New subespecies of Conepatus semistriatus23
ANARTIA
Publicación del Museo de Biología de la Universidad del Zulia
ISSN 1315-642X (impresa) / ISSN 2665-0347 (digital)
https://doi.org/10.5281/zenodo.13755829 / Anartia, 38 (junio 2024): 23-44
http://zoobank.org/urn:lsid:zoobank.org:pub:75873A56-5383-42BE-8F68-F02E0571F35E
A new subspecies of Conepatus semistriatus (Boddaert, 1784)
(Mammalia, Carnivora, Mephitidae) from Venezuela, and the
first known case of insular dwarfism in living skunks
Una subespecie nueva de Conepatus semistriatus (Boddaert, 1784)
(Mammalia, Carnivora, Mephitidae) de Venezuela, y el primer caso conocido
de enanismo insular en mofetas vivientes
Jesús Molinari1
, María R. Abarca-Medina2 & Belkis A. Rivas-Rodríguez 1
1
Departamento de Biología, Facultad de Ciencias, Universidad de Los Andes, Mérida 5101, Venezuela
2
Departmento de Ecología, Instituto Venezolano de Investigaciones Científicas, Caracas 1020, Venezuela
Correspondencia: jmvault@gmail.com
(Received: 21-06-2024 / Accepted: 24-07-2024 / On line: 12-09-2024)
ABSTRACT
The Striped Hog-nosed Skunk, Conepatus semistriatus (Boddaert, 1784), is distributed from southern Mexico to northern
Peru. It is one of the least studied carnivorans of the New World. Not even basic morphometric information exists for the
species. This study assesses the taxonomic status of the Striped Hog-nosed Skunk occurring on Margarita Island, off the
coast of eastern Venezuela. Multivariate analyses of 25 cranial measurements revealed the Margaritan skunk to be much
dwarfed with respect to its nearby mainland congener, thus complying with the ‘island rule’, which predicts that under in-
sular conditions large continental animals become smaller, and small continental animals become larger. The craniometric
differences between the Margaritan and mainland skunks are of similar or greater magnitude to those found in interspe-
cific comparisons within genera of several carnivoran families. Geographic variation in cranial size was not evident within
the Margaritan sample, or within the mainland sample, analyzed separately. The Margaritan form is sufficiently distinct to
be deemed a new subspecies, thus it is here formally described as C. semistriatus elieceri ssp. nov. This skunk becomes the
seventh mammal known to be endemic to Margarita Island. Its conservation status is undetermined.
Key words: Caribbean islands, endemism, geographic variation, island rule, Margarita Island, morphometry, Striped Hog-
nosed Skunk.
RESUMEN
El zorrillo rayado, Conepatus semistriatus (Boddaert, 1784), se distribuye desde el sur de México hasta el norte de Perú. Es
uno de los carnívoros menos estudiados del Nuevo Mundo. Ni siquiera se ha publicado información morfométrica básica
sobre la especie. Este trabajo evalúa el estatus taxonómico del zorrillo rayado de la Isla de Margarita, frente a la costa oriental
de Venezuela. Los análisis multivariados de 25 medidas craneanas demuestran que el zorrillo margariteño es mucho más
pequeño que su congénere continental más cercano, cumpliendo así con la “regla de las islas”, la cual predice que bajo condi-
ciones insulares los animales continentales grandes reducen y los animales continentales pequeños incrementan su tamaño
corporal. Las diferencias craneométricas entre los zorrillos de Margarita y tierra firme son de magnitud similar o mayor a las
encontradas al efectuar comparaciones interespecíficas dentro de géneros de varias familias de carnívoros. No se encontró
variación geográfica evidente en el tamaño del cráneo dentro de la muestra de Margarita, ni dentro de la de tierra firme,
analizadas por separado. La forma margariteña es lo suficientemente distinta para ser considerada una subespecie nueva,
ANARTIA
Publicación del Museo de Biología de la Universidad del Zulia
ISSN 1315-642X (impresa) / ISSN 2665-0347 (digital)
https://doi.org/10.5281/zenodo.13755829 / Anartia, 38 (junio 2024): 23-44
http://zoobank.org/urn:lsid:zoobank.org:pub:75873A56-5383-42BE-8F68-F02E0571F35E
A new subspecies of Conepatus semistriatus (Boddaert, 1784)
(Mammalia, Carnivora, Mephitidae) from Venezuela, and the
first known case of insular dwarfism in living skunks
Una subespecie nueva de Conepatus semistriatus (Boddaert, 1784)
(Mammalia, Carnivora, Mephitidae) de Venezuela, y el primer caso conocido
de enanismo insular en mofetas vivientes
Jesús Molinari1
, María R. Abarca-Medina2 & Belkis A. Rivas-Rodríguez 1
1
Departamento de Biología, Facultad de Ciencias, Universidad de Los Andes, Mérida 5101, Venezuela
2
Departmento de Ecología, Instituto Venezolano de Investigaciones Científicas, Caracas 1020, Venezuela
Correspondencia: jmvault@gmail.com
(Received: 21-06-2024 / Accepted: 24-07-2024 / On line: 12-09-2024)
ABSTRACT
The Striped Hog-nosed Skunk, Conepatus semistriatus (Boddaert, 1784), is distributed from southern Mexico to northern
Peru. It is one of the least studied carnivorans of the New World. Not even basic morphometric information exists for the
species. This study assesses the taxonomic status of the Striped Hog-nosed Skunk occurring on Margarita Island, off the
coast of eastern Venezuela. Multivariate analyses of 25 cranial measurements revealed the Margaritan skunk to be much
dwarfed with respect to its nearby mainland congener, thus complying with the ‘island rule’, which predicts that under in-
sular conditions large continental animals become smaller, and small continental animals become larger. The craniometric
differences between the Margaritan and mainland skunks are of similar or greater magnitude to those found in interspe-
cific comparisons within genera of several carnivoran families. Geographic variation in cranial size was not evident within
the Margaritan sample, or within the mainland sample, analyzed separately. The Margaritan form is sufficiently distinct to
be deemed a new subspecies, thus it is here formally described as C. semistriatus elieceri ssp. nov. This skunk becomes the
seventh mammal known to be endemic to Margarita Island. Its conservation status is undetermined.
Key words: Caribbean islands, endemism, geographic variation, island rule, Margarita Island, morphometry, Striped Hog-
nosed Skunk.
RESUMEN
El zorrillo rayado, Conepatus semistriatus (Boddaert, 1784), se distribuye desde el sur de México hasta el norte de Perú. Es
uno de los carnívoros menos estudiados del Nuevo Mundo. Ni siquiera se ha publicado información morfométrica básica
sobre la especie. Este trabajo evalúa el estatus taxonómico del zorrillo rayado de la Isla de Margarita, frente a la costa oriental
de Venezuela. Los análisis multivariados de 25 medidas craneanas demuestran que el zorrillo margariteño es mucho más
pequeño que su congénere continental más cercano, cumpliendo así con la “regla de las islas”, la cual predice que bajo condi-
ciones insulares los animales continentales grandes reducen y los animales continentales pequeños incrementan su tamaño
corporal. Las diferencias craneométricas entre los zorrillos de Margarita y tierra firme son de magnitud similar o mayor a las
encontradas al efectuar comparaciones interespecíficas dentro de géneros de varias familias de carnívoros. No se encontró
variación geográfica evidente en el tamaño del cráneo dentro de la muestra de Margarita, ni dentro de la de tierra firme,
analizadas por separado. La forma margariteña es lo suficientemente distinta para ser considerada una subespecie nueva,
Molinari, Abarca-Medina & Rivas-Rodríguez24
que aquí se describe formalmente como C. semistriatus elieceri ssp. nov. Esta mofeta se convierte en el séptimo mamífero
endémico de la Isla de Margarita. Su estado de conservación es indeterminado.
Palabras clave: endemismo, Isla de Margarita, Islas del Caribe, morfometría, regla de las islas, variación geográfica, zorrillo
rayado.
Rica, the skunks occurring in southern Mexico and Central
America, and in South America from Venezuela and Co-
lombia to Bolivia, have never been revised taxonomically:
not even basic morphometric information exists for them.
Current classifications for skunks from this ample region
are largely based on the viewpoints of twentieth-century
authorities (Cabrera 1958, Hall 1981), who did not use
modern study methods. Such classifications remain widely
used in national and regional species lists (e.g., Solari et al.
2013, Pacheco et al. 2021, Boher-Bentti et al. 2023, Tirira
et al. 2023), distributional analyses (Meza-Joya et al. 2018,
Castillo & Caruso 2024), and conservation assessments
(Belant et al. 2009, Hernández‑Sánchez et al. 2022).
Hall (1981) assigned the populations from Veracruz
and Campeche to C. semistriatus conepatl (Gmelin, 1788),
those from Yucatan and Quintana Roo to Nicaragua to
C. s. yucatanicus Goldman, 1943, and those from Costa
Rica and western Panama to C. s. trichurus Thomas, 1905.
Cabrera (1958) assigned the populations from Colombia
and Venezuela to C. s. semistriatus, those from Ecuador to
C. s. quitensis (Humboldt, 1812), those from northwest-
ern Peru to C. s. zorrino Thomas, 1901, those from north-
eastern Peru to C. s. taxinus Thomas, 1924, and those from
southern Peru and Bolivia to C. rex Thomas, 1898. Except
for the reallocation of C. rex to C. chinga, this classification
was maintained by Wozencraft (2005). Rodrigues (2013)
departed markedly from these schemes by applying the
name C. conepatl to the Mexican and Central American
populations, and the name C. quitensis to Southern Co-
lombian, Ecuadorian, and Peruvian populations. The three
Central American subspecies, and three South American
subspecies (C. s. semistriatus, C. s. zorrino, C. s. taxinus),
are currently included in C. semistriatus; the other two
(C. s. quitensis, C. c. rex) in C. chinga (ASM 2024).
Skunks are remarkable among carnivorans for their
poor capacity to cross water barriers: for example, skunks
do not occur to the south of the Orinoco and the north
of the Amazon Rivers, and the Uruguay River has been
implicated in the genetic differentiation between Argen-
tinian and Uruguayan skunks (Rodrigues 2013); the only
known insular skunks are those occurring on the Chan-
nel Islands of California (Van Gelder 1959, McDonough
et al. 2022), and on Margarita Island of Venezuela (Bisbal
1983). The colonization of these islands by skunks likely
required the formation of ecologically suitable land bridg-
INTRODUCTION
The family Mephitidae accounts for 0.8% of the generic
and 4.4% of the specific diversity of living carnivorans (or-
der Carnivora); it includes two species of stink badgers
(genus Mydaus F. G. Cuvier, 1821) occurring in insular
Malaysia, Indonesia and the Philippines, and 12 species of
skunks (genera Mephitis E. Geoffroy Saint-Hilaire & F. G.
Cuvier, 1795; Spilogale Gray, 1865; and Conepatus Gray,
1837) occurring in North, Central, and South America
(ASM 2024). Fossil skunks are known from the late Mio-
cene to Pliocene (13.5–3 Mya) of Eurasia, and from the
Late Miocene (9 Mya) of North America (Wang & Qiu
2004, Wang et al. 2005).
Skunks are charismatic mammals that are renowned
for their unusual coloration, serving as a warning, and for
their smelly spray, used in self-defense. They have been
regarded as symbols of cuteness, confidence, pacifism,
humor, and luck (Miller 2015). Descented skunks are be-
coming popular pets. Despite their appeal, the scientific
knowledge of skunks shows substantial gaps. They have
been best studied in North America, but even there ma-
jor discoveries have been made in the last decades: thus,
Conepatus mesoleucus (Lesson, 1865), long considered a
separate species, was shown to be not distinct from the
American Hog-nosed Skunk, C. leuconotus (Lichten-
stein, 1832) (Dragoo et al. 2003); conversely, the Spotted
Skunks, genus Spilogale, were shown to possess a greater
species diversity than traditionally recognized (Mc-
Donough et al. 2022). In the other hemisphere, the Jarita-
caca, C. amazonicus (Lichtenstein, 1838), long considered
a subspecies of the Striped Hog-nosed Skunk, C. semis-
triatus (Boddaert, 1785) (Cabrera 1958), was shown to
be more closely related to the Molina’s Hog-nosed Skunk,
C. chinga (Molina, 1782), from southern South America
(Rodrigues 2013, Schiaffini et al. 2013); and based on
morphometric and genetic analyses, C. humboldtii Gray,
1837 was considered a synonym of the latter (Schiaffini et
al. 2013), though this conclusion has been disputed also
based on genetic data (Rodrigues 2013). Both C. amazo-
nicus and C. humboldtii are currently deemed conspecific
with C. chinga (ASM 2024).
Except for Spilogale (McDonough et al. 2022) and
C. leuconotus (Dragoo et al. 2003, Dragoo & Sheffield
2009), whose southern distributional limits reach Costa
que aquí se describe formalmente como C. semistriatus elieceri ssp. nov. Esta mofeta se convierte en el séptimo mamífero
endémico de la Isla de Margarita. Su estado de conservación es indeterminado.
Palabras clave: endemismo, Isla de Margarita, Islas del Caribe, morfometría, regla de las islas, variación geográfica, zorrillo
rayado.
Rica, the skunks occurring in southern Mexico and Central
America, and in South America from Venezuela and Co-
lombia to Bolivia, have never been revised taxonomically:
not even basic morphometric information exists for them.
Current classifications for skunks from this ample region
are largely based on the viewpoints of twentieth-century
authorities (Cabrera 1958, Hall 1981), who did not use
modern study methods. Such classifications remain widely
used in national and regional species lists (e.g., Solari et al.
2013, Pacheco et al. 2021, Boher-Bentti et al. 2023, Tirira
et al. 2023), distributional analyses (Meza-Joya et al. 2018,
Castillo & Caruso 2024), and conservation assessments
(Belant et al. 2009, Hernández‑Sánchez et al. 2022).
Hall (1981) assigned the populations from Veracruz
and Campeche to C. semistriatus conepatl (Gmelin, 1788),
those from Yucatan and Quintana Roo to Nicaragua to
C. s. yucatanicus Goldman, 1943, and those from Costa
Rica and western Panama to C. s. trichurus Thomas, 1905.
Cabrera (1958) assigned the populations from Colombia
and Venezuela to C. s. semistriatus, those from Ecuador to
C. s. quitensis (Humboldt, 1812), those from northwest-
ern Peru to C. s. zorrino Thomas, 1901, those from north-
eastern Peru to C. s. taxinus Thomas, 1924, and those from
southern Peru and Bolivia to C. rex Thomas, 1898. Except
for the reallocation of C. rex to C. chinga, this classification
was maintained by Wozencraft (2005). Rodrigues (2013)
departed markedly from these schemes by applying the
name C. conepatl to the Mexican and Central American
populations, and the name C. quitensis to Southern Co-
lombian, Ecuadorian, and Peruvian populations. The three
Central American subspecies, and three South American
subspecies (C. s. semistriatus, C. s. zorrino, C. s. taxinus),
are currently included in C. semistriatus; the other two
(C. s. quitensis, C. c. rex) in C. chinga (ASM 2024).
Skunks are remarkable among carnivorans for their
poor capacity to cross water barriers: for example, skunks
do not occur to the south of the Orinoco and the north
of the Amazon Rivers, and the Uruguay River has been
implicated in the genetic differentiation between Argen-
tinian and Uruguayan skunks (Rodrigues 2013); the only
known insular skunks are those occurring on the Chan-
nel Islands of California (Van Gelder 1959, McDonough
et al. 2022), and on Margarita Island of Venezuela (Bisbal
1983). The colonization of these islands by skunks likely
required the formation of ecologically suitable land bridg-
INTRODUCTION
The family Mephitidae accounts for 0.8% of the generic
and 4.4% of the specific diversity of living carnivorans (or-
der Carnivora); it includes two species of stink badgers
(genus Mydaus F. G. Cuvier, 1821) occurring in insular
Malaysia, Indonesia and the Philippines, and 12 species of
skunks (genera Mephitis E. Geoffroy Saint-Hilaire & F. G.
Cuvier, 1795; Spilogale Gray, 1865; and Conepatus Gray,
1837) occurring in North, Central, and South America
(ASM 2024). Fossil skunks are known from the late Mio-
cene to Pliocene (13.5–3 Mya) of Eurasia, and from the
Late Miocene (9 Mya) of North America (Wang & Qiu
2004, Wang et al. 2005).
Skunks are charismatic mammals that are renowned
for their unusual coloration, serving as a warning, and for
their smelly spray, used in self-defense. They have been
regarded as symbols of cuteness, confidence, pacifism,
humor, and luck (Miller 2015). Descented skunks are be-
coming popular pets. Despite their appeal, the scientific
knowledge of skunks shows substantial gaps. They have
been best studied in North America, but even there ma-
jor discoveries have been made in the last decades: thus,
Conepatus mesoleucus (Lesson, 1865), long considered a
separate species, was shown to be not distinct from the
American Hog-nosed Skunk, C. leuconotus (Lichten-
stein, 1832) (Dragoo et al. 2003); conversely, the Spotted
Skunks, genus Spilogale, were shown to possess a greater
species diversity than traditionally recognized (Mc-
Donough et al. 2022). In the other hemisphere, the Jarita-
caca, C. amazonicus (Lichtenstein, 1838), long considered
a subspecies of the Striped Hog-nosed Skunk, C. semis-
triatus (Boddaert, 1785) (Cabrera 1958), was shown to
be more closely related to the Molina’s Hog-nosed Skunk,
C. chinga (Molina, 1782), from southern South America
(Rodrigues 2013, Schiaffini et al. 2013); and based on
morphometric and genetic analyses, C. humboldtii Gray,
1837 was considered a synonym of the latter (Schiaffini et
al. 2013), though this conclusion has been disputed also
based on genetic data (Rodrigues 2013). Both C. amazo-
nicus and C. humboldtii are currently deemed conspecific
with C. chinga (ASM 2024).
Except for Spilogale (McDonough et al. 2022) and
C. leuconotus (Dragoo et al. 2003, Dragoo & Sheffield
2009), whose southern distributional limits reach Costa
New subespecies of Conepatus semistriatus25
es to the mainland, which may have occurred during a gla-
cial period (Van Gelder 1959, Molinari 2007).
As in other mammals, it is debated whether carniv-
orans follow the ‘island rule’, according to which on
islands large continental animals become smaller, and
small continental animals become larger (Meiri et al.
2004, Lyras et al. 2010, Molinari 2023a). The Channel
Islands skunk, Spilogale gracilis amphialus Dickey, 1929,
despite being sufficiently differentiated genetically to be
deemed a full species (Floyd et al. 2011, McDonough et
al. 2022), is indistinct morphometrically (Van Gelder
1959), thus it does not follow the island rule. No data
exist for the Margaritan skunk, but based on this rule it
could be predicted to be small because its nearby main-
land relative is large.
The fauna of Margarita Island (Fig. 1) includes 16 spe-
cies of bats, and 13 species of nonvolant mammals (Smith
& Genoways 1974, Linares 1998). None of the bats is en-
demic, but specimens of the Allen’s Common Moustached
bat, Pteronotus fuscus ( J. A. Allen, 1911), from the island
show ‘a notable reduction in overall size’, and ‘are some-
what paler (Sayal Brown) in color than adjacent mainland
populations’ (Smith 1972). Six of the nonvolant mammals
are endemic, namely:
1. The Margaritan Robinson’s Mouse Opossum, Mar-
mosa robinsoni robinsoni Bangs, 1898, about which
and other insular forms of the species it has been
concluded that they are ‘nothing more than large
island forms with habitat-correlated coat color dif-
ferences’ (Rossi et al. 2010) [no genetic information
exists for Margaritan specimens to test this hypoth-
esis].
2. The Margaritan Cottontail Rabbit, Sylvilagus flori-
danus margaritae Miller, 1898. Unstudied taxonom-
ically using modern methods.
3. The Margaritan Red-tailed Squirrel, Syntheosciurus
granatensis nesaeus (G. M. Allen, 1902), which is
‘easily distinguished from mainland samples’ (Vivo
& Carmignotto 2015).
4. The Yellow Speckled Tree-rat, Pattonomys flavidus
(Hollister, 1914), which is ‘cranially distinctive and
readily diagnosable’ (Emmons 2005).
5. The Margaritan Brown Capuchin monkey, Sapa-
jus apella margaritae Hollister, 1914, about which
based on the 800 km separation from its southern
Venezuelan conspecifics, Linares (1998) and Groves
(2001) suggested that it might have been introduced
to the island in Pre-Columbian times. Groves also
Figure 1. Map of Venezuela east of the Andes and north of the Orinoco River showing the geographic origin of the specimens of the
new subspecies (red dots) and C. s. semistriatus (blue diamonds) included in this study. The insert in the top left shows the relative posi-
tions of Margarita Island, its two divisions (Margarita Oriental, also known as Paraguachoa; and the Macanao Peninsula), the Cubagua
and Coche Islands, and the Araya Peninsula. For the geographic coordinates of localities, see Materials and methods.
es to the mainland, which may have occurred during a gla-
cial period (Van Gelder 1959, Molinari 2007).
As in other mammals, it is debated whether carniv-
orans follow the ‘island rule’, according to which on
islands large continental animals become smaller, and
small continental animals become larger (Meiri et al.
2004, Lyras et al. 2010, Molinari 2023a). The Channel
Islands skunk, Spilogale gracilis amphialus Dickey, 1929,
despite being sufficiently differentiated genetically to be
deemed a full species (Floyd et al. 2011, McDonough et
al. 2022), is indistinct morphometrically (Van Gelder
1959), thus it does not follow the island rule. No data
exist for the Margaritan skunk, but based on this rule it
could be predicted to be small because its nearby main-
land relative is large.
The fauna of Margarita Island (Fig. 1) includes 16 spe-
cies of bats, and 13 species of nonvolant mammals (Smith
& Genoways 1974, Linares 1998). None of the bats is en-
demic, but specimens of the Allen’s Common Moustached
bat, Pteronotus fuscus ( J. A. Allen, 1911), from the island
show ‘a notable reduction in overall size’, and ‘are some-
what paler (Sayal Brown) in color than adjacent mainland
populations’ (Smith 1972). Six of the nonvolant mammals
are endemic, namely:
1. The Margaritan Robinson’s Mouse Opossum, Mar-
mosa robinsoni robinsoni Bangs, 1898, about which
and other insular forms of the species it has been
concluded that they are ‘nothing more than large
island forms with habitat-correlated coat color dif-
ferences’ (Rossi et al. 2010) [no genetic information
exists for Margaritan specimens to test this hypoth-
esis].
2. The Margaritan Cottontail Rabbit, Sylvilagus flori-
danus margaritae Miller, 1898. Unstudied taxonom-
ically using modern methods.
3. The Margaritan Red-tailed Squirrel, Syntheosciurus
granatensis nesaeus (G. M. Allen, 1902), which is
‘easily distinguished from mainland samples’ (Vivo
& Carmignotto 2015).
4. The Yellow Speckled Tree-rat, Pattonomys flavidus
(Hollister, 1914), which is ‘cranially distinctive and
readily diagnosable’ (Emmons 2005).
5. The Margaritan Brown Capuchin monkey, Sapa-
jus apella margaritae Hollister, 1914, about which
based on the 800 km separation from its southern
Venezuelan conspecifics, Linares (1998) and Groves
(2001) suggested that it might have been introduced
to the island in Pre-Columbian times. Groves also
Figure 1. Map of Venezuela east of the Andes and north of the Orinoco River showing the geographic origin of the specimens of the
new subspecies (red dots) and C. s. semistriatus (blue diamonds) included in this study. The insert in the top left shows the relative posi-
tions of Margarita Island, its two divisions (Margarita Oriental, also known as Paraguachoa; and the Macanao Peninsula), the Cubagua
and Coche Islands, and the Araya Peninsula. For the geographic coordinates of localities, see Materials and methods.
Molinari, Abarca-Medina & Rivas-Rodríguez26
thought it to be more allied to S. a. fatuellus (Lin-
naeus, 1766) from Colombia than to S. a. apella
(Linnaeus, 1758). However, the latter subspecies
is now known to occur in Guyana, and in the Ori-
noco Delta (Boher-Bentti & Cordero-Rodríguez
2000), thus could potentially range or have ranged
more widely along the coastal forests of northeastern
Venezuela, which reach the Araya Peninsula south
of Margarita, making a natural colonization of the
island possible through a land bridge connecting it
to the mainland during a glacial period [no genetic
information exists for Margaritan specimens to test
these hypotheses].
6. The Margaritan White-tailed Deer, Odocoileus mar-
garitae Osgood, 1910, deemed a distinct species due
to its very small size and highly distinctive cranial
characters (Molinari 2007). Based on D-loop se-
quences, the separation of this deer from its main-
land congeners has been estimated to have occurred
at least 118,000 years ago (Moscarella 2001).
Over the course of decades, specimens of Conepatus
from both mainland Venezuela and Margarita Island,
many of them roadkills salvaged by selfless scientific col-
lectors, have slowly accumulated in zoological museums.
The point has been reached in which the material available
is satisfactory for statistical analyses. We take the oppor-
tunity that this represents to conduct the first morpho-
logical and morphometric assessment of members of the
C. semistriatus group, with the main objective of clarifying
the taxonomic status of the Margaritan populations.
