 47 new sequences for cytb were combined with 24 pre- digital calipers to the nearest of 0.01 mm following defini-
viously published sequences representing P. sp. 2, P. bon- tions provided by Hershkovitz (1962), Myers (1989) and
ariensis, P. caprinus, P. darwini, P. limatus, P. vaccarum, Myers et al. (1990): total length of the skull, TLS; con-
P. xanthopygus s.s., P. x. chilensis, P. x. posticalis and dyloincisive length, CIL; basal length, BL; palatal length,
P. x. rupestris for phylogenetic analysis. Sequences were PL; diastema length, DL; palatal bridge, PB; maxillary too-
aligned using Clustal Omega (Sievers et al., 2011), and op- throw length, MTRL; bullar length less tube, BLLT; bul-
timization of phylogenetic inferences was performed by lar breadth, BuB; incisive foramina length, IFL; alveolar
RAxML (Stamatakis, 2006), under the GTR+Γ substitution width 1 (across external side of both M1), AW1; alveolar
model selected by AIC in jModeltest2 (Darriba et al., 2012), width 2 (across external side of both M3), AW2; zygo-
using five batches of 100 searches, partitioned by codon posi- matic length, ZL; zygomatic plate depth, ZP; zygomatic
tion, and random seed values for initial trees. To infer sup- breadth, ZB; braincase breadth, BB; interorbital constric-
port for clades, 1,000 bootstrap replicates were conducted. tion, IOC; mid-­rostral width, RW2; nasal length, NL; ros-
Patristic distances within and between clades for cytb were tral length, RL; orbital length, OL; occipital condyle width,
estimated in Mega X (Kumar et al., 2018). OCW. In addition, we recorded the following mandible
JAYAT et al.   
| 5

measurements: mandible length, ML; mandibular toothrow 3 | RESULTS

length, mTRL. A total of 604 specimens were assigned to
five age classes (age class 1 = 188; age class 2 = 191; age 3.1 | Taxonomic assignments
class 3 = 129; age class 4 = 61; age class 5 = 35) based on
tooth wear following the criteria figured and described by Populations of Phyllotis vary little in skull and molar mor-
Jayat et al. (2016). phology. Nonetheless, several qualitative characters allow
Descriptive morphometric and univariate comparisons us to allocate the specimens to each one of the four clades
(ANOVA several sample test and Tukey's pairwise com- recovered by Ojeda et al. (2021) in the central Andes, from
parison) for the specimens assigned to each one of the main southwestern Bolivia to central western Argentina. Below we
clades recovered in the phylogenetic analysis developed by describe, from north to south, the morphological characteris-
Ojeda et al. (2021) and for samples of each one of the nomi- tics distinguishing each of these clades.
nal forms were carried out with the software PAST (Hammer
et al., 2001). Specimens of age classes 2 and 3 were pooled
(to obtain largest samples between available specimens) in 3.1.1 | The clade of
tests of significant body and skull size differences (for both, P. x. posticalis-­P. x. rupestris
p ≤ .05 and p ≤ .01).
In order to reduce the dimensionality of morphometric We found qualitative morphological evidence for recognizing
data and explore the differences in size of the skull among three taxa of the P. xanthopygus species complex that occur
samples, we used a Principal Component Analysis (sPCA) in sympatry in some localities, in the Puna and Altos Andes
for individuals in age classes 2 and 3. For this analysis, PCs arid ecoregions of northwestern Argentina, in Jujuy and Salta
were extracted from the variance–­covariance matrix and provinces (Figure 1). One of these groups, assigned to the
computed by using variables after transformations to Log clade of P. x. posticalis-­P. x. rupestris (Ojeda et al., 2021),
10. We also conducted a Discriminant Function Analysis is characterized by specimens with small and delicate skulls,
(sDFA) using the same set of specimens with the aim to with the interorbital region with rounded (not sharp-­edged)
evaluate significant differences among predefined groups borders, the frontoparietal suture mostly ‘U’ shaped, narrow
and check for misclassified specimens. Subsequently, we rostrum (Figure 2), and by a simplified enamel molar patterns
explored differences in the shape of the skull among sam- on the upper molar series, which are characterized by a very
ples using a ‘size-­ free’ Principal Component Analysis shallow paraflexus on M2 and no trace of the mesoloph com-
(sfPCA) and a ‘size-­free’ Discriminant Function Analysis plex (sensu Barbière et al., 2019) on M1 and M2 (Figure 3).
(sfDFA), including all individuals without missing data The clade of P. caprinus.
(and pooling specimens of the five age classes). To reduce Other specimens registered in the extreme north of our
the effect of size, Mosimann shape variables were calcu- study area (Figure 1) show morphological state charac-
lated through geometric mean transformation of data prior ters that closely match to those described for P. caprinus.
to statistical analyses (as developed by Meachen-­Samuels Specimens assigned to this clade have large and heavy skulls,
& Van Valkenburgh, 2009). Statistical significance of with a flat, sharp-­edged, long-­waisted interorbital region, a
the PCs were evaluated following the Broken-­ stick test predominantly ‘V’ shaped frontoparietal suture and a com-
(Frontier, 1976; Jackson, 1993). All multivariate statisti- paratively heavy rostrum (Figure 2). All specimens assigned
cal analyses, based on partial datasets (e.g. only specimens to this clade come from the type locality of P. caprinus or
without missing values), were performed with the software nearby areas in Jujuy Province (Figure 1).
PAST (Hammer et al., 2001). The clade of P. vaccarum.
To visualize geographic variation in the univariate space From the western limit between Jujuy and Salta prov-
and search for clinical patterns or morphometric discontinu- inces to southern Mendoza Province (Figure 1), we found
ities within the clade of P. vaccarum of Ojeda et al. (2021), qualitative and quantitative morphologic characters for rec-
we constructed Dice-­Leraas diagrams (modified from Hubbs ognizing two other taxa of the P. xanthopygus complex. One
& Hubbs, 1953) showing the median, the 25%–­75% quar- of them corresponds to individuals that are representative
tiles and the minimum and maximum values (calculated on of the clade of P. vaccarum (Ojeda et al., 2021), which is
the base of the geometric mean of each individual). We used sympatric, in the arid cold Puna ecoregion of southwest-
specimens of age classes 2 and 3, including only those locali- ern Jujuy and northwestern Salta provinces, with specimens
ties with more than 10 specimens (in some cases, we grouped corresponding to the form here referred to P. x. postica-
geographically close localities to achieve this minimum lis-­P. x. rupestris (Ojeda et al., 2021, Figure 1). Populations
sample size; see Libardi & Percequillo, 2016 for a similar assigned to the clade of P. vaccarum differ from those of the
analysis). clade of P. x. posticalis-­P. x. rupestris by having larger and
6
|    JAYAT et al.

F I G U R E 2 Dorsal view of the skull showing some morphological F I G U R E 3 Occlusal molar view showing the simplified enamel
character states differentiating specimens of the clade of molar pattern (very shallow or absent paraflexus on M2 and no trace
P. x. posticalis-­P. x. rupestris (left: MACN-­Ma 29596) and P. caprinus of mesoloph complex on M1 and M2) on the upper molar series of
(right: MACN-­Ma 29425) in northwestern Argentina. Notice the specimens of the clade of P. x. posticalis-­P. x. rupestris (left: MACN-­
smaller and more delicate skulls, the rounded (not sharp-­edged) Ma 29594), in comparison with the more complex pattern generally
interorbital region, the frontoparietal suture mostly ‘U’ shaped and the observed in specimens of the clade of P. vaccarum (right: MACN-­Ma
narrow rostrum in the specimen of P. x. posticalis-­P. x. rupestris. Scale 29565). Photographs are not in scale
bar = 10 mm

