Tremarctos ornatus (Carnivora: Ursidae)

Abstract

Tremarctos ornatus (F.G. Cuvier, 1825) is a tremarctine bear commonly known as the Andean bear. It is a medium-sized bear with black to dark red-brown pelage with dense, long, coarse fur; creamy white marks occur on the chin, neck, and chest, and often white to creamy marks occur on the face, around the muzzle, and eyes. It is distributed in the tropical Andes of Venezuela, Colombia, Ecuador, Perú, Bolivia, and northern Argentina in South America. T. ornatus is catalogued as “Vulnerable” (VU) by the International Union for Conservation of Nature and Natural Resources and is included in CITES Appendix I. Main threats include habitat loss and fragmentation, illegal killing, human–bear conflicts, and most likely climate change.

Resumen

 Tremarctos ornatus (F.G. Cuvier, 1825) es un oso tremractino, conocido comúnmente como oso andino. Es un oso de tamaño medio, con pelaje denso, largo y grueso, de color negro a rojizo-café. Tiene marcas de color blanco - crema en el cuello y pecho, en la cara, alrededor del hocico y los ojos. La especie se distribuye en los Andes tropicales de Venezuela, Colombia, Ecuador, Perú, Bolivia y el norte de Argentina en Sur América. T. ornatus está catalogado como “Vulnerable” (VU) por la Unión Internacional para la Conservación de la Naturaleza y los Recursos Naturales y está incluida en el Apéndice I de CITES. Sus principales amenazas incluyen la pérdida y fragmentación de hábitats, cacería ilegal, conflictos humano – oso andino y probablemente el cambio climático.

TremarctosGervais, 1855

• Ursus: Lesson, 1827:129. Part; not Linnaeus, 1758.

• TremarctosGervais, 1855:20. Type species Ursus ornatus F. Cuvier, 1825, by monotypy.

• Helarctos: Gray, 1864:698. Part; not Horsfield, 1825.

• NearctosGray, 1873:182. Type species Helarctos ornatus (=Ursus ornatus F. Cuvier, 1825), by monotypy.

Context and Content. Order Carnivora, suborder Caniformia, family Ursidae, subfamily Tremarctinae

Tremarctos ornatus (F. G. Cuvier, 1825)

Spectacled Bear

• Ursus ornatus F. G. Cuvier, 1825:57. Type locality “Cordilléres du Chili,” restricted to “las montañas al Este de Trujillo, departamento de la Libertad, Perú,” by Cabrera (1957:242).

• Ursus frugilegusTschudi, 1844:90. Type locality unknown; stated as Perú, probably near Lima by Allen (1942:396; see “Nomenclatural Notes”).

• Ursus fructilegusSchinz, 1845:34. Type locality “In Peru.”

• Helarctos ornatusGray, 1864:698. Type locality “South America, Cordilleras.”

• Ursus nasutusSclater, 1868:72, Plate VIII. Type locality not given (description based on a purchased specimen); stated as Simitarra River, Upper Magdalena, Colombia by María (1924).

• Nearctos ornatus: Gray, 1873:182. Name combination.

• Tremarctos ornatus: Gill, 1874:15. First use of current name combination.

• Tremarctos ornatus majoriThomas, 1902: 216. Type locality “Southern Ecuador, probably the province of Azuay.”

• Ursus ornatus thomasiHornaday, 1911:748. Type locality “Andes of southern Colombia.”

• Tremarctos lasalleiMaría, 1924:115. Type locality “región de Arauca,” Colombia.

Context and Content. Context as for genus. Tremarctos ornatus is considered monotypic (Kitchener 2010).

Nomenclatural Notes. The monophyletic origin of Ursidae with the giant panda (Ailuropoda melanoleuca) as a sister clade is widely accepted (Agnarsson et al. 2010). Tremarctos ornatus and its related fossil forms are placed in a separate subfamily, Tremarctinae or “short-faced” bears, but this taxonomic classification is not universally accepted (Thenius 1976; Wilson and Reeder 2005; Garshelis 2009). Five specific or subspecific names have been applied to T. ornatus, based on claw length, facial markings, and body proportions. Allen (1942) refers to U. frugilegus, Tschudi 1844:90, U. ornatus thomasiHornaday 1911:748, and T. lasalleiMaría 1924:115 as synonyms of T. ornatus majori. Allen believed U. frugilegus to be inseparable from the species described by Cuvier. In a review of the materials collected by J. Tschudi in Peru (1838–1942) it was concluded that Tschudi’s description of U. frugilegus was based on discussions with local hunters rather than an actual specimen (Serrano-Villavicencio et al. 2020). Tschudi made no reference to a collected specimen, only commenting that “due to the local climate, preserving skins of this species was a difficult task”; as of 2020 no type material had been located (Serrano-Villavicencio et al. 2020:916).

Because all geographical forms were described by different authors based only on physical variations, Cabrera (1957) assigned all geographical forms as inadmissible, referring all variations to Tremarctos ornatusCuvier, 1825. T. ornatus, is treated as a monotypic species, although considerable phenotypic variation is present within the species (Krause et al. 2008; Agnarsson et al. 2010; Kitchener 2010).

The etymological origin of the generic name came from Trema (Greek meaning hole) and arktos (Greek meaning bear) and refers to an unusual hole in the humerus. The specific name, ornatus (Latin meaning dress—Gotch 1979), refers to the light-colored patches of fur surrounding the eyes, muzzle, chest, and throat. These characters contrast with the black-colored fur, which in some cases creates the appearance of eyeglasses (also called spectacles), and serves as the basis for one of its common English names: spectacled bear (Pérez-Torres 2001), whereas the other common English name is Andean bear (Velez-Liendo and García-Rangel 2017).

In several Andean indigenous cultures, the species is known as “Jukumari,” which means bear (Paisley and Saunders 2010). In Quechua, the word uku means hole, so the word probably was used as the “bear with holes in the eyes” (Pérez-Torres 2001). Throughout its range, T. ornatus is known by other common names, such as Mashíramo, Oso Frontino, Oso Salvaje, Oso Real, Uí, Oso Careto, Iznachi, Manaba, Puca mate, and Ucucu (Mondolfi 1989; Goldstein 2002; Castellanos et al. 2016a).