MATERIALS AND METHODS
Taxonomic categories
We adhere to a version of Simpson’s (1951) Evolution-
ary Species Concept that has been expanded to include
operational criteria (Molinari 2023b), according to which
a species is ‘a phyletic lineage (ancestral-descendent se-
quence of populations) evolving independently of others,
with its own separate and unitary evolutionary role and
tendencies, and diagnosably distinct based on heritable
morphological characters, genetic markers, or both’. We
deem a subspecies to be ‘a collection of populations occu-
pying a distinct geographic range and diagnosably distinct
from other such conspecific populations based on herita-
ble morphological characters’ (Patten & Unitt 2002, Mo-
linari 2023b).
Institutions
The collection acronyms mentioned in this study are:
AMNH, American Museum of Natural History, New
York, USA; BMNH, Natural History Museum, London,
UK; CFA, Colección de Mastozoología, Fundación de
Historia Natural “Félix de Azara”, Buenos Aires, Argen-
tina; CVULA, Colección de Vertebrados de la Universi-
dad de Los Andes, Mérida, Venezuela; EBRG, Museo de
la Estación Biológica Rancho Grande, Maracay, Venezu-
ela; FMNH, Field Museum of Natural History, Chicago,
USA; IAvH, Instituto de Investigación de Recursos Bi-
ológicos Alexander von Humboldt, Villa de Leyva, Co-
lombia; ICN, Instituto de Ciencias Naturales, Universi-
dad Nacional de Colombia, Bogotá, Colombia; MACN,
Colección Mastozoología, Museo Argentino de Ciencias
Naturales, Buenos Aires, Argentina; MBUCV, Museo
de Biología de la Universidad Central de Venezuela, Ca-
racas, Venezuela; MCNG, Museo de Ciencias Natura-
les de Guanare, Guanare, Venezuela; MCZ, Museum of
Comparative Zoology, Harvard University, Cambridge,
USA; MEPN, Museo de la Escuela Politécnica Nacional,
Quito, Ecuador; MHNLS, Museo de Historia Natural La
Salle, Caracas, Venezuela; MHNM, Museo Nacional de
Historia Natural, Montevideo, Uruguay (specimens with
this acronym were obtained as an exchange and will be re-
catalogued in the CVULA); MNHN, Muséum national
d’Histoire naturelle, Paris; MZUFV, Museu de Zoologia
João Moojen, Viçosa, Brazil; QCAZ, Museo de Zoología
de la Pontificia Universidad Católica del Ecuador, Quito,
Ecuador; RBMC, Centro de Visitantes de la Reserva Bi-
ológica Montecano, Paraguaná Peninsula, Venezuela;
USNM, National Museum of Natural History, Washing-
ton, USA; UV, Colección de mamíferos de la Universidad
del Valle, Cali, Colombia. All specimens were examined
physically, except for a few from the CFA, MACN, MCZ,
MZUFV, and UV collections, which were examined pho-
tographically.
Measurements
Unless otherwise indicated, cranial measurements
(Fig. 2A) were taken with a digital caliper. Letter and
number combinations (e.g., M1–M3) refer to opposite
points of the skull used as landmarks. A total of 25 skull
measurements were obtained, as follows:
1. Basilar length* (M1–M3), distance from the pos-
teriormost margin of the first upper incisors to the
anteriormost margin of the foramen magnum.
* Measurements marked with an asterisk after Van Gelder (1968).
thought it to be more allied to S. a. fatuellus (Lin-
naeus, 1766) from Colombia than to S. a. apella
(Linnaeus, 1758). However, the latter subspecies
is now known to occur in Guyana, and in the Ori-
noco Delta (Boher-Bentti & Cordero-Rodríguez
2000), thus could potentially range or have ranged
more widely along the coastal forests of northeastern
Venezuela, which reach the Araya Peninsula south
of Margarita, making a natural colonization of the
island possible through a land bridge connecting it
to the mainland during a glacial period [no genetic
information exists for Margaritan specimens to test
these hypotheses].
6. The Margaritan White-tailed Deer, Odocoileus mar-
garitae Osgood, 1910, deemed a distinct species due
to its very small size and highly distinctive cranial
characters (Molinari 2007). Based on D-loop se-
quences, the separation of this deer from its main-
land congeners has been estimated to have occurred
at least 118,000 years ago (Moscarella 2001).
Over the course of decades, specimens of Conepatus
from both mainland Venezuela and Margarita Island,
many of them roadkills salvaged by selfless scientific col-
lectors, have slowly accumulated in zoological museums.
The point has been reached in which the material available
is satisfactory for statistical analyses. We take the oppor-
tunity that this represents to conduct the first morpho-
logical and morphometric assessment of members of the
C. semistriatus group, with the main objective of clarifying
the taxonomic status of the Margaritan populations.
MATERIALS AND METHODS
Taxonomic categories
We adhere to a version of Simpson’s (1951) Evolution-
ary Species Concept that has been expanded to include
operational criteria (Molinari 2023b), according to which
a species is ‘a phyletic lineage (ancestral-descendent se-
quence of populations) evolving independently of others,
with its own separate and unitary evolutionary role and
tendencies, and diagnosably distinct based on heritable
morphological characters, genetic markers, or both’. We
deem a subspecies to be ‘a collection of populations occu-
pying a distinct geographic range and diagnosably distinct
from other such conspecific populations based on herita-
ble morphological characters’ (Patten & Unitt 2002, Mo-
linari 2023b).
Institutions
The collection acronyms mentioned in this study are:
AMNH, American Museum of Natural History, New
York, USA; BMNH, Natural History Museum, London,
UK; CFA, Colección de Mastozoología, Fundación de
Historia Natural “Félix de Azara”, Buenos Aires, Argen-
tina; CVULA, Colección de Vertebrados de la Universi-
dad de Los Andes, Mérida, Venezuela; EBRG, Museo de
la Estación Biológica Rancho Grande, Maracay, Venezu-
ela; FMNH, Field Museum of Natural History, Chicago,
USA; IAvH, Instituto de Investigación de Recursos Bi-
ológicos Alexander von Humboldt, Villa de Leyva, Co-
lombia; ICN, Instituto de Ciencias Naturales, Universi-
dad Nacional de Colombia, Bogotá, Colombia; MACN,
Colección Mastozoología, Museo Argentino de Ciencias
Naturales, Buenos Aires, Argentina; MBUCV, Museo
de Biología de la Universidad Central de Venezuela, Ca-
racas, Venezuela; MCNG, Museo de Ciencias Natura-
les de Guanare, Guanare, Venezuela; MCZ, Museum of
Comparative Zoology, Harvard University, Cambridge,
USA; MEPN, Museo de la Escuela Politécnica Nacional,
Quito, Ecuador; MHNLS, Museo de Historia Natural La
Salle, Caracas, Venezuela; MHNM, Museo Nacional de
Historia Natural, Montevideo, Uruguay (specimens with
this acronym were obtained as an exchange and will be re-
catalogued in the CVULA); MNHN, Muséum national
d’Histoire naturelle, Paris; MZUFV, Museu de Zoologia
João Moojen, Viçosa, Brazil; QCAZ, Museo de Zoología
de la Pontificia Universidad Católica del Ecuador, Quito,
Ecuador; RBMC, Centro de Visitantes de la Reserva Bi-
ológica Montecano, Paraguaná Peninsula, Venezuela;
USNM, National Museum of Natural History, Washing-
ton, USA; UV, Colección de mamíferos de la Universidad
del Valle, Cali, Colombia. All specimens were examined
physically, except for a few from the CFA, MACN, MCZ,
MZUFV, and UV collections, which were examined pho-
tographically.
Measurements
Unless otherwise indicated, cranial measurements
(Fig. 2A) were taken with a digital caliper. Letter and
number combinations (e.g., M1–M3) refer to opposite
points of the skull used as landmarks. A total of 25 skull
measurements were obtained, as follows:
1. Basilar length* (M1–M3), distance from the pos-
teriormost margin of the first upper incisors to the
anteriormost margin of the foramen magnum.
* Measurements marked with an asterisk after Van Gelder (1968).
New subespecies of Conepatus semistriatus27
2. Condylobasal length* (F1–F2), distance from the
anteriormost margin of the premaxillae to the poste-
riormost margin of both occipital condyles.
3. Zygomatic breadth* (C1–C2), maximum distance
across the outer margins of the zygomatic arches.
4. Mastoid breadth* (D1–D2), maximum distance
across the mastoid processes.
5. Interorbital breadth* (A1–A2), distance across the
frontal bones at the level of the frontomaxillary sutures.
6. Postorbital breadth* (B1–B2), minimum distance
across the frontal bones posteriorly to the fronto-
maxillary sutures.
7. Palatilar length* (M1–M2), minimum distance
from the posteriormost margin of the first upper in-
* Measurements marked with an asterisk after Van Gelder (1968).
Figure 2. A) Points between which skull measurements were taken: from top to bottom, dorsal, lateral and ventral views of the cra-
nium, and lateral view of the mandible of a specimen of C. s. semistriatus (MHNLS 3671), condylobasal length = 81.5 mm, and left
half of the palate of a specimen of the new subspecies (CVULA 9124), condylobasal length = 73.0 mm. B) Skull of the holotype of
new subspecies (EBRG 3138): from top to bottom, dorsal, lateral and ventral views of the cranium, and lateral view of the mandible.
The scales are based on condylobasal lengths and zygomatic breadths (in ventral view). The horizontal scale (right) applies to MHNLS
3671 and EBRG 3138. The vertical scale (left) applies to CVULA 9124.
2. Condylobasal length* (F1–F2), distance from the
anteriormost margin of the premaxillae to the poste-
riormost margin of both occipital condyles.
3. Zygomatic breadth* (C1–C2), maximum distance
across the outer margins of the zygomatic arches.
4. Mastoid breadth* (D1–D2), maximum distance
across the mastoid processes.
5. Interorbital breadth* (A1–A2), distance across the
frontal bones at the level of the frontomaxillary sutures.
6. Postorbital breadth* (B1–B2), minimum distance
across the frontal bones posteriorly to the fronto-
maxillary sutures.
7. Palatilar length* (M1–M2), minimum distance
from the posteriormost margin of the first upper in-
* Measurements marked with an asterisk after Van Gelder (1968).
Figure 2. A) Points between which skull measurements were taken: from top to bottom, dorsal, lateral and ventral views of the cra-
nium, and lateral view of the mandible of a specimen of C. s. semistriatus (MHNLS 3671), condylobasal length = 81.5 mm, and left
half of the palate of a specimen of the new subspecies (CVULA 9124), condylobasal length = 73.0 mm. B) Skull of the holotype of
new subspecies (EBRG 3138): from top to bottom, dorsal, lateral and ventral views of the cranium, and lateral view of the mandible.
The scales are based on condylobasal lengths and zygomatic breadths (in ventral view). The horizontal scale (right) applies to MHNLS
3671 and EBRG 3138. The vertical scale (left) applies to CVULA 9124.
Molinari, Abarca-Medina & Rivas-Rodríguez28
cisors to the posterior margin of the bony palate. If
a nasal spine was present postero-centrally, this mea-
surement was taken laterally to it.
8. Precanine length (F1 to I1–I2), minimum distance
from the anteriormost margin of the premaxillae to
the imaginary line connecting the anteriormost mar-
gins of the canines. This measurement is obtained
digitally based on a photograph of the palate.
9. Canine to notch length (I1–I2 to L1–L2), mini-
mum distance from the imaginary line connecting
the anteriormost edges of the canines to the imagi-
nary line connecting the vertices of the notches be-
tween the metacone and the hypocone of each mo-
lar. This measurement is obtained digitally based on
a photograph of the palate.
10. Post-notch length (L1–L2 to M2), minimum dis-
tance from the imaginary line connecting the ver-
tices of the notches between the metacone and the
hypocone of each molar and the posterior margin of
the bony palate. This measurement is obtained digi-
tally based on a photograph of the palate.
11. Postpalatal length* (M2–M3), maximum distance
from anteriormost posterior margin of the bony pal-
ate to the anteriormost margin of the foramen mag-
num.
12. 12. Heigth of the cranium* (E1–E2), distance from
the basicranium at the center of the junction of the
basisphenoid and basioccipital bones, and the upper
surface of the parietal bones, excluding the sagittal
crest.
13. Length of the maxillary toothrow* (G1–G3), dis-
tance from the anteriormost margin of the canine to
the posteriormost margin of the molar at the alveo-
lar levels.
14. Width across incisors* (H1–H2), distance across the
latero-labial margins of both third upper incisors.
15. Width across canines* ( J1–J2), distance across the
latero-labial margins of both upper canines.
16. Width across molars* (K1–K2), distance across the
latero-labial margins of both upper molars.
17. Diameter of the canine* (G1–G2), distance from
the anteriormost to the posteriormost margin of
one of the two upper canines at the alveolar level.
On crania in which both canines were missing, this
measurement was guessed based on the innermost
margins of the alveolus.
18. Length of PM3 (R1–R2), distance from the antero-
labial to the postero-lingual margin of the third up-
per premolar at the alveolar level (PM1 and PM2 are
absent in Conepatus).
19. Length of PM4* (S1–S2), distance from the antero-
to the postero-labial margin of the fourth upper pre-
molar at the alveolar level.
20. Length of the molar (S2–S3), maximum distance
from the labial junction of the fourth upper premo-
lar (PM4) and the molar to the posteriormost mar-
gin of the molar along the parastyle-metacone axis.
21. Width of the molar (S2–S4), maximum distance
from the labial junction of the fourth upper premo-
lar (PM4) and the molar to the postero-lingual mar-
gin of the molar.
22. Width of the interpterygoid fossa* (N1–N2), maxi-
mum distance across the tips of the hamuli of the
pterygoids.
23. Length of the lower carnassial (P1–P2), maximum
distance from the anterior to the posterior margin of
the lower carnassial.
24. Height of the coronoid* (Q1–Q2), distance from
the highest point of the coronoid process to the low-
est point of the angular process of the mandible.
25. Length of the mandible* (O1–O2), maximum dis-
tance from the mandibular symphysis to the posteri-
ormost margin of the angular process.
In the case of breath measurements, when the measur-
ing landmark was available on one side of the cranium and
missing on the other, as for example in specimens with
one intact and one broken zygomatic arch, the minimum
distance from the available landmark to the longitudinal
midline of the cranium was measured, and the resulting
value was multiplied by two. This was done to minimize
the proportion of missing measurements.
A posteriori, it was found that conventional linear
measurements were not sufficient to reflect some visually
evident shape differences, particularly with regard to the
proportions of the rostrum. For this reason, an additional
measurement was added, here referred to as rostral angle.
Adobe Photoshop CS6 was used to measure on photo-
graphs of crania in lateral view the angle formed by two
lines, both starting in the posteriormost alveolar margin
of PM3, one of them crossing the antero-central margin of
the nasal bones, and the other crossing the alveolar margin
of the canine.
Morphometric analyses
Adult specimens of Conepatus have been defined as
those possessing a fully erupted permanent dentition, and
* Measurements marked with an asterisk after Van Gelder (1968).
cisors to the posterior margin of the bony palate. If
a nasal spine was present postero-centrally, this mea-
surement was taken laterally to it.
8. Precanine length (F1 to I1–I2), minimum distance
from the anteriormost margin of the premaxillae to
the imaginary line connecting the anteriormost mar-
gins of the canines. This measurement is obtained
digitally based on a photograph of the palate.
9. Canine to notch length (I1–I2 to L1–L2), mini-
mum distance from the imaginary line connecting
the anteriormost edges of the canines to the imagi-
nary line connecting the vertices of the notches be-
tween the metacone and the hypocone of each mo-
lar. This measurement is obtained digitally based on
a photograph of the palate.
10. Post-notch length (L1–L2 to M2), minimum dis-
tance from the imaginary line connecting the ver-
tices of the notches between the metacone and the
hypocone of each molar and the posterior margin of
the bony palate. This measurement is obtained digi-
tally based on a photograph of the palate.
11. Postpalatal length* (M2–M3), maximum distance
from anteriormost posterior margin of the bony pal-
ate to the anteriormost margin of the foramen mag-
num.
12. 12. Heigth of the cranium* (E1–E2), distance from
the basicranium at the center of the junction of the
basisphenoid and basioccipital bones, and the upper
surface of the parietal bones, excluding the sagittal
crest.
13. Length of the maxillary toothrow* (G1–G3), dis-
tance from the anteriormost margin of the canine to
the posteriormost margin of the molar at the alveo-
lar levels.
14. Width across incisors* (H1–H2), distance across the
latero-labial margins of both third upper incisors.
15. Width across canines* ( J1–J2), distance across the
latero-labial margins of both upper canines.
16. Width across molars* (K1–K2), distance across the
latero-labial margins of both upper molars.
17. Diameter of the canine* (G1–G2), distance from
the anteriormost to the posteriormost margin of
one of the two upper canines at the alveolar level.
On crania in which both canines were missing, this
measurement was guessed based on the innermost
margins of the alveolus.
18. Length of PM3 (R1–R2), distance from the antero-
labial to the postero-lingual margin of the third up-
per premolar at the alveolar level (PM1 and PM2 are
absent in Conepatus).
19. Length of PM4* (S1–S2), distance from the antero-
to the postero-labial margin of the fourth upper pre-
molar at the alveolar level.
20. Length of the molar (S2–S3), maximum distance
from the labial junction of the fourth upper premo-
lar (PM4) and the molar to the posteriormost mar-
gin of the molar along the parastyle-metacone axis.
21. Width of the molar (S2–S4), maximum distance
from the labial junction of the fourth upper premo-
lar (PM4) and the molar to the postero-lingual mar-
gin of the molar.
22. Width of the interpterygoid fossa* (N1–N2), maxi-
mum distance across the tips of the hamuli of the
pterygoids.
23. Length of the lower carnassial (P1–P2), maximum
distance from the anterior to the posterior margin of
the lower carnassial.
24. Height of the coronoid* (Q1–Q2), distance from
the highest point of the coronoid process to the low-
est point of the angular process of the mandible.
25. Length of the mandible* (O1–O2), maximum dis-
tance from the mandibular symphysis to the posteri-
ormost margin of the angular process.
In the case of breath measurements, when the measur-
ing landmark was available on one side of the cranium and
missing on the other, as for example in specimens with
one intact and one broken zygomatic arch, the minimum
distance from the available landmark to the longitudinal
midline of the cranium was measured, and the resulting
value was multiplied by two. This was done to minimize
the proportion of missing measurements.
A posteriori, it was found that conventional linear
measurements were not sufficient to reflect some visually
evident shape differences, particularly with regard to the
proportions of the rostrum. For this reason, an additional
measurement was added, here referred to as rostral angle.
Adobe Photoshop CS6 was used to measure on photo-
graphs of crania in lateral view the angle formed by two
lines, both starting in the posteriormost alveolar margin
of PM3, one of them crossing the antero-central margin of
the nasal bones, and the other crossing the alveolar margin
of the canine.
Morphometric analyses
Adult specimens of Conepatus have been defined as
those possessing a fully erupted permanent dentition, and
* Measurements marked with an asterisk after Van Gelder (1968).
New subespecies of Conepatus semistriatus29
a (visibly) fused and obliterated basisphenoid-basioccip-
ital suture (Van Gelder 1968). These simplified criteria
may be insufficient. In skunks, subadult-juvenile speci-
mens already show a permanent dentition (Van Gelder
1959), and the basisphenoid-basioccipital suture may stay
unfused past and estimated age of eight months (Mead
1967, J. Molinari personal observation). Moreover, cra-
nial sutures may appear unfused in fully adult specimens
whose skeletal remains have been exposed to weather for
a long time. For this reason, to judge whether specimens
were adult, we added three criteria. First, as illustrated for
other carnivores (García‐Perea 1996), young skunks pos-
sess two ridges along the parietal surfaces that are widely
apart from the dorsal midline of the cranium. During
growth, these ridges migrate centrally until they converge
in the sagittal line. Irrespective of whether a well-devel-
oped sagittal crest was present or not, a much reduced
distance between both ridges was deemed an indicator of
adulthood. Second, adult skunks score low in postorbital
breadth relatively to interorbital breadth, thus a marked
constriction in the postorbital region was deemed an indi-
cator of adulthood. Third, owing to their partly hypogean
diets, meaning that food often comes mixed with mineral
particles; wild skunks suffer substantial dental wear along
their lives. Thus the possession of markedly worn teeth,
or the absence of teeth accompanied by scarring or oblit-
eration of the alveoli, was deemed an indicator of adult-
hood. The morphometric analyses performed in this study
include only specimens judged to be adult based on the
combination of these criteria.
We divided the sample into two geographic groups:
Margarita Island, and mainland (Fig. 1). In the case of the
latter, we only included specimens from Venezuela east of
the Andes) (note that Conepatus has not been recorded in
cis-Andean Colombia, Venezuela south of the Orinoco,
and the Guianas) under the assumption that the ancestors
of the Margaritan populations came from this region. For
the multivariate assessment of the data, we used Principal
Component Analysis (PCA) and Linear Discriminant
Analysis (LDA), which were computed using PAST, ver-
sion 4.16 (Hammer 2024). For univariate tests (chi-square
and correlation analysis), we used SPSS version 17.
To give equal weight to all measurements, the PCA was
performed on the correlation matrix. Missing measure-
ments were imputed through regression analysis, using
known values for the same measurement as predictors.
This was done separately for each geographic group. In
PCA, variables often have positive loadings on Compo-
nent 1 (PC1), and both positive and negative loadings in
the remaining components. Thus PC1 has been interpret-
ed as a univariate measure of multivariate size, and PC2–
PCn as univariate measures of multivariate shape (e.g.,
Jolicoeur & Mosimann 1960, Gutiérrez & Molinari 2008,
Molinari et al. 2023). To quantify the relative contribu-
tion of the measurements to overall differences in shape,
we calculated their communalities, which equal the sum
of the squared loadings of the measurement in the com-
ponents of interest (McGarigal et al. 2000), which in our
case were PC2–PC16. We excluded PC17–PC25 because
they explain a very small fraction of total variance. In PCA
performed on the correlation matrix, the total communal-
ity of any variable (i.e., the sum of its squared loadings in
PC1–PCn) equals unity (McGarigal et al., 2000).
To determine whether the western (Macanao Penin-
sula) and eastern (Paraguachoa) Margaritan populations
differ in skull size, and whether the mainland populations
show west-to-east clinal variation in skull size, we calculat-
ed the Pearson correlation coefficients (r) between: 1) the
Component 1 (PC1) scores of the specimens in the PCA;
and 2) the geographic longitudes at which they were col-
lected. We also carried out this procedure using condylo-
basal lengths instead of PC1 scores.
Material examined
We carefully georeferenced specimen localities using
Google Earth Pro (https://www.google.com/earth/), and
the combination of various sources of information, includ-
ing collector’s field notes, museum databases, zoological
and botanical literature, and maps. The list of specimens
used in the comparisons between the new subspecies and
cis-Andean C. s. semistriatus, and the localities (Fig. 1) in
which they were collected, is as follows:
C. semistriatus elieceri ssp. nov. (n = 22).— VENEZU-
ELA: Nueva Esparta, Margarita Oriental (Paraguachoa),
Cerro El Tamoco, 4 km E Santa Ana, 11.07°, -63.89°, 400
m (EBRG 3138*, MBUCV 5274*); Margarita Oriental,
Parque Nacional Cerro El Copey, 3 km SW La Asun-
ción, 11.01°, -63.89°, 500 m (MBUCV 5275*); Margarita
Oriental, Río Tacarigua, 1.5 km SSE Santa Ana, 11.06°,
-63.92°, 35 m (EBRG 3137*); Península de Macanao, near
San Francisco, 11.01°, -64.29°, 160 m (CVULA 5770*,
5771*, CVULA 8539*, 8540*, 8541*, 9121*, 9122*, 9123,
9124*, 9125*, 9126*, 9127, 9128*, 9129*, EBRG 18978*);
Península de Macanao, Quebrada la Montaña, 12 km
W Boca de Río, 10.95°, -64.29°, 12 m (EBRG 18981*,
18983*, 18984*).
C. s. semistriatus (n = 39).— VENEZUELA: Aragua,
0.5 km NW Cata, 10.47°, -67.74°, 35 m (EBRG 21079*);
Estación Biológica Rancho Grande, Parque Nacional
Henri Pittier, 14 km NW Maracay, 10.35°, -67.68°, 1150
m (AMNH 144821, EBRG 197*, 274*, 276*, 803*). Ba-
rinas, Reserva Forestal Ticoporo, 15 km S Socopó, 8.10°,
a (visibly) fused and obliterated basisphenoid-basioccip-
ital suture (Van Gelder 1968). These simplified criteria
may be insufficient. In skunks, subadult-juvenile speci-
mens already show a permanent dentition (Van Gelder
1959), and the basisphenoid-basioccipital suture may stay
unfused past and estimated age of eight months (Mead
1967, J. Molinari personal observation). Moreover, cra-
nial sutures may appear unfused in fully adult specimens
whose skeletal remains have been exposed to weather for
a long time. For this reason, to judge whether specimens
were adult, we added three criteria. First, as illustrated for
other carnivores (García‐Perea 1996), young skunks pos-
sess two ridges along the parietal surfaces that are widely
apart from the dorsal midline of the cranium. During
growth, these ridges migrate centrally until they converge
in the sagittal line. Irrespective of whether a well-devel-
oped sagittal crest was present or not, a much reduced
distance between both ridges was deemed an indicator of
adulthood. Second, adult skunks score low in postorbital
breadth relatively to interorbital breadth, thus a marked
constriction in the postorbital region was deemed an indi-
cator of adulthood. Third, owing to their partly hypogean
diets, meaning that food often comes mixed with mineral
particles; wild skunks suffer substantial dental wear along
their lives. Thus the possession of markedly worn teeth,
or the absence of teeth accompanied by scarring or oblit-
eration of the alveoli, was deemed an indicator of adult-
hood. The morphometric analyses performed in this study
include only specimens judged to be adult based on the
combination of these criteria.