heavier skulls, more robust molar series, and more complex to the west in the higher elevations of Chile (Steppan &
enamel molar patterns, which are characterized by a gen- Ramírez, 2015; Storz et al., 2020).
erally very deep paraflexus on M2 and by the presence of
a poorly developed but clearly visible mesoloph complex
on M1 and M2 (Figure 3). This clade is morphologically 3.1.2 | The clade of P. sp. 2
cohesive and encompasses several of the nominal forms
described for the region (i.e. oreigenus with type locality We recorded specimens of this clade in sympatry (and in
in Laguna Blanca, north-­central Catamarca Province; ricar- syntopy), in at least one locality, with populations of the
dulus with type locality in Otro Cerro, southern Catamarca clade of P. vaccarum in southwestern Mendoza Province
Province; and darwini vaccarum with type locality in Punta (Figure 1). The skulls of specimens of both clades are very
de Vacas, northwestern Mendoza Province). We examined similar, but the molars in P. sp. 2 are appreciably less robust
type, paratypes or topotypes of all these nominal forms and and show a more simplified enamel molar pattern, without
several geographic intermediate populations from west- the deep paraflexus on M2 and the mesoloph complex on M1
ern Argentina (in La Rioja, Salta, San Juan, San Luis, and and/or M2 observed in specimens of the clade of P. vacca-
Tucumán provinces) and did not find constant qualitative rum (Figure 6). We also observed some integumental char-
morphologic differences in the skull and enamel molars pat- acters (see the discussion section) that differentiate both taxa
terns to distinguish among them (Figures 4 and 5). In south- and facilitate the assignment of these specimens in sympatric
western Mendoza and northwestern Neuquén provinces areas.
(Figure 1), representatives of the clade of P. vaccarum are DNA samples from throughout the range of P. sp. 2, as
sympatric with populations corresponding to the clade of P. delimited by the morphological characters, and sequenced
sp. 2 (Ojeda et al., 2021), which represent an undescribed for cytb to test the species limits, greatly expanded the
species that will be treated in depth in following paragraphs range for DNA data beyond that in Ojeda et al. (2021). The
and in the discussion section of this study. Cytb sequence phylogeny (Figure 7) clearly corroborates the prior results,
data indicate that the clade of P. vaccarum extends further with the 47 new sequences from all 10 localities assigned
JAYAT et al.   
| 7

F I G U R E 4 Dorsal, ventral and lateral view of the skull, and lateral view of the mandible of topotype specimens of the nominal forms
Phyllotis oreigenus (left and top: MACN-­Ma 29565), P. ricardulus (centre: MACN-­Ma 29605) and P. darwini vaccarum (right and bottom: CML
4480). Scale bar = 10 mm

to P. sp. 2 by morphology forming a well-­differentiated mi- populations representing each clade for all the metric state
tochondrial clade along with the eight sequences from the characters analysed (Supplementary Material S2). Tukey's
four localities from Ojeda et al. (2021). Genetic distances pairwise comparisons indicated one external (W) and sev-
(p) ranged from 0% to 3.5% within the clade of P. sp. 2 eral skull (e.g. TSL, CIL, MTRL, BLLT, BuB, IFL, AW1,
with a mean of 1.5%, and 8.8%–­11.5% among members of AW2, ZL, ZB, mTRL) measurements that significantly
this clade and the other taxa of the P. xanthopygus complex differed among most of these main clades (Supplementary
discussed here. Material S2).
Specimens of the clade of P. x. posticalis-­P. x. rupestris
were the smallest among the studied samples. This holds for
3.2 | Morphometric analyses all the studied characters with only two exceptions, BLLT
and the BuB, which averaged larger in this sample compared
Descriptive morphometrics comparing five of the eight main with P. caprinus (Table 1). This clade differed in 14 or more
clades obtained by Ojeda et al. (2021) are summarized in characters when compared with populations of the other
Table 1. The ANOVA showed a good separation among clades (Supplementary Material S2).
8
|    JAYAT et al.

F I G U R E 5 Occlusal molar view

showing the enamel molar pattern (well-­
developed paraflexus on M2 and visible
mesoloph complex on M1 and M2) on
the upper molar series of specimens of
the nominal forms Phyllotis oreigenus
(left: MACN-­Ma 29565), P. ricardulus
(centre: MACN-­Ma 29601) and P. darwini
vaccarum (right: CML 4478)

Material S2). This species was morphometrically very similar

to samples of specimens of the clade of P. vaccarum, but they
differed in four measurements (Supplementary Material S2).
Phyllotis caprinus shows, in average, the smaller ears (E) and
bullar measurements (both, BLLT and BuB), but the broader in-
terorbital constriction (IOC) of all the studied samples (Table 1).
With the exception of P. caprinus, samples of the clade of
P. vaccarum clearly separate from other clades. Specimens of
this clade were intermediate in size (smaller than P. bonar-
iensis and the clade of P. sp. 2 and larger than the clade of
P. x. posticalis-­P. x. rupestris) (Table 1), differing by 20 or
more morphometric state characters when compared with the
other clades (Supplementary Material S2). Univariate statis-
tical analyses showed scarce differentiation among the nomi-
nal forms included within this clade (Supplementary Material
S3). Only the nominal form ricardulus could be somewhat
separated from the nominal forms oreigenus and vaccarum
(and from populations representing the entire geographic
range of the clade of P. vaccarum in western Argentina). This
nominal form averaged significantly smaller than samples of
the clade of P. vaccarum (in nine skull measurements), and
topotype specimens of the oreigenus (eight skull measure-
F I G U R E 6 Occlusal molar view showing the simplified enamel
ments) and the vaccarum (one external and six skull mea-
molar pattern (very shallow or absent paraflexus on M2 and no trace
of mesoloph complex on M1 and M2) on the upper molar series of
surements) nominal forms.
specimens of the clade of P. sp. 2 (left: CMI 7638), in comparison Samples of specimens here assigned to the clade of
with the more complex pattern generally observed in specimens of the P. sp. 2 were comparatively large sized, averaging larger
clade of P. vaccarum (right: MACN-­Ma 29565). Photographs are not for most measurement when compared with all clades
in scale excepting P. bonariensis (Table 1). This last species was
the most similar, but it averaged significantly larger for
Samples representatives of P. caprinus were intermediate 13 measurements, including both measurements of the
in size, showing several morphometric state characters sig- length (e.g. TLS, CIL, PL, DL, PB) and the width (e.g.
nificantly differing from P. bonariensis, the clade of P. sp. 2., AW1, ZB, IOC, RW2, OCW) of the skull (Supplementary
and the clade of P. x. posticalis-­P. x. rupestris (Supplementary Material S2).
JAYAT et al.   
| 9

F I G U R E 7 Maximum likelihood
phylogeny of cytb sequences including all
new and published samples for P. sp. 2 and
a representative sampling for other taxa in
the Phyllotis xanthopygus species complex.
Individuals are identified by Genbank
accession number or for newly sequenced
individuals, by the collector's number (see
Supplementary Material S1). Localities are
indicated in (parentheses), corresponding
to those listed in the Specimens Examined
(Supplementary Material S1). Numbers
above nodes are bootstrap percentages
for values >50%. Scale bar is expected
substitution rate
 10
|
TABLE 1 External and craniodental measurements for specimens (age classes 2 and 3) of five of the eight main clades obtained by Ojeda et al. (2021) for the Phyllotis xanthopygus complex

P. bonariensis P. caprinus P. x. posticalis-­P. x. rupestris P. vaccarum P. sp. 2

  