DIAGNOSIS

Tremarctos ornatus (Fig. 1) is a medium-sized bear (head–body length 1,190–1,740 mm—Mondolfi 1971; Bininda-Emonds 1998) and is larger than the sun bear (Helarctos malayanus, head–body length 1,000–1,400 mm—Fitzgerald and Krausman 2002), but is similar in size to the American black bear (Ursus americanus, head–body length 1,430–1,570 mm—Lariviere 2001). T. ornatus has the shortest muzzle (9–129 mm) of all bear species and the shortest mean (± SD) skull length of any bear species (201 ± 6.45 mm) only comparable with the sun bear (213 ± 13.15 mm—Christiansen 2007; Stucchi and Figueroa 2013). Claws are short on front and rear feet (Peyton 1980, 1999; Garshelis 2009) in comparison with the sun bear that has strongly curved and pointed claws (Fitzgerald and Krausman 2002). The mandible of T. ornatus has a unique trait not present in other ursid species: a deep fossa called the premasseteric fossa (Mondolfi 1983).

Fig. 1.

—Adult male Tremarctos ornatus in the Chingaza National Protected Area, Colombia, July 2016. Used with permission of the photographer, Luis Guillermo Linares-Romero.

GENERAL CHARACTERS

Tremarctos ornatus is a medium-sized bear with rounded ears, a plantigrade stance, front limbs longer than the hindlimbs, nonretractable curved claws, and a short tail (Mondolfi 1971; Nowak 1999; Peyton 1999; Garshelis 2009; García-Rangel 2012). Pelage is black to dark red-brown with dense, long, and coarse fur (Mondolfi 1971; Garshelis 2009). Body coat color varies from black or blackish brown to blackish red (Mondolfi 1989; Nowak 1999; Garshelis 2009). Individual bears usually exhibit white, yellowish or creamy marks around the eyes that continue to the muzzle, cheeks, throat, neck, and chest (María 1924; Allen 1942; Garshelis 2009; García-Rangel 2012). These marks are highly variable across the geographic range, not only in color but also size, and can range from prominent to completely absent (Allen 1942; Mondolfi 1971; Garshelis 2009; Réyes et al. 2017), and they are the most distinctive phenotypic character for T. ornatus. When present, facial marks are more conspicuous in young, whereas in adults the facial marks can become thinner with age; they do not indicate family relationships between individuals (Van Horn et al. 2014b, 2015).

As with all bears, T. ornatus is sexually dimorphic; adult females are 67% as large as males (Mondolfi 1989; Garshelis 2009; García-Rangel 2012). Furthermore, male skulls are larger than female skulls (skull length [mean ± SD] males = 236.8 ± 12.0 mm; females = 202.2 ± 21.8 mm) and males have a well-developed lambdoid crest and a prominent sagittal crest that is absent or reduced in females (Fig. 2; Emslie 1995; Stucchi and Figueroa 2013).

Fig. 2.

—Dorsal, ventral, and lateral views of the skull and lateral view of the mandible of an adult female Tremarctos ornatus from Nariño, Colombia (FNMH [Field Museum of Natural History] 88488). Photo used with permission of the Field Museum of Natural History, Chicago, Illinois. Condylobasal length is 193 mm.

Head–body length for T. ornatus varies between 1,100 and 2,200 mm (Bininda-Emonds 1998; Nowak 1999; Garshelis 2009; García-Rangel 2012). Mean external measurements (mm; ± SD; n—Mondolfi 1971) for males from Venezuela were: total length (1,485 ± 360; 2), tail length (100; 1), ear length (100; 1), and mean weight (91 ± 68 kg; 2). Colombian males in the wild have a mean head–body length of (1,350 ± 219; 3), tail length (733 ± 20; 3), ear length (103 ± 18; 3), and weight (105 ± 60 kg; 3). Females have a mean head–body length (1,260 ± 28; 2), tail length (80 ± 14; 2), ear length (100; 1), and mean weight (37.5 ± 2 kg; 2—Rodríguez et al. 2013). Head–body length from one adult male from Ecuador was 1,312 mm, tail length 150 mm, and ear length 64 mm. Head–body length from one juvenile female in Ecuador was 270 mm and ear length 52 mm (Tirira 2009). T. ornatus from Bolivia weighed 34 kg (male subadult) and 30 kg (“undersized” adult male), but no other measurements were reported (Paisley 2001). No information is available for T. ornatus in Peru, except that Peyton (1980) estimated weights between 20 and 120 kg.

Derived from the oldest fossil specimen and four modern T. ornatus, skull and mandible measurements from T. ornatus (mm; mean ± SD—Stucchi et al. 2009) were: condylobasal length 223.16 ± 14.02, median palatal length 103.2 ± 7.9, palatal breadth at M2 41.28 ± 2.72, breadth at the labial C (upper canine) alveoli 57.62 ± 2.31, breadth at the lingual C alveoli 34.4 ± 2.06, zygomatic breadth 159.82 ± 6.9, frontal breadth 84.74 ± 8.45, least breadth between the orbits 62.22 ± 6.98, facial length 114.44 ± 2.61, mandible length 164.54 ± 6.66, height of the coronoid process 95.18 ± 7.87, basal length of the coronoid process 48.65 ± 6.64, height of the horizontal ramus at M2 35.58 ± 1.53.

Tremarctos ornatus lacks a diastema, unlike the insectivorous (e.g., the sloth bear, Melursus ursinus) and carnivorous (e.g., the polar bear, Ursus maritimus) bears (Figueirido et al. 2009; Stucchi and Figueroa 2013). The dental formula is i 3/3, c 1/1, p 4/4, m 2/3, total 42 (Ramsay 2003; García-Rangel 2012) and includes the shortest incisors in the Ursidae. The canines are bladed and an extra lateral cusp, located between the trigonid and taloned, occurs on m1 (Emslie 1995; Sacco and Van Valkenburgh 2004; Stucchi and Figueroa 2013).

The humerus of T. ornatus can be recognized by the presence of a fistula situated above the internal epicondyle (Cabrera and Yepes 1940). Davis (1958) described the tarsal ligaments of T. ornatus in comparison with human tarsal ligaments, detailing that the tarsus of T. ornatus is more flexible but is a less stable structure than in humans. T. ornatus has an enlarged radial sesamoid called “false thumb,” a feature shared with the panda. The function of this thumb is still debated (Salesa et al. 2006).

DISTRIBUTION

Tremarctos ornatus is distributed in the South American tropical Andes region (Fig. 3) in an elevational range between 200 and 4,250 m (Peyton 1980, 1981, 1999; García-Rangel 2012). During the Pleistocene, the distributional range was larger (Stock 1950). Currently the distribution is an elongated and narrow range that is about 200–650 km wide and 4,600 km long (Peyton 1999; Kattan et al. 2004; García-Rangel 2012), including six countries: Venezuela, Colombia, Ecuador, Perú, Bolivia, and Argentina (Mondolfi 1989; Peyton 1999; Garshelis 2009; García-Rangel 2012; Cosse et al. 2014; Velez-Liendo and García-Rangel 2017). T. ornatus potentially occurs in Panama, but there are no records to support its presence (Allen 1942; Hershkovitz 1957; Mondolfi 1971, 1989; Goldstein et al. 2008).