We divided the sample into two geographic groups:
Margarita Island, and mainland (Fig. 1). In the case of the
latter, we only included specimens from Venezuela east of
the Andes) (note that Conepatus has not been recorded in
cis-Andean Colombia, Venezuela south of the Orinoco,
and the Guianas) under the assumption that the ancestors
of the Margaritan populations came from this region. For
the multivariate assessment of the data, we used Principal
Component Analysis (PCA) and Linear Discriminant
Analysis (LDA), which were computed using PAST, ver-
sion 4.16 (Hammer 2024). For univariate tests (chi-square
and correlation analysis), we used SPSS version 17.
To give equal weight to all measurements, the PCA was
performed on the correlation matrix. Missing measure-
ments were imputed through regression analysis, using
known values for the same measurement as predictors.
This was done separately for each geographic group. In
PCA, variables often have positive loadings on Compo-
nent 1 (PC1), and both positive and negative loadings in
the remaining components. Thus PC1 has been interpret-
ed as a univariate measure of multivariate size, and PC2–
PCn as univariate measures of multivariate shape (e.g.,
Jolicoeur & Mosimann 1960, Gutiérrez & Molinari 2008,
Molinari et al. 2023). To quantify the relative contribu-
tion of the measurements to overall differences in shape,
we calculated their communalities, which equal the sum
of the squared loadings of the measurement in the com-
ponents of interest (McGarigal et al. 2000), which in our
case were PC2–PC16. We excluded PC17–PC25 because
they explain a very small fraction of total variance. In PCA
performed on the correlation matrix, the total communal-
ity of any variable (i.e., the sum of its squared loadings in
PC1–PCn) equals unity (McGarigal et al., 2000).
To determine whether the western (Macanao Penin-
sula) and eastern (Paraguachoa) Margaritan populations
differ in skull size, and whether the mainland populations
show west-to-east clinal variation in skull size, we calculat-
ed the Pearson correlation coefficients (r) between: 1) the
Component 1 (PC1) scores of the specimens in the PCA;
and 2) the geographic longitudes at which they were col-
lected. We also carried out this procedure using condylo-
basal lengths instead of PC1 scores.
Material examined
We carefully georeferenced specimen localities using
Google Earth Pro (https://www.google.com/earth/), and
the combination of various sources of information, includ-
ing collector’s field notes, museum databases, zoological
and botanical literature, and maps. The list of specimens
used in the comparisons between the new subspecies and
cis-Andean C. s. semistriatus, and the localities (Fig. 1) in
which they were collected, is as follows:
C. semistriatus elieceri ssp. nov. (n = 22).— VENEZU-
ELA: Nueva Esparta, Margarita Oriental (Paraguachoa),
Cerro El Tamoco, 4 km E Santa Ana, 11.07°, -63.89°, 400
m (EBRG 3138*, MBUCV 5274*); Margarita Oriental,
Parque Nacional Cerro El Copey, 3 km SW La Asun-
ción, 11.01°, -63.89°, 500 m (MBUCV 5275*); Margarita
Oriental, Río Tacarigua, 1.5 km SSE Santa Ana, 11.06°,
-63.92°, 35 m (EBRG 3137*); Península de Macanao, near
San Francisco, 11.01°, -64.29°, 160 m (CVULA 5770*,
5771*, CVULA 8539*, 8540*, 8541*, 9121*, 9122*, 9123,
9124*, 9125*, 9126*, 9127, 9128*, 9129*, EBRG 18978*);
Península de Macanao, Quebrada la Montaña, 12 km
W Boca de Río, 10.95°, -64.29°, 12 m (EBRG 18981*,
18983*, 18984*).
C. s. semistriatus (n = 39).— VENEZUELA: Aragua,
0.5 km NW Cata, 10.47°, -67.74°, 35 m (EBRG 21079*);
Estación Biológica Rancho Grande, Parque Nacional
Henri Pittier, 14 km NW Maracay, 10.35°, -67.68°, 1150
m (AMNH 144821, EBRG 197*, 274*, 276*, 803*). Ba-
rinas, Reserva Forestal Ticoporo, 15 km S Socopó, 8.10°,
Molinari, Abarca-Medina & Rivas-Rodríguez30
-70.84°, 175 m (EBRG 15721*). Cojedes, Hato Los Ca-
ballos, km 68 carretera Tinaco-El Baúl, 54 km SE Tinaco,
9.29°, -68.17°, 85 m (MHNLS 6496*, 6497*); Hato Pi-
ñero, near El Baúl, 8.93°, -68.08°, 68 m (MCNG 990*);
Río Portuguesa, 11 km SSW El Baúl, 8.87°, -68.33°, 70 m
(MHNLS 1225). Guárico, Chaguaramas, 9.33°, -66.26°,
183 m (CVULA 1510); Hato Santa Bárbara, 10 km S El
Socorro, 8.90°, -65.74°, 135 m (MBUCV 3775); Zara-
za, 9.36°, -65.34°, 75 m (AMNH 135481*). Miranda,
Agua Blanca, Parque Nacional Guatopo, 10.07°, -66.47°,
400 m (CVULA 761*); El Junquito–Colonia Tovar road,
10.43°, -67.17°, 2075 m (MHNLS 3412*); Estación Ex-
perimental Río Negro, 5.5 km W Río Negro, 10.33°,
-66.20°, 45 m (MBUCV 3032*, 3964*, 4007*, 4025*,
4145*); Near Guarenas, 10.44°, -66.52°, 280 m (MBUCV
3968*); Toma de agua, Río Marasmita, 0.3 km NE Ca-
paya, 10.43°, -66.27°, 95 m (MHNLS 3671*). Mona-
gas, Caicara de Maturín, 9.82°, -63.61°, 190 m (USNM
296626*); Hato Mata de Bejuco, 47 Km SSE Maturín,
9.32°, -62.93°, 23 m (EBRG 3190, 3190, 3191, USNM
388241, 388242, 388243, 388244*); Río Guarapiche,
Cachipo sector, 9.89°, -63.06°, 20 m (MHNLS 10772*).
Portuguesa, Autopista José Antonio Páez, 3.8 km S Gua-
nare, 9.01°, -69.75°, 153 m (CVULA 8371). Sucre, 16 km
Chacopata, towards Cariaco, 10.66°, -63.71°, 10 m (CVU-
LA 8538*); Mount Turimiquire, 10.12°, -63.87°, 2300 m
(FMNH 38061*); Río Neveri, 24 km WSW Cumanacoa,
10.18°, -64.13°, 425 m (AMNH 69609). Vargas, Canales
de Naiguatá, vertiente norte del Parque Nacional El Ávila,
10.58°, -66.73°, 800 m (EBRG 3009, MHNLS 8727*).
Yaracuy, Puente Yaracuy, Nirgua-Chivacoa road, 10.14°,
-68.80°, 210 m (EBRG 490).
The specimens whose skull measurements were used for
multivariate analyses are indicated with an asterisk. The
remaining specimens consist of immature skulls with or
without study skin, or study skins only. Many other speci-
mens of Conepatus were examined for the comparisons
performed in the Taxonomy section (Appendix 1).
RESULTS
Measurements
The skull (not including imputed values) measurements
of the holotype, the Margarita Island sample, and the
mainland sample, are provided in table 1. Except for the
width of interpterygoid fossa (Fig. 2A), the new subspe-
cies averages smaller than its mainland relative in all mea-
surements. No overlap was observed between both forms
in six measurements (basilar length, condylobasal length,
mastoid breadth, interorbital breadth, postpalatal length,
and width across molars). The skull of the holotype is lon-
ger than the average for its subspecies (Fig. 2B, Table 1).
Expressed as the mean ± SD (min–max) [n], the rostral
angle (see Methods) for the new subspecies was 43.8° ±
3.4° (38.3°–48.7) [16], and that of C. s. semistriatus was
52.7° ± 1.8° (50.8°–57.0°) [21].
Morphometric analyses
The summary results of the PCA performed on the cor-
relation matrix are shown in Appendix 2. PC1 explained
69.44% of total variance. All measurements, except for
the weakly negative width of interpterygoid fossa, have
positive loadings. Thus PC1 can be viewed as an axis pri-
marily reflecting multivariate size. The remaining compo-
nents show even mixtures of negative and positive load-
ings, indicating that they can be viewed as comparisons
of shape. Based on the communality values on the second
to sixteenth axes, measurements weakly associated with
skull volume, such as precanine length, width of interpt-
erygoid fossa, length of PM4, and post-notch length have
the greatest influence on overall differences, whereas the
opposite is true for measurements strongly associated with
skull volume, such as mastoid breadth, basilar length, and
condylobasal length.
Based on specimen scores (Fig. 3), the two forms segre-
gate sharply in the first axis of the PCA, with no overlap
observed between the 95% confidence intervals of their
scores. On the contrary, they overlap broadly in the second
axis, as they do in the remaining 23 axes (not shown). To
determine whether despite this finding the PCA reflects
interspecific ‘shape’ differences, we performed a LDA on
the specimen scores in the second to twenty-fifth axes. This
analysis produced a single axis (eigenvalue = 0.11) explain-
ing 100% of the variance. The resulting discriminant func-
tion was able to classify correctly 14 of the 20 specimens
of the new subspecies, and 17 of the 25 specimens of C. s.
semistriatus. A chi-square analysis revealed these propor-
tions to differ significantly (p < 0.05) from the expected
proportions should there be no differences between both
forms. Thus the PCA did detect a differentiation in shape,
albeit an incomplete and diffuse one.
For the new subspecies, the Pearson correlation coeffi-
cient between PC1 scores and geographic longitude was
r = 0.23 (p = 0.34). For C. s. semistriatus it was r = -0.05
(p = 0.80). The results using condylobasal length instead
of PC1 were similar: r = 0.19 (p = 0.43) and r = -0.17
(p = 0.41), respectively. Hence, with respect to skull size,
there is not differentiation between the populations of
western (Macanao Peninsula) and eastern (Paraguachoa)
Margarita, which are connected by a narrow isthmus, and
there is not west to east clinal variation within the popula-
tions of mainland Venezuela east of the Andes.
-70.84°, 175 m (EBRG 15721*). Cojedes, Hato Los Ca-
ballos, km 68 carretera Tinaco-El Baúl, 54 km SE Tinaco,
9.29°, -68.17°, 85 m (MHNLS 6496*, 6497*); Hato Pi-
ñero, near El Baúl, 8.93°, -68.08°, 68 m (MCNG 990*);
Río Portuguesa, 11 km SSW El Baúl, 8.87°, -68.33°, 70 m
(MHNLS 1225). Guárico, Chaguaramas, 9.33°, -66.26°,
183 m (CVULA 1510); Hato Santa Bárbara, 10 km S El
Socorro, 8.90°, -65.74°, 135 m (MBUCV 3775); Zara-
za, 9.36°, -65.34°, 75 m (AMNH 135481*). Miranda,
Agua Blanca, Parque Nacional Guatopo, 10.07°, -66.47°,
400 m (CVULA 761*); El Junquito–Colonia Tovar road,
10.43°, -67.17°, 2075 m (MHNLS 3412*); Estación Ex-
perimental Río Negro, 5.5 km W Río Negro, 10.33°,
-66.20°, 45 m (MBUCV 3032*, 3964*, 4007*, 4025*,
4145*); Near Guarenas, 10.44°, -66.52°, 280 m (MBUCV
3968*); Toma de agua, Río Marasmita, 0.3 km NE Ca-
paya, 10.43°, -66.27°, 95 m (MHNLS 3671*). Mona-
gas, Caicara de Maturín, 9.82°, -63.61°, 190 m (USNM
296626*); Hato Mata de Bejuco, 47 Km SSE Maturín,
9.32°, -62.93°, 23 m (EBRG 3190, 3190, 3191, USNM
388241, 388242, 388243, 388244*); Río Guarapiche,
Cachipo sector, 9.89°, -63.06°, 20 m (MHNLS 10772*).
Portuguesa, Autopista José Antonio Páez, 3.8 km S Gua-
nare, 9.01°, -69.75°, 153 m (CVULA 8371). Sucre, 16 km
Chacopata, towards Cariaco, 10.66°, -63.71°, 10 m (CVU-
LA 8538*); Mount Turimiquire, 10.12°, -63.87°, 2300 m
(FMNH 38061*); Río Neveri, 24 km WSW Cumanacoa,
10.18°, -64.13°, 425 m (AMNH 69609). Vargas, Canales
de Naiguatá, vertiente norte del Parque Nacional El Ávila,
10.58°, -66.73°, 800 m (EBRG 3009, MHNLS 8727*).
Yaracuy, Puente Yaracuy, Nirgua-Chivacoa road, 10.14°,
-68.80°, 210 m (EBRG 490).
The specimens whose skull measurements were used for
multivariate analyses are indicated with an asterisk. The
remaining specimens consist of immature skulls with or
without study skin, or study skins only. Many other speci-
mens of Conepatus were examined for the comparisons
performed in the Taxonomy section (Appendix 1).
RESULTS
Measurements
The skull (not including imputed values) measurements
of the holotype, the Margarita Island sample, and the
mainland sample, are provided in table 1. Except for the
width of interpterygoid fossa (Fig. 2A), the new subspe-
cies averages smaller than its mainland relative in all mea-
surements. No overlap was observed between both forms
in six measurements (basilar length, condylobasal length,
mastoid breadth, interorbital breadth, postpalatal length,
and width across molars). The skull of the holotype is lon-
ger than the average for its subspecies (Fig. 2B, Table 1).
Expressed as the mean ± SD (min–max) [n], the rostral
angle (see Methods) for the new subspecies was 43.8° ±
3.4° (38.3°–48.7) [16], and that of C. s. semistriatus was
52.7° ± 1.8° (50.8°–57.0°) [21].
Morphometric analyses
The summary results of the PCA performed on the cor-
relation matrix are shown in Appendix 2. PC1 explained
69.44% of total variance. All measurements, except for
the weakly negative width of interpterygoid fossa, have
positive loadings. Thus PC1 can be viewed as an axis pri-
marily reflecting multivariate size. The remaining compo-
nents show even mixtures of negative and positive load-
ings, indicating that they can be viewed as comparisons
of shape. Based on the communality values on the second
to sixteenth axes, measurements weakly associated with
skull volume, such as precanine length, width of interpt-
erygoid fossa, length of PM4, and post-notch length have
the greatest influence on overall differences, whereas the
opposite is true for measurements strongly associated with
skull volume, such as mastoid breadth, basilar length, and
condylobasal length.
Based on specimen scores (Fig. 3), the two forms segre-
gate sharply in the first axis of the PCA, with no overlap
observed between the 95% confidence intervals of their
scores. On the contrary, they overlap broadly in the second
axis, as they do in the remaining 23 axes (not shown). To
determine whether despite this finding the PCA reflects
interspecific ‘shape’ differences, we performed a LDA on
the specimen scores in the second to twenty-fifth axes. This
analysis produced a single axis (eigenvalue = 0.11) explain-
ing 100% of the variance. The resulting discriminant func-
tion was able to classify correctly 14 of the 20 specimens
of the new subspecies, and 17 of the 25 specimens of C. s.
semistriatus. A chi-square analysis revealed these propor-
tions to differ significantly (p < 0.05) from the expected
proportions should there be no differences between both
forms. Thus the PCA did detect a differentiation in shape,
albeit an incomplete and diffuse one.
For the new subspecies, the Pearson correlation coeffi-
cient between PC1 scores and geographic longitude was
r = 0.23 (p = 0.34). For C. s. semistriatus it was r = -0.05
(p = 0.80). The results using condylobasal length instead
of PC1 were similar: r = 0.19 (p = 0.43) and r = -0.17
(p = 0.41), respectively. Hence, with respect to skull size,
there is not differentiation between the populations of
western (Macanao Peninsula) and eastern (Paraguachoa)
Margarita, which are connected by a narrow isthmus, and
there is not west to east clinal variation within the popula-
tions of mainland Venezuela east of the Andes.
New subespecies of Conepatus semistriatus31
Table 1. Summary statistics for the skull measurements (mm) of adult Conepatus specimens used in this study. The values
are expressed as mean ± SD (min–max) [n]. Only specimens from Margarita Island (C. semistriatus elieceri ssp. nov.) and
the mainland of Venezuela east of the Andes (C. s. semistriatus) are included (Fig. 1).
Measurement (mm) Holotype (EBRG 3138) C. s. elieceri ssp. nov. C. s. semistriatus
Basilar length 65.7 63.8 ± 2.1 (60.0–66.4) [18] 73.8 ± 3.1 (67.7–79.2) [24]
Condylobasal length 73.0 71.9 ± 2.0 (68.3–74.7) [18] 82.4 ± 3.1 (77.9–88.8) [24]
Zygomatic breadth 50.5 47.4 ± 2.4 (43.4–50.9) [15] 54.6 ± 2.8 (50.0–59.4) [24]
Mastoid breadth 39.7 38.6 ± 1.3 (36.4–40.9) [20] 45.0 ± 1.6 (42.4–47.9) [24]
Interorbital breadth 22.8 23.4 ± 0.7 (22.3–24.5) [20] 27.0 ± 1.4 (24.9–29.9) [24]
Postorbital breadth 21.3 21.3 ± 0.7 (20.1–22.5) [20] 23.3 ± 0.9 (21.8–25.5) [24]
Palatilar length 30.4 29.5 ± 1.1 (26.7–30.7) [18] 33.5 ± 1.5 (30.4–36.8) [25]
Post-notch length 5.2 4.2 ± 0.6 (3.1–5.2) [18] 4.4 ± 0.9 (2.5–5.9) [25]
Notch to canine length 22.3 21.7 ± 0.9 (19.6–23.3) [18] 25.4 ± 1.4 (23.3–28.6) [25]
Precanine length 4.7 5.4 ± 0.4 (4.7–6.2) [20] 5.8 ± 0.4 (5.0–6.9) [25]
Postpalatal length 33.9 33.4 ± 1.0 (31.5–35.1) [18] 38.8 ± 2.2 (36.0–43.8) [24]
Heigth of cranium 27.1 26.5 ± 0.7 (25.2–27.8) [19] 29.9 ± 1.9 (25.5–35.1) [24]
Length of maxillary toothrow 23.0 22.9 ± 0.8 (21.3–25.3) [20] 26.3 ± 1.0 (24.8–27.9) [25]
Width across incisors 12.1 11.7 ± 0.6 (10.0–12.7) [20] 13.1 ± 0.7 (11.4–14.6) [25]
Width across canines 19.5 18.4 ± 0.7 (17.2–19.5) [20] 21.7 ± 1.0 (19.5–23.4) [25]
Width across molars 31.1 30.3 ± 0.6 (29.2–31.6) [20] 34.9 ± 1.1 (32.6–37.5) [24]
Diameter of canine 4.8 4.4 ± 0.3 (3.8–5.1) [20] 5.4 ± 0.4 (4.5–6.2) [25]
Length of PM3 4.3 3.9 ± 0.5 (2.2–4.5) [20] 4.7 ± 0.4 (3.9–5.7) [25]
Length of PM4 7.8 7.9 ± 0.3 (7.2–8.4) [20] 8.9 ± 0.5 (7.9–9.8) [25]
Length of molar 7.7 8.0 ± 0.4 (7.3–8.6) [19] 9.5 ± 0.7 (8.5–11.7) [25]
Width of molar 10.0 10.4 ± 0.4 (9.6–11.2) [19] 11.9 ± 0.6 (10.5–12.8) [25]
Width of interpterygoid fossa 6.4 8.6 ± 0.9 (6.4–9.7) [16] 8.0 ± 1.2 (6.1–10.5) [23]
Length of lower carnassial 10.6 10.7 ± 0.3 (10.2–11.1) [12] 11.5 ± 0.3 (11.0–12.2) [23]
Height of coronoid 22.9 24.5 ± 1.3 (22.5–27.8) [19] 26.6 ± 1.9 (23.5–29.7) [21]
Length of mandible 49.0 48.8 ± 2.2 (44.6–53.0) [19] 53.9 ± 2.6 (49.8–58.7) [21]
A LDA performed directly on the skull measurements
also produced a single axis (eigenvalue = 49.96), thus the
results are presented as one histogram for each subspecies
on the same axis (Fig. 4). A normal curve (new subspecies,
mean = 7.7, SD = 0.9; C. s. semistriatus, mean = 6.2, SD =
1.1) was fitted to each histogram. The means are separated
by more than 12 (rightwards) and 16 (leftwards) standard
deviations. Hence, the discriminant function obtained
should be able to classify correctly all specimens of both
subspecies even if much larger samples were available.
The discriminant function is: specimen score = 68.54
+ (loading 1 × measurement 1) + (loading 2 × measure-
ment 2) + … + (loading 25 × measurement 25). The mean
of the loadings on the single axis is 0.30. The loadings be-
low the mean are: length of pm4, 2.10; width across inci-
sors, 1.89; basilar length, 1.27; condylobasal length, 1.19;
Table 1. Summary statistics for the skull measurements (mm) of adult Conepatus specimens used in this study. The values
are expressed as mean ± SD (min–max) [n]. Only specimens from Margarita Island (C. semistriatus elieceri ssp. nov.) and
the mainland of Venezuela east of the Andes (C. s. semistriatus) are included (Fig. 1).
Measurement (mm) Holotype (EBRG 3138) C. s. elieceri ssp. nov. C. s. semistriatus
Basilar length 65.7 63.8 ± 2.1 (60.0–66.4) [18] 73.8 ± 3.1 (67.7–79.2) [24]
Condylobasal length 73.0 71.9 ± 2.0 (68.3–74.7) [18] 82.4 ± 3.1 (77.9–88.8) [24]
Zygomatic breadth 50.5 47.4 ± 2.4 (43.4–50.9) [15] 54.6 ± 2.8 (50.0–59.4) [24]
Mastoid breadth 39.7 38.6 ± 1.3 (36.4–40.9) [20] 45.0 ± 1.6 (42.4–47.9) [24]
Interorbital breadth 22.8 23.4 ± 0.7 (22.3–24.5) [20] 27.0 ± 1.4 (24.9–29.9) [24]
Postorbital breadth 21.3 21.3 ± 0.7 (20.1–22.5) [20] 23.3 ± 0.9 (21.8–25.5) [24]
Palatilar length 30.4 29.5 ± 1.1 (26.7–30.7) [18] 33.5 ± 1.5 (30.4–36.8) [25]
Post-notch length 5.2 4.2 ± 0.6 (3.1–5.2) [18] 4.4 ± 0.9 (2.5–5.9) [25]
Notch to canine length 22.3 21.7 ± 0.9 (19.6–23.3) [18] 25.4 ± 1.4 (23.3–28.6) [25]
Precanine length 4.7 5.4 ± 0.4 (4.7–6.2) [20] 5.8 ± 0.4 (5.0–6.9) [25]
Postpalatal length 33.9 33.4 ± 1.0 (31.5–35.1) [18] 38.8 ± 2.2 (36.0–43.8) [24]
Heigth of cranium 27.1 26.5 ± 0.7 (25.2–27.8) [19] 29.9 ± 1.9 (25.5–35.1) [24]
Length of maxillary toothrow 23.0 22.9 ± 0.8 (21.3–25.3) [20] 26.3 ± 1.0 (24.8–27.9) [25]
Width across incisors 12.1 11.7 ± 0.6 (10.0–12.7) [20] 13.1 ± 0.7 (11.4–14.6) [25]
Width across canines 19.5 18.4 ± 0.7 (17.2–19.5) [20] 21.7 ± 1.0 (19.5–23.4) [25]
Width across molars 31.1 30.3 ± 0.6 (29.2–31.6) [20] 34.9 ± 1.1 (32.6–37.5) [24]
Diameter of canine 4.8 4.4 ± 0.3 (3.8–5.1) [20] 5.4 ± 0.4 (4.5–6.2) [25]
Length of PM3 4.3 3.9 ± 0.5 (2.2–4.5) [20] 4.7 ± 0.4 (3.9–5.7) [25]
Length of PM4 7.8 7.9 ± 0.3 (7.2–8.4) [20] 8.9 ± 0.5 (7.9–9.8) [25]
Length of molar 7.7 8.0 ± 0.4 (7.3–8.6) [19] 9.5 ± 0.7 (8.5–11.7) [25]
Width of molar 10.0 10.4 ± 0.4 (9.6–11.2) [19] 11.9 ± 0.6 (10.5–12.8) [25]
Width of interpterygoid fossa 6.4 8.6 ± 0.9 (6.4–9.7) [16] 8.0 ± 1.2 (6.1–10.5) [23]
Length of lower carnassial 10.6 10.7 ± 0.3 (10.2–11.1) [12] 11.5 ± 0.3 (11.0–12.2) [23]
Height of coronoid 22.9 24.5 ± 1.3 (22.5–27.8) [19] 26.6 ± 1.9 (23.5–29.7) [21]
Length of mandible 49.0 48.8 ± 2.2 (44.6–53.0) [19] 53.9 ± 2.6 (49.8–58.7) [21]
A LDA performed directly on the skull measurements
also produced a single axis (eigenvalue = 49.96), thus the
results are presented as one histogram for each subspecies
on the same axis (Fig. 4). A normal curve (new subspecies,
mean = 7.7, SD = 0.9; C. s. semistriatus, mean = 6.2, SD =
1.1) was fitted to each histogram. The means are separated
by more than 12 (rightwards) and 16 (leftwards) standard
deviations. Hence, the discriminant function obtained
should be able to classify correctly all specimens of both
subspecies even if much larger samples were available.