n X̅ ± SD r n X̅ ± SD r n X̅ ± SD r n X̅ ± SD r n X̅ ± SD r
TBL 5 255 ± 9.37 240–­262 8 225 ± 16.40 202–­245 57 224 ± 18.07 129–­253 92 235 ± 22.48 120–­274 14 240 ± 15.14 216–­280
T 5 124 ± 9.97 110–­135 8 118 ± 16.22 82–­133 57 114 ± 8.33 95–­129 92 120 ± 13.50 80–­152 14 115 ± 7.43 102–­135
HF 5 27 ± 1.14 25–­28 11 27 ± 2.31 21–­29 56 25 ± 1.62 19–­29 94 28 ± 1.68 23–­32 15 30 ± 2.50 25–­33
E 5 24 ± 0.71 23–­25 11 22 ± 1.42 19–­24 56 25 ± 1.87 20–­30 94 25 ± 1.61 21–­30 15 25 ± 1.45 22–­27
W 5 72 ± 12.58 62–­87 9 38 ± 10.78 22–­59 49 41 ± 7.64 25–­55 83 47 ± 11.02 27–­79 14 55 ± 14.29 36–­80
TLS 6 33.04 ± 1.38 31.74–­35.24 11 30.47 ± 1.91 28.00–­34.68 65 28.93 ± 1.97 26.51–­30.99 99 30.02 ± 1.20 27.26–­32.70 15 31.43 ± 0.94 30.29–­33.56
CIL 6 30.95 ± 1.55 29.35–­33.61 13 27.70 ± 2.01 24.06–­31.92 66 26.54 ± 1.03 23.83–­28.73 101 27.78 ± 1.30 24.82–­30.70 15 28.88 ± 0.93 27.77–­30.97
BL 6 28.52 ± 1.49 26.98–­31.20 13 25.23 ± 2.03 21.61–­29.56 64 24.43 ± 1.03 21.62–­26.43 101 25.50 ± 1.28 22.82–­28.89 15 26.62 ± 0.81 25.39–­28.31
PL 6 17.47 ± 0.64 16.87–­18.44 14 15.27 ± 0.95 13.23–­17.24 68 14.75 ± 0.58 13.44–­15,93 101 15.64 ± 0.74 13.97–­17.38 16 16.30 ± 0.53 15.58–­17.35
DL 6 8.86 ± 0.49 8.33–­9.71 15 7.62 ± 0.65 6.18–­8.84 69 7.42 ± 0.39 6.37–­8.20 102 7.69 ± 0.45 6.65–­8.83 16 8.08 ± 0.47 7.57–­8.99
PB 6 6.44 ± 0.29 6.12–­6.82 14 5.42 ± 0.29 4.97–­5.92 68 5.22 ± 0.29 4.35–­6.26 101 5.62 ± 0.36 4.63–­6.61 16 5.76 ± 0.28 5.14–­6.29
MTRL 6 5.97 ± 0.25 5.57–­6.20 15 5.27 ± 0.26 4.85–­5.76 70 5.00 ± 0.16 4.59–­5.30 103 5.42 ± 0.28 4.81–­6.16 16 5.85 ± 0.19 5.59–­6.20
BLLT 6 5.96 ± 0.34 5.42–­6.32 12 5.15 ± 0.34 4.60–­5.73 68 5.41 ± 0.26 4.58–­6.00 101 5.59 ± 0.35 4.91–­6.32 15 6.10 ± 0.24 5.67–­6.51
BuB 6 5.08 ± 0.22 4.67–­5.28 12 4.33 ± 0.34 3.70–­4.84 68 4.72 ± 0.16 4.44–­5.15 101 4.90 ± 0.24 4.26–­5.39 15 5.24 ± 0.19 4.86–­5.57
IFL 6 7.99 ± 0.53 7.20–­8.77 15 6.99 ± 0.54 5.80–­8.03 69 6.84 ± 0.42 5.34–­7.63 102 7.07 ± 0.46 5.92–­8.19 16 7.41 ± 0.31 6.99–­8.20
AW1 6 6.25 ± 0.30 5.94–­6.62 15 5.73 ± 0.17 5.39–­6.07 68 5.73 ± 0.16 5.33–­6.10 103 5.81 ± 0.22 5.32–­6.37 16 5.95 ± 0.20 5.55–­6.32
AW2 6 5.68 ± 0.19 5.35–­5.86 15 5.18 ± 0.25 4.76–­5.57 68 5.00 ± 0.18 4.59–­5.31 103 5.24 ± 0.25 4.76–­5.79 16 5.43 ± 0.23 5.09–­5.84
ZL 6 17.06 ± 0.78 16.03–­18.38 15 15.80 ± 0.82 14.45–­17.35 68 14.95 ± 0.60 13.26–­16.15 103 15.73 ± 0.72 14.18–­17.68 16 16.56 ± 0.59 15.67–­17.49
ZP 6 3.75 ± 0.14 3.60–­3.97 15 3.34 ± 0.27 2.78–­3.74 70 3.26 ± 0.25 2.51–­3.72 103 3.41 ± 0.27 2.79–­4.13 16 3.59 ± 0.18 3.20–­3.87
ZB 6 17.18 ± 0.76 16.50–­18.50 14 15.44 ± 0.78 14.06–­16.92 66 14.83 ± 0.54 13.38–­16.15 103 15.54 ± 0.61 14.20–­16.78 16 15.99 ± 0.48 15.01–­16.54
BB 6 14.38 ± 0.33 13.85–­14.74 15 13.52 ± 0.40 12.88–­14.26 68 13.31 ± 0.34 12.65–­14.23 102 13.70 ± 0.36 12.81–­14.46 15 13.94 ± 0.31 13.41–­14.46
IOC 6 4.44 ± 0.19 4.18–­4.69 15 4.53 ± 0.29 4.22–­5.19 69 4.16 ± 0.15 3.85–­4.59 103 4.31 ± 0.19 3.92–­4.82 16 4.14 ± 0.18 3.85–­4.44
RW2 6 5.95 ± 0.33 5.58–­6.48 15 5.06 ± 0.40 4.18–­5.91 68 4.88 ± 0.30 4.25–­5.55 103 5.01 ± 0.28 4.43–­5.72 16 5.10 ± 0.28 4.67–­5.63
NL 6 14.45 ± 0.61 13.38–­15.10 13 12.79 ± 0.72 11.68–­14.07 68 12.4 ± 0.60 10.76–­13.85 101 12.64 ± 0.75 11.03–­14.59 16 13.77 ± 0.56 12.96–­14.75
RL 6 12.96 ± 0.53 12.58–­13.88 13 11.64 ± 0.77 10.41–­13.44 68 11.10 ± 0.48 10.10–­12.45 101 11.56 ± 0.66 10.15–­13.09 16 12.30 ± 0.47 11.67–­13.33
OL 6 11.21 ± 0.41 10.70–­11.80 15 10.14 ± 0.60 9.23–­11.17 69 9.50 ± 0.42 8.27–­10.39 103 10.19 ± 0.41 9.02–­11.17 16 10.74 ± 0.38 10.26–­11.42
OCW 6 7.40 ± 0.21 7.12–­7.66 13 7.09 ± 0.19 6.78–­7.46 65 6.75 ± 0.20 6.28–­7.25 102 7.08 ± 0.24 6.60–­7.78 15 7.07 ± 0.20 6.79–­7.39
ML 6 17.51 ± 0.82 16.33–­18.72 15 15.59 ± 0.91 14.23–­17.50 68 15.22 ± 0.53 13.90–­16.32 103 15.98 ± 0.73 14.23–­17.89 16 16.92 ± 0.62 16.15–­18.54
mTRL 6 5.90 ± 0.16 5.74–­6.10 15 5.34 ± 0.27 5.00–­5.91 69 5.03 ± 0.18 4.63–­5.38 103 5.36 ± 0.23 4.78–­5.96 16 5.77 ± 0.16 5.44–­5.96
Note: Measurement abbreviations are listed in Materials and Methods.
Abbreviations: n, sample size; r, range; SD, standard deviation; X̅, mean.
JAYAT et al.
JAYAT et al.   
| 11

The first 3 principal components of the sPCA accounted of the P. x. posticalis-­P. x. rupestris, the P. vaccarum, and
for 76.55% of the variability in the data set, but only the the P. sp. 2 clades, mainly by metric characters related to
first (64.21% of the variance explained) was judged statis- the bulla (BLLT and BuB), the molar row series (MTRL
tically significant by the Broken-­ stick test (Table 2). PC and mTRL) and the rostrum (RW2) (Table 2 and Figure 8).
I was a size component, as all the variables had equal sign Phyllotis caprinus shows smaller bullae and molar series but
and loaded heavily on this axis (we confirm this by regress- a broader rostrum. In summary, the bivariate plot of PC I and
ing PC I scores against several total length measures). This PC II shows a good separation among samples of all studied
PC mainly separated samples of the small-­sized P. x. pos- clades with the only exception of P. vaccarum, which is more
ticalis-­P. x. rupestris on the negative side, from the larger variable and widely overlaps with specimens of the P. x. pos-
P. sp. 2 (and P. bonariensis) located well on the positive ticalis-­P. x. rupestris and the P. sp. 2 clades (Figure 8).
side. Samples of P. caprinus and specimens representing The sDFA (Table 2) completely separated the sam-
the clade of P. vaccarum occupied an intermediate position, ples of P. x. posticalis-­P. x. rupestris, P. caprinus, P. sp. 2
widely overlapping on this PC with other samples (excepting and P. bonariensis. Specimens of the clade of P. vaccarum
P. bonariensis) (Figure 8). PC II mostly separated specimens slightly overlapped with those of P. caprinus and P. x. pos-
of P. caprinus (well on the negative side), from the samples ticalis-­P. x. rupestris, but heavily overlapped with samples