Fig. 3.

—Distribution of Tremarctos ornatus based on Velez-Liendo and García-Rangel (2017).

In Venezuela, T. ornatus is widely distributed in the Andes in discontinuous forested mountain areas in the western portion of the country (Lara, Táchira, Mérida, Portuguesa, Zulia, and Trujillo States—Osgood 1911; Mondolfi 1971, 1989; Goldstein 1992; Bisbal 1993; Soriano et al. 1999; Sánchez-Mercado et al. 2014).

In Colombia, T. ornatus has been reported from three biogeographic regions (Caribbean, Andes, and Pacific) and 20 departments (Alberico et al. 2000; Jorgenson and Sandoval-A 2005; Solari et al. 2013; Vela-Vargas et al. 2014; Cáceres-Martínez et al. 2016; Rodríguez et al. 2019a). In Ecuador, T. ornatus is distributed in the Sierra, High Amazon, and Andes regions, and is present in both ranges of the Andes and the Condor and Cucutú subranges (Suarez 1988; Castellanos 2011; Zapata-Ríos and Branch 2016). In Peru, T. ornatus is found in all three ranges of the Andes, with 973 records of presence distributed in 17 departments on an elevation range between 400 and 3,692 m (Márquez and Pacheco 2010; Falconi 2019; Falconi et al. 2020). In addition to Andean forest and páramos, T. ornatus lives in dry and humid tropical areas of less than 1,000 m elevation (Figueroa 2012; Figueroa and Stucchi 2013; Filipczyková et al. 2016; Appleton et al. 2018).

In Bolivia, T. ornatus is known from four departments: La Paz, Santa Cruz, Chuquisaca Cochabamba, and Tarija (Salazar and Anderson 1990; Anderson 1997; Martínez et al. 2008; Albarracín et al. 2013). The southern distributional extent of T. ornatus is the southern portion of Tarija department in Bolivia, coinciding with the northwestern border of Argentina (Vargas and Azurduy 2006). Derived from 18 unconfirmed sightings in northwestern Argentina, del Moral and Bracho (2005) suggested the presence of T. ornatus in the country. After several years of debate (Del Moral and Bracho 2005, 2009; Rumiz et al. 2012), Cosse et al. (2014) confirmed its presence in Jujuy and Salta provinces in northern Argentina using noninvasive samples (hair and feces) to generate genetic identification.

The potential presence of T. ornatus in Panama is based on a skull and traditional knowledge from indigenous communities (Hershkovitz 1957). Later, field surveys at the Serranía de Pirre (along the border with Colombia) were unable to confirm the presence of T. ornatus in Panama (Goldstein et al. 2008).

FOSSIL RECORD

The earliest tremarctine bears are known from late Miocene fossils in North America (Tedford and Martin 2001). Four genera of tremarctines are currently recognized: Plionarctos (extinct, two species), Arctodus (extinct, two species), Arctotherium (extinct, five species), and Tremarctos with two species, including the only extant species, T. ornatus (Soibelzon et al. 2005, 2008; Mitchell et al. 2016). The most complete fossil specimen includes skull, mandible vertebrae, ribs, hindlimbs, and forelimbs and the oldest record of T. ornatus, an adult male from Chaquil Cave, Amazonas department, Peru, dated 5,980 ± 50 radiocarbon years ago, is reported by Stucchi et al. (2009). Additional records are from two archeological sites in Colombia dated from 4,030 to 2,725 years ago, in Cundinamarca department (Urrego-Correal 1990), and one fossil record from Peru dated from 1,500 years ago in Cajamarca department (Florez 1975). These fossil records consist only of small pieces of bones (ulna, canine, metacarpus), but are the only specimens known. Skeletal remains from 700 BCE to 250 CE were excavated at La Chimba site in Ecuador (Stahl and Athens 2001).

Tremarctine is considered monophyletic by Mitchel et al. (2016), but mtDNA phylogenetic analyses suggest that Arctodus (North America) and Arctotherium (South America) lineages are not sister taxa (from a monophyletic clade), as suggested by Trajano and Ferrarezzi (1995). Tremarctos and Arctotherium are hypothesized to be sister taxa, diverging in the Pliocene. The estimated divergence estimation for Arctodus and Tremarctos is 5.66 million years ago (Krause et al. 2008). The Plionarctos linage is ancestral to Arctodus, Arctotherium, and Tremarctos but formed as a paraphyletic stem group (Tedford and Martin 2001).

Based on three fossils of an ursine bear from the late Miocene in Nebraska (United States) described as Aurorarctos tirawa gen. et sp. nov. (two mandibles, 14 isolated teeth, and one partial humerus), Jiangzuo and Flynn (2020) hypothesized the monophyly of tremarctine bears, but the clade name Arctotheriini is used to represent the crown + stem-group lineages. Tremarctinae is not eliminated but is just not used as a clade name because Jiangzuo and Flynn recognize Ursinae as the group containing not only the Ursus lineages (Ursini), but the tremarctines and the early ursine Aurorarctos.

FORM AND FUNCTION

Form.—

The skull of Tremarctos ornatus is wide and heavy with thick boney walls and high bone density (Fig. 2; Christiansen 2008a; Stucchi and Figueroa 2013). These features could be a response to an omnivorous diet with high consumption of plants (Christiansen and Wroe 2007; Christiansen 2008b; Jiangzuo and Flynn 2020). T. ornatus possesses blade-shaped canines (bite force [mean ± SD] at the canine tip = 607.52 ± 143.77 N) with enlarged molars, characteristic traits of omnivorous species (Sacco and Van Valkenburgh 2004; Christiansen and Wroe 2007). The zygomaticomandibularis of T. ornatus is relatively larger than in other bear species showing a moderate bite force in comparison with specialized bears, such as the giant panda (Ailuropoda melanoleuca—Davis 1955; Súarez 1985; Peyton 1999; Christiansen and Wroe 2007; García-Rangel 2012). In contrast, the superficial masseter is smaller than in other carnivores and differs from flesh-eating species in having a shorter, concave-shaped mandible that allows grinding and cutting movements (Davis 1955; Figueirido et al. 2009). These bone, teeth, and muscle characteristics are correlated with a predominantly herbivorous diet (Davis 1955; Christiansen and Wroe 2007; Figueirido et al. 2009; García-Rangel 2012). The tongue is wide (35 mm). The lips are large and flexible, as in all Ursidae (Davis 1955; García-Rangel 2012). Brain weight from a captive individual was 240 g (Hirata 1987).