The discriminant function is: specimen score = 68.54
+ (loading 1 × measurement 1) + (loading 2 × measure-
ment 2) + … + (loading 25 × measurement 25). The mean
of the loadings on the single axis is 0.30. The loadings be-
low the mean are: length of pm4, 2.10; width across inci-
sors, 1.89; basilar length, 1.27; condylobasal length, 1.19;
Molinari, Abarca-Medina & Rivas-Rodríguez32
zygomatic breadth, 0.84; width of interpterygoid fossa,
0.75; diameter of canine, 0.51; height of coronoid, 0.47;
height of cranium, 0.44; postorbital breadth, 0.36; length
of lower carnassial, 0.28; length of molar, 0.24; and length
of mandible, 0.28. The loadings above the mean are: notch
to canine length, 0.32; width across molars, 0.54; post-
notch length, 0.72; palatilar length, 1.02; length of pm3,
1.30; mastoid breadth, 1.43; width across canines, 1.59;
length of maxillary toothrow, 1.69; interorbital breadth,
1.70; precanine length, 2.07; postpalatal length, 2.12; and
width of molar, 2.60.
TAXONOMY
Conepatus semistriatus elieceri ssp. nov.
http://zoobank.org/urn:lsid:zoobank.org:act:3363DEF5-5EAA-4BF7-
ADA4-5BF710E457B7
Margaritan Hog-nosed Skunk
Zorrillo rayado margariteño
Holotype (Figs. 2B and 5A)
An adult female (EBRG 3138), consisting of cranium,
mandibles, and study skin.
Type locality
Venezuela, Estado Nueva Esparta, Margarita Island,
Cerro El Tamoco, 4 km E Santa Ana, 11.07° N, 63.89° W,
400 m.
Paratypes
We designate as paratypes three specimens of unknown
sex: MBUCV 5274, from the type locality; EBRG 18983,
from Quebrada la Montaña, 12 km W Boca de Río,
10.946° N, 64.287° W, 12 m; and CVULA 9121, from
near San Francisco, 11.01° N, 64.29° W, 160 m.
Measurements of the type material
The measurements of the holotype are provided in
Table 1. The skull measurements (mm) of the paratypes
(MBUCV 5274, EBRG 18983, CVULA 9121) are:
basilar length, 63.4, 65.0, 62.3; condylobasal length, 71.8,
73.0, 70.6; zygomatic breadth, 46.6, 50.1, 44.0; mastoid
breadth, 39.7, 40.0, 36.6; interorbital breadth, 23.5, 24.3,
Figure 3. Specimen scores on the first two axes of the Princi-
pal Components Analysis (PCA) of the cranial measurements.
New subspecies (red dots). C. s. semistriatus (blue diamonds).
The ellipses enclosing the specimen symbols represent the 95%
confidence intervals.
Figure 4. Results of the Linear Discriminant Analysis of the cranial measurements. A single axis explaining 100% of variance was ob-
tained. The scores of the specimens of the new subspecies (left, red) and C. s. semistriatus (right, blue) are represented as two histograms
along the axis.
zygomatic breadth, 0.84; width of interpterygoid fossa,
0.75; diameter of canine, 0.51; height of coronoid, 0.47;
height of cranium, 0.44; postorbital breadth, 0.36; length
of lower carnassial, 0.28; length of molar, 0.24; and length
of mandible, 0.28. The loadings above the mean are: notch
to canine length, 0.32; width across molars, 0.54; post-
notch length, 0.72; palatilar length, 1.02; length of pm3,
1.30; mastoid breadth, 1.43; width across canines, 1.59;
length of maxillary toothrow, 1.69; interorbital breadth,
1.70; precanine length, 2.07; postpalatal length, 2.12; and
width of molar, 2.60.
TAXONOMY
Conepatus semistriatus elieceri ssp. nov.
http://zoobank.org/urn:lsid:zoobank.org:act:3363DEF5-5EAA-4BF7-
ADA4-5BF710E457B7
Margaritan Hog-nosed Skunk
Zorrillo rayado margariteño
Holotype (Figs. 2B and 5A)
An adult female (EBRG 3138), consisting of cranium,
mandibles, and study skin.
Type locality
Venezuela, Estado Nueva Esparta, Margarita Island,
Cerro El Tamoco, 4 km E Santa Ana, 11.07° N, 63.89° W,
400 m.
Paratypes
We designate as paratypes three specimens of unknown
sex: MBUCV 5274, from the type locality; EBRG 18983,
from Quebrada la Montaña, 12 km W Boca de Río,
10.946° N, 64.287° W, 12 m; and CVULA 9121, from
near San Francisco, 11.01° N, 64.29° W, 160 m.
Measurements of the type material
The measurements of the holotype are provided in
Table 1. The skull measurements (mm) of the paratypes
(MBUCV 5274, EBRG 18983, CVULA 9121) are:
basilar length, 63.4, 65.0, 62.3; condylobasal length, 71.8,
73.0, 70.6; zygomatic breadth, 46.6, 50.1, 44.0; mastoid
breadth, 39.7, 40.0, 36.6; interorbital breadth, 23.5, 24.3,
Figure 3. Specimen scores on the first two axes of the Princi-
pal Components Analysis (PCA) of the cranial measurements.
New subspecies (red dots). C. s. semistriatus (blue diamonds).
The ellipses enclosing the specimen symbols represent the 95%
confidence intervals.
Figure 4. Results of the Linear Discriminant Analysis of the cranial measurements. A single axis explaining 100% of variance was ob-
tained. The scores of the specimens of the new subspecies (left, red) and C. s. semistriatus (right, blue) are represented as two histograms
along the axis.
New subespecies of Conepatus semistriatus33
23.0; postorbital breadth, 20.7, 22.5, 20.5; palatilar length,
29.3, 30.3, 28.7; post-notch length, 5.0, 3.9, 4.0; notch to
canine length, 22.6, 23.3, 20.7; precanine length, 5.6, 5.1,
5.2; postpalatal length, 32.9, 33.5, 32.9; height of cranium,
25.6, 27.8, 25.2; length of maxillary toothrow, 22.1, 23.7,
22.4; width across incisors, 11.3, 12.1, 11.6; width across
canines, 17.2, 19.3, 18.1; width across molars, 29.9, 30.8,
29.8; diameter of canine, 4.2, 5.1, 4.0; length of PM3, 3.6,
4.5, 3.2; length of PM4, 7.2, 8.0, 7.8; length of molar, 7.3,
7.8, 8.0; width of molar, 9.7, 10.2, 10.6; width of interpt-
erygoid fossa, 8.2, 8.9, 9.7; length of lower carnassial, —,
10.6, 10.5; height of coronoid, 27.8, 25.0, 22.5; length of
mandible, 49.1, 48.6, 44.6.
Diagnosis
Small for a South American member of the C. semistri-
atus group. The rostrum is relatively low (rostral angle less
than 50°, see Methods) in lateral view (Fig. 2B). On dorsal
view of the cranium, the anteriormost margins of the nasal
bones do not project fully over the palatal plane, leaving
the anterior region of the premaxillae and the incisive and
interincisive foramina fully exposed. In most specimens,
the anterior opening of the infraorbital foramen is single
(not divided) in at least one side of the cranium. The na-
sal spine is short or absent, and the nasal septum (part of
the vomer bone) does not extend rearwards, thus does not
form a keel, past the postpalatal shelf. The hypoglossal
(condyloid) foramen is well separated from the posterior
lacerate foramen.
Description
As other Venezuelan, Colombian, and Central Ameri-
can members of the C. semistriatus group, the holotype
possesses two white dorsal stripes, joined only on the
head, each of them broader anteriorly and narrower pos-
teriorly, and with little separation between them (Fig. 5A).
The type specimen is unusual in possessing numerous dark
spots on the dorsal stripes, and a fully dark tail with short
hairs all along (Fig. 5A). The hairs of the dorsal stripes are
longer than those of the back. The pelage is brown and the
skin is yellowish. As characteristic of Conepatus, the area
around the nose is bare, the ears are much reduced, and the
claws on the forefeet are long.
The cranial size (Table 1) is medium for the genus:
condylobasal length is more than 70 mm, and less than
77 mm. The skull is much higher in the temporal than in
the frontal region. The rostrum is narrow and short in dor-
sal view (Fig. 2B). In older adults, a low (usually 1–2 mm)
sagittal crest is present. The zygomatic arches vary from
moderately to strongly bowed upwards. The tympanic bul-
lae are small, and not inflated. The dental formula is: inci-
sors 3/3, canines 1/1, premolars 2/3, and molars 1/2 on
each side, for a total of 32 teeth. The single upper molar is
much enlarged; P1 is absent, as in all skunks; P2 is absent,
as in most Conepatus. There is a distinct notch between
the metacone and the hypocone of the upper molars. The
lower edges of the mandibles are bowed downwards. Most
specimens have a single mental foramen (few have one or
two small accessory foramina near the mental foramen) on
each mandible. The coronoid process has a narrow tip, and
is bowed forward both anteriorly and posteriorly: these
characteristics of the mandible are more marked in other
Margaritan specimens than in the holotype (Fig. 2B). In
m1, the talonid is much longer than trigonid.
Comparisons (Table 1, Figs. 2 and 5)
We compare the new subspecies first with its geograph-
ic neighbor, and likely closest relative, then with species of
Conepatus distributed from north to south in the Nearctic
and Neotropical regions.
Compared to C. s. semistriatus from the Venezuelan
mainland east of the Andes (this study), the new subspe-
cies is much smaller, with no overlap in several cranial mea-
surements. Its rostrum is proportionally shorter, narrower,
and lower, with a rostral angle of less than 50°, as opposed
to more than 50°. The coronoid process is narrow-tipped,
and much bowed forwards, as opposed to broad tipped,
and moderately bowed or straight. In the new subspe-
cies, each mandible usually has a single mental foramen,
sometimes surrounded by one or two poorly developed
accessory foramina, as opposed to one mental foramen ac-
companied by one to four, often well-developed, accessory
foramina. Little is known about the external appearance
of the new subspecies, but the study skin of the holotype,
which is the only one available, has two features not ob-
served in C. s. semistriatus, namely numerous dark spots
on the dorsal stripes, and a dark tail, as opposed to totally
white stripes, and a tail having at least a white tip (in most
cases the distal half or two-thirds are white). The tail also
has shorter hairs.
Compared to C. leuconotus, the new subspecies possess-
es a hypoglossal foramen well separated from the posterior
lacerate foramen, as opposed to confluent with it. In m1,
the talonid is much longer than trigonid, as opposed to
the talonid slightly longer than the trigonid. Dorsally, the
pelage shows two white stripes, as opposed to a single and
central white stripe.
Compared to Central American C. semistriatus, on
dorsal view of the cranium, the incisive and interincisive
foramina of the new subspecies are fully exposed, as op-
posed to partly covered by the nasals, which extend farther
frontally. The precanine and postdental regions are pro-
23.0; postorbital breadth, 20.7, 22.5, 20.5; palatilar length,
29.3, 30.3, 28.7; post-notch length, 5.0, 3.9, 4.0; notch to
canine length, 22.6, 23.3, 20.7; precanine length, 5.6, 5.1,
5.2; postpalatal length, 32.9, 33.5, 32.9; height of cranium,
25.6, 27.8, 25.2; length of maxillary toothrow, 22.1, 23.7,
22.4; width across incisors, 11.3, 12.1, 11.6; width across
canines, 17.2, 19.3, 18.1; width across molars, 29.9, 30.8,
29.8; diameter of canine, 4.2, 5.1, 4.0; length of PM3, 3.6,
4.5, 3.2; length of PM4, 7.2, 8.0, 7.8; length of molar, 7.3,
7.8, 8.0; width of molar, 9.7, 10.2, 10.6; width of interpt-
erygoid fossa, 8.2, 8.9, 9.7; length of lower carnassial, —,
10.6, 10.5; height of coronoid, 27.8, 25.0, 22.5; length of
mandible, 49.1, 48.6, 44.6.
Diagnosis
Small for a South American member of the C. semistri-
atus group. The rostrum is relatively low (rostral angle less
than 50°, see Methods) in lateral view (Fig. 2B). On dorsal
view of the cranium, the anteriormost margins of the nasal
bones do not project fully over the palatal plane, leaving
the anterior region of the premaxillae and the incisive and
interincisive foramina fully exposed. In most specimens,
the anterior opening of the infraorbital foramen is single
(not divided) in at least one side of the cranium. The na-
sal spine is short or absent, and the nasal septum (part of
the vomer bone) does not extend rearwards, thus does not
form a keel, past the postpalatal shelf. The hypoglossal
(condyloid) foramen is well separated from the posterior
lacerate foramen.
Description
As other Venezuelan, Colombian, and Central Ameri-
can members of the C. semistriatus group, the holotype
possesses two white dorsal stripes, joined only on the
head, each of them broader anteriorly and narrower pos-
teriorly, and with little separation between them (Fig. 5A).
The type specimen is unusual in possessing numerous dark
spots on the dorsal stripes, and a fully dark tail with short
hairs all along (Fig. 5A). The hairs of the dorsal stripes are
longer than those of the back. The pelage is brown and the
skin is yellowish. As characteristic of Conepatus, the area
around the nose is bare, the ears are much reduced, and the
claws on the forefeet are long.
The cranial size (Table 1) is medium for the genus:
condylobasal length is more than 70 mm, and less than
77 mm. The skull is much higher in the temporal than in
the frontal region. The rostrum is narrow and short in dor-
sal view (Fig. 2B). In older adults, a low (usually 1–2 mm)
sagittal crest is present. The zygomatic arches vary from
moderately to strongly bowed upwards. The tympanic bul-
lae are small, and not inflated. The dental formula is: inci-
sors 3/3, canines 1/1, premolars 2/3, and molars 1/2 on
each side, for a total of 32 teeth. The single upper molar is
much enlarged; P1 is absent, as in all skunks; P2 is absent,
as in most Conepatus. There is a distinct notch between
the metacone and the hypocone of the upper molars. The
lower edges of the mandibles are bowed downwards. Most
specimens have a single mental foramen (few have one or
two small accessory foramina near the mental foramen) on
each mandible. The coronoid process has a narrow tip, and
is bowed forward both anteriorly and posteriorly: these
characteristics of the mandible are more marked in other
Margaritan specimens than in the holotype (Fig. 2B). In
m1, the talonid is much longer than trigonid.
Comparisons (Table 1, Figs. 2 and 5)
We compare the new subspecies first with its geograph-
ic neighbor, and likely closest relative, then with species of
Conepatus distributed from north to south in the Nearctic
and Neotropical regions.
Compared to C. s. semistriatus from the Venezuelan
mainland east of the Andes (this study), the new subspe-
cies is much smaller, with no overlap in several cranial mea-
surements. Its rostrum is proportionally shorter, narrower,
and lower, with a rostral angle of less than 50°, as opposed
to more than 50°. The coronoid process is narrow-tipped,
and much bowed forwards, as opposed to broad tipped,
and moderately bowed or straight. In the new subspe-
cies, each mandible usually has a single mental foramen,
sometimes surrounded by one or two poorly developed
accessory foramina, as opposed to one mental foramen ac-
companied by one to four, often well-developed, accessory
foramina. Little is known about the external appearance
of the new subspecies, but the study skin of the holotype,
which is the only one available, has two features not ob-
served in C. s. semistriatus, namely numerous dark spots
on the dorsal stripes, and a dark tail, as opposed to totally
white stripes, and a tail having at least a white tip (in most
cases the distal half or two-thirds are white). The tail also
has shorter hairs.
Compared to C. leuconotus, the new subspecies possess-
es a hypoglossal foramen well separated from the posterior
lacerate foramen, as opposed to confluent with it. In m1,
the talonid is much longer than trigonid, as opposed to
the talonid slightly longer than the trigonid. Dorsally, the
pelage shows two white stripes, as opposed to a single and
central white stripe.
Compared to Central American C. semistriatus, on
dorsal view of the cranium, the incisive and interincisive
foramina of the new subspecies are fully exposed, as op-
posed to partly covered by the nasals, which extend farther
frontally. The precanine and postdental regions are pro-
Molinari, Abarca-Medina & Rivas-Rodríguez34
portionally longer with respect to the rest of the palate.
The anterior opening of the infraorbital foramen is usually
single, as opposed to double or triple. Each mandible usu-
ally has a single mental foramen, as opposed to one mental
foramen accompanied by up to six accessory foramina. In
contrast with the holotype of the new subspecies, speci-
mens of Central American C. semistriatus possess distally
white (one-third to more than one-half the length) tails.
Compared to Ecuadorian and Peruvian C. semistriatus
(C. s. taxinus, and C. s. zorrino) and C. chinga (C. c. qui-
tensis, C. c. rex), the new subspecies is smaller. On dorsal
view of the cranium, its incisive and interincisive foramina
are fully exposed, as opposed to partly or totally covered
by the nasals. Its postorbital region is not as narrowly con-
stricted, its sagittal crest and nasal spine are less developed,
and its upper molars and fourth upper premolars (P4) are
proportionally larger. The holotype of the new subspecies
has dorsal stripes that are less separated from each other
than those of Ecuadorian and Peruvian specimens.
Compared to C. chinga amazonicus and southern
South American members of the C. chinga-group, the
new subspecies has a short or absent nasal spine with the
Figure 5. Comparison of study skins. A) holotype of C. semistriatus elieceri ssp. nov. (EBRG 3138). B) C. s. semistriatus (FMNH
38061). C) C. s. semistriatus (EBRG 3009). The scale (bottom) applies to the three skins. Note the uniformly dark tail, and the dark
spots on the dorsal stripes of the study skin of the holotype, which is the only one available for the new subspecies.
portionally longer with respect to the rest of the palate.
The anterior opening of the infraorbital foramen is usually
single, as opposed to double or triple. Each mandible usu-
ally has a single mental foramen, as opposed to one mental
foramen accompanied by up to six accessory foramina. In
contrast with the holotype of the new subspecies, speci-
mens of Central American C. semistriatus possess distally
white (one-third to more than one-half the length) tails.
Compared to Ecuadorian and Peruvian C. semistriatus
(C. s. taxinus, and C. s. zorrino) and C. chinga (C. c. qui-
tensis, C. c. rex), the new subspecies is smaller. On dorsal
view of the cranium, its incisive and interincisive foramina
are fully exposed, as opposed to partly or totally covered
by the nasals. Its postorbital region is not as narrowly con-
stricted, its sagittal crest and nasal spine are less developed,
and its upper molars and fourth upper premolars (P4) are
proportionally larger. The holotype of the new subspecies
has dorsal stripes that are less separated from each other
than those of Ecuadorian and Peruvian specimens.
Compared to C. chinga amazonicus and southern
South American members of the C. chinga-group, the
new subspecies has a short or absent nasal spine with the
Figure 5. Comparison of study skins. A) holotype of C. semistriatus elieceri ssp. nov. (EBRG 3138). B) C. s. semistriatus (FMNH
38061). C) C. s. semistriatus (EBRG 3009). The scale (bottom) applies to the three skins. Note the uniformly dark tail, and the dark
spots on the dorsal stripes of the study skin of the holotype, which is the only one available for the new subspecies.
New subespecies of Conepatus semistriatus35
nasal septum fully inside the nasal cavity, as opposed to a
well-developed nasal spine with a nasal septum that typi-
cally (C. c. amazonicus), or sometimes (southern South
American material), continues into a keel posterior to the
nasal spine. Its zygomatic arches are from moderately to
strongly bowed upwards, as opposed to moderately bowed
or almost straight. It has a less inflated tympanic bulla.
Compared only to C. c. amazonicus, the new subspecies
is smaller. Unlike the holotype of the new subspecies, C. c.
amazonicus typically has a predominantly white tail. Com-
pared only to southern South American specimens, the
new subspecies is larger. On dorsal view of the cranium,
its incisive and interincisive foramina are fully exposed, as
opposed to partly or, more typically, covered by the nasals,
which extend anteriorly causing the nasal cavity to open
frontally through a comparatively small and rounded ori-
fice. The holotype of the new subspecies has narrowly-sep-
arated dorsal stripes, as opposed to widely-separated dor-
sal stripes (often reduced or absent). The C. semistriatus
from Venezuela, Colombia, and Central America always
have well-developed dorsal stripes. This seems to be also de
case of C. c. amazonicus.
Distribution (Fig. 1)
Endemic to Margarita Island, Venezuela. Occurring in
both geographic subdivisions (Macanao Peninsula, Para-
guachoa) of the island.
Etymology
The epithet elieceri, a masculine noun in the genitive
case, honors the Venezuelan researcher Eliécer E. Gutiér-
rez, in recognition of his outstanding contributions to
Neotropical mammalogy.
DISCUSSION
Cranially, species of the genus Conepatus lack highly
distinctive characters, and can be distinguished from each
other mainly based on size and proportions. In addition, at
the intraspecific level they are variable in coloration (Van
Gelder 1968, Dragoo et al. 2003, Schiaffini et al. 2013,
Teta et al. 2020, Ferguson et al. 2022). Perhaps it is for
these reasons that Hershkovitz (1959) suggested all mem-
bers of the genus to be conspecific, which the genetic data
now available (Dragoo et al. 2003, Schiaffini et al. 2013,
Rodrigues 2013) show not to be the case. In this context,
the finding that the Margaritan skunk is 100% distin-
guishable from its nearby mainland congener, not only
morphometrically but also to the naked eye, is remarkable.
In fact, the differences in raw measurements, and PCA
and LDA scores, between both forms are similar or greater
to those found in interspecific comparisons within genera
of several carnivoran families (e.g., Van Valkenburgh &
Wayne 1994, Taylor & Matheson 1999, Bertrand & Mor-
isot 2012, Bornholdt et al. 2013, Srinivas & Jhala 2021).
We are assigning the subspecies rank to this skunk based
on the findings that it is highly diagnosable, and that its
morphometric differences are large; we are not assigning
the species rank because such differences involve mainly
size (Molinari 2023b).
Middle-sized carnivorans often evolve dwarfism after
colonizing islands, either by over-water dispersal or gla-
cial-period vicariance. Examples include insular endemics
such as the Channel Islands fox, Urocyon littoralis (Baird,
1858) (Coonan et al. 2010), the Pygmy Raccoon, Procyon
pygmaeus Merriam, 1901 (Villa-Meza et al. 2011), the
Dwarf Coati, Nasua nelsoni Merriam, 1901 (Cuarón et al.
2009), the Cozumel Island fox, genus Urocyon Baird, 1857
(Gompper et al. 2006), and the Margaritan Hog-nosed
Skunk, C. s. elieceri (this study). The latter is the first known
case of insular dwarfism in extant skunks: a dwarfed fossil
skunk, Promephitis majori Pilgrim, 1933, has been found
on the Greek island of Samos (Pilgrim 1933). Exceptions
include the Tres Marías Raccoon, Procyon lotor insularis
Merriam, 1898 (Wilson 1991), and the extinct Falkland
Islands wolf, Dusicyon australis (Kerr, 1792) (Lyras et al.
2010), which are (or were) similar-sized to their mainland
ancestors. Thus, the Margaritan skunk is not alone among
carnivorans in following the island rule (Foster 1964, Van
Valen 1973).
The suite of morphological, functional, and behavioral
changes often observed in insular organisms has been re-
ferred to as the ‘island syndrome’ (Adler & Levins 1994,
Baeckens & Damme 2020). These changes may involve
traits such as body size (considered by the island rule),
cranial shape, limb proportions, coloration, diet, and es-
cape behavior (Adler & Levins 1994, Sánchez-Villagra et
al. 2016, van der Geer 2019, Baeckens & Damme 2020).
To some degree, C. s. elieceri also complies with the island
syndrome since it differs from C. s. semistriatus in a cranial
shape parameter (rostral angle) that could not be quanti-
fied using linear measurements.