T A B L E 2 Results of the sPCA and sDFA comparing five of the eight main clades obtained by Ojeda et al. (2021) for the Phyllotis
xanthopygus complex

Eigenvectors Eigenvectors

Variables PC 1* PC 2 PC 3 DF 1 DF 2 DF 3 DF 4
TLS 0.20189 −0.091675 0.018215 0.006791 −0.00080839 0.0073229 0.00065257
CIL 0.22862 −0.11153 0.021123 0.0076825 −0.0011583 0.007479 −0.0014684
BL 0.23718 −0.10804 −0.017943 0.0075585 −0.000164 0.0081219 −0.001697
PL 0.24086 −0.060011 0.07054 0.0091748 −0.00032107 0.0068221 −0.0030643
DL 0.25007 −0.26457 −0.11006 0.0070834 −0.0001997 0.010655 −0.0033772
PB 0.23512 0.14808 0.44351 0.011039 −0.00049466 0.0069655 −0.0085175
MTRL 0.22986 0.38844 0.23825 0.012853 0.002658 0.0061106 0.0010651
BLLT 0.18466 0.42653 −0.45067 0.0072513 0.0092084 0.0065521 0.0013531
BuB 0.17913 0.3782 −0.24409 0.0073746 0.011615 0.002334 −0.0037874
IFL 0.24283 −0.26767 −0.25328 0.0064484 0.0009399 0.0099851 −0.0016779
AW1 0.093554 −0.018907 0.035335 0.0030208 5.39E−01 0.0053949 −0.0023438
AW2 0.15597 0.17874 0.082756 0.0074285 −0.0005764 0.0047833 −0.0017979
ZL 0.21383 0.02365 −0.025498 0.0080517 1.89E−01 0.005583 0.00071986
ZP 0.26197 −0.024232 −0.22837 0.0075098 0.0035559 0.0069328 −0.0030443
ZB 0.18422 −0.051331 0.099502 0.007143 −0.00086808 0.0062946 −0.0034899
BB 0.095676 0.044527 0.080576 0.004472 0.00026256 0.0027563 −0.0018798
IOC 0.038667 −0.1387 0.39915 0.0032077 −0.0076132 −0.0014234 −0.0045258
RW2 0.20408 −0.39312 0.05642 0.0047758 −0.0038956 0.01309 −0.0059822
NL 0.23231 −0.11651 −0.21014 0.0060147 0.0023477 0.013055 0.0029131
RL 0.24461 −0.071999 −0.067777 0.0076722 0.0002206 0.0092537 0.00065939
OL 0.22583 −0.032701 0.14077 0.010559 −0.00072364 0.0056227 −0.00027935
OCW 0.10848 0.024653 0.21529 0.0059913 −0.0023479 0.00071538 −0.0033741
ML 0.22335 0.0079482 0.0024213 0.0085466 0.0018427 0.0064958 −1.06E−01
mTRL 0.1918 0.3035 0.2163 0.010947 0.00097811 0.0069765 0.0030469
Eigenvalue 0.00982538 0.00104763 0.000839976 2.6777 1.1085 0.73487 0.40623
% variance 64.21 6.85 5.49 54.34 22.50 14.91 8.244
Note: Phyllotis bonariensis (N = 6), P. caprinus (N = 8), clade of P. x. posticalis-­P. x. rupestris (N = 59), clade of P. sp. 2 (N = 15) and clade of P. vaccarum
(N = 97). Loadings of the variables, eigenvalues and proportion of the variance explained for the first 3 principal components (PC) and the first 4 discriminant
functions (DF). Results are based on log10-­transformed craniodental variables. See ‘Material and Methods’ for variable abbreviations.
12
|    JAYAT et al.

F I G U R E 8 Individual specimen
scores based on log-­transformed values of
24 cranial measurements projected onto
the first and second principal components
of the sPCA (top), and the first and second
discriminant functions (DF) of the sDFA
(bottom) extracted from analysis of
specimens (age classes 2 and 3) of Phyllotis
bonariensis (turquoise dots, N = 6),
P. caprinus (orange dots, N = 8), clade of
P. x. posticalis-­P. x. rupestris (light blue
dots, N = 59), clade of P. sp. 2 (red dots,
N = 15) and clade of P. vaccarum (yellow
dots, N = 97). Character loadings and the
variance explained by each of the first
three principal components and each of
the four discriminant functions appear in
Table 2

F I G U R E 9 Individual specimen
scores based on log-­transformed values
of 24 cranial measurements (Mosimann
shape variables) projected onto the first
and second principal components of the
sfPCA (top) and the first and second
discriminant functions (DF) of the sfDFA
(bottom) extracted from analysis of
specimens (all age classes) of Phyllotis
bonariensis (turquoise dots, N = 6),
P. caprinus (orange dots, N = 16), clade of
P. x. posticalis-­P. x. rupestris (light blue
dots, N = 155), clade of P. sp. 2 (red dots,
N = 47) and clade of P. vaccarum (yellow
dots, N = 283). Character loadings and
the variance explained by each of the first
three principal components and each of
the four discriminant functions appear in
Table 3
JAYAT et al.   
| 13

referred to the clade of P. sp. 2 and P. bonariensis (Figure 8). already documented (e.g. Pearson, 1958; Steppan, 1998; Teta
DF I (54.34% of the variance explained) segregated the sam- et al., 2018). At least in part, this morphological homogene-
ples mostly by the molar tooth rows (upper and lower), the ity explains the assignment of most of these populations (i.e.
orbital length (OL) and the palatal bridge (PB). The DF II nominal forms) to the several subspecies of P. xanthopygus
also separated well P. caprinus (on the negative side) from s.l. In this work, we also documented those morphometric
the other samples. Metric characters related to the bulla, the similarities for the populations of the central Andes. Despite
rostrum and the interorbital region were the most influential strong genetic divergences that indicate a significant period
(Table 2). The percentage of correct classifications following of isolation and divergence (having been suggested began be-
the jackknifed confusion matrix of this analysis was 80.54% tween 2.83 and 4.05 MA AP in the Pliocene; Riverón, 2011),
(Supplementary Material S4). Most of the specimens of the this low morphological variability could be explained by
clades of P. x. posticalis-­P. x. rupestris (93%), P. sp. 2 (93%), strong stabilizing selection favouring niche conservatism and
P. vaccarum (72%) and P. bonariensis (83%) were correctly phenotypic stasis, resulting in cryptic species. Nonetheless,
classified, but P. caprinus obtained intermediate values (63%). there are some differences in size and shape of the skull that
The sfPCA showed a large overlap for most of the studied may be appropriate to establish morphometric limits. In the
samples, but specimens corresponding to P. caprinus (lower following paragraphs we discuss these morphometric differ-
left quadrant) and the clade of P. sp. 2 (upper left quadrant) ences and other sources of evidence (qualitative morphology
appeared as well separated each other on the bivariate plot of of the skull and molars, cytogenetic data and geographic dis-
PC I and PC II (Figure 9). This separation was mostly influ- tribution patterns), which could be viewed as indicators of
enced by the molar series (MTRL and mTRL) and the bullae the specific status of these forms in light of the mitochondrial
(BLLT and BuB), which were larger in P. sp. 2, and the in- phylogeny.
terorbital constriction (IOC) and the mid-­rostral width, which
were wider in P. caprinus (Table 3). The first principal com-
ponent (the only judged statistically significant according to 4.2 | The clade of
the Broken Stick test) of this analysis summarized 44.71% of P. x. posticalis-­P. x. rupestris
the explained variance.
Similar results to that showed by the sfPCA were ob- According to the phylogenetic analyses of Ojeda et al. (2021;
tained with the sfDFA analysis, but in this case P. caprinus see also Albright, 2004 and Steppan et al., 2007), representa-
separated much better from P. sp. 2 (specimens of P. capri- tives of this clade in northwestern Argentina correspond to
nus occupy mostly the right upper quadrat and those of the a southern sub-­clade widely distributed from west-­central
clade of P. sp. 2 the left lower quadrat, of the multivariate Bolivia to extreme northwestern Argentina. We did not
space) and the rest of the samples, which overlapped heavily measure specimens of this clade from Bolivia or Peru, pre-
(Figure 9). This analysis also revealed the influence of the cluding us from resolving the status of Argentinean popu-
metric characters related to the bulla, the molar series and lations; nevertheless, we are confident that Argentinean
the interorbital region as determinant for the separation of specimens represent a morphological cohesive sample and
P. caprinus (which showed the smallest bullae and molar a different species from other clades of the eastern Andean
series, and the broadest interorbital constriction revealed flanks of western Argentina. Mean pairwise genetic dis-
by previous analyses; Table 3). The percentage of correct tances between P. x. posticalis-­P. x. rupestris and other
classifications following the jackknifed confusion matrix of clades of the P. xanthopygus complex range from 9.3% to
the sfPCA was 73.37% (Supplementary Material S4). Most 10.6% (see Table 1 in Ojeda et al., 2021), coinciding with (or
of the specimens of the clades of P. x. posticalis-­P. x. rup- well above of) the values observed among other species of
estris (86%), P. sp. 2 (83%) and P. bonariensis (83%) were Phyllotis (e.g. Jayat et al., 2016; Ojeda et al., 2021; Rengifo
correctly classified, but the other samples obtained inter- & Pacheco, 2017). Specimens of this clade from Jujuy
mediate values (65% for the clade of P. vaccarum and 63% Province show chromosome complements with 2n = 38,
for P. caprinus). FNa = 70–­71, including acrocentric autosomes with different
patterns of heterochromatin, which are not present in sam-
ples from Catamarca and northern Mendoza here assigned
4 | D IS C U SSION to the clade of P. vaccarum (with 2n = 38, FNa =72 and a
karyotype with all biarmed chromosomes and lower amount
4.1 | Species limits in the P. xanthopygus of heterochromatin). Specimens of P. sp. 2 have 2n = 38,
species complex of the central Andes FNa = 71–­72 and present more heterochromatin than mem-
bers of the clade of P. x. posticalis-­P. x. rupestris, including
The morphometric similarity (size and shape of the skull) in an exclusive acrocentric autosome mostly heterochromatic
populations of the P. xanthopygus species complex has been (Labaroni et al., 2014). Furthermore, our morphometric
14
|    JAYAT et al.