In locomotion, T. ornatus uses only five ligaments in the talocrural articulation and a well-developed, interarticular meniscus (14 mm long by 3 mm wide), which implies lateral thrust of the astragalus against the fibula in strong inversion, such as during climbing (Davis 1958). The radial sesamoid is mediolaterally flattened, whereas the distal tip is scarcely developed, like a blunt protuberance (Salesa et al. 2006).

Length of the baculum is 10.47 mm, height at the level of the basal end is 6.1 mm, height at the level of the distal end is 3.8 mm, and maximum width is 7.9 mm; these measurements were derived from one subadult male killed in Venezuela (Mondolfi 1983). The baculum is nearly straight, gradually tapered from base to tip, and ends in a slightly enlarged blunt tip (Mondolfi 1983).

The adaptive value of the large white marks on the muzzle, upper chest, and neck, and the large white circles around the eyes is unknown; however, these regions might modulate intraspecific aggression and dominance (Caro 2009). All bears have epipharyngeal pouches. Based on the examination from one individual of T. ornatus (Forstenpointner and Weissengruber 2000), these are a single pouch and two smaller sacs that are outfoldings of the lateral pharyngeal wall. It is thought that these could play a role in the production of vocalizations via the movement of the neck and enable management of the air column (Forstenpointner and Weissengruber 2000; Weissengruber et al. 2001).

Ovaries in Tremarctos ornatus are located on the sublumbar region anterior of the kidneys (Enciso and Vásquez 2007). The mean length of the right ovary is 18.5 ± 2 mm (n = 2), 11.5 ± 2 mm width (n = 2), and 9 ± 1 mm thickness (n = 2). The mean length of the left is 19 mm (n = 2), width is 10.5 ± 2 mm (n = 2), and thickness is 8.5 ± 0.7 mm (n = 2—Lengwinat et al. 2001; Enciso and Vásquez 2007). Macroscopically, the placenta (placenta discoidalis) is disk-shaped with a length of 120 mm, width of 95 mm, thickness of 5 mm, and mass of 55.4 g (Michel et al. 1983).

Scrotum length and height are 465.0 and 629.5 mm, respectively. Testes are oval oblique to the dorsocaudal major axis (right testis 386 by 241.5 mm, left testis 476 by 241.5 mm—Sanchez-Arbouin and Nassar-Montoya 1997). Sperm measurements (mean ± SD) were: head length 5.34 ± 0.15 µm, head width 3.72 ± 0.05 µm, area of head 16.84 ± 0.55 µm², head perimeter 16.07 ± 0.32 µm, acrosome 67.70 ± 3.38%, area if intermedium piece 2.06 ± 0.22 µm², and intermedium piece width 0.95 ± 0.09 µm (Enciso et al. 2006). Sperm motility in males was recorded between 50% and 70%, with 10–15% abnormalities (Sanchez-Arbouin and Nassar-Montoya 1997; Enciso et al. 2006).

Function.—

The polar bear and Tremarctos ornatus have the lowest means for corpuscular volume (MCV) of red blood cells and the highest blood albumin level of all bears (Seal et al. 1970). Based on a sample of 62 bears of all species, T. ornatus possesses the smallest corpuscular volume with an average of 55 fl (n = 40—Seal et al. 1967; Castellanos et al. 2010). Blood cell size could be a response to seasonal changes and food availability for T. ornatus and the polar bear (Seal et al. 1967; Castellanos et al. 2010).

Male individuals (n = 23) had higher mean serum protein levels (155.6 g/l) than did females (n = 23, 132.8 g/l—Castellanos et al. 2010). Hematological values (mean ± SD) for combined sexes of captive, reintroduced, and wild animals were: cholesterol 7.98 ± 1.97 mmol/l (n = 33), total protein 77.30 ± 15.30 g/l (n= 27), triglycerides 7.33 ± 1.64 mmol/l (n = 34), blood urea nitrogen 4.77 ± 1.51 mmol/l (n = 29), glutamic oxalic transaminase 30.23 ± 20.68 U/l (n = 38), glutamic pyruvic transaminase 21.43 ± 19.41 U/l (n = 39), alkaline phosphatase 97.57 ± 58.21 U/l (n = 31), calcium 1.87 ± 0.28 mmol/l (n = 32), phosphorus 1.68 ± 0.70 mmol/l (n = 35), glucose 3.56 ± 1.46 mmol/l (n = 18), urea 9.85 ± 4.48 mmol/l (n = 38), hematocrit 0.43 ± 0.05 l/l (n = 45), hemoglobin 144.45 ± 20.89 g/l (n = 46), leukocytes 9.11 ± 2.98 × 10⁹/l (n = 44), erythrocytes 7.87 ± 1.52 × 10¹²/l (n = 37), segmented leukocytes 6.47 ± 0.94 × 10⁹/l (n = 45), lymphocytes 2.20 ± 0.77 × 10⁹/l (n = 44), monocytes 0.13 ± 0.13 × 10⁹/l (n = 46), eosinophiles 0.16 ± 0.22 × 10⁹/l (n = 43), basophiles 0.01 ± 0.05 × 10⁹/l (n = 43), band cells 0.027 ± 0.11 × 10⁹/l (n = 40), mean cellular hemoglobin concentration 334.17 ± 43.10 g/l (n = 36), mean cellular hemoglobin 18.33 ± 3.03 pg (n = 38—Nassar-Montoya et al. 1997; Castellanos et al. 2010).

ONTOGENY AND REPRODUCTION

Length at birth is 225–280 mm and weight is 300–500 g (Saporiti 1949; Roth 1964; Dathe 1967; Castellanos et al. 2016a). Neonates are black and toothless at birth, with closed eyes. Eyes open completely by day 31 (Saporiti 1949). The only record of growth rate for young Tremarctos ornatus was presented by Saporiti (1949) as 50 mm/day.

Most of the information available for reproduction of T. ornatus has been obtained from captive individuals (Saporiti 1949; Roth 1964; Gensch 1965; Bloxam 1977; Michel et al. 1983; Kuhme 1991; Lengwinat et al. 2001; García-Rangel 2012). T. ornatus is a polyestrous species with facultative seasonal reproduction (Saporiti 1949; Mondolfi 1971; Spady et al. 2007; Enciso 2013) and is capable of embryonic diapause (Lengwinat et al. 2001; Knauf et al. 2003; Enciso 2013). After diapause, gestation length is short and difficult to calculate (Michel et al. 1983; Rosenthal 1987; Spady et al. 2007; García-Rangel 2012).

Captive females present 3–4 phases of ovarian activity per year (Enciso 2013). Wild females have an interval of 9 months between estrous cycles with a maximum of three cycles in 24 months (Spady et al. 2007). Duration of estrus is estimated at 5 days (Spady et al. 2007) but can vary depending on the latitude and associated photoperiod cycle of the facility (Knauf et al. 2003). Captive females from Brazil and Colombia had estrous cycles that lasted 3–10 days (Sanchez-Arbouin and Nassar-Montoya 1997; Enciso 2013).