Along the last 500,000 years, on four times global sea
level was more than 100 m below the present level: about
20,000 (Last Glacial Maximum), 140,000 (Penultimate
Glacial Period), 250,000, and 340,000 YBP (Waelbroeck
et al. 2002, Siddall et al. 2007, Lichter et al. 2010). At such
times, many continental-shelf islands were connected to
the mainland by land bridges, offering the opportunity
of colonization to mainland populations. Animals can be
expected to cross ecologically suitable land bridges every
time that they are formed, and pre-established insular
nasal septum fully inside the nasal cavity, as opposed to a
well-developed nasal spine with a nasal septum that typi-
cally (C. c. amazonicus), or sometimes (southern South
American material), continues into a keel posterior to the
nasal spine. Its zygomatic arches are from moderately to
strongly bowed upwards, as opposed to moderately bowed
or almost straight. It has a less inflated tympanic bulla.
Compared only to C. c. amazonicus, the new subspecies
is smaller. Unlike the holotype of the new subspecies, C. c.
amazonicus typically has a predominantly white tail. Com-
pared only to southern South American specimens, the
new subspecies is larger. On dorsal view of the cranium,
its incisive and interincisive foramina are fully exposed, as
opposed to partly or, more typically, covered by the nasals,
which extend anteriorly causing the nasal cavity to open
frontally through a comparatively small and rounded ori-
fice. The holotype of the new subspecies has narrowly-sep-
arated dorsal stripes, as opposed to widely-separated dor-
sal stripes (often reduced or absent). The C. semistriatus
from Venezuela, Colombia, and Central America always
have well-developed dorsal stripes. This seems to be also de
case of C. c. amazonicus.
Distribution (Fig. 1)
Endemic to Margarita Island, Venezuela. Occurring in
both geographic subdivisions (Macanao Peninsula, Para-
guachoa) of the island.
Etymology
The epithet elieceri, a masculine noun in the genitive
case, honors the Venezuelan researcher Eliécer E. Gutiér-
rez, in recognition of his outstanding contributions to
Neotropical mammalogy.
DISCUSSION
Cranially, species of the genus Conepatus lack highly
distinctive characters, and can be distinguished from each
other mainly based on size and proportions. In addition, at
the intraspecific level they are variable in coloration (Van
Gelder 1968, Dragoo et al. 2003, Schiaffini et al. 2013,
Teta et al. 2020, Ferguson et al. 2022). Perhaps it is for
these reasons that Hershkovitz (1959) suggested all mem-
bers of the genus to be conspecific, which the genetic data
now available (Dragoo et al. 2003, Schiaffini et al. 2013,
Rodrigues 2013) show not to be the case. In this context,
the finding that the Margaritan skunk is 100% distin-
guishable from its nearby mainland congener, not only
morphometrically but also to the naked eye, is remarkable.
In fact, the differences in raw measurements, and PCA
and LDA scores, between both forms are similar or greater
to those found in interspecific comparisons within genera
of several carnivoran families (e.g., Van Valkenburgh &
Wayne 1994, Taylor & Matheson 1999, Bertrand & Mor-
isot 2012, Bornholdt et al. 2013, Srinivas & Jhala 2021).
We are assigning the subspecies rank to this skunk based
on the findings that it is highly diagnosable, and that its
morphometric differences are large; we are not assigning
the species rank because such differences involve mainly
size (Molinari 2023b).
Middle-sized carnivorans often evolve dwarfism after
colonizing islands, either by over-water dispersal or gla-
cial-period vicariance. Examples include insular endemics
such as the Channel Islands fox, Urocyon littoralis (Baird,
1858) (Coonan et al. 2010), the Pygmy Raccoon, Procyon
pygmaeus Merriam, 1901 (Villa-Meza et al. 2011), the
Dwarf Coati, Nasua nelsoni Merriam, 1901 (Cuarón et al.
2009), the Cozumel Island fox, genus Urocyon Baird, 1857
(Gompper et al. 2006), and the Margaritan Hog-nosed
Skunk, C. s. elieceri (this study). The latter is the first known
case of insular dwarfism in extant skunks: a dwarfed fossil
skunk, Promephitis majori Pilgrim, 1933, has been found
on the Greek island of Samos (Pilgrim 1933). Exceptions
include the Tres Marías Raccoon, Procyon lotor insularis
Merriam, 1898 (Wilson 1991), and the extinct Falkland
Islands wolf, Dusicyon australis (Kerr, 1792) (Lyras et al.
2010), which are (or were) similar-sized to their mainland
ancestors. Thus, the Margaritan skunk is not alone among
carnivorans in following the island rule (Foster 1964, Van
Valen 1973).
The suite of morphological, functional, and behavioral
changes often observed in insular organisms has been re-
ferred to as the ‘island syndrome’ (Adler & Levins 1994,
Baeckens & Damme 2020). These changes may involve
traits such as body size (considered by the island rule),
cranial shape, limb proportions, coloration, diet, and es-
cape behavior (Adler & Levins 1994, Sánchez-Villagra et
al. 2016, van der Geer 2019, Baeckens & Damme 2020).
To some degree, C. s. elieceri also complies with the island
syndrome since it differs from C. s. semistriatus in a cranial
shape parameter (rostral angle) that could not be quanti-
fied using linear measurements.
Along the last 500,000 years, on four times global sea
level was more than 100 m below the present level: about
20,000 (Last Glacial Maximum), 140,000 (Penultimate
Glacial Period), 250,000, and 340,000 YBP (Waelbroeck
et al. 2002, Siddall et al. 2007, Lichter et al. 2010). At such
times, many continental-shelf islands were connected to
the mainland by land bridges, offering the opportunity
of colonization to mainland populations. Animals can be
expected to cross ecologically suitable land bridges every
time that they are formed, and pre-established insular
Molinari, Abarca-Medina & Rivas-Rodríguez36
populations can be expected to retain at least part of their
acquired differences even if genetic exchange takes place
with more recent immigrants. Thus insular animals that
are classifiable as endemic species or subspecies, such as the
Margaritan skunk, are likely to have colonized the islands
and started their separate evolution earlier than the Last
Glacial Maximum. Examples exist for insular mammals
of the Caribbean: the Pygmy Three-toed Sloth, Bradypus
pygmaeus Anderson & Handley, 2001, from the Escudo de
Veraguas Island (Panama) was estimated to have diverged
at least 4.3 Mya (Ruiz-García et al. 2017); the pygmy rac-
coon and the dwarf coati from Cozumel Island (Mexico)
were estimated to have been isolated since 50,000 YBP, and
the fox from this island ‘for a minimum of 5000–13000
years, and perhaps far longer’ (McFadden 2004, Gompper
et al. 2006); as already noted, the Margaritan White-tailed
Deer was estimated to have diverged at least 118,000 YBP
(Moscarella 2001, Molinari 2007).
As areas of endemism, the Margarita and Cozumel Is-
lands show similarities. Both have about the same number
of endemic species and subspecies, and are at a similar dis-
tances from the mainland. However, Margarita is twice as
large (1,071 opposed to 478 km2
); and with mountains as
high as 760 m (Cerro Macanao to the west) and 960 m
(Cerro El Copey to the east) it has a greater environmental
diversity than Cozumel, which has a maximum elevation
of about 15 m. Margarita has a more complex biogeo-
graphic history: it has been emerged, intermittently con-
nected to the mainland during glacial periods, for more
than one million years, whereas Cozumel was under wa-
ter about 120,000 YBP; relatively to the mainland, Mar-
garita was from 20 to 50 km to the west 1 Mya, whereas
Cozumel has not shifted its position over the last 200,000
years (Gompper et al. 2006, Molinari 2007). More atten-
tion needs to be given to the protection of the Margaritan
flora and fauna. The conservation status of the Margaritan
skunk is undetermined, but the abundant material of the
subspecies in museum collections suggests that it is com-
mon, that it suffers a high mortality rate, or both.
ACKNOWLEDGEMENTS
We are grateful to the museum staff that facilitated
examination of specimens under their care, namely Rob-
ert Voss and Nancy Simmons (AMNH), Paula Jenkins
and Roberto Portela Miguez (BMNH), Pascual Soriano
and Johnny Murillo (CVULA), the late Francisco Bisbal,
and Javier Sánchez (EBRG), Bruce Patterson (FMNH),
Yaneth Muñoz-Saba (ICN), Mercedes Salazar-Candelle
and Carmen Ferreira-Marques (MBUCV), Alexis Araujo
(MCNG), Luis Albuja (MEPN), Jacques Cuisin, Chris-
tiane Denys, Jean-Marc Pons, Anne Previato, and Géral-
dine Veron (MNHN), Santiago F. Burneo and M. Alejan-
dra Camacho (QCAZ), and Michael D. Carleton, Nicole
Edmison, Alfred L. Gardner, Linda K. Gordon, Esther
Langan, and Darrin Lunde (USNM). Verónica Vargas
Muñoz (UV) sent photographs of two study skins and one
skull of Colombian specimens. Guilherme Siniciato Terra
Garbino (MZUFV) sent photographs and measurements
of the skulls of two Brazilian specimens.
REFERENCES
Adler, G. H. & R. Levins. 1994. The island syndrome in ro-
dent populations. Quarterly Review of Biology 69: 473–490.
https://doi.org/10.1086/418744
ASM [American Society of Mammalogists]. 2024. Mammal
diversity database, version 1.12.1. https://doi.org/10.5281/
zenodo.10595931
Baeckens, S. & R. Damme. 2020. The island syndrome. Cur-
rent Biology 30: 338–339. https://doi.org/10.1016/j.
cub.2020.03.029
Belant, J. L., J. Schipper & J. Conroy. 2009. The conservation
status of small carnivores in the Americas. Small Carnivore
Conservation 41: 3–8.
Bertrand, A. S. & A. Morisot. 2012. Neotropical spotted cat
species discrimination using morphometrics. Natureza
& Conservação 10: 40–44. https://doi.org/10.4322/nat-
con.2012.007
Bisbal, F. J. 1983. Dos nuevos mamíferos para la Isla de Margari-
ta, Venezuela. Acta Científica Venezolana 34: 366–367.
Boher-Bentti, S. & G. A. Cordero-Rodríguez. 2000. Distribu-
tion of brown capuchin monkeys (Cebus apella) in Venezu-
ela: a piece of the puzzle. Neotropical Primates 8: 152–153.
Boher-Bentti, S., M. Salazar-Candelle & C. Ferreira-Marques.
2023. Mamíferos de Venezuela: lista actualizada 2023 y
comentarios taxonómicos. Anartia 36: 7–35. https://doi.
org/10.5281/zenodo.10433912
Bornholdt, R., K. Helgen, K. P. Koepfli, L. Oliveira, M. Lucherini
& E. Eizirik. 2013. Taxonomic revision of the genus Galictis
(Carnivora: Mustelidae): Species delimitation, morphologi-
cal diagnosis, and refined mapping of geographical distribu-
tion. Zoological Journal of the Linnean Society 167: 449–472.
https://doi.org/10.1111/j.1096-3642.2012.00859.x
Cabrera, A. 1958. Catálogo de los mamíferos de América del
Sur. Revista del Museo Argentino de Ciencias Naturales “Ber-
nardino Rivadavia”, Ciencias Zoológicas 4: 1–308.
Castillo, D. F. & N. C. Caruso. 2024. Potential distribution
and conservation of the hog-nosed skunk (genus Conepatus,
Mammalia: Mephitidae). Journal for Nature Conservation
77: 126519. https://doi.org/10.1016/j.jnc.2023.126519
Coonan, T. J., C. A. Schwemm & D. K. Garcelon. 2010. Decline
and recovery of the island fox - a case study for population re-
covery. Cambridge, United Kingdom: Cambridge University
Press, xi + 212 p.
populations can be expected to retain at least part of their
acquired differences even if genetic exchange takes place
with more recent immigrants. Thus insular animals that
are classifiable as endemic species or subspecies, such as the
Margaritan skunk, are likely to have colonized the islands
and started their separate evolution earlier than the Last
Glacial Maximum. Examples exist for insular mammals
of the Caribbean: the Pygmy Three-toed Sloth, Bradypus
pygmaeus Anderson & Handley, 2001, from the Escudo de
Veraguas Island (Panama) was estimated to have diverged
at least 4.3 Mya (Ruiz-García et al. 2017); the pygmy rac-
coon and the dwarf coati from Cozumel Island (Mexico)
were estimated to have been isolated since 50,000 YBP, and
the fox from this island ‘for a minimum of 5000–13000
years, and perhaps far longer’ (McFadden 2004, Gompper
et al. 2006); as already noted, the Margaritan White-tailed
Deer was estimated to have diverged at least 118,000 YBP
(Moscarella 2001, Molinari 2007).
As areas of endemism, the Margarita and Cozumel Is-
lands show similarities. Both have about the same number
of endemic species and subspecies, and are at a similar dis-
tances from the mainland. However, Margarita is twice as
large (1,071 opposed to 478 km2
); and with mountains as
high as 760 m (Cerro Macanao to the west) and 960 m
(Cerro El Copey to the east) it has a greater environmental
diversity than Cozumel, which has a maximum elevation
of about 15 m. Margarita has a more complex biogeo-
graphic history: it has been emerged, intermittently con-
nected to the mainland during glacial periods, for more
than one million years, whereas Cozumel was under wa-
ter about 120,000 YBP; relatively to the mainland, Mar-
garita was from 20 to 50 km to the west 1 Mya, whereas
Cozumel has not shifted its position over the last 200,000
years (Gompper et al. 2006, Molinari 2007). More atten-
tion needs to be given to the protection of the Margaritan
flora and fauna. The conservation status of the Margaritan
skunk is undetermined, but the abundant material of the
subspecies in museum collections suggests that it is com-
mon, that it suffers a high mortality rate, or both.
ACKNOWLEDGEMENTS
We are grateful to the museum staff that facilitated
examination of specimens under their care, namely Rob-
ert Voss and Nancy Simmons (AMNH), Paula Jenkins
and Roberto Portela Miguez (BMNH), Pascual Soriano
and Johnny Murillo (CVULA), the late Francisco Bisbal,
and Javier Sánchez (EBRG), Bruce Patterson (FMNH),
Yaneth Muñoz-Saba (ICN), Mercedes Salazar-Candelle
and Carmen Ferreira-Marques (MBUCV), Alexis Araujo
(MCNG), Luis Albuja (MEPN), Jacques Cuisin, Chris-
tiane Denys, Jean-Marc Pons, Anne Previato, and Géral-
dine Veron (MNHN), Santiago F. Burneo and M. Alejan-
dra Camacho (QCAZ), and Michael D. Carleton, Nicole
Edmison, Alfred L. Gardner, Linda K. Gordon, Esther
Langan, and Darrin Lunde (USNM). Verónica Vargas
Muñoz (UV) sent photographs of two study skins and one
skull of Colombian specimens. Guilherme Siniciato Terra
Garbino (MZUFV) sent photographs and measurements
of the skulls of two Brazilian specimens.
REFERENCES
Adler, G. H. & R. Levins. 1994. The island syndrome in ro-
dent populations. Quarterly Review of Biology 69: 473–490.
https://doi.org/10.1086/418744
ASM [American Society of Mammalogists]. 2024. Mammal
diversity database, version 1.12.1. https://doi.org/10.5281/
zenodo.10595931
Baeckens, S. & R. Damme. 2020. The island syndrome. Cur-
rent Biology 30: 338–339. https://doi.org/10.1016/j.
cub.2020.03.029
Belant, J. L., J. Schipper & J. Conroy. 2009. The conservation
status of small carnivores in the Americas. Small Carnivore
Conservation 41: 3–8.
Bertrand, A. S. & A. Morisot. 2012. Neotropical spotted cat
species discrimination using morphometrics. Natureza
& Conservação 10: 40–44. https://doi.org/10.4322/nat-
con.2012.007
Bisbal, F. J. 1983. Dos nuevos mamíferos para la Isla de Margari-
ta, Venezuela. Acta Científica Venezolana 34: 366–367.
Boher-Bentti, S. & G. A. Cordero-Rodríguez. 2000. Distribu-
tion of brown capuchin monkeys (Cebus apella) in Venezu-
ela: a piece of the puzzle. Neotropical Primates 8: 152–153.
Boher-Bentti, S., M. Salazar-Candelle & C. Ferreira-Marques.
2023. Mamíferos de Venezuela: lista actualizada 2023 y
comentarios taxonómicos. Anartia 36: 7–35. https://doi.
org/10.5281/zenodo.10433912
Bornholdt, R., K. Helgen, K. P. Koepfli, L. Oliveira, M. Lucherini
& E. Eizirik. 2013. Taxonomic revision of the genus Galictis
(Carnivora: Mustelidae): Species delimitation, morphologi-
cal diagnosis, and refined mapping of geographical distribu-
tion. Zoological Journal of the Linnean Society 167: 449–472.
https://doi.org/10.1111/j.1096-3642.2012.00859.x
Cabrera, A. 1958. Catálogo de los mamíferos de América del
Sur. Revista del Museo Argentino de Ciencias Naturales “Ber-
nardino Rivadavia”, Ciencias Zoológicas 4: 1–308.
Castillo, D. F. & N. C. Caruso. 2024. Potential distribution
and conservation of the hog-nosed skunk (genus Conepatus,
Mammalia: Mephitidae). Journal for Nature Conservation
77: 126519. https://doi.org/10.1016/j.jnc.2023.126519
Coonan, T. J., C. A. Schwemm & D. K. Garcelon. 2010. Decline
and recovery of the island fox - a case study for population re-
covery. Cambridge, United Kingdom: Cambridge University
Press, xi + 212 p.
New subespecies of Conepatus semistriatus37
Cuarón, A. D., D. Valenzuela-Galván, D. García-Vasco, M.
E. Copa, S. Bautista, H. Mena, D. Martínez-Godínez, C.
González-Baca, L. A. Bojórquez-Tapia, L. Barraza, P. C. De
Grammont, F. Galindo-Maldonado, M. A. Martínez-Mo-
rales, E. Vázquez-Domínguez, E, Andresen, J. Benítez-Malvi-
do, D. Pérez-Salicrup, K. W. McFadden & M. E. Gompper.
2009. Conservation of the endemic dwarf carnivores of
Cozumel Island, Mexico. Small Carnivore Conservation 41:
15–21.
Dragoo, J. W., R. L. Honeycutt & D. J. Schmidly. 2003. Tax-
onomic status of white-backed hog-nosed skunks, genus
Conepatus (Carnivora: Mephitidae). Journal of Mam-
malogy 84: 159–176. https://doi.org/10.1644/1545-
1542(2003)084%3C0159:TSOWBH%3E2.0.CO;2
Dragoo, J. W. & S. R. Sheffield. 2009. Conepatus leuconotus
(Carnivora: Mephitidae). Mammalian Species 827: 1–8.
https://doi.org/10.1644/827.1
Emmons, L. H. 2005. A revision of the genera of arboreal Echi-
myidae (Rodentia: Echimyidae, Echimyinae), with descrip-
tions of two new genera. pp. 247–310. In: Lacey, E. A. & P.
Myers (eds). Mammalian diversification: from chromosomes
to phylogeography. Berkeley, USA: University of California
Publications in Zoology. https://doi.org/10.1525/califor-
nia/9780520098534.003.0009
Ferguson, A. W., R. E. Strauss & R. C. Dowler. 2022. Beyond
black and white: Addressing colour variation in the context
of local environmental conditions for the aposematic North
American Hog-nosed skunk. pp. 107–130. In: Do Linh San,
E., J. J. Sato, J. L. Belant & M. J. Somers (eds). Small carni-
vores: Evolution, ecology, behaviour, and conservation. New
York, USA: Wiley.
Floyd, C. H., D. H. Van Vuren, K. R. Crooks, K. L. Jones, D. K.
Garcelon, N. M. Belfiore, J. W. Dragoo & B. May. 2011. Ge-
netic differentiation of island spotted skunks, Spilogale graci-
lis amphiala. Journal of Mammalogy 92: 148–158. https://
doi.org/10.1644/09-MAMM-A-204.1
Foster, J. B. 1964. Evolution of mammals on islands. Nature 202:
234–235. https://doi.org/10.1038/202234a0
García‐Perea, R. 1996. Patterns of postnatal development in
skulls of lynxes, genus Lynx (Mammalia: Carnivora). Jour-
nal of Morphology 229: 241–254. https://doi.org/10.1002/
( S I C I ) 1 0 9 7 - 4 6 8 7 ( 1 9 9 6 0 9 ) 2 2 9 : 3 % 3 C 2 4 1 : : A I D -
JMOR1%3E3.0.CO;2-1
Gompper, M. E., A. E. Petrites & R. L. Lyman. 2006. Cozumel
Island fox (Urocyon sp.) dwarfism and possible divergence
history based on subfossil bones. Journal of Zoology 270:
72–77. https://doi.org/10.1111/j.1469-7998.2006.00119.x
Groves, C. P. 2001. Primate Taxonomy. Washington, USA:
Smithsonian Institution Press, 350 pp.
Gutiérrez, E. E. & J. Molinari. 2008. Morphometrics and tax-
onomy of bats of the genus Pteronotus (subgenus Phyllodia)
in Venezuela. Journal of Mammalogy 89: 292–305. https://
doi.org/10.1644/06-MAMM-A-452R.1
Hall, E. R. 1981. The mammals of North America, 2nd ed. New
York, USA: Wiley, xv + vi + 1181 pp.
Hammer, Ø. 2024. PAST, Paleontological Statistics, Version
4.16, Reference manual. Oslo, Norway: University of Oslo.
Hernández-Sánchez, A., A. Santos-Moreno & G. Pérez-Irineo.
2022. The Mephitidae in the Americas: A review of the cur-
rent state of knowledge and future research priorities. Mam-
malian Biology 102: 307–320. https://doi.org/10.1007/
s42991-022-00249-z
Hershkovitz, P. 1959. Nomenclature and taxonomy of the Neo-
tropical mammals described by Olfers, 1818. Journal of Mam-
malogy 40: 337–353. https://doi.org/10.2307/1376558
Jolicoeur, P. & J. R. Mosimann. 1960. Size and shape variation in
the painted turtle: A principal component analysis. Growth
24: 339–354.
Lichter, M., D. Zviely, M. Klein & D. Sivan. 2010. Sea-level
changes in the Mediterranean: Past, present, and future –a
review. pp. 3–17. In: Seckbach, J., R. Einav & A. Israel (eds).
Seaweeds and their role in globally changing environments (cel-
lular origin, life in extreme habitats and astrobiology). Dordre-
cht, Netheerlands: Springer. https://doi.org/10.1007/978-
90-481-8569-6_1
Linares, O. J. 1998. Mamíferos de Venezuela. Caracas, Venezuela:
Sociedad Conservacionista Audubon de Venezuela, 691 pp.
Lyras, G. A., A. A. Van Der Geer & L. Rook. 2010. Body size
of insular carnivores: evidence from the fossil record. Journal
of Biogeography 37: 1007–1021. https://doi.org/10.1111/
j.1365-2699.2010.02312.x
McDonough, M. M., A. W. Ferguson, R. C. Dowler, M. E.
Gompper & J. E. Maldonado. 2022. Phylogenomic systemat-
ics of the spotted skunks (Carnivora, Mephitidae, Spilogale):
Additional species diversity and Pleistocene climate change
as a major driver of diversification. Molecular Phylogenet-
ics and Evolution 167: 107266. https://doi.org/10.1016/j.
ympev.2021.107266
McFadden, K. W. 2004. The ecology, evolution, and natural his-
tory of the dwarf carnivores of Cozumel Island, Mexico. New
York, USA: Columbia University, 148 pp. [PhD Thesis]
McGarigal, K., S. Cushman & S. Stafford. 2000. Multivariate
statistics for wildlife and ecology research. New York, USA:
Springer, xiii + 283 pp.
Mead, R. A. 1967. Age determination in the spotted skunk.
Journal of Mammalogy 48: 606–616. https://doi.
org/10.2307/1377584
Meiri, S., T. Dayan & D. Simberloff. 2004. Body size of insular
carnivores: Little support for the island rule. American Natu-
ralist 163: 469-479. https://doi.org/10.1086/382229
Meza-Joya, F. L., E. Ramos, F. Cediel, V. Martínez-Arias, J.
Colmenares & D. Cardona. 2018. Predicted distribu-
tions of two poorly known small carnivores in Colombia:
The greater grison and striped hog-nosed skunk. Mastozo-
ología Neotropical 25: 89–105. https://doi.org/10.31687/
saremMN.18.25.1.0.09
Miller, A. 2015. Skunk. London, United Kingdom: Reaktion
Books, 199 pp.
Molinari, J. 2007. Variación geográfica en los venados de cola
blanca (Cervidae, Odocoileus) de Venezuela, con énfasis en
Cuarón, A. D., D. Valenzuela-Galván, D. García-Vasco, M.
E. Copa, S. Bautista, H. Mena, D. Martínez-Godínez, C.
González-Baca, L. A. Bojórquez-Tapia, L. Barraza, P. C. De
Grammont, F. Galindo-Maldonado, M. A. Martínez-Mo-
rales, E. Vázquez-Domínguez, E, Andresen, J. Benítez-Malvi-
do, D. Pérez-Salicrup, K. W. McFadden & M. E. Gompper.
2009. Conservation of the endemic dwarf carnivores of
Cozumel Island, Mexico. Small Carnivore Conservation 41:
15–21.