T A B L E 3 Results of the sfPCA and sfDFA comparing five of the eight main clades obtained by Ojeda et al. (2021) for the Phyllotis
xanthopygus complex

Eigenvectors Eigenvectors

Variables PC 1* PC 2 PC 3 DF 1 DF 2 DF 3 DF 4
TLS −0.064486 −0.00081786 0.09055 0.0005501 0.0014951 −0.0025847 −0.0018964
CIL −0.14443 −0.014142 0.10604 −0.00033915 0.0030881 −0.0031464 0.00019611
BL −0.17718 0.00015537 0.089086 −0.00036487 0.0020992 −0.0038998 0.00061931
PL −0.14729 0.054239 0.16222 −0.0023649 0.0024609 −0.000774 0.0021849
DL −0.28704 −0.087896 0.079574 0.00080222 0.0023844 −0.0054639 0.0050455
PB 0.032972 0.13568 0.45638 −0.003321 0.0029763 0.0035788 0.0072166
MTRL 0.21893 0.39578 0.10228 −0.0077591 −0.0022133 0.0020275 −0.0023089
BLLT 0.17843 0.3551 −0.5252 −0.00054067 −0.010588 0.0036351 −0.0024004
BuB 0.19548 0.24465 −0.28461 −0.0014576 −0.0096625 0.010565 −0.0011718
IFL −0.2404 0.022381 0.0039828 −0.00073701 0.0012578 −0.0039838 0.0022323
AW1 0.20388 −0.22688 −0.16201 0.0071951 −0.0028616 0.0043505 0.00060222
AW2 0.15842 −0.14289 −0.14568 0.0022153 −0.00090324 0.0022975 0.0005136
ZL −0.077017 −0.00257 −0.02344 −0.00028538 0.0022044 −0.00023174 −0.0018018
ZP −0.35194 −0.18375 −0.49253 −0.0020028 −0.001167 −0.0060887 −0.0013915
ZB 0.0080186 −0.098001 0.033125 0.001098 0.0012575 −7.54E−05 0.0010577
BB 0.24481 −0.13611 −0.01532 0.0047537 −0.0017854 0.0063639 −0.002689
IOC 0.40055 −0.49845 0.088899 0.0094794 0.0054394 0.0094494 −0.0034131
RW2 −0.1057 −0.30612 −0.049949 0.0029669 0.0020517 −0.0067643 0.0071927
NL −0.2201 0.11725 −0.03161 −4.89E−05 −0.0027306 −0.010025 −0.000524
RL −0.21935 0.1219 0.063147 −0.0021651 0.0006078 −0.006809 −0.0010116
OL −0.037728 0.019585 0.16407 −0.0038804 0.0037381 0.00080492 −0.00018664
OCW 0.25364 −0.13822 0.1041 0.0032537 0.0029731 0.0083603 −0.0027789
ML −0.079334 0.051403 0.063133 −0.0021905 −0.00022496 −0.0023266 −0.0022353
mTRL 0.25686 0.31782 0.12376 −0.0048572 −0.0018976 0.00073995 −0.0030516
Eigenvalue 0.0041704 0.00104178 0.00070736 1.3505 0.66717 0.31121 0.13745
% variance 44.71 11.17 7.58 54.76 27.05 12.62 5.57
Note: Phyllotis bonariensis (N = 6), P. caprinus (N = 16), clade of P. x. posticalis-­P. x. rupestris (N = 155), clade of P. sp. 2 (N = 47) and clade of P. vaccarum
(N = 283). Loadings of the variables, eigenvalues and proportion of the variance explained for the first 3 principal components (PC) and the first 4 discriminant
functions (DF). Results are based on log10-­transformed craniodental variables. See ‘Material and Methods’ for variable abbreviations.

analysis shows that representative specimens of this clade 4.3 | The clade of P. caprinus
are, on average, the smallest of all studied populations in the
eastern Andean slopes of Argentina (Table 1, Figures 8 and P. caprinus was described on the basis of external and skull
9). More importantly, significant size differences, for several (qualitative and quantitative) characters by Pearson (1958).
external and cranial traits, separate representatives of this Shortly after its original description, Hershkovitz (1962)
clade from the sympatric P. caprinus and P. vaccarum clades considered this nominal form to be a subspecies of the poly-
(Supplementary Material S2), and there are even some shape typic P. darwini (along with all taxa currently allocated to
differences in the skull with P. caprinus (Figure 9). The the P. xanthopygus complex), but most of the subsequent au-
rounded interorbital region, the frontoparietal suture mostly thors retained it as different species (e.g. Cabrera, 1961; Jayat
‘U’ shaped, the narrow rostrum and the simplified enamel et al., 2016; Steppan, 1998; Steppan and Ramírez, 2015). The
molar pattern on the upper molar series are qualitative char- close phylogenetic relationship among P. caprinus and other
acters that also support its differentiation (Figures 2 and 3). nominal forms included in the P. xanthopygus species com-
Despite these differences, the taxonomic status of this form plex was first mentioned by Jayat et al. (2016) based on molec-
still needs further studies, including evaluating samples of ular grounds. The phylogenetic analyses conducted by Ojeda
northern Bolivia and Peru. et al. (2021) consistently (in the Bayesian and the Maximum
JAYAT et al.   
| 15