Gestation varies from 120 to 254 days (Saporiti 1949; Gensch 1965; Bloxam 1977; Kuhme 1991; Castellanos 2015). Castellanos (2015) reported the shortest gestation period of 120–125 days, similar to the period reported for the sun bear (Frederick et al. 2012). Litter size varies from 1 to 3 young. Sexual maturity in females is attained at 4–7 years of age (Saporiti 1949; Dathe 1967; Bloxam 1977; Rosenthal 1987).

Tremarctos ornatus displays seasonal reproductive activity in wild individuals that corresponds to resource availability, but in captive individuals is associated with photoperiod (Appleton et al. 2018). Births are concentrated in autumn in tropical latitudinal zones (< 23.5°N or S), whereas in mid-temperate latitudinal zones (35–55°N or S) most births occur during the winter (Peyton 1980; Rosenthal 1987; Spady et al. 2007; García-Rangel 2012; Appleton et al. 2018).

ECOLOGY

Population characteristics.—

The few attempts at population estimates for Tremarctos ornatus have been derived from species such as the American black bear or from genetic analyses; some local approaches have been performed. The first approximation for population estimation was made by Peyton et al. (1998), resulting in a total wild population of 18,000–65,000 individuals across the entire range. Later, based on genetic analyses Ruiz-Garcia (2003) estimated a global population size of 19,000–24,000 individuals. The majority of density estimations published before 2011 and mentioned in this account are discussed in detail by Garshelis (2011). Field-based density estimates for T. ornatus are preliminary and characterize sites from across its geographical range. In Venezuela, extrapolating from American black bear density data, Yerena and Torres (1994) estimated 0.04 individuals/km², values that Cáceres-Martínes et al. (2018) validated with field data in northern Colombia. Density was estimated at 0.03 individuals/km² in the eastern range (Cordillera Oriental) of the Andes in Colombia (Réyes et al. 2018; Rodríguez et al. 2019b; Rodríguez et al. 2020). Viteri (2007) calculated densities between 0.03 and 0.07 individuals/km² and Molina et al. (2017) determined a similar value of 0.074 individuals/km², whereas an estimate of 0.039 individuals/km² was obtained by Morrell (2018) in northern Ecuador. Density calculations in Bolivia suggested slightly lower densities of 0.04–0.06 individuals/km² (Ríos-Uzeda et al. 2007).

The greatest distance recorded for daily movements by an individual is 15 km in Bolivia (Rechberger et al. 2001). Daily movements between 800 and 2,435 m in elevation are common in Bolivia (Paisley 2001).

No records of longevity exist in wild individuals, but in captivity, T. ornatus can live as long as 40 years (Castellanos et al. 2016a). Sexual maturity in captivity is attained at 3–7 years (mean age for females = 4 years, mean age for males = 5 years—Rodríguez-Clark and Sánchez-Mercado 2006; García-Rangel 2012; Arias-Bernal and Yarto-Jaramillo 2019). Annual survival rates for neonates in captive populations are low (59–64%—Rodríguez-Clark and Sánchez-Mercado 2006). Fertility declines after 15 years of age and decreases rapidly after 25 years of age (Rosenthal 1987; Rodríguez-Clark and Sánchez-Mercado 2006).

Space use.—

Tremarctos ornatus uses several habitat types, from scrub desert to subalpine páramos at elevations from 200 to 4,170 m (Peyton 1980, 1981). T. ornatus shows a preference for higher elevation forest types such as montane humid forest, elfin, and Andean cloud forests (Mondolfi 1989; Cuesta et al. 2003; Ríos-Uzeda et al. 2006; Vela-Vargas et al. 2014; Filipczyková et al. 2016), high-altitude grasslands (Goldstein 2002), páramos (Yerena and Torres 1994; Cuesta et al. 2003), pajonales, Polylepis forest (Azurduy and Velez-Liendo 2001), and oak forests (Otálora-Ardila 2003). Use of lower elevation habitats is rare, but Figueroa (2012) recorded seasonal use of tropical Amazon forest in Peru and tropical dry coastal forests (Appleton et al. 2018; Kleiner et al. 2018), with elevation ranges between 140 and 1,300 m.

Home ranges of two males in Bolivia were 6.6 and 7.4 km² (Paisley 2001). In Ecuador, home range estimates for a reintroduced male were larger than the average of two females (male = 61 km², female = 4.1 km²—Castellanos 2003; Castellanos et al. 2005). Home range nearest-neighbor convex hull (K-NNCH) estimates for wild males (n = 3; 59 ± 4.33 km²) were on average three times larger than for females (n = 5; 15 ± 5.35 km²—Castellanos 2011).

Diet.—

Feeding strategies of Tremarctos ornatus vary based on four habitat characteristics: fruiting events, resource availability, geographical movements, and local migration (Figueroa 2013a; Gonzales et al. 2016; Cáceres-Martínez et al. 2020). Although its main diet is based on plant consumption, with 314 plant species identified (Chung 2006; García-Rangel 2012; Figueroa 2013a; 2013b), T. ornatus is considered to be an opportunistic omnivore (Figueroa 2013b; Gonzales et al. 2016), with high preference for consumption of plant meristematic tissues and fruits (Peyton 1980; Garcia and Betancur 2002; Goldstein 2004; Troya et al. 2004). Bromeliads (Bromeliaceae), palms (Araceae), bamboo (Poaceae), and fruits from Ericaceae and Lauraceae are the principal food items identified in the diet (García-Rangel 2012; Figueroa 2013a; Cáceres-Martínez et al. 2020). In dry areas of northern Perú, T. ornatus frequently consumes cacti, including: Browningia microsperma, Melocactus peruvianus, Echinopsis pachanoi, Neoralmondia arequipensis gigantea, Opuntia ficus-indica, Echinocactus, and Mammillaria (Figueroa and Stucchi 2008). T. ornatus is considered an agent to disperse seeds of species such as Nectandra cf. cuneatocordata, Symplocos cf. cernuay, and Gaultheria vaccinioides (Rivadeneira-Canedo 2008), and other species of Lauraceae (Peyton 1987) and Styrax ovatus (Styracaceae—Young 1990). T. ornatus consumes domestic crops such as corn (Zea mays), sugar cane (Saccharum officinarum), and plantain (Musa paradisiaca), generating conflicts with human communities (Peyton 1980; Peyton et al. 1998; Jorgenson and Sandoval-A 2005; Figueroa 2015a; Albarracín and Aliaga-Rossel 2018; Escobar-Lasso et al. 2020). In northern Colombia, T. ornatus showed a greater niche breadth during the dry season (July–August and December–February) than in the wet season (March–June and September–November) and low overlap in the diet as evaluated using the Pianka Index (O = 0.12—Cáceres-Martínez et al. 2020).