Dragoo, J. W., R. L. Honeycutt & D. J. Schmidly. 2003. Tax-
onomic status of white-backed hog-nosed skunks, genus
Conepatus (Carnivora: Mephitidae). Journal of Mam-
malogy 84: 159–176. https://doi.org/10.1644/1545-
1542(2003)084%3C0159:TSOWBH%3E2.0.CO;2
Dragoo, J. W. & S. R. Sheffield. 2009. Conepatus leuconotus
(Carnivora: Mephitidae). Mammalian Species 827: 1–8.
https://doi.org/10.1644/827.1
Emmons, L. H. 2005. A revision of the genera of arboreal Echi-
myidae (Rodentia: Echimyidae, Echimyinae), with descrip-
tions of two new genera. pp. 247–310. In: Lacey, E. A. & P.
Myers (eds). Mammalian diversification: from chromosomes
to phylogeography. Berkeley, USA: University of California
Publications in Zoology. https://doi.org/10.1525/califor-
nia/9780520098534.003.0009
Ferguson, A. W., R. E. Strauss & R. C. Dowler. 2022. Beyond
black and white: Addressing colour variation in the context
of local environmental conditions for the aposematic North
American Hog-nosed skunk. pp. 107–130. In: Do Linh San,
E., J. J. Sato, J. L. Belant & M. J. Somers (eds). Small carni-
vores: Evolution, ecology, behaviour, and conservation. New
York, USA: Wiley.
Floyd, C. H., D. H. Van Vuren, K. R. Crooks, K. L. Jones, D. K.
Garcelon, N. M. Belfiore, J. W. Dragoo & B. May. 2011. Ge-
netic differentiation of island spotted skunks, Spilogale graci-
lis amphiala. Journal of Mammalogy 92: 148–158. https://
doi.org/10.1644/09-MAMM-A-204.1
Foster, J. B. 1964. Evolution of mammals on islands. Nature 202:
234–235. https://doi.org/10.1038/202234a0
García‐Perea, R. 1996. Patterns of postnatal development in
skulls of lynxes, genus Lynx (Mammalia: Carnivora). Jour-
nal of Morphology 229: 241–254. https://doi.org/10.1002/
( S I C I ) 1 0 9 7 - 4 6 8 7 ( 1 9 9 6 0 9 ) 2 2 9 : 3 % 3 C 2 4 1 : : A I D -
JMOR1%3E3.0.CO;2-1
Gompper, M. E., A. E. Petrites & R. L. Lyman. 2006. Cozumel
Island fox (Urocyon sp.) dwarfism and possible divergence
history based on subfossil bones. Journal of Zoology 270:
72–77. https://doi.org/10.1111/j.1469-7998.2006.00119.x
Groves, C. P. 2001. Primate Taxonomy. Washington, USA:
Smithsonian Institution Press, 350 pp.
Gutiérrez, E. E. & J. Molinari. 2008. Morphometrics and tax-
onomy of bats of the genus Pteronotus (subgenus Phyllodia)
in Venezuela. Journal of Mammalogy 89: 292–305. https://
doi.org/10.1644/06-MAMM-A-452R.1
Hall, E. R. 1981. The mammals of North America, 2nd ed. New
York, USA: Wiley, xv + vi + 1181 pp.
Hammer, Ø. 2024. PAST, Paleontological Statistics, Version
4.16, Reference manual. Oslo, Norway: University of Oslo.
Hernández-Sánchez, A., A. Santos-Moreno & G. Pérez-Irineo.
2022. The Mephitidae in the Americas: A review of the cur-
rent state of knowledge and future research priorities. Mam-
malian Biology 102: 307–320. https://doi.org/10.1007/
s42991-022-00249-z
Hershkovitz, P. 1959. Nomenclature and taxonomy of the Neo-
tropical mammals described by Olfers, 1818. Journal of Mam-
malogy 40: 337–353. https://doi.org/10.2307/1376558
Jolicoeur, P. & J. R. Mosimann. 1960. Size and shape variation in
the painted turtle: A principal component analysis. Growth
24: 339–354.
Lichter, M., D. Zviely, M. Klein & D. Sivan. 2010. Sea-level
changes in the Mediterranean: Past, present, and future –a
review. pp. 3–17. In: Seckbach, J., R. Einav & A. Israel (eds).
Seaweeds and their role in globally changing environments (cel-
lular origin, life in extreme habitats and astrobiology). Dordre-
cht, Netheerlands: Springer. https://doi.org/10.1007/978-
90-481-8569-6_1
Linares, O. J. 1998. Mamíferos de Venezuela. Caracas, Venezuela:
Sociedad Conservacionista Audubon de Venezuela, 691 pp.
Lyras, G. A., A. A. Van Der Geer & L. Rook. 2010. Body size
of insular carnivores: evidence from the fossil record. Journal
of Biogeography 37: 1007–1021. https://doi.org/10.1111/
j.1365-2699.2010.02312.x
McDonough, M. M., A. W. Ferguson, R. C. Dowler, M. E.
Gompper & J. E. Maldonado. 2022. Phylogenomic systemat-
ics of the spotted skunks (Carnivora, Mephitidae, Spilogale):
Additional species diversity and Pleistocene climate change
as a major driver of diversification. Molecular Phylogenet-
ics and Evolution 167: 107266. https://doi.org/10.1016/j.
ympev.2021.107266
McFadden, K. W. 2004. The ecology, evolution, and natural his-
tory of the dwarf carnivores of Cozumel Island, Mexico. New
York, USA: Columbia University, 148 pp. [PhD Thesis]
McGarigal, K., S. Cushman & S. Stafford. 2000. Multivariate
statistics for wildlife and ecology research. New York, USA:
Springer, xiii + 283 pp.
Mead, R. A. 1967. Age determination in the spotted skunk.
Journal of Mammalogy 48: 606–616. https://doi.
org/10.2307/1377584
Meiri, S., T. Dayan & D. Simberloff. 2004. Body size of insular
carnivores: Little support for the island rule. American Natu-
ralist 163: 469-479. https://doi.org/10.1086/382229
Meza-Joya, F. L., E. Ramos, F. Cediel, V. Martínez-Arias, J.
Colmenares & D. Cardona. 2018. Predicted distribu-
tions of two poorly known small carnivores in Colombia:
The greater grison and striped hog-nosed skunk. Mastozo-
ología Neotropical 25: 89–105. https://doi.org/10.31687/
saremMN.18.25.1.0.09
Miller, A. 2015. Skunk. London, United Kingdom: Reaktion
Books, 199 pp.
Molinari, J. 2007. Variación geográfica en los venados de cola
blanca (Cervidae, Odocoileus) de Venezuela, con énfasis en
Molinari, Abarca-Medina & Rivas-Rodríguez38
O. margaritae, la especie enana de la Isla de Margarita. Me-
moria de la Fundación La Salle de Ciencias Naturales 167:
29–72.
Molinari, J. 2023a. A global assessment of the ‘island rule’ in bats
based on functionally distinct measures of body size. Journal
of Biogeography 50: 1179–1190. https://doi.org/10.1111/
jbi.14624
Molinari, J. 2023b. A bare-bones scheme to choose between
the species, subspecies, and ‘evolutionarily significant unit’
categories in taxonomy and conservation. Journal for Na-
ture Conservation 72: 126335. https://doi.org/10.1016/j.
jnc.2023.126335
Molinari, J., E. E. Gutiérrez & B. K. Lim. 2023. Systematics and
biogeography of Anoura cultrata (Mammalia, Chiroptera,
Phyllostomidae): A morphometric, niche modeling, and
genetic perspective, with a taxonomic reappraisal of the ge-
nus. Zootaxa 5297: 151–188. https://doi.org/10.11646/
zootaxa.5297.2.1
Moscarella, R. A. 2001. Filogeografía y genética de la conserva-
ción del venado caramerudo de Venezuela. Caracas, Venezuela:
Universidad Simón Bolívar, 87 pp. [MSc Thesis]
Pacheco, V., S. Diaz, L. Graham-Angeles, M. Flores-Quispe, G.
Calizaya-Mamani, D. Ruelas & P. Sánchez-Vendizú. 2021.
Lista actualizada de la diversidad de los mamíferos del Perú
y una propuesta para su actualización. Revista Peruana de
Biología 28: e21019. http://dx.doi.org/10.15381/rpb.
v28i4.21019
Patten, M. A., & P. Unitt. 2002. Diagnosability versus mean
differences of Sage Sparrow subspecies. Auk 119: 26–35.
https://doi.org/10.1093/auk/119.1.26
Pilgrim, G. E. 1933. A fossil skunk from Samos. American Mu-
seum Novitates 663: 1–15.
Rodrigues, M. L. F. 2013. História evolutiva de Conepatus (Car-
nivora: Mephitidae): padrões biogeográficos de diversificação,
investigação filogenética e revisão taxonómica do gênero. Por-
to Alegre, Brazil: Pontifícia Universidade Católica do Río
Grande do Sul, 162 pp. [PhD Thesis]
Rossi, R. V., R. S. Voss & D. P. Lunde. 2010. A revision of the
didelphid marsupial Genus Marmosa Part 1. The species in
Tate’s ‘mexicana’ and ‘mitis’ sections and other closely related
forms. Bulletin of the American Museum of Natural History
334: 1–83. https://doi.org/10.1206/334.1
Ruíz-García, M., D. Chacón, T. Plese, I. Schuler & J. M. Shos-
tell. 2017. Mitogenomics phylogenetic relationships of the
current sloth’s genera and species (Bradypodidae and Meg-
alonychidae). Mitochondrial DNA Part A 29: 281–299.
http://dx.doi.org/10.1080/24701394.2016.1275602
Sánchez-Villagra, M. R., Geiger, M., & R. A. Schneider. 2016.
The taming of the neural crest: a developmental perspective
on the origins of morphological covariation in domesticated
mammals. Royal Society Open Science 3: 160107. http://
dx.doi.org/10.1098/rsos.160107
Schiaffini, M. I., M. Gabrielli, F. J. Prevosti, Y. P. Cardoso, D.
Castillo, R. Bo, E. Casanave & M. Lizarralde. 2013. Taxo-
nomic status of southern South American Conepatus (Car-
nivora: Mephitidae). Zoological Journal of the Linnean Soci-
ety 167: 327–344. https://doi.org/10.1111/zoj.12006
Siddall, M., J. Chappell & E. K. Potter. 2006. Eustatic sea lev-
el during past interglacials. pp. 75–92. In: Sirocko, F., M.
Claussen, T. Litt & M. F. Sanchez-Goni (eds). The climate
of past interglacials. Amsterdam, The Netherlands: Elsevier.
https://doi.org/10.1016/S1571-0866(07)80032-7
Simpson, G. G. 1951. The species concept. Evolution 5: 285–
298. https://doi.org/10.1111/j.1558-5646.1951.tb02788.x
Smith, J. D. 1972. Systematics of the chiropteran family Mor-
moopidae. Miscellaneous Publication, Museum of Natural
History, University of Kansas 56: 1–132.
Smith, J. D. & H. H. Genoways. 1974. Bats of Margarita island,
Venezuela, with zoogeographic comments. Bulletin of the
Southern California of Sciences 73: 64–79.
Solari, S., Y. Muñoz-Saba, J. V. Rodríguez-Mahecha, T. R. De-
fler, H. E. Ramírez-Chaves & F. Trujillo. 2013. Riqueza,
endemismo y conservación de los mamíferos de Colombia.
Mastozoología Neotropical 20: 301–365.
Srinivas, Y. & Y. Jhala. 2021. Morphometric variation in wolves
and golden jackal in India (Mammalia, Carnivora). Biodi-
versity Data Journal 9: e67677. https://doi.org/10.3897/
BDJ.9.e67677
Taylor, M. E. & J. Matheson. 1999. A craniometric comparison
of the African and Asian mongooses in the genus Herpestes
(Carnivora: Herpestidae). Mammalia 63: 449–464. https://
doi.org/10.1515/mamm.1999.63.4.449
Teta, P., G. D’Elía, P. Jayat, G. S. Libardi, J. A. Oliveira, R.
Moratelli, A. R. Percequillo, J. R. Prado, P. E. Ortiz, N.
Hurtado, M. E. Schiaffini, E. F. Abreu, Jr., E. A. Chiquito,
A. L. Giménez & J. Torres. 2020. On the distinction and
availability of the new taxa proposed by Agnolin et al.,
(2019). Mastozoología Neotropical 27: 155–171. https://doi.
org/10.31687/saremMN.20.27.1.0.20
Tirira, D. G., J. Brito, S. F. Burneo, C. M. Pinto & J. A. Salas.
2023. Mamíferos del Ecuador: lista oficial actualizada de espe-
cies. Quito, Ecuador: Asociación Ecuatoriana de Mastozoo-
logía, 83 pp.
van der Geer, A. A. E. 2019. Effect of isolation on coat colour
polymorphism of Polynesian rats in Island Southeast Asia
and the Pacific. PeerJ 7:e6894. http://doi.org/10.7717/
peerj.6894
Van Gelder, R. G. 1959. A taxonomic revision of the spotted
skunks (genus Spilogale). Bulletin of the American Museum of
Natural History 117: 229–392.
Van Gelder, R. G. 1968. The genus Conepatus (Mammalia, Mus-
telidae): variation within a population. American Museum
Novitates 2322: 1–37.
Van Valen, L. 1973. Body size and numbers of plants
and animals. Evolution 27: 27–35. https://doi.
org/10.1111/j.1558-5646.1973.tb059 14.x
Van Valkenburgh, B. & R. K. Wayne. 1994. Shape diver-
gence associated with size convergence in sympatric East
African jackals. Ecology 75: 1567–1581. https://doi.
org/10.2307/1939618
O. margaritae, la especie enana de la Isla de Margarita. Me-
moria de la Fundación La Salle de Ciencias Naturales 167:
29–72.
Molinari, J. 2023a. A global assessment of the ‘island rule’ in bats
based on functionally distinct measures of body size. Journal
of Biogeography 50: 1179–1190. https://doi.org/10.1111/
jbi.14624
Molinari, J. 2023b. A bare-bones scheme to choose between
the species, subspecies, and ‘evolutionarily significant unit’
categories in taxonomy and conservation. Journal for Na-
ture Conservation 72: 126335. https://doi.org/10.1016/j.
jnc.2023.126335
Molinari, J., E. E. Gutiérrez & B. K. Lim. 2023. Systematics and
biogeography of Anoura cultrata (Mammalia, Chiroptera,
Phyllostomidae): A morphometric, niche modeling, and
genetic perspective, with a taxonomic reappraisal of the ge-
nus. Zootaxa 5297: 151–188. https://doi.org/10.11646/
zootaxa.5297.2.1
Moscarella, R. A. 2001. Filogeografía y genética de la conserva-
ción del venado caramerudo de Venezuela. Caracas, Venezuela:
Universidad Simón Bolívar, 87 pp. [MSc Thesis]
Pacheco, V., S. Diaz, L. Graham-Angeles, M. Flores-Quispe, G.
Calizaya-Mamani, D. Ruelas & P. Sánchez-Vendizú. 2021.
Lista actualizada de la diversidad de los mamíferos del Perú
y una propuesta para su actualización. Revista Peruana de
Biología 28: e21019. http://dx.doi.org/10.15381/rpb.
v28i4.21019
Patten, M. A., & P. Unitt. 2002. Diagnosability versus mean
differences of Sage Sparrow subspecies. Auk 119: 26–35.
https://doi.org/10.1093/auk/119.1.26
Pilgrim, G. E. 1933. A fossil skunk from Samos. American Mu-
seum Novitates 663: 1–15.
Rodrigues, M. L. F. 2013. História evolutiva de Conepatus (Car-
nivora: Mephitidae): padrões biogeográficos de diversificação,
investigação filogenética e revisão taxonómica do gênero. Por-
to Alegre, Brazil: Pontifícia Universidade Católica do Río
Grande do Sul, 162 pp. [PhD Thesis]
Rossi, R. V., R. S. Voss & D. P. Lunde. 2010. A revision of the
didelphid marsupial Genus Marmosa Part 1. The species in
Tate’s ‘mexicana’ and ‘mitis’ sections and other closely related
forms. Bulletin of the American Museum of Natural History
334: 1–83. https://doi.org/10.1206/334.1
Ruíz-García, M., D. Chacón, T. Plese, I. Schuler & J. M. Shos-
tell. 2017. Mitogenomics phylogenetic relationships of the
current sloth’s genera and species (Bradypodidae and Meg-
alonychidae). Mitochondrial DNA Part A 29: 281–299.
http://dx.doi.org/10.1080/24701394.2016.1275602
Sánchez-Villagra, M. R., Geiger, M., & R. A. Schneider. 2016.
The taming of the neural crest: a developmental perspective
on the origins of morphological covariation in domesticated
mammals. Royal Society Open Science 3: 160107. http://
dx.doi.org/10.1098/rsos.160107
Schiaffini, M. I., M. Gabrielli, F. J. Prevosti, Y. P. Cardoso, D.
Castillo, R. Bo, E. Casanave & M. Lizarralde. 2013. Taxo-
nomic status of southern South American Conepatus (Car-
nivora: Mephitidae). Zoological Journal of the Linnean Soci-
ety 167: 327–344. https://doi.org/10.1111/zoj.12006
Siddall, M., J. Chappell & E. K. Potter. 2006. Eustatic sea lev-
el during past interglacials. pp. 75–92. In: Sirocko, F., M.
Claussen, T. Litt & M. F. Sanchez-Goni (eds). The climate
of past interglacials. Amsterdam, The Netherlands: Elsevier.
https://doi.org/10.1016/S1571-0866(07)80032-7
Simpson, G. G. 1951. The species concept. Evolution 5: 285–
298. https://doi.org/10.1111/j.1558-5646.1951.tb02788.x
Smith, J. D. 1972. Systematics of the chiropteran family Mor-
moopidae. Miscellaneous Publication, Museum of Natural
History, University of Kansas 56: 1–132.
Smith, J. D. & H. H. Genoways. 1974. Bats of Margarita island,
Venezuela, with zoogeographic comments. Bulletin of the
Southern California of Sciences 73: 64–79.
Solari, S., Y. Muñoz-Saba, J. V. Rodríguez-Mahecha, T. R. De-
fler, H. E. Ramírez-Chaves & F. Trujillo. 2013. Riqueza,
endemismo y conservación de los mamíferos de Colombia.
Mastozoología Neotropical 20: 301–365.
Srinivas, Y. & Y. Jhala. 2021. Morphometric variation in wolves
and golden jackal in India (Mammalia, Carnivora). Biodi-
versity Data Journal 9: e67677. https://doi.org/10.3897/
BDJ.9.e67677
Taylor, M. E. & J. Matheson. 1999. A craniometric comparison
of the African and Asian mongooses in the genus Herpestes
(Carnivora: Herpestidae). Mammalia 63: 449–464. https://
doi.org/10.1515/mamm.1999.63.4.449
Teta, P., G. D’Elía, P. Jayat, G. S. Libardi, J. A. Oliveira, R.
Moratelli, A. R. Percequillo, J. R. Prado, P. E. Ortiz, N.
Hurtado, M. E. Schiaffini, E. F. Abreu, Jr., E. A. Chiquito,
A. L. Giménez & J. Torres. 2020. On the distinction and
availability of the new taxa proposed by Agnolin et al.,
(2019). Mastozoología Neotropical 27: 155–171. https://doi.
org/10.31687/saremMN.20.27.1.0.20
Tirira, D. G., J. Brito, S. F. Burneo, C. M. Pinto & J. A. Salas.
2023. Mamíferos del Ecuador: lista oficial actualizada de espe-
cies. Quito, Ecuador: Asociación Ecuatoriana de Mastozoo-
logía, 83 pp.
van der Geer, A. A. E. 2019. Effect of isolation on coat colour
polymorphism of Polynesian rats in Island Southeast Asia
and the Pacific. PeerJ 7:e6894. http://doi.org/10.7717/
peerj.6894
Van Gelder, R. G. 1959. A taxonomic revision of the spotted
skunks (genus Spilogale). Bulletin of the American Museum of
Natural History 117: 229–392.
Van Gelder, R. G. 1968. The genus Conepatus (Mammalia, Mus-
telidae): variation within a population. American Museum
Novitates 2322: 1–37.
Van Valen, L. 1973. Body size and numbers of plants
and animals. Evolution 27: 27–35. https://doi.
org/10.1111/j.1558-5646.1973.tb059 14.x
Van Valkenburgh, B. & R. K. Wayne. 1994. Shape diver-
gence associated with size convergence in sympatric East
African jackals. Ecology 75: 1567–1581. https://doi.
org/10.2307/1939618
New subespecies of Conepatus semistriatus39
Villa-Meza, A., R. Avila-Flores, A. D. Cuarón & D. Valenzuela-
Galván. 2011. Procyon pygmaeus (Carnivora: Procyonidae).
Mammalian Species 877: 87–93. https://doi.org/10.1644/877.1
Vivo, M. & A. P. Carmignotto. 2015. Family Sciuridae G.
Fischer, 1817. pp. 1–48. In: Patton, J. L., U. F. Pardiñas & G.
D’Elía (eds). Mammals of South America, volume 2. Rodents.
Chicago, USA: University of Chicago Press.
Waelbroeck, C., L. Labeyrie, E. Michel, J. C. Duplessy, J. F. Mc-
Manus, K. Lambeck, E. Balbon & M. Labracherie. 2002.
Sea-level and deep water temperature changes derived from
benthonic foraminifera isotopic records. Quaternary Sci-
ence Reviews 21: 295–305. https://doi.org/10.1016/S0277-
3791(01)00101-9
Wang, X. & Z. Qiu. 2004. Late Miocene Promephitis (Car-
nivora, Mephitidae) from China. Journal of Vertebrate Pa-
leontology 24: 721–731. https://doi.org/10.1671/0272-
4634(2004)024[0721:LMPCMF]2.0.CO;2
Wang, X., D. P. Whistler & G. T. Takeuchi. 2005. A new
basal skunk Martinogale (Carnivora, Mephitinae) from
late Miocene Dove Spring Formation, California, and
origin of new world mephitines. Journal of Vertebrate Pa-
leontology 25: 936–949. https://doi.org/10.1671/0272-
4634(2005)025[0936:ANBSMC]2.0.CO;2
Wilson, D. E. 1991. Mammals of the Tres Marías Islands. Bul-
letin of the American Museum of Natural History 206: 214–
250.
Wozencraft, W. C. 2005. Order Carnivora. pp. 532–562. In:
Wilson, D. E. & D. M. Reeder (eds) Mammal species of the
world: A taxonomic and geographic reference. Washington,
USA: Smithsonian Institution Press, 2142 pp.
Villa-Meza, A., R. Avila-Flores, A. D. Cuarón & D. Valenzuela-
Galván. 2011. Procyon pygmaeus (Carnivora: Procyonidae).
Mammalian Species 877: 87–93. https://doi.org/10.1644/877.1
Vivo, M. & A. P. Carmignotto. 2015. Family Sciuridae G.
Fischer, 1817. pp. 1–48. In: Patton, J. L., U. F. Pardiñas & G.
D’Elía (eds). Mammals of South America, volume 2. Rodents.
Chicago, USA: University of Chicago Press.
Waelbroeck, C., L. Labeyrie, E. Michel, J. C. Duplessy, J. F. Mc-
Manus, K. Lambeck, E. Balbon & M. Labracherie. 2002.
Sea-level and deep water temperature changes derived from
benthonic foraminifera isotopic records. Quaternary Sci-
ence Reviews 21: 295–305. https://doi.org/10.1016/S0277-
3791(01)00101-9
Wang, X. & Z. Qiu. 2004. Late Miocene Promephitis (Car-
nivora, Mephitidae) from China. Journal of Vertebrate Pa-
leontology 24: 721–731. https://doi.org/10.1671/0272-
4634(2004)024[0721:LMPCMF]2.0.CO;2
Wang, X., D. P. Whistler & G. T. Takeuchi. 2005. A new
basal skunk Martinogale (Carnivora, Mephitinae) from
late Miocene Dove Spring Formation, California, and
origin of new world mephitines. Journal of Vertebrate Pa-
leontology 25: 936–949. https://doi.org/10.1671/0272-
4634(2005)025[0936:ANBSMC]2.0.CO;2
Wilson, D. E. 1991. Mammals of the Tres Marías Islands. Bul-
letin of the American Museum of Natural History 206: 214–
250.
Wozencraft, W. C. 2005. Order Carnivora. pp. 532–562. In:
Wilson, D. E. & D. M. Reeder (eds) Mammal species of the
world: A taxonomic and geographic reference. Washington,
USA: Smithsonian Institution Press, 2142 pp.