Likelihood analyses) placed specimens of this clade as more Riverón, 2011; Steppan et al., 2007). Ojeda et al. (2021)
closely related to representatives of the clade of P. xanthopy- corroborated this relationship after adding specimens of the
gus s.s. (samples from Patagonia, south of the Limay river) nominal form ricardulus to the already included rupestris
than to specimens of the sympatric P. x. posticalis-­P. x. ru- and vaccarum, plus specimens from additional localities of
pestris and the P. vaccarum clades. Mean pairwise genetic western Argentina. Genetic p-­distances between P. limatus
distances among this and the other clades are large, between and the clade of P. vaccarum range from 2.6% to 3.7% and
7.7% and 10.2% (Ojeda et al., 2021), well in line with the are the smallest distances among the eight main clades ob-
values observed among other Phyllotis species (e.g. Jayat tained by Ojeda et al. (2021). Steppan (1998) first argued for
et al., 2016; Ojeda et al., 2021; Rengifo & Pacheco, 2017). the specific status of P. limatus on distinctive morphology
The karyotype described for this taxon is nearly identical to (e.g. uniquely quite narrow and deep incisors, short to mod-
other populations of the P. xanthopygus complex, having a erate molar row, light coloration, belly frequently white),
2n = 38, FNa = 72, with 36 size-­graded biarmed autosomes, monophyly of the mitochondrial lineage and geographic
a large submetacentric X, and a small metacentric Y chro- distribution (P. limatus are mainly restricted to western
mosome (Pearson & Patton, 1976). However, our morpho- Andean areas). Genetic sampling was however very limited
logical results are in line with the phylogenetic and genetic in Steppan (1998). Kuch et al. (2002) and Albright (2004)
evidence, supporting the specific status of this nominal form. suggested a recent speciation for P. limatus, which could be
Pearson (1958: pg. 434) noted the broad rostrum and interor- an independent lineage diverging in the last ~140,000 years.
bital breadth of this species when compared to other Phyllotis We examined specimens of P. limatus and corroborated its
species in general and with P. xanthopygus (=P. darwini) in distinctiveness from specimens of the clade of P. vaccarum
particular. We confirm the applicability of these metric char- with regard to the former's quite narrow and deep incisors.
acters in separating P. caprinus from most other samples here Specimens referred to P. vaccarum from western Argentina
studied, but also note the smaller ears, bullae, and molars of and central Chile have differences in the amount of consti-
this species (Table 1). As mentioned previously, representa- tutive heterochromatin of their karyotypes when compared
tives of this form could be separated from sympatric popula- with individuals of P. x. posticalis-­P. x. rupestris and P. sp.
tions of the clade of P. x. posticalis-­P. x. rupestris through 2 (cf. Labaroni et al., 2014; Walker et al., 1991). Mean pair-
several qualitative (see Pearson, 1958; Steppan and Ramírez, wise genetic distances between specimens of the clade of
2015; and Figures 2 and 3) and quantitative state characters P. vaccarum and those of other clades within the complex
(Table 1 and Supplementary Material S2). It is morphomet- of P. xanthopygus (excluding limatus) range from 7.9% to
rically more similar to specimens of the clade of P. vacca- 9.7% (see Table 1 in Ojeda et al., 2021), values that are in
rum. In addition to the morphometric characters separating line with other recognized species. Samples of specimens
them from most other samples, there are a few qualitative corresponding to this clade are intermediate in size between
morphological characters (e.g. the shape of the interorbital the populations here analysed, being on average larger than
region and the frontoparietal suture) that could be useful in samples of the sympatric P. x. posticalis-­P. x. rupestris
separating both forms. Furthermore, all multivariate analyses and somewhat smaller compared with the sympatric P. sp.
including size consistently separate these two forms at dif- 2 (Table 1). Although specimens of this clade overlapped
ferent ends (of the PCII and of the DF II) of the multivariate with samples of the P. x. posticalis-­P. x. rupestris and the P.
morphometric space (Figures 8 and 9). In short, our analysis sp. 2 clades in the univariate statistic comparisons, several
and several other sources of evidence indicate that the con- metric characters can be useful in segregate them (Table 1
sideration of this nominal form as a valid species is the more and Supplementary Material S2). The PCA and the DFA
strongly supported taxonomic hypothesis. This geographi- also show this overlap (both, in size and shape of the skull),
cally restricted form inhabits a relatively small geographic but the confusion matrix of the sDFA successfully classified
area in shrubby habitats on the eastern Andean slopes of more than 72% of its specimens. A more detailed quantitative
northwestern Argentina and southern Bolivia, between 2,100 and qualitative comparison of this form with representatives
and 4,500 m (Steppan & Ramírez, 2015; Figure 1) and is sep- of P. sp. 2 will be developed in the following paragraphs.
arate by more than 1,600 km of its sister P. xanthopygus s.s. This clade includes specimens from several populations
in the Patagonian steppe. on eastern Andean slopes (from northern Chile and north-
western Argentina south to northeastern Neuquén Province,
Argentina), three of them coming from the type localities
4.4 | The clade of P. vaccarum of the nominal forms ricardulus, oreigenus, and vaccarum.
Although we documented some morphometric differences
Phylogenies based on cytb have placed specimens of this in the univariate analysis for the nominal form ricardulus,
clade (alternatively referred as x. vaccarum, or x. rupestris in size and shape multivariate analyses showed that specimens
previous studies) as sister to P. limatus (e.g. Albright, 2004; of this nominal form (and those representing oreigenus
16
|    JAYAT et al.

and vaccarum) are well inside the convex hull polygons of not detected in other clades (Labaroni et al., 2014). Teta
other populations of this clade (sPCA and sfPCA analysis) et al. (2018) observed the morphometric similarities of pop-
or cannot be correctly classified consistently (sDFA and ulations of this clade with that of populations belonging to
sfDFA) from populations coming from different geographic other clades of the P. xanthopygus complex. Notwithstanding,
areas (Supplementary Material S3). Furthermore, we do specimens of this clade were distinguishable from other spe-
not find differences in qualitative skull characters between cies of this complex on univariate and multivariate analy-
them (Figures 4 and 5) or sharp morphometric discontinui- ses (Figures 8 and 9; Table 1 and Supplementary Material
ties among populations along latitudinal or environmental S3). These differences, complemented and integrated
gradients (Supplementary Material S3) to diagnose them as with other sources of evidence (e.g. genealogical relation-
species. All evidence analysed suggests that this clade must ships, genetic divergence, and karyotype characteristics),
be recognized at the species level and vaccarum appears as strongly indicate that this clade represents a distinct and
the most appropriate name for this species. Pearson (1958) unnamed species. Here, we describe and compare this new
used the name rupestris for a short tailed, small sized and entity with related species of the P. xanthopygus complex
pale form, distributed from southern Peru to northwestern (Supplementary Material S6). The new name was regis-
Argentina (in Jujuy Province); in contrast, individuals we tered in the Official Register of Zoological Nomenclature
assign to P. vaccarum are large-­bodied, long tailed and with (ZooBank) with number LSID urn:lsid:zoobank.
generally more richly coloured pelage, being distributed from org:pub:E39D9E98-­CC02-­4D2E-­AC32-­3573136FE6EE.
northern Chile and southern Jujuy, south to central Chile and
west-­central Argentina (cf. Pearson, 1958). Steppan (1998)
argued that the location of the rupestris type is uncertain be- 4.6 | Final remarks
cause the original description gives the location as ‘un trou
de rocher des hautes montagnes de Cobija’ (‘a rock hole in Although we think that accumulated evidence (i.e. molecu-
the high mountains of Cobija;’ Gervais 1841:51) and could lar, morphologic, cytogenetic) has contributed substantially
therefore either be near the coast in the coastal ranges east of to species delimitation of populations of the genus Phyllotis
Cobija (as argued by Heshkovitz, 1964) or high in the Andes distributed on western Andean slops of central Argentina, un-
near San Pedro de Atacama (as argued by Pearson, 1958). doubtedly there is a need to include additional approaches to
Both possible locations are at the extreme northwestern limit further refine our knowledge. These must include the study
of this species, rather than near the core of the distribution, of nuclear DNA sequences (the mtDNA gene tree may de-
but are also within the range of P. limatus, a species distin- part from the species tree due to lineage sorting of ancestral
guished principally by incisor shape, a trait not measured polymorphisms or later introgression), environmental niche
by Gervais. The type of rupestris is lost (Steppan, 1998), preferences and the biogeographic history of these popu-
making it impossible from the available evidence to deter- lations, among others. Similarly, there is a need for assess
mine which species that animal was a member of. Ojeda populations from the western side of the Andes, in particular
et al. (2021) also highlighted the convenience, in terms of Chilean populations, which could help to adequately deline-
taxonomic stability, of referring populations of specimens ate the species diversity of this complex species group. Small
representing this clade to vaccarum rather than rupestris. mammal diversity studies in Andean ecosystems are urgently
Consequently, we recognize vaccarum as the oldest avail- needed in light of major threats (e.g. habitat degradation as-
able name that can be assigned to this taxon (see the taxo- sociated to open mining operations, biological invasions, cat-
nomic discussion in Ojeda et al., 2021 and further comments tle overgrazing, exotic species and climate change; Ceballos
on the taxonomic account of this species in Supplementary & Ehrlich, 2006; Nuñez et al., 2009; Reborati, 2005; Zapata-­
Material S5). Here we formally recognize the specific status Ríos & Branch, 2016). Even when the systematics of the leaf-­
of this form, including a taxonomic account, an emended eared mice is one of the most studied in South America, the
diagnosis, a thorough analysis of morphologic variations description of the new species, P. pehuenche new sp., high-
and a geographic distribution description for the species lights a common theme in Neotropic mammalogy. That is,
(Supplementary Material S5). the need to fulfil the Linnean deficit through exploration and
description of the tremendous diversity of small mammals
within a phylogenetic and integrative taxonomic framework,
4.5 | The clade of P. sp. 2 particularly for those complexes of almost morphologically
cryptic species.
Evidence for the recognition of this clade was first provided
by phylogenetic analysis of molecular data (Albright, 2004; ACKNOWLEDGEMENTS
Riverón, 2011; Ojeda et al., 2021). In addition, specimens We thank R. Gonzalez, F. Barbiere, P. Cuello and A.
of P. sp. 2 have distinctive chromosome characteristics Tarquino for helping us during fieldwork. Ulyses F. Pardiñas
JAYAT et al.   
| 17