Different species of mammals were identified in scat samples from T. ornatus, with small rodents (Akodon, Microryzomys, and Cavia cf. tschudii) being most abundant (Gonzales et al. 2016). Other mammals registered in scats are medium-sized rodents, such as the Central American agouti (Dasyprocta punctata), nine-banded armadillo (Dasypus novemcinctus), coatis (Nasua nasua and Nasuella olivacea), and vicuñas (Vicugna vicugna—Cáceres-Martínez et al. 2020; Hernani-Lineros et al. 2020). Attacks on mountain tapirs (Tapirus pinchaque) have been reported in Colombia (Rodríguez et al. 2014) and scavenging on white-tailed deer (Odocoileus virginianus) reported in Peru (Van Horn et al. 2014b). T. ornatus is considered to be a facultative carnivore that preys upon and scavenges both live animals and dead domestic animals, such as donkeys, cows, and sheep (Goldstein 2002; Goldstein et al. 2006; Figueroa 2013b; Parra-Romero et al. 2019). Fragments of plastic have also been recorded in scats at Tamá National Park, Colombia (Cáceres-Martínez et al. 2015).

Diseases and parasites.—

In captive conditions, Tremarctos ornatus tends to develop alopecia syndrome, leading to partial or complete loss of hair due to unknown causes (Nassar-Montoya et al. 1997; Owen et al. 2009; Nicolau et al. 2018; Arias-Bernal and Yarto-Jaramillo 2019). The syndrome mostly seems to affect females in captive social housing conditions and has not been reported in wild individuals (Barbon 2013; Van Horn et al. 2019). Multiple types of neoplasia have been reported, including transitional cell carcinoma (Chandrasekar et al. 2016) and thymoma (Cooper 1999), and other types of carcinomas (Nassar-Montoya et al. 1997; Arias-Bernal and Yarto-Jaramillo 2019).

Nematodes such as Baylisascaris transfuga, Baylisascaris venezuelensis, Strongyloides, Ancylostoma, and Ascaris have been detected in both captive (Schaul 2006; Luzuriaga 2014) and wild individuals (Figueroa 2015b; Guerrero and Castellanos 2016; Pérez Mata et al. 2016). Castellanos et al. (2005) reported the potential death of one reintroduced T. ornatus in Ecuador due to tick fever caused by the hemoparasite Babesia. Distemper virus (Morbillivirus) resulted in the death in Europe of one captive T. ornatus cub (Schönbauer et al. 1984). Recorded protozoa parasites in wild and captive individuals include Blastocystis, Cryptosporidium, Giardia, Coccidia, and Entamoeba (Luzuriaga 2014; Figueroa 2015b). Only one record of ticks has been reported in T. ornatus (Trichodectes ferrissi—Cardozo-de-Almeida et al. 2003).

Interspecific interactions.—

Interactions with both vertebrates and invertebrates have been reported. Tremarctos ornatus and mountain tapirs are frequently recorded in sympatric conditions in high-altitude ecosystems, such as Andean forest and páramo (Castellanos 2014), but one record of an attack by a T. ornatus on a mountain tapir exists (Rodríguez et al. 2014). When feral dogs (Canis familiaris) are present in areas where T. ornatus is present, occupancy rates of T. ornatus decrease and its diurnal activity patterns change too (Zapata-Ríos and Branch 2016, 2018). Dung beetles (Uroxys brachialis, Uroxys deppressifrons, and Uroxys nebulinus: Scarabaeidae) fed on the dung of T. ornatus in the forest at La Planada Natural Reserve, southern Colombia (Escobar 2003).

HUSBANDRY

Tremarctos ornatus has been exhibited in captivity since 1903 when the first individual was displayed at Amsterdam Zoo (Cooper 1999). The captive population worldwide for 2017 was 255 individuals (129 males, 114 females, and 12 individuals with no reported sex) housed in 163 institutions (Anonymous 2017; Hall 2017).

Traditional chemical restraint protocols include ketamine (3–8 mg/kg body weight) and xylazine (2 mg/kg) and Yohimbine (Reversine S.A., Australia, New South Wales) as the reversal agent (0.1–0.25 mg/kg—Castellanos et al. 2016a). Protocols used at Bioparque Wakatá, Colombia, include ketamine (5.6–7.3 mg/kg), dexmedetomidine (0.02–0.035 mg/kg), and midazolam (0.27–1.01 mg/kg) with the reversal agent atipamazole (0.2 mg/kg), flumazenil (0.01 mg/kg) via intravenous, n = 5; tiletamine–zolazepam (6–6.5 mg/kg), ketamine (2–6.2 mg/kg) with reversal agent flumazenil (0.01 mg/kg) via intravenous, n = 6; ketamine (4 mg/kg), medetomidine (0.04 mg/kg), and midazolam (0.1 mg/kg) with reversal agent atipamezole (0.24 mg/kg), flumazenil (0.01 mg/kg), n = 2 (Paisley 2001; Bourne et al. 2010; Arias-Bernal and Yarto-Jaramillo 2019).

Captive feeding protocols vary (Rosenthal 1987). At La Planada Natural Reserve (Colombia), captive individuals were fed with a corn soup base (milk, raw sugar honey, vegetables, calcium, eggs, meat, and liver emulsion). At Cologne Zoo (Germany), fruits and vegetables were offered, including carrots, apples, pears, and grapes. Based on wild diets, a solid diet called “chapo” was formulated for rehabilitation purposes (Castellanos et al. 2016a). For a 60-kg bear, daily quantities were: 3 kg of mixed fruits (banana, papaya, and seasonal staples), guava (500 g), oat (340 g), wheatgerm (260 g), and mineralized salt (30 g). As the animal rehabilitates, the formula is adjusted over time with increasing amounts of wild food items (Castellanos et al. 2016a).

In a feeding trial, two captive animals consumed 1.6% of their body weight in dry matter daily, with an average digestibility of 60.5%. Apparent digestion coefficients for neutral detergent fiber, crude protein, soluble sugars, and crude fat were 12.8%, 70%, 80.3%, and 64%, respectively. Gastrointestinal transit was measured at 8–24 h, leading researchers to hypothesize that wild T. ornatus consumed high-fiber foods, whereas captive bears tended to select low-fiber ingredients in their diets (Goldman et al. 2001).

Cage requirements for captive T. ornatus are listed in Bloxam (1977), with an outdoor space of 267 m², a swimming pool 32.5 m² in diameter and a depth of 1.5 m. Sleeping dens should be on average 3.2 m in lenght by 2.65 m in width by 2.4 m in height (Bloxam 1977; Peel et al. 1979). Requirements for rehabilitation enclosures were reviewed extensively in Castellanos et al. (2016a).