Molinari, Abarca-Medina & Rivas-Rodríguez40
APPENDIX 1
Additional specimens of Conepatus examined
3.11.5.10, 3.11.5.11). Río Negro, Estancia Huanu-Luan
(FMNH 34193*); Huanuluan (MCZ 19110–19114);
Pichi Mahuida (BMNH 27.6.21.1–27.6.21.9); Pilca-
niyeu (BMNH 19.1.5.2, 19.1.5.22, 20.11.4.3); S shore
of Lake Nahuel Huapi (BMNH 3.11.5.9). Salta, Cachi
(BMNH 6.5.8.12–6.5.8.14, 6.5.8.16); Metán (La Caña-
da) (BMNH 34.11.4.14); Metán (La Represa) (BMNH
34.11.4.10–34.11.4.12, 34.11.4.8, 34.11.4.9). San Juan,
Pedernal (BMNH 21.6.19.2–21.6.19.4). Santa Cruz,
40 km SW Puerto Santa Cruz (AMNH 17446*); Estancia
Alta Vista, Lago Argentino (BMNH 28.12.11.8); Puer-
to Santa Cruz (CFA 9926); Río Chico, Departamento
Corpen Aike (AMNH 25669*); Río Gallegos (USNM
264479); Santa Cruz (BMNH 99.2.4.2); unknown loca-
lity (MNHN 1883-163*). Santa Fe, Esperanza (BMNH
1.2.4.5, 1.2.4.7); Las Rosas (BMNH 17.5.2.2–17.5.2.4);
San Cristóbal (BMNH 17.5.3.5). Santiago del Este-
ro, Clodomira (CFA 9459); Lavalle (AMNH 41530*,
41531*, 41532, 41533, 41534*); Roversi (BMNH
34.11.4.13); Villa La Punta (CFA 10803). Tucumán,
Tafi Viejo (AMNH 41529*). BOLIVIA: Cochabam-
ba, 32 km S Tiraque (USNM 271410*); 43 km ESE of
Iquisivi (BMNH 2.1.1.11*); Cochabamba (BMNH
2.1.1.15*); El Choro (BMNH 2.1.1.12–2.1.1.14); Tuj-
ma, near Mizque (AMNH 39011*). La Paz, 34 km NNE
Nevado Sajama (BMNH 98.3.16.4*); 5 km E Ulla Ulla
(AMNH 247712); Nevado Sajama (BMNH 3.2.9.2).
Oruro, Pampa Aulliaga (BMNH 2.2.2.11). Santa Cruz,
Comarapa (BMNH 34.9.2.53). Tarija, Carlazo (BMNH
26.1.1.2, 26.1.1.3); Tapecua, 13.5 k, WSW Palos Blancos
(AMNH 264464*). Santa Cruz, 5 km SE Tita (AMNH
260327*). BRAZIL: Bahia, Lamarão (BMNH 3.9.5.45,
3.9.5.46, 3.9.5.47*, 3.9.5.48). Goias, Anapolis (AMNH
133948*). Mato Grosso do Sul, Maracaju (AMNH
133946*). Minas Gerais, Arinos (MZUFV 3471*); Itape-
cerica (MZUFV 4422*); Rio Jordão (BMNH 1.11.3.24,
1.11.3.25*). Piauí, Central Piauí (MCZ 24828). Unk-
nown state (BMNH 68.a*). Rio Grande do Sul, Quinta
(AMNH 235512*, 235513*, AMNH 235514*); Santana
da Boa Vista (AMNH 235993*); Uruguaiana (AMNH
235994*). CHILE: Araucania, 2.5 km NNE Angol
(AMNH 93324); Maquehue, 9.5 km WSW Temuco
(AMNH 33290*, 33291*). Talca?, San Rafael? (BMNH
45.11.18.17). Arica y Parinacota, Parinacota (USNM
391849, 391850). Magallanes, Punta Arenas (AMNH
130053, 130068, 130095); Strait of Magellan (BMNH
Specimens are grouped according to current classifica-
tion (ASM 2024), which must be deemed provisional be-
cause the genus has never been reviewed. Adult specimens
whose skulls were examined are indicated with an asterisk.
All other specimens consist of immature skulls with or
without study skin, or study skins only.
C. chinga (n = 272).— ARGENTINA: Buenos Aires,
25 de Mayo (CFA 9752); Bonifacio (BMNH 17.9.15.3);
Bonifacio (Laguna Alsina) (BMNH 17.9.15.1*); Estación
Ombucta (BMNH 28.12.11.9); Estancia Los Angeles,
64 Mi SE Azul (USNM 331066); Estancia Los Ingleses,
near Mar de Ajó (BMNH 20.2.7.10, 20.2.7.8, 20.2.7.9,
9.12.1.16, 9.12.1.17, 9.12.1.18*); Gándara (MACN
29.912); Mar del Plata (BMNH 13.2.24.1, 16.10.3.6);
N of Necochea (USNM 172792*); near Henderson
(FMNH 24354*); San José Ranch, 32 km SSW San
Blas (USNM 171957, 171958, 172791). Catamarca,
‘Otro Cerro’, Sierra de Ambato, aprox. 35 km N Chum-
bicha (BMNH 19.2.7.1*, 19.2.7.2, 19.2.7.3*, 20.3.17.2);
Belén (BMNH 34.11.4.6, 34.11.4.7). Chubut, 100 km
NW Comodoro Rivadavia (AMNH 94317*); Barran-
cas Blancas, 5.5 km al SW Puerto Madryn (MNHN
1897-1244*); La Concepción (Laguna) (BMNH
28.12.11.6, 28.12.11.7); Lago Colhue Huapi (AMNH
94328, 94330*); Pico Salamanca (BMNH 28.12.11.3–
28.12.11.5); Rawson (MACN 28.72); Río Chubut
(BMNH 99.2.22.10); Sarmiento (AMNH 94329*); unk-
nown locality (BMNH 96.10.7.3); Valle del Lago Blan-
co (BMNH 3.7.9.13–3.7.9.15). Córdoba, Cruz del Eje
(BMNH 2.2.5.2, 2.2.5.3*, 2.2.5.4–2.2.5.6, 2.2.5.9); El
Carrizal, Villa Dolores (BMNH 17.6.29.5–17.6.29.7);
Noetinger (BMNH 17.1.25.8, 17.1.25.9); unknown loca-
lity (AMNH 36933). Corrientes, Manantiales (MACN
13.714). Entre Ríos, Paraná (AMNH 36932). Formosa,
‘Formosa, Paso de las Niñas, Río Teuco’ [39 km SSW In-
geniero Suárez, Río Teuco-Bermejo] (MACN 47.119);
Fortín Nuevo Pilcomayo (MACN 43.58). Jujuy, Alfar-
cito (BMNH 21.11.1.3); Humahuaca (MACN 26.182);
Maimara (BMNH 12.12.12.1*, 12.12.12.2). La Rioja,
Desiderio Tello (USNM 172793*, 172794). Mendoza,
Colonia Alvear (BMNH 10.9.12.2); Tupungato (BMNH
21.7.5.3*). Neuquén, Chos Malal (BMNH 26.10.11.1);
Collón Curá (BMNH 27.6.4.68); 16 km SE of La Rin-
conada (BMNH 27.5.1.1*); San Martín de los Andes
(BMNH 27.5.1.119); Sierra de Pil Pil (BMNH 27.5.1.2).
‘Patagonia’, unknown locality (BMNH 1899.2.41,
APPENDIX 1
Additional specimens of Conepatus examined
3.11.5.10, 3.11.5.11). Río Negro, Estancia Huanu-Luan
(FMNH 34193*); Huanuluan (MCZ 19110–19114);
Pichi Mahuida (BMNH 27.6.21.1–27.6.21.9); Pilca-
niyeu (BMNH 19.1.5.2, 19.1.5.22, 20.11.4.3); S shore
of Lake Nahuel Huapi (BMNH 3.11.5.9). Salta, Cachi
(BMNH 6.5.8.12–6.5.8.14, 6.5.8.16); Metán (La Caña-
da) (BMNH 34.11.4.14); Metán (La Represa) (BMNH
34.11.4.10–34.11.4.12, 34.11.4.8, 34.11.4.9). San Juan,
Pedernal (BMNH 21.6.19.2–21.6.19.4). Santa Cruz,
40 km SW Puerto Santa Cruz (AMNH 17446*); Estancia
Alta Vista, Lago Argentino (BMNH 28.12.11.8); Puer-
to Santa Cruz (CFA 9926); Río Chico, Departamento
Corpen Aike (AMNH 25669*); Río Gallegos (USNM
264479); Santa Cruz (BMNH 99.2.4.2); unknown loca-
lity (MNHN 1883-163*). Santa Fe, Esperanza (BMNH
1.2.4.5, 1.2.4.7); Las Rosas (BMNH 17.5.2.2–17.5.2.4);
San Cristóbal (BMNH 17.5.3.5). Santiago del Este-
ro, Clodomira (CFA 9459); Lavalle (AMNH 41530*,
41531*, 41532, 41533, 41534*); Roversi (BMNH
34.11.4.13); Villa La Punta (CFA 10803). Tucumán,
Tafi Viejo (AMNH 41529*). BOLIVIA: Cochabam-
ba, 32 km S Tiraque (USNM 271410*); 43 km ESE of
Iquisivi (BMNH 2.1.1.11*); Cochabamba (BMNH
2.1.1.15*); El Choro (BMNH 2.1.1.12–2.1.1.14); Tuj-
ma, near Mizque (AMNH 39011*). La Paz, 34 km NNE
Nevado Sajama (BMNH 98.3.16.4*); 5 km E Ulla Ulla
(AMNH 247712); Nevado Sajama (BMNH 3.2.9.2).
Oruro, Pampa Aulliaga (BMNH 2.2.2.11). Santa Cruz,
Comarapa (BMNH 34.9.2.53). Tarija, Carlazo (BMNH
26.1.1.2, 26.1.1.3); Tapecua, 13.5 k, WSW Palos Blancos
(AMNH 264464*). Santa Cruz, 5 km SE Tita (AMNH
260327*). BRAZIL: Bahia, Lamarão (BMNH 3.9.5.45,
3.9.5.46, 3.9.5.47*, 3.9.5.48). Goias, Anapolis (AMNH
133948*). Mato Grosso do Sul, Maracaju (AMNH
133946*). Minas Gerais, Arinos (MZUFV 3471*); Itape-
cerica (MZUFV 4422*); Rio Jordão (BMNH 1.11.3.24,
1.11.3.25*). Piauí, Central Piauí (MCZ 24828). Unk-
nown state (BMNH 68.a*). Rio Grande do Sul, Quinta
(AMNH 235512*, 235513*, AMNH 235514*); Santana
da Boa Vista (AMNH 235993*); Uruguaiana (AMNH
235994*). CHILE: Araucania, 2.5 km NNE Angol
(AMNH 93324); Maquehue, 9.5 km WSW Temuco
(AMNH 33290*, 33291*). Talca?, San Rafael? (BMNH
45.11.18.17). Arica y Parinacota, Parinacota (USNM
391849, 391850). Magallanes, Punta Arenas (AMNH
130053, 130068, 130095); Strait of Magellan (BMNH
Specimens are grouped according to current classifica-
tion (ASM 2024), which must be deemed provisional be-
cause the genus has never been reviewed. Adult specimens
whose skulls were examined are indicated with an asterisk.
All other specimens consist of immature skulls with or
without study skin, or study skins only.
C. chinga (n = 272).— ARGENTINA: Buenos Aires,
25 de Mayo (CFA 9752); Bonifacio (BMNH 17.9.15.3);
Bonifacio (Laguna Alsina) (BMNH 17.9.15.1*); Estación
Ombucta (BMNH 28.12.11.9); Estancia Los Angeles,
64 Mi SE Azul (USNM 331066); Estancia Los Ingleses,
near Mar de Ajó (BMNH 20.2.7.10, 20.2.7.8, 20.2.7.9,
9.12.1.16, 9.12.1.17, 9.12.1.18*); Gándara (MACN
29.912); Mar del Plata (BMNH 13.2.24.1, 16.10.3.6);
N of Necochea (USNM 172792*); near Henderson
(FMNH 24354*); San José Ranch, 32 km SSW San
Blas (USNM 171957, 171958, 172791). Catamarca,
‘Otro Cerro’, Sierra de Ambato, aprox. 35 km N Chum-
bicha (BMNH 19.2.7.1*, 19.2.7.2, 19.2.7.3*, 20.3.17.2);
Belén (BMNH 34.11.4.6, 34.11.4.7). Chubut, 100 km
NW Comodoro Rivadavia (AMNH 94317*); Barran-
cas Blancas, 5.5 km al SW Puerto Madryn (MNHN
1897-1244*); La Concepción (Laguna) (BMNH
28.12.11.6, 28.12.11.7); Lago Colhue Huapi (AMNH
94328, 94330*); Pico Salamanca (BMNH 28.12.11.3–
28.12.11.5); Rawson (MACN 28.72); Río Chubut
(BMNH 99.2.22.10); Sarmiento (AMNH 94329*); unk-
nown locality (BMNH 96.10.7.3); Valle del Lago Blan-
co (BMNH 3.7.9.13–3.7.9.15). Córdoba, Cruz del Eje
(BMNH 2.2.5.2, 2.2.5.3*, 2.2.5.4–2.2.5.6, 2.2.5.9); El
Carrizal, Villa Dolores (BMNH 17.6.29.5–17.6.29.7);
Noetinger (BMNH 17.1.25.8, 17.1.25.9); unknown loca-
lity (AMNH 36933). Corrientes, Manantiales (MACN
13.714). Entre Ríos, Paraná (AMNH 36932). Formosa,
‘Formosa, Paso de las Niñas, Río Teuco’ [39 km SSW In-
geniero Suárez, Río Teuco-Bermejo] (MACN 47.119);
Fortín Nuevo Pilcomayo (MACN 43.58). Jujuy, Alfar-
cito (BMNH 21.11.1.3); Humahuaca (MACN 26.182);
Maimara (BMNH 12.12.12.1*, 12.12.12.2). La Rioja,
Desiderio Tello (USNM 172793*, 172794). Mendoza,
Colonia Alvear (BMNH 10.9.12.2); Tupungato (BMNH
21.7.5.3*). Neuquén, Chos Malal (BMNH 26.10.11.1);
Collón Curá (BMNH 27.6.4.68); 16 km SE of La Rin-
conada (BMNH 27.5.1.1*); San Martín de los Andes
(BMNH 27.5.1.119); Sierra de Pil Pil (BMNH 27.5.1.2).
‘Patagonia’, unknown locality (BMNH 1899.2.41,
New subespecies of Conepatus semistriatus41
66.a). Valdivia, Riñihue (FMNH 24350*). Valparaíso,
Cerro Castillo, Viña del Mar (BMNH 10.7.23.1); Cu-
raumilla Farm-Coast Hills (BMNH 0.10.2.2). COLOM-
BIA: Nariño, Vereda El Espino, 5.2 km SW Túquerres
(UV 13287*). ECUADOR: Bolívar, Sinche (= Hacienda
Sinche), 6 km NNE Guaranda (AMNH 67085*, BMNH
99.9.9.7*). Chimborazo, Hacienda Alao (MCZ 52661);
Volcán Chimborazo (QCAZ 642). Imbabura, Hacienda
La Vega, 5 km ESE San Pablo del Lago (FMNH 125113*);
Volcán Imbabura (QCAZ 2046*). Napo, Baeza (MEPN
8169*); Cuyuja (QCAZ 726* = ‘638’); Probably Volcán
Antisana (1923 expedition, H. E. Anthony and G. H.
H. Tate) (AMNH 66244*); Volcán Antisana (AMNH
66719*). Pastaza, Río Pastaza, Mera (MNHN 1932-
2884*). Pichincha, 40 km S Quito (AMNH 187838*);
Alóag (AMNH 66722*); Cumbre del Monte Quitolo-
ma (QCAZ 8338); El Castillo, vía Esmeraldas (FMNH
44336*); Hacienda Antisanilla, 33 km SE Quito (AMNH
63577, 66721*); Mindo (MEPN 8305*); Northwest side
of Mindo (MCZ 27341); Pichincha Volcano (AMNH
36462*, 36463*); Quito (AMNH 36464–36466, BMNH
99.2.18.13*); San José de Minas (QCAZ 640); Santa Rosa
above Río Pita (AMNH 66720*); unknown locality
(BMNH 34.9.10.81*). Tungurahua, Montaña de Run-
tún, near Baños de Agua Santa (MCZ 38732). Unknown
province (AMNH 66723, 66724, MEPN 2862, 6864,
MNHN 1904-774). PARAGUAY: Boquerón, 50 km
WSW Fortín Madrejón (AMNH 248467*, 248468*,
248469*, 248470*); Guachalla, Río Pilcomayo, 5.4 km
SW San Agustín (FMNH 54329*, 54330*). PERU: An-
cash, Carpa (AMNH 238425*). Arequipa, 2 km NE Yura
Viejo (FMNH 106007*); 2.9 km NW Sumbay (FMNH
49720*); 5.3 km NW Salinas Moche (FMNH 49732,
49733*, 49734*); Caylloma (BMNH 3.8.4.1*, FMNH
49721, 49722, 49723*); Sumbay (BMNH 0.10.1.2).
Cajamarca, Celendín (BMNH 26.4.1.116); Hacienda
Limón, 60 km NE Cajamarca (FMNH 19680*). Ha-
cienda Taulis (AMNH 73123). Callao, Callao (BMNH
0.5.7.34*). Cusco, Chospyoc (BMNH 22.1.1.19); Chos-
pyoc, Río Huarocondo (USNM 194322); Ocobamba
Valley (BMNH 22.1.1.20, USNM 194319*, 194320*);
Orca, Near Calca (USNM 194324*); San Miguel Bridge,
near Matchu Picchu (USNM 194323*). Huánuco, 4 km
E Ambo (FMNH 24355*, FMNH 24356*). Lima, Near
Huarochiri (USNM 176320*); Surco (BMNH 0.5.7.35–
0.5.7.37). Puno, 2.2 km ESE Huacullani (FMNH
52486); Azángaro (‘Sangero’) (BMNH 1.1.1.10); Ha-
cienda Checayani, near Azángaro (MNHN 1957-1293*,
1957-1294, 1957-1295, 1957-1296, 1957-1297*, 1970-
301*, 1970-302*); Hacienda Collacachi, 12 km SSE
Puno (FMNH 49724*, 49725*, 49726*, 49727*, 49728,
49729*, 49730*, 49731*); West shore of Lake Titicaca
(MCZ 5257–5259). URUGUAY: Artigas, 6 km NNW
Belén (in neighboring Departamento de Salto) (AMNH
205833–205835). Cerro Largo, 20 km NW Paso del
Dragón (AMNH 205839); Estancia Las Marías, 6 km SE
Melo (AMNH 205837, 205838). Lavalleja, 12 km WSW
Zapicán (AMNH 205843, 205844). Paysandú, Arro-
yo Negro stream, 15 km S Paysandú (AMNH 205849,
205866). Rocha, Rocha, 24 km N San Vicente De Casti-
llos (USNM 259436*). Treinta y Tres, 16 km SSW Tacua-
rí River mouth (AMNH 205895). Unknown department
(BMNH 91.4.24.4, MHNM 3382*, 4298*, 4299*).
C. leuconotus (n = 21).— Honduras: Francisco Mora-
zán, El Caliche (AMNH 127569*). Olancho, Catacamas
(AMNH 128125*). MEXICO: Jalisco, Garabatos (Tepa-
titlán de Morelos) (AMNH 25171*); La Estancia (Aran-
das) (AMNH 25178*). Oaxaca, La Concepción, 11 km
NE San Miguel Tenango (AMNH 145973*); San Pedro
Tapanatepec (AMNH 176665*, 176668*). Sinaloa, Escui-
napa de Hidalgo (AMNH 24707*). Veracruz, 11 km NW
Alvarado (AMNH 172187*); 39 km S Veracruz (AMNH
204288*, 204289*); Córdoba (AMNH 30526*). NI-
CARAGUA: Jinotega, San Rafael del Norte (AMNH
29282). UNITED STATES: Arizona, Near Fort Verde
(AMNH 1921*, 1922*). ‘California’, erroneous state
(BMNH 55.12.24.221). New Mexico, Cliff (Grant Co.)
(AMNH 127112*); Gila (Grant Co.) (AMNH 127110*).
Texas, Juniper Canyon, Chisos Mountains (AMNH
136415*); Rockport (Aransas Co.) (AMNH 5130, 5883).
C. semistriatus (n = 111).— BELIZE: Belize Dis-
trict, 3.55 mi Northern Hwy (FMNH 58560*). Cayo,
Red Creek, Before Santa Elena (FMNH 121557). Stann
Creek, Stann Creek Valley (FMNH 63902). COLOM-
BIA: Cesar, Colonia Agrícola de Caracolicito, 9.3 km
E Pamparejo (USNM 281452*, 281453, 281454); El
Orinoco, Río Cesar, 37 km SSW Valledupar (USNM
281455, 281456*). Córdoba, Catival, upper Río San
Jorge (FMNH 68904, 68905). Cundinamarca, Choachí
(MCZ 27218, MCZ 27219); Finca El Soche, 4.0 km E
Granada (UV 8103); Laguna de Fúquene (ICN 283); Las
Balsillas (currently Bogotá) (AMNH 38423*, 38424*).
La Guajira, Las Marimondas, 4 km ESE Conejo (USNM
281464*, 281465*); Sierra Negra, 8.5 km ENE Villanueva
(USNM 281457*, 281458*, 281459*, 281460*); Villa-
nueva (USNM 281461*, 281462*, 281463*). Magdalena,
Bonda (AMNH 14632); Cuchilla de San Lorenzo (IAvH-
M 1759). Norte de Santander, Parque Nacional Natural
El Tamá, Maraña site (IAvH-M 3117). COSTA RICA:
Cártago, Ricardo Jiménez Ranch, Irazú Volcano (AMNH
19206*). Limón, Jiménez (AMNH 2794). Puntarenas,
Pozo Azul de Pirrís, plains of the Río Grande de Pirrís
66.a). Valdivia, Riñihue (FMNH 24350*). Valparaíso,
Cerro Castillo, Viña del Mar (BMNH 10.7.23.1); Cu-
raumilla Farm-Coast Hills (BMNH 0.10.2.2). COLOM-
BIA: Nariño, Vereda El Espino, 5.2 km SW Túquerres
(UV 13287*). ECUADOR: Bolívar, Sinche (= Hacienda
Sinche), 6 km NNE Guaranda (AMNH 67085*, BMNH
99.9.9.7*). Chimborazo, Hacienda Alao (MCZ 52661);
Volcán Chimborazo (QCAZ 642). Imbabura, Hacienda
La Vega, 5 km ESE San Pablo del Lago (FMNH 125113*);
Volcán Imbabura (QCAZ 2046*). Napo, Baeza (MEPN
8169*); Cuyuja (QCAZ 726* = ‘638’); Probably Volcán
Antisana (1923 expedition, H. E. Anthony and G. H.
H. Tate) (AMNH 66244*); Volcán Antisana (AMNH
66719*). Pastaza, Río Pastaza, Mera (MNHN 1932-
2884*). Pichincha, 40 km S Quito (AMNH 187838*);
Alóag (AMNH 66722*); Cumbre del Monte Quitolo-
ma (QCAZ 8338); El Castillo, vía Esmeraldas (FMNH
44336*); Hacienda Antisanilla, 33 km SE Quito (AMNH
63577, 66721*); Mindo (MEPN 8305*); Northwest side
of Mindo (MCZ 27341); Pichincha Volcano (AMNH
36462*, 36463*); Quito (AMNH 36464–36466, BMNH
99.2.18.13*); San José de Minas (QCAZ 640); Santa Rosa
above Río Pita (AMNH 66720*); unknown locality
(BMNH 34.9.10.81*). Tungurahua, Montaña de Run-
tún, near Baños de Agua Santa (MCZ 38732). Unknown
province (AMNH 66723, 66724, MEPN 2862, 6864,
MNHN 1904-774). PARAGUAY: Boquerón, 50 km
WSW Fortín Madrejón (AMNH 248467*, 248468*,
248469*, 248470*); Guachalla, Río Pilcomayo, 5.4 km
SW San Agustín (FMNH 54329*, 54330*). PERU: An-
cash, Carpa (AMNH 238425*). Arequipa, 2 km NE Yura
Viejo (FMNH 106007*); 2.9 km NW Sumbay (FMNH
49720*); 5.3 km NW Salinas Moche (FMNH 49732,
49733*, 49734*); Caylloma (BMNH 3.8.4.1*, FMNH
49721, 49722, 49723*); Sumbay (BMNH 0.10.1.2).
Cajamarca, Celendín (BMNH 26.4.1.116); Hacienda
Limón, 60 km NE Cajamarca (FMNH 19680*). Ha-
cienda Taulis (AMNH 73123). Callao, Callao (BMNH
0.5.7.34*). Cusco, Chospyoc (BMNH 22.1.1.19); Chos-
pyoc, Río Huarocondo (USNM 194322); Ocobamba
Valley (BMNH 22.1.1.20, USNM 194319*, 194320*);
Orca, Near Calca (USNM 194324*); San Miguel Bridge,
near Matchu Picchu (USNM 194323*). Huánuco, 4 km
E Ambo (FMNH 24355*, FMNH 24356*). Lima, Near
Huarochiri (USNM 176320*); Surco (BMNH 0.5.7.35–
0.5.7.37). Puno, 2.2 km ESE Huacullani (FMNH
52486); Azángaro (‘Sangero’) (BMNH 1.1.1.10); Ha-
cienda Checayani, near Azángaro (MNHN 1957-1293*,
1957-1294, 1957-1295, 1957-1296, 1957-1297*, 1970-
301*, 1970-302*); Hacienda Collacachi, 12 km SSE
Puno (FMNH 49724*, 49725*, 49726*, 49727*, 49728,
49729*, 49730*, 49731*); West shore of Lake Titicaca
(MCZ 5257–5259). URUGUAY: Artigas, 6 km NNW
Belén (in neighboring Departamento de Salto) (AMNH
205833–205835). Cerro Largo, 20 km NW Paso del
Dragón (AMNH 205839); Estancia Las Marías, 6 km SE
Melo (AMNH 205837, 205838). Lavalleja, 12 km WSW
Zapicán (AMNH 205843, 205844). Paysandú, Arro-
yo Negro stream, 15 km S Paysandú (AMNH 205849,
205866). Rocha, Rocha, 24 km N San Vicente De Casti-
llos (USNM 259436*). Treinta y Tres, 16 km SSW Tacua-
rí River mouth (AMNH 205895). Unknown department
(BMNH 91.4.24.4, MHNM 3382*, 4298*, 4299*).