graciously provided tissues. We also thank the critical review Jackson, D. A. (1993). Stopping rules in principal components analysis:
for three anonymous reviewers on the first submission of A comparison of heuristical and statistical approaches. Ecology, 74,
2204–­2214. https://doi.org/10.2307/1939574
the manuscript. Financial support on this research was pro-
Jayat, J. P., D’Elia, G., Pardiñas, U. F. J., & Namen, J. G. (2007). A new
vided by the National Council for Science and Technology species of Phyllotis (Rodentia, Cricetidae, Sigmodontine) from the
of Argentina, CONICET (PIP 11220150100258CO), upper montane forest of the Yungas of northwestern Argentina. In: D.
Agencia Nacional de Promoción Científica y Tecnológica A. Kelt, E. P. Lessa, J. Salazar-­Bravo, & J. L. Patton (eds.). The quint-
(PICT 1636 and PICT 2012-­0050) and the National Science essential naturalist: honoring the life and legacy of Oliver P. Pearson,
Foundation (DEB-­0108422, DEB-­0841447, DEB-­1754748 pp. 775–­798. University of California, Publications in Zoology, 134.
to SJS). We are grateful by the support of our research https://doi.org/10.1525/calif​ornia/​97805​20098​596.003.0022
Jayat, J. P., Ortiz, P. E., González, F. R., & D’Elía, G. (2016). Taxonomy
institutions, universities and the Dirección de Recursos
of the Phyllotis osilae species group in Argentina; the status of the
Renovables, Mendoza, for collection permits (N° 461-­1-­04-­
“Rata de los nogales” (Phyllotis nogalaris Thomas, 1921; Rodentia:
03873. RES 405). Cricetidae). Zootaxa, 4083, 397–­ 417. https://doi.org/10.11646/​
zoota​xa.4083.3.5
ORCID Kuch, M., Rohland, N., Betancourt, J. L., Latorre, C., Steppan, S., &
J. Pablo Jayat https://orcid.org/0000-0002-6838-2987 Poinar, H. N. (2002). Molecular analysis of a 11 700–­year–­old ro-
dent midden from the Atacama Desert, Chile. Molecular Ecology,
R E F E R E NC E S 11, 913–­924. https://doi.org/10.1046/j.1365-­294X.2002.01492.x
Kumar, S., Stecher, G., Li, M., Knyaz, C., & Tamura, K. (2018). MEGA
Albright, J. C. (2004). Phylogeography of the sigmodontine rodent,
X: Molecular evolutionary genetics analysis across computing plat-
Phyllotis xanthopygus, and a test of the sensitivity of nested clade
forms. Molecular Biology and Evolution, 35(6), 1547–­1549. https://
analysis to elevation–­based alternative distances. M.S. Thesis. The
doi.org/10.1093/molbe​v/msy096
Florida State University. 46 pp.
Labaroni, C. A., Malleret, M. M., Novillo, A., Ojeda, A., Rodriguez, D.,
Barbière, F., Ronez, C., Ortiz, P. E., Martin, R. A., & Pardiñas, U. F. J.
Cuello, P., Ojeda, R., Marti, D., & Lanzone, C. (2014). Karyotypic
(2019). A new nomenclatural system for the study of sigmodontine
variation in the Andean rodent Phyllotis xanthopygus (Waterhouse,
rodent molars: First step towards an integrative phylogeny of fossil
1837) (Rodentia, Cricetidae, Sigmodontinae). Comparatype
and living cricetids. Biological Journal of the Linnean Society, 127,
Cytogenetics, 8, 369–­381. https://doi.org/10.3897/CompC​ytogen.
224–­244. https://doi.org/10.1093/bioli​nnean/​blz021
v8i4.8115
Cabrera, A. (1926). Dos roedores nuevos de las montañas de Catamarca.
Libardi, G. S., & Percequillo, A. R. (2016). Variation of cranioden-
Revista Chilena De Historia Natural, 30, 319–­321.
tal traits in russet rats Euryoryzomys russatus (Wagner, 1848)
Cabrera, A. (1961). Catálogo de los mamíferos de América del Sur.
(Rodentia: Cricetidae: Sigmodontinae) from Eastern Atlantic
Revista del Museo Argentino de Ciencias Naturales “Bernardino
Forest. Zoologischer Anzeiger –­A Journal of Comparative Zoology,
Rivadavia”. Ciencias Zoológicas, 4, 309–­732.
262, 57–­74. https://doi.org/10.1016/j.jcz.2016.03.005
Ceballos, G., & Ehrlich, P. R. (2006). Global mammal distributions, bio-
Maddison, W. P., & Maddison, D. R. (2017). Mesquite: A modular sys-
diversity hotspots, and conservation. Proceedings of the National
tem for evolutionary analysis. Available Online at: http://mesqu​itepr​
Academy of Sciences of the United States of America, 103(51),
oject.org
19374–­19379. https://doi.org/10.1073/pnas.06093​34103
Meachen-­Samuels, J., & Van Valkenburgh, B. (2009). Craniodental
Crespo, J. A. (1964). Descripción de una nueva subespecie de roedor
indicators of prey size preference in the Felidae. Biological
filotino. Neotrópica, 10, 99–­101.
Journal of the Linnean Society, 96, 784–­ 799. https://doi.
Darriba, D., Taboada, G. L., Doallo, R., & Posada, D. (2012). jModelT-
org/10.1111/j.1095-­8312.2008.01169.x
est 2: More models, new heuristics and parallel computing. Nature
Myers, P. (1989). A preliminary revision of the varius group of Akodon
Methods, 9, 772. https://doi.org/10.1038/nmeth.2109
(A. dayi, dolores, molinae, neocenus, simulator, toba and varius).
Ferro, L. I., Martínez, J. J., & Barquez, R. M. (2010). A new species
In K. H. Redford, & J. F. Eisenberg (ed.), Advances in neotropical
of Phyllotis (Rodentia, Cricetidae, Sigmodontinae) from Tucumán
mammalogy (pp. 5–­54). Sandhill Crane Press, Inc.
Province, Argentina. Mammalian Biology, 75, 523–­537. https://doi.
Myers, P., Patton, J. L., & Smith, M. F. (1990). A review of the boliv-
org/10.1016/j.mambio.2009.09.005
iensis group of Akodon (Muridae: Sigmodontinae) with emphasis
Frontier, S. (1976). Etude de la decroissance des valeurs propres dans
on Peru and Bolivia. Miscellaneous Publications of the Museum of
une analyse en composantes principales: Comparaison avec le mod-
Zoology, University of Michigan, 177, 1–­89.
ele du baton brise. Journal of Experimental Marine Biology and
Nuñez, M. N., Solman, S. A., & Cabre, M. F. (2009). Regional cli-
Ecology, 25, 67–­75. https://doi.org/10.1016/0022-­0981(76)90076​-­9
mate change experiments over southern South America. II: Climate
Hammer, O., Harper, D. A. T., & Ryan, P. D. (2001). Past:
change scenarios in the late twenty-­first century. Climate Dynamics,
Paleontological statistics software package for education and data
32, 1081–­1095. https://doi.org/10.1007/s0038​2-­008-­0449-­8
analysis. Palaeontologia Electronica, 4, 1–­9.
Ojeda, A. A., Teta, P., Jayat, J. P., Lanzone, C., Cornejo, P., Novillo, A.,
Hershkovitz, P. (1962). Evolution of Neotropical cricetine rodents
& Ojeda, R. A. (2021). Phylogenetic relationships among cryptic
(Muridae) with special reference to the phyllotine group. Fieldiana
species of the Phyllotis xanthopygus complex (Rodentia, Cricetidae).
Zoology, 46, 1–­524. https://doi.org/10.5962/bhl.title.2781
Zoologica Scripta, 50, 269–­281. https://doi.org/10.1111/zsc.12472
Hubbs, C. L., & Hubbs, C. (1953). An improved graphical analysis and
Pacheco, V., Rengifo, E. M., & Vivas, D. (2014). Una nueva especie
comparison of series of samples. Systematic Zoology, 2(2), 49–­56.
de ratón orejón del género Phyllotis Waterhouse, 1837 (Rodentia:
https://doi.org/10.2307/2411659
18
|    JAYAT et al.