BEHAVIOR

Grouping behavior.—

Tremarctos ornatus is generally solitary, but forms temporary pairs during mating periods (Mondolfi 1971; Peyton 1999; Castellanos et al. 2005; García-Rangel 2012; Appleton et al. 2018). When food resources are concentrated, T. ornatus sometimes forms rudimentary social groups to exploit crops or crop waste (Figueroa 2015a), or to scavenge livestock (Paisley 2001; Parra-Romero et al. 2019).

Reproductive behavior.—

Descriptions of mating behaviors are scarce and focus mostly on captive individuals. Normal behavior includes nonaggressive fights and multiple copulations (2–8 copulations) over intervals of 8–60 min (Bloxam 1977).

During pregnancy, females construct and use dens. The den is about 2 m long and 1 m wide with a sheltered depression in the middle that is 1.5 m long, 800 mm wide, and 500 mm deep (Castellanos 2010). The area surrounding the den is clean with no feces or strong odors present in order to avoid attracting potential predators. The den is used for about 9–12 weeks after the birth (Mondolfi 1971; Castellanos et al. 2016a). Dens may be used throughout the year as the young mature (Castellanos 2010). While nursing, females may leave maternal dens for up to 48 h to seek food. The length of time away from the young may be due to anthropogenic effects of habitat loss, with greater times away in areas with greater habitat loss. Adult females may even abandon young for extended periods of time due to lack of resources (Castellanos et al. 2019).

Communication.—

Males apply chemical scent-marks to a variety of tree species for self-advertisement and potentially to mark territories (although this is an untested hypothesis), whereas females do not scent-mark perhaps due to risk of cannibalism by males (Filipczyková et al. 2016). In dry areas of Peru, Tremarctos ornatus concentrates tree rubbing areas close to waterholes (Kleiner et al. 2018).

Sound spectrograms show that vocalizations are composed of tonal and atonal elements with ranges from 0.01 to 7 kHz. Two reintroduced T. ornatus, produced six types of sounds that were characterized as follows: guttural sounds when individuals were curious (kurrrrr, tuutucttt), screams when the animals fought for food, small chuffs when animals attacked other individuals, and whining sounds (eggmmmmm) when the animals were found in the forest canopy. A male produced a particular sound when masturbating (MMrnnMMrnn). This unique sound could be heard at least 30 m away (Castellanos 2003; Castellanos et al. 2005). Captive cubs produced a particular vocalization called humming, which had a main frequency < 2 kHz and was produced on average over 3.6 s, associated with comfort or contentment (Peters et al. 2007).

Communication between mother and young was recorded by Elowson (1988) in captive conditions. Trill calls (n = 23) from the mother were recorded with a peak frequency (mean ± SD) of 418 ± 39 Hz and growls (n = 2; 175 ± 35 Hz). Cub sounds were characterized into five types including trills (n = 33; 375 ± 104 Hz), yelp (n = 15; 817 ± 331 Hz), whimper (n = 39; 347 ± 181 Hz), squeals (long and short) with frequency peaks, respectively, of 2,000 ± 1,471 and 1,493 ± 718 Hz, and screams (long and broad [n = 15], short and broad [n = 20], long narrow [n = 9]) with frequencies of 1,312 ± 648 Hz, 1,279 ± 726 Hz, and 328 ± 86 Hz, respectively.

Miscellaneous behavior.—

Tremarctos ornatus is mostly diurnal with activity peaks between 600 and 1000 h (Paisley and Garshelis 2006; Parra-Romero et al. 2019). In Ecuador, activity patterns changed from a bimodal distribution of activity in absence of feral dogs to more diurnal activity concentrated between 1200 and 1400 h if feral dog packs were present (Zapata-Ríos and Branch 2016).

Cannibalism is rare, but two cases have been reported in Ecuador. In one instance, a wild T. ornatus was shot by a hunter and later consumed by other bears (Castellanos 2006). Another incident occurred at Cayambe Coca National Park (Ecuador), where a reintroduced T. ornatus female was found dead with signs of predation attributed to a large male bear (Castellanos et al. 2016b).

Tree nests consist of a platform and an oval depression at the top of the nest; different tree species, such as Benchesia, Ficus, Cedrela, Clusia, and Podocarpus are used. Nest locations have been associated with consumption of fruit, resting, and areas close to livestock (Peyton 1980; Goldstein 1991, 2002; Peyton et al. 1998).

GENETICS

Cytogenetics.—

Although all the Ursine bears have 74 chromosomes and Ailuropodinae has 42, the karyotype of Tremarctos ornatus is completely different (Nash and O’Brien 1987; Garshelis 2009). T. ornatus has a diploid number (2n) of 52 largely biarmed chromosomes, with 16 pairs of metacentric autosomes and nine pairs of acrocentric autosomes (Newnham and Davidson 1966, 1967; Nash and O’Brien 1987). The X chromosome is relatively large and metacentric, and the Y chromosome is smaller and acrocentric (Newnham and Davidson 1967).

Tremarctos ornatus diverged from an ancestor with the ursine karyotype (2n = 74) about 10.91 million years ago (9.93–11.89 million years ago) and differs from that of the presumed ancestor by two fusions, two fissions, 11 centric fusions, and one inversion (Nash and O’Brien 1987; Tian et al. 2004; Wienberg 2004; Yu et al. 2007; Garshelis 2009).

Molecular genetics.—

Mitochondrial genome length was 16,766 base pairs in Tremarctos ornatus (Yu et al. 2007). Fecal sampling using the PCR amelogemin gene with primers SE47-SE48 had a 100% success rate in determining the sex of the individuals (Caselli and Maturrano 2016).

Genetic diversity parameters of microsatellite data showed that T. ornatus had a moderate mean number of alleles per microsatellite locus and moderate expected heterozygosity (H_e 0.382 ± 0.298—Ruiz-Garcia 2003; Ruiz-Garcia et al. 2003; Ruiz Garcia et al. 2005; Viteri 2007; Zhang et al. 2007). Haplotype diversity calculations for the Ecuador population were 0.5–0.705 (Viteri and Waits 2009; Cueva et al. 2018) and overall for South America were calculated as 0.980 ± 0.016 (Arias-Vásquez 2017). Recent studies revealed that T. ornatus is divided into two evolutionary significant units (ESUs) with an overall haplotype diversity of 0.914 ± 0.013 and nucleotide diversity (π) of 0.0191 ± 0.0126, average microsatellite H was 0.55, lower than in other bear species (Ruiz-García et al. 2020a).