C. leuconotus (n = 21).— Honduras: Francisco Mora-
zán, El Caliche (AMNH 127569*). Olancho, Catacamas
(AMNH 128125*). MEXICO: Jalisco, Garabatos (Tepa-
titlán de Morelos) (AMNH 25171*); La Estancia (Aran-
das) (AMNH 25178*). Oaxaca, La Concepción, 11 km
NE San Miguel Tenango (AMNH 145973*); San Pedro
Tapanatepec (AMNH 176665*, 176668*). Sinaloa, Escui-
napa de Hidalgo (AMNH 24707*). Veracruz, 11 km NW
Alvarado (AMNH 172187*); 39 km S Veracruz (AMNH
204288*, 204289*); Córdoba (AMNH 30526*). NI-
CARAGUA: Jinotega, San Rafael del Norte (AMNH
29282). UNITED STATES: Arizona, Near Fort Verde
(AMNH 1921*, 1922*). ‘California’, erroneous state
(BMNH 55.12.24.221). New Mexico, Cliff (Grant Co.)
(AMNH 127112*); Gila (Grant Co.) (AMNH 127110*).
Texas, Juniper Canyon, Chisos Mountains (AMNH
136415*); Rockport (Aransas Co.) (AMNH 5130, 5883).
C. semistriatus (n = 111).— BELIZE: Belize Dis-
trict, 3.55 mi Northern Hwy (FMNH 58560*). Cayo,
Red Creek, Before Santa Elena (FMNH 121557). Stann
Creek, Stann Creek Valley (FMNH 63902). COLOM-
BIA: Cesar, Colonia Agrícola de Caracolicito, 9.3 km
E Pamparejo (USNM 281452*, 281453, 281454); El
Orinoco, Río Cesar, 37 km SSW Valledupar (USNM
281455, 281456*). Córdoba, Catival, upper Río San
Jorge (FMNH 68904, 68905). Cundinamarca, Choachí
(MCZ 27218, MCZ 27219); Finca El Soche, 4.0 km E
Granada (UV 8103); Laguna de Fúquene (ICN 283); Las
Balsillas (currently Bogotá) (AMNH 38423*, 38424*).
La Guajira, Las Marimondas, 4 km ESE Conejo (USNM
281464*, 281465*); Sierra Negra, 8.5 km ENE Villanueva
(USNM 281457*, 281458*, 281459*, 281460*); Villa-
nueva (USNM 281461*, 281462*, 281463*). Magdalena,
Bonda (AMNH 14632); Cuchilla de San Lorenzo (IAvH-
M 1759). Norte de Santander, Parque Nacional Natural
El Tamá, Maraña site (IAvH-M 3117). COSTA RICA:
Cártago, Ricardo Jiménez Ranch, Irazú Volcano (AMNH
19206*). Limón, Jiménez (AMNH 2794). Puntarenas,
Pozo Azul de Pirrís, plains of the Río Grande de Pirrís
Molinari, Abarca-Medina & Rivas-Rodríguez42
(AMNH 19205*). San José, Escazú (AMNH 135269,
135271*, 137282*, 137283*); La Hondura, 22.5 km NE
San José (AMNH 135270*); Santa Teresa (currently San
José) (AMNH 141858*). Unknown province (USNM
19646, 19647, 61205*, 61275*). ECUADOR: Sto. Do-
mingo de los Tsáchilas, Río Palenque Science Center,
1.7 km SSE Consumulo (USNM 568103*). MEXICO:
Quintana Roo, La Vega, on mainland coast opposite Isla
Cancún (USNM 108502*, 108503*). Unknown state
(BMNH 2001.5). Veracruz, Achotal (FMNH 13825*);
Catemaco (USNM 65762*, 65763*, AMNH 172190);
Paso Nuevo (AMNH 17201*, 17202); Pérez (USNM
132512*). Yucatán, Mérida (USNM 8610*). NICA-
RAGUA: Chontales, Villa Somoza (= Villa Sandino)
(USNM 337832*). Jinotega, Hacienda La Trampa, 16 km
E, 5.5 km N Jinotega (USNM 338870*). Río San Juan,
La Esperanza, 9.5 km SE San Carlos (USNM 361359*).
PANAMA: Bocas del Toro, Sibube (USNM 335773*).
Chiriquí, 3.2 km NE El Volcán (USNM 332037*); Bo-
querón (AMNH 18900); Boquete (BMNH 4.7.6.5*,
MCZ 10115, 10116); Cerro Punta (USNM 324236);
Progreso (USNM 363346*). PERU: Amazonas, Cha-
chapoyas (BMNH 24.7.11.10*, 24.7.11.11–24.7.11.13,
24.7.11.8, 24.7.11.9). La Libertad, Chicama Valley
(USNM 172857*); Menocucho, 23.5 km ENE Trujillo
(FMNH 19976*); San Pedro de Lloc (AMNH 73220).
Lambayeque, Eten (BMNH 0.3.1.39*). VENEZUE-
LA: Falcón, 20.5 km SSE Tucacas (USNM 372745*);
3.5–6.0 km NE Capatárida (EBRG 3309*, 3310*,3311*,
3312*, 443290*, 443291*, 443295*); Capatárida
(USNM 443285*, 443286*, 443289*, 443293*, 443294*,
443296*); Dabajuro to Mene de Mauroa road, 25.6 km
WSW Dabajuro (CVULA 8544*); Muaco, 3.5 km ENE
Vela de Coro (MBUCV 4170*); Península de Paragua-
ná, 5.5 km WNW Adícora (CVULA 8545*); Península
de Paraguaná, Cueva de Piedra Honda (EBRG 20237*,
20241*); Península de Paraguaná, Near Moruy, 15 km
SSW Pueblo Nuevo (USNM 443414*); Península de
Paraguaná, Reserva Biológica de Monte Cano (RBMC
unnumbered); Pueblo Nuevo to Adícora road (CVULA
8542); Urumaco to Dabajuro road, 17 km WSW Uru-
maco (CVULA 8543*). Mérida, Casa del Ángel del Sol,
9.7 km ENE Mérida (CVULA 9130*); El Mirabel, 3 km
SSE La Azulita (CVULA 6210*); La Hechicera, 3 km
NNW Mérida (CVULA 8537*); Mountains W of Méri-
da (FMNH 22202*, 22203); Near Chorrera de las Gon-
zález, 5.1 km WSW Jají (CVULA 1025); unknown loca-
lity (AMNH 21634, AMNH 21635, BMNH 5.2.5.10*,
5.4.5.4*, 5.4.5.5*. Yaracuy, El Hacha (AMNH 32073).
Zulia, El Rosario, 39 Km WNW Encontrados (USNM
443576*); Río Aurare, 12.5 km ESE Maracaibo (FMNH
18770*).
(AMNH 19205*). San José, Escazú (AMNH 135269,
135271*, 137282*, 137283*); La Hondura, 22.5 km NE
San José (AMNH 135270*); Santa Teresa (currently San
José) (AMNH 141858*). Unknown province (USNM
19646, 19647, 61205*, 61275*). ECUADOR: Sto. Do-
mingo de los Tsáchilas, Río Palenque Science Center,
1.7 km SSE Consumulo (USNM 568103*). MEXICO:
Quintana Roo, La Vega, on mainland coast opposite Isla
Cancún (USNM 108502*, 108503*). Unknown state
(BMNH 2001.5). Veracruz, Achotal (FMNH 13825*);
Catemaco (USNM 65762*, 65763*, AMNH 172190);
Paso Nuevo (AMNH 17201*, 17202); Pérez (USNM
132512*). Yucatán, Mérida (USNM 8610*). NICA-
RAGUA: Chontales, Villa Somoza (= Villa Sandino)
(USNM 337832*). Jinotega, Hacienda La Trampa, 16 km
E, 5.5 km N Jinotega (USNM 338870*). Río San Juan,
La Esperanza, 9.5 km SE San Carlos (USNM 361359*).
PANAMA: Bocas del Toro, Sibube (USNM 335773*).
Chiriquí, 3.2 km NE El Volcán (USNM 332037*); Bo-
querón (AMNH 18900); Boquete (BMNH 4.7.6.5*,
MCZ 10115, 10116); Cerro Punta (USNM 324236);
Progreso (USNM 363346*). PERU: Amazonas, Cha-
chapoyas (BMNH 24.7.11.10*, 24.7.11.11–24.7.11.13,
24.7.11.8, 24.7.11.9). La Libertad, Chicama Valley
(USNM 172857*); Menocucho, 23.5 km ENE Trujillo
(FMNH 19976*); San Pedro de Lloc (AMNH 73220).
Lambayeque, Eten (BMNH 0.3.1.39*). VENEZUE-
LA: Falcón, 20.5 km SSE Tucacas (USNM 372745*);
3.5–6.0 km NE Capatárida (EBRG 3309*, 3310*,3311*,
3312*, 443290*, 443291*, 443295*); Capatárida
(USNM 443285*, 443286*, 443289*, 443293*, 443294*,
443296*); Dabajuro to Mene de Mauroa road, 25.6 km
WSW Dabajuro (CVULA 8544*); Muaco, 3.5 km ENE
Vela de Coro (MBUCV 4170*); Península de Paragua-
ná, 5.5 km WNW Adícora (CVULA 8545*); Península
de Paraguaná, Cueva de Piedra Honda (EBRG 20237*,
20241*); Península de Paraguaná, Near Moruy, 15 km
SSW Pueblo Nuevo (USNM 443414*); Península de
Paraguaná, Reserva Biológica de Monte Cano (RBMC
unnumbered); Pueblo Nuevo to Adícora road (CVULA
8542); Urumaco to Dabajuro road, 17 km WSW Uru-
maco (CVULA 8543*). Mérida, Casa del Ángel del Sol,
9.7 km ENE Mérida (CVULA 9130*); El Mirabel, 3 km
SSE La Azulita (CVULA 6210*); La Hechicera, 3 km
NNW Mérida (CVULA 8537*); Mountains W of Méri-
da (FMNH 22202*, 22203); Near Chorrera de las Gon-
zález, 5.1 km WSW Jají (CVULA 1025); unknown loca-
lity (AMNH 21634, AMNH 21635, BMNH 5.2.5.10*,
5.4.5.4*, 5.4.5.5*. Yaracuy, El Hacha (AMNH 32073).
Zulia, El Rosario, 39 Km WNW Encontrados (USNM
443576*); Río Aurare, 12.5 km ESE Maracaibo (FMNH
18770*).
New subespecies of Conepatus semistriatus43
APPENDIX 2
Results of the Principal Component Analysis (PCA). Measurement loadings and percent variance explained by compo-
nents are shown. Components 17 to 25 are omitted because together they explain less than 2% of total variance and 5% of
total ‘shape’ (PC2 to PC25) variance. The measurements are ordered according to their communality values.
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 Communality
(PC2–PC16)
Precanine length 0.12 0.16 -0.52 0.58 0.05 -0.14 0.23 -0.14 0.96
Width of interpterygoid fossa -0.07 0.15 0.70 0.50 -0.07 0.19 0.28 0.06 0.95
Length of PM4 0.20 -0.22 0.07 -0.03 -0.16 0.08 0.09 0.53 0.92
Post-notch length 0.06 0.65 0.06 -0.09 0.47 -0.08 -0.08 0.33 0.91
Heigth of cranium 0.20 0.17 0.13 -0.27 0.15 0.02 0.11 -0.35 0.91
Length of PM3 0.17 0.12 -0.11 0.23 -0.13 0.69 -0.31 -0.10 0.88
Postorbital breadth 0.19 -0.14 0.25 0.21 0.13 -0.21 -0.46 -0.26 0.87
Width across incisors 0.20 0.00 -0.08 0.23 0.19 0.03 -0.27 0.26 0.87
Height of coronoid 0.17 0.18 0.02 0.14 -0.55 -0.45 -0.07 -0.03 0.84
Length of mandible 0.20 0.23 -0.07 -0.05 -0.31 -0.15 0.13 0.22 0.79
Diameter of canine 0.21 -0.04 0.07 -0.14 -0.19 0.01 -0.34 0.17 0.72
Length of lower carnassial 0.21 -0.15 0.03 0.09 0.18 -0.07 0.40 0.17 0.67
Interorbital breadth 0.22 0.11 0.18 0.00 -0.01 -0.05 0.01 -0.23 0.60
Zygomatic breadth 0.22 0.06 0.10 -0.16 0.12 -0.10 0.02 -0.12 0.59
Width of molar 0.19 -0.35 0.07 -0.05 0.23 -0.07 0.09 0.10 0.50
Postpalatal length 0.22 -0.01 0.06 -0.15 -0.14 0.13 0.25 -0.20 0.48
Length of molar 0.19 -0.35 -0.08 0.19 0.24 -0.08 0.02 -0.08 0.46
Notch to canine length 0.22 -0.02 -0.13 -0.10 -0.13 0.17 0.04 -0.02 0.44
Length of maxillary toothrow 0.22 -0.13 0.01 -0.05 -0.04 0.12 0.05 0.12 0.43
Width across canines 0.23 0.01 0.07 0.03 0.01 -0.13 -0.18 0.14 0.34
Width across molars 0.23 -0.07 0.14 0.11 0.05 -0.05 -0.07 -0.03 0.28
Palatilar length 0.23 0.13 -0.16 -0.01 0.10 0.14 0.02 0.07 0.21
Mastoid breadth 0.23 0.06 0.03 -0.04 0.07 -0.10 0.00 -0.19 0.15
Basilar length 0.23 0.06 -0.04 -0.09 -0.04 0.16 0.14 -0.07 0.15
Condylobasal length 0.23 0.07 -0.02 -0.09 -0.01 0.08 0.17 -0.08 0.10
EIGENVALUE 17.36 1.65 1.24 0.76 0.72 0.63 0.52 0.36
% TOTAL (PC1–PC25) VARIANCE 69.44 6.61 4.95 3.05 2.87 2.50 2.10 1.43
% SHAPE (PC2–PC25) VARIANCE 0.00 21.63 16.18 9.98 9.38 8.20 6.86 4.68
APPENDIX 2
Results of the Principal Component Analysis (PCA). Measurement loadings and percent variance explained by compo-
nents are shown. Components 17 to 25 are omitted because together they explain less than 2% of total variance and 5% of
total ‘shape’ (PC2 to PC25) variance. The measurements are ordered according to their communality values.
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 Communality
(PC2–PC16)
Precanine length 0.12 0.16 -0.52 0.58 0.05 -0.14 0.23 -0.14 0.96
Width of interpterygoid fossa -0.07 0.15 0.70 0.50 -0.07 0.19 0.28 0.06 0.95
Length of PM4 0.20 -0.22 0.07 -0.03 -0.16 0.08 0.09 0.53 0.92
Post-notch length 0.06 0.65 0.06 -0.09 0.47 -0.08 -0.08 0.33 0.91
Heigth of cranium 0.20 0.17 0.13 -0.27 0.15 0.02 0.11 -0.35 0.91
Length of PM3 0.17 0.12 -0.11 0.23 -0.13 0.69 -0.31 -0.10 0.88
Postorbital breadth 0.19 -0.14 0.25 0.21 0.13 -0.21 -0.46 -0.26 0.87
Width across incisors 0.20 0.00 -0.08 0.23 0.19 0.03 -0.27 0.26 0.87
Height of coronoid 0.17 0.18 0.02 0.14 -0.55 -0.45 -0.07 -0.03 0.84
Length of mandible 0.20 0.23 -0.07 -0.05 -0.31 -0.15 0.13 0.22 0.79
Diameter of canine 0.21 -0.04 0.07 -0.14 -0.19 0.01 -0.34 0.17 0.72
Length of lower carnassial 0.21 -0.15 0.03 0.09 0.18 -0.07 0.40 0.17 0.67
Interorbital breadth 0.22 0.11 0.18 0.00 -0.01 -0.05 0.01 -0.23 0.60
Zygomatic breadth 0.22 0.06 0.10 -0.16 0.12 -0.10 0.02 -0.12 0.59
Width of molar 0.19 -0.35 0.07 -0.05 0.23 -0.07 0.09 0.10 0.50
Postpalatal length 0.22 -0.01 0.06 -0.15 -0.14 0.13 0.25 -0.20 0.48
Length of molar 0.19 -0.35 -0.08 0.19 0.24 -0.08 0.02 -0.08 0.46
Notch to canine length 0.22 -0.02 -0.13 -0.10 -0.13 0.17 0.04 -0.02 0.44
Length of maxillary toothrow 0.22 -0.13 0.01 -0.05 -0.04 0.12 0.05 0.12 0.43
Width across canines 0.23 0.01 0.07 0.03 0.01 -0.13 -0.18 0.14 0.34
Width across molars 0.23 -0.07 0.14 0.11 0.05 -0.05 -0.07 -0.03 0.28
Palatilar length 0.23 0.13 -0.16 -0.01 0.10 0.14 0.02 0.07 0.21
Mastoid breadth 0.23 0.06 0.03 -0.04 0.07 -0.10 0.00 -0.19 0.15
Basilar length 0.23 0.06 -0.04 -0.09 -0.04 0.16 0.14 -0.07 0.15
Condylobasal length 0.23 0.07 -0.02 -0.09 -0.01 0.08 0.17 -0.08 0.10
EIGENVALUE 17.36 1.65 1.24 0.76 0.72 0.63 0.52 0.36
% TOTAL (PC1–PC25) VARIANCE 69.44 6.61 4.95 3.05 2.87 2.50 2.10 1.43
% SHAPE (PC2–PC25) VARIANCE 0.00 21.63 16.18 9.98 9.38 8.20 6.86 4.68
Molinari, Abarca-Medina & Rivas-Rodríguez44
PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 Communality
(PC2–PC16)
Precanine length 0.20 -0.28 0.29 0.08 0.03 0.07 0.07 0.08 0.96
Width of interpterygoid fossa 0.00 0.05 0.15 0.07 0.05 0.17 -0.03 0.01 0.95
Length of PM4 0.20 -0.34 -0.12 0.21 -0.09 -0.33 0.24 0.39 0.92
Post-notch length 0.23 0.20 0.00 0.03 -0.04 -0.17 -0.13 0.03 0.91
Heigth of cranium -0.11 0.04 0.10 -0.06 0.32 0.14 0.52 0.46 0.91
Length of PM3 0.21 0.09 -0.28 -0.21 0.08 -0.09 0.06 -0.01 0.88
Postorbital breadth 0.08 0.06 0.01 0.01 -0.58 0.04 0.23 0.09 0.87
Width across incisors -0.74 -0.06 0.01 0.11 0.18 -0.02 0.11 -0.13 0.87
Height of coronoid -0.04 0.31 -0.16 0.00 0.23 -0.09 -0.18 0.23 0.84
Length of mandible 0.07 0.18 -0.14 -0.09 -0.20 0.16 0.46 -0.44 0.79
Diameter of canine 0.09 -0.08 0.65 -0.08 -0.05 0.16 -0.21 0.02 0.72
Length of lower carnassial -0.23 0.03 -0.18 -0.42 -0.31 0.09 -0.14 0.14 0.67
Interorbital breadth -0.09 -0.20 -0.12 0.53 0.03 -0.32 -0.09 -0.23 0.60
Zygomatic breadth 0.17 -0.46 -0.26 -0.02 0.06 0.33 -0.20 -0.23 0.59
Width of molar 0.28 0.30 0.16 0.10 0.28 -0.04 0.00 -0.10 0.50
Postpalatal length -0.10 -0.01 0.27 -0.18 -0.11 -0.36 -0.01 -0.21 0.48
Length of molar 0.15 0.37 -0.08 0.06 0.12 -0.15 0.02 -0.12 0.46
Notch to canine length -0.12 0.15 -0.10 0.20 -0.12 0.24 -0.37 0.26 0.44
Length of maxillary toothrow 0.08 0.05 -0.07 0.29 0.10 0.50 0.09 -0.13 0.43
Width across canines 0.02 -0.23 0.06 -0.35 0.28 -0.03 0.07 -0.08 0.34
Width across molars 0.07 -0.06 -0.13 -0.31 0.25 -0.04 -0.22 0.00 0.28
Palatilar length -0.05 0.16 0.07 0.13 -0.15 0.14 -0.08 0.16 0.21
Mastoid breadth 0.03 -0.14 -0.16 -0.01 -0.11 -0.06 -0.11 0.15 0.15
Basilar length -0.08 0.08 0.18 -0.03 -0.08 -0.14 -0.06 -0.07 0.15
Condylobasal length -0.02 0.04 0.10 0.05 -0.10 -0.10 -0.02 -0.12 0.10
EIGENVALUE 0.32 0.26 0.20 0.18 0.16 0.15 0.12 0.12
% TOTAL (PC1–PC25) VARIANCE 1.28 1.02 0.79 0.72 0.62 0.60 0.48 0.46
% SHAPE (PC2–PC25) VARIANCE 4.20 3.35 2.60 2.37 2.03 1.97 1.56 1.52
APPENDIX 2. (Continuation)
PC9 PC10 PC11 PC12 PC13 PC14 PC15 PC16 Communality
(PC2–PC16)
Precanine length 0.20 -0.28 0.29 0.08 0.03 0.07 0.07 0.08 0.96
Width of interpterygoid fossa 0.00 0.05 0.15 0.07 0.05 0.17 -0.03 0.01 0.95
Length of PM4 0.20 -0.34 -0.12 0.21 -0.09 -0.33 0.24 0.39 0.92
Post-notch length 0.23 0.20 0.00 0.03 -0.04 -0.17 -0.13 0.03 0.91
Heigth of cranium -0.11 0.04 0.10 -0.06 0.32 0.14 0.52 0.46 0.91
Length of PM3 0.21 0.09 -0.28 -0.21 0.08 -0.09 0.06 -0.01 0.88
Postorbital breadth 0.08 0.06 0.01 0.01 -0.58 0.04 0.23 0.09 0.87
Width across incisors -0.74 -0.06 0.01 0.11 0.18 -0.02 0.11 -0.13 0.87
Height of coronoid -0.04 0.31 -0.16 0.00 0.23 -0.09 -0.18 0.23 0.84
Length of mandible 0.07 0.18 -0.14 -0.09 -0.20 0.16 0.46 -0.44 0.79
Diameter of canine 0.09 -0.08 0.65 -0.08 -0.05 0.16 -0.21 0.02 0.72
Length of lower carnassial -0.23 0.03 -0.18 -0.42 -0.31 0.09 -0.14 0.14 0.67
Interorbital breadth -0.09 -0.20 -0.12 0.53 0.03 -0.32 -0.09 -0.23 0.60
Zygomatic breadth 0.17 -0.46 -0.26 -0.02 0.06 0.33 -0.20 -0.23 0.59
Width of molar 0.28 0.30 0.16 0.10 0.28 -0.04 0.00 -0.10 0.50
Postpalatal length -0.10 -0.01 0.27 -0.18 -0.11 -0.36 -0.01 -0.21 0.48
Length of molar 0.15 0.37 -0.08 0.06 0.12 -0.15 0.02 -0.12 0.46
Notch to canine length -0.12 0.15 -0.10 0.20 -0.12 0.24 -0.37 0.26 0.44
Length of maxillary toothrow 0.08 0.05 -0.07 0.29 0.10 0.50 0.09 -0.13 0.43
Width across canines 0.02 -0.23 0.06 -0.35 0.28 -0.03 0.07 -0.08 0.34
Width across molars 0.07 -0.06 -0.13 -0.31 0.25 -0.04 -0.22 0.00 0.28
Palatilar length -0.05 0.16 0.07 0.13 -0.15 0.14 -0.08 0.16 0.21
Mastoid breadth 0.03 -0.14 -0.16 -0.01 -0.11 -0.06 -0.11 0.15 0.15
Basilar length -0.08 0.08 0.18 -0.03 -0.08 -0.14 -0.06 -0.07 0.15
Condylobasal length -0.02 0.04 0.10 0.05 -0.10 -0.10 -0.02 -0.12 0.10
EIGENVALUE 0.32 0.26 0.20 0.18 0.16 0.15 0.12 0.12
% TOTAL (PC1–PC25) VARIANCE 1.28 1.02 0.79 0.72 0.62 0.60 0.48 0.46
% SHAPE (PC2–PC25) VARIANCE 4.20 3.35 2.60 2.37 2.03 1.97 1.56 1.52
APPENDIX 2. (Continuation)