Cricetidae) del norte del Perú. Therya, 5, 81–­ 508. https://doi. South America, Volume 2, Rodents (pp. 535–­555), The University of
org/10.12933/​thery​a-­14-­185 Chicago Press Chicago and London. https://doi.org/10.7208/chica​
Pearson, O. P. (1958). A taxonomic revision of the rodent genus go/97802​26169​606.001.0001
Phyllotis. University of California, Publications in Zoology, 56, Steppan, S. J., Ramirez, O., Banbury, J., Huchon, D., Pacheco, V.,
391–­496. Walker, L. I., & Spotorno, A. E. (2007). A molecular reappraisal
Pearson, O. P. (1972). New Information on ranges and relationships of the systematics of the leaf–­eared mice Phyllotis and their rela-
within the rodent genus Phyllotis in Peru and Ecuador. Journal of tives. In D. A. Kelt, E. P. Lessa, J. Salazar-­Bravo, & J. L. Patton
Mammalogy, 53, 677–­686. https://doi.org/10.2307/1379206 (Eds.), The Quintessential Naturalist: Honoring the Life and Legacy
Pearson, O. P., & Patton, J. L. (1976). Relationships among South of Oliver P. Pearson (pp. 799–­ 826). University of California,
American phyllotine rodents based on chromosome analysis. Journal Publications of Zoology. https://doi.org/10.1525/calif​ ornia/​
97805​
of Mammalogy, 57, 339–­350. https://doi.org/10.2307/1379693 20098​596.001.0001
Reborati, C. (2005). Situación ambiental en las ecorregiones Puna Storz, J. F., Quiroga-­Carmona, M., Opazo, J. C., Bowen, T., Farson,
y Altos Andes. In: A. Brown, U. Martinez Ortiz, M. Acerbi, & J. M., Steppan, S. J., & D'Elia, G. (2020). Discovery of the world's
Corcuera (eds.). La Situación Ambiental Argentina 2005 (pp. 32–­ highest-­dwelling mammal. Proceedings of the National Academy of
47), Fundación Vida Silvestre Argentina. Sciences of the United States of America, 117, 18169–­18171. https://
Reig, O. A. (1978). Roedores cricétidos del Plioceno superior de la doi.org/10.1073/pnas.20052​65117
provincia de Buenos Aires (Argentina). Publicaciones Del Museo Teta, P., Jayat, J. P., Lanzone, C., Novillo, A., Ojeda, A., & Ojeda, R.
Municipal De Ciencias Naturales De Mar Del Plata Lorenzo A. (2018). Geographic variation in quantitative skull traits and
Scaglia, 2, 164–­190. systematics of southern populations of the leaf-­eared mice of the
Rengifo, E. M., & Pacheco, V. (2015). Taxonomic revision of the Andean Phyllotis xanthopygus complex (Cricetidae, Phyllotini) in southern
leaf–­eared mouse, Phyllotis andium Thomas 1912 (Rodentia: South America. Zootaxa, 4446, 68–­ 80. https://doi.org/10.11646/​
Cricetidae), with the description of a new species. Zootaxa, 4018, zoota​xa.4446.1.5
349–­380. https://doi.org/10.11646/​zoota​xa.4018.3.2 Thomas, O. (1912). New bats and rodents from S. America. The Annals
Rengifo, E. M., & Pacheco, V. (2017). Phylogenetic position of the and Magazine of Natural History, 8(10), 403–­411.
Ancash leaf-­eared mouse Phyllotis definitus Osgood 1915 (Rodentia: Thomas, O. (1919). On small mammals from “Otro Cerro”, north-­east-
Cricetidae). Mammalia, 82, 153–­ 166. https://doi.org/10.1515/ ern Rioja, collected by Sr. E. Budin. The Annals and Magazine of
mamma​lia-­2016-­0138 Natural History, 9(3), 489–­500.
Riverón, S. (2011). Estructura poblacional e historia demográfica Walker, L. I., Spotorno, A. E., & Arrau, J. (1984). Cytogenetic and re-
del “pericote patagónico” Phyllotis xanthopygus (Rodentia: productive studies of two nominal subspecies of Phyllotis darwini
Sigmodontinae) en Patagonia Argentina. Ph. D. Dissertation (pp. and their experimental hybrids. Journal of Mammalogy, 65, 220–­
100). Uruguay: Universidad de la República. 230. https://doi.org/10.2307/1381161
Sievers, F., Wilm, A., Dineen, D., Gibson, T. J., Karplus, K., Li, W., Walker, L. I., Spotorno, A. E., & Sans, J. (1991). Genome size variation
Lopez, R., McWilliam, H., Remmert, M., Söding, J., Thompson, and its phenotypic consequences in Phyllotis rodents. Hereditas,
J. D., & Higgins, D. G. (2011). Fast, scalable generation of high-­ 115, 99–­107. https://doi.org/10.1111/j.1601-­5223.1991.tb035​42.x
quality protein multiple sequence alignments using Clustal Omega. Zapata-­Ríos, G., & Branch, L. C. (2016). Altered activity patterns and
Molecular Systematics and Biology, 7, 539. https://doi.org/10.1038/ reduced abundance of native mammals in sites with feral dogs in
msb.2011.75 the high Andes. Biological Conservation, 193, 9–­16. https://doi.
Stamatakis, A. (2006). RAxML-­VI-­HPC: Maximum likelihood-­based org/10.1016/j.biocon.2015.10.016
phylogenetic analyses with thousands of taxa and mixed models.
Bioinformatics, 22, 2688–­2690. https://doi.org/10.1093/bioin​forma​
tics/btl446 SUPPORTING INFORMATION
Steppan, S. J. (1993). Phylogenetic relationships among the Phyllotini Additional supporting information may be found online in
(Rodentia: Sigmodontinae) using morphological characters. Journal the Supporting Information section.
of Mammalian Evolution, 1, 187–­ 213. https://doi.org/10.1007/
BF010​24707
Steppan, S. J. (1995). Revision of the tribe Phyllotini (Rodentia: How to cite this article: Jayat, J. P., Teta, P.,
Sigmodontinae), with a phylogenetic hypothesis for the Ojeda, A. A., Steppan, S. J., Osland, J. M., Ortiz, P. E.,
Sigmodontinae. Fieldiana Zoology, New Series, 80, 1–­112. https://
Novillo, A., Lanzone, C., & Ojeda, R. A. (2021). The
doi.org/10.5962/bhl.title.3336
Phyllotis xanthopygus complex (Rodentia, Cricetidae)
Steppan, S. J. (1998). Phylogenetic relationships and species limits
within Phyllotis (Rodentia: Sigmodontinae): Concordance between in central Andes, systematics and description of a new
mtDNA sequence and morphology. Journal of Mammalogy, 7, 573–­ species. Zoologica Scripta, 00, 1–­18. https://doi.
593. https://doi.org/10.2307/1382988 org/10.1111/zsc.12510
Steppan, S. J., & Ramírez, O. (2015). Genus Phyllotis Waterhouse, 1837.
In: J. L. Patton, U. F. J. Pardiñas, & G. D’Elía (eds.). Mammals of