For the 551 individuals in captivity, expected heterozygosity was calculated as H_e = 0.97, with an inbreeding coefficient of 0.20 (Rodríguez-Clark and Sánchez-Mercado 2006). These values are the result of a pedigree analysis, considering only live, reproductive individuals and their direct ancestors, these results are based on software analyses and not derived from any genetic sampling approximation.

Population genetics.—

Population genetics for Tremarctos ornatus is poorly known. Populations do not appear to be in Hardy–Weinberg equilibrium and a high level of homozygosity and genetic isolation exists among populations (Ruiz-Garcia 2003; Ruiz Garcia et al. 2005).

Genetic structure of T. ornatus in Ecuador showed high haplotypic diversity (H_d = 0.845 ± 0.033), whereas microsatellite diversity was moderate (H = 0.584 ± 0.188), demonstrating that genetic structure of the species in Ecuador is homogeneous and has not been affected by the presence of the Andes range, suggesting that T. ornatus in Ecuador should be considered as an unique management unit (Ruiz-García et al. 2020c). The same study was performed in Colombia (Ruiz-García et al. 2020b) and showed a similar haplotypic diversity in a sample of 108 wild individuals (H_d = 0.895 ± 0.018) with higher microsatellite diversity than in Ecuador (H = 0.645 ± 0.211).

Arias-Vasquez (2017) identified three well-differentiated phylogeographic groups: 1) southern Colombia, 2) northern Andes (Venezuela, Colombia, Ecuador, and northern Peru), and 3) southern Andes (southern Peru and Bolivia). Based on a sample of 294 wild individuals, two evolutionary significant units were identified: a Northern Andean Clade (Venezuela, Colombia, Ecuador, and middle-northern Peru) and the Southern Andean Clade (southern Peru and northern and middle area of the Bolivian Andes—Ruiz-García et al. 2020a). Ruiz-García et al. (2003) estimated the effective population size as N_e = 19,681 individuals across the range of the species.

CONSERVATION

Tremarctos ornatus is assessed as “Vulnerable” (VU) by the International Union for Conservation of Nature and Natural Resources Red List of Threatened Species, maintaining the same category since 1982, and it has been included in CITES (Convention on Trade in Endangered Species of Wild Fauna and Flora) Appendix I since 1975 (Velez-Liendo and García-Rangel 2017). The population trend is estimated to be decreasing, with about 2,500–10,000 individuals across its geographic distribution (Velez-Liendo and García-Rangel 2017). All countries have implemented hunting bans; however, enforcement often is poor (García-Rangel 2012; Figueroa 2014). Nevertheless, conservation action plans have been proposed for Venezuela, Colombia, and Ecuador, a national assessment has been prepared for Bolivia, and education and conservation programs have been conducted in Bolivia, Peru, Colombia, and Venezuela (Ministerio Del Medio Ambiente 2001; Rodríguez et al. 2003; SERFOR 2016; Velez-Liendo and García-Rangel 2017; Parques Nacionales Naturales de Colombia and Wildlife Conservation Society 2018; Ministerio del Ambiente y Agua del Ecuador 2020). Most programs, however, encounter implementation challenges due to a shortage of funding.

The most significant threats for T. ornatus across its range include habitat loss and fragmentation, illegal killing, and conflicts with humans, whether real or perceived (Kattan et al. 2004; Sánchez-Mercado et al. 2008; Jampel 2016; Zukowski and Ormsby 2016; Velez-Liendo and García-Rangel 2017; Bazantes-Chamorro et al. 2018; Parra-Romero et al. 2019). Nonetheless, climate change also is a critical threat for T. ornatus, especially through the impact on the upper elevation ecosystems that it uses (Velez-Liendo and García-Rangel 2017). Current estimates of habitat loss have identified Venezuela as the country with the greatest projected loss, followed by Peru, Colombia, and Ecuador (Kattan et al. 2004; Sánchez-Mercado et al. 2014; Velez-Liendo and García-Rangel 2017; Bazantes-Chamorro et al. 2018; Guerrero-Casado and Zambrano 2020). Habitat loss is mainly due to inappropriate agricultural practices, expansion of the agricultural and cattle frontier, ineffective land and agrarian reforms and violence control, expansion of illegal crops, and mining and oil exploitation. Roads and other human infrastructure also fragment the habitat (Armenteras et al. 2003, 2011; Dávalos et al. 2011; García-Rangel 2012; Velez-Liendo et al. 2014; Chadid et al. 2015; Velez-Liendo and García-Rangel 2017; Zapata-Ríos and Branch 2018). Illegal killing by livestock owners is mostly due to retaliation following depredation, but T. ornatus also is harvested for illegal commercial trade and for use in traditional medicine. In Perú illegal trade of bear fat and body parts was recorded in 27 rural markets in 14 regions of the country with a range of prices between $0.90 (20 ml of fat), $14.50 (180 ml of blood), and $43.00 (skin—Figueroa 2014). No estimates exist for the number of individuals killed, but substantial numbers may be taken (Figueroa 2014; Sánchez-Mercado et al. 2014; Velez-Liendo and García-Rangel 2017; Andrade et al. 2019).

Overall, conservation of T. ornatus represents a significant challenge for all the countries across its geographic distribution. Not only are the threats continuing to increase along its geographic range, the intrinsic vulnerability of the specific habitats in which T. ornatus thrives represents a major challenge to preserve (Crespo-Gascón and Guerrero-Casado 2019; Guerrero-Casado and Zambrano 2020).

ACKNOWLEDGMENTS

Thanks to the following: Dr. Noe de la Sancha for his assistance with the pictures of the skull from the Field Museum of Natural History and Angela Hurtado for her assistance on the photo editing process; Luis Guillermo Linares for allowing use of the photograph of a wild specimen from Chingaza National Protected Area in Colombia; Dr. Peter Lurz for his help in the translation of several German documents; Dr. Jairo Pérez-Torres for facilitating information about the species; and Dr. Pamela R. Owen by her comments and clarifications on the fossil record. JPJ wishes to acknowledge the following individuals for their support over the years: Jorge Ignacio “El Mono” Hernández-Camacho (deceased), Federico Medem-Medem (deceased), Bernard Peyton, Ernesto Barriga-Bonilla, José Vicente Rodríguez-Mahecha, and Amanda Barrera de Jorgenson. IMV-V acknowledges the financial support provided by the COLCIENCIAS-Fulbright program in Colombia and the College of Agriculture and Life Sciences of the University of Arizona. The findings and conclusions in this article are those of the author(s) and do not necessarily represent the views of the U.S. Fish and Wildlife Service. The authors acknowledge the comments from two anonymous reviewers and Dr. Meredith Hamilton.

Version of Record, first published online July 15, 2021, with fixed content and layout in compliance with Art. 8.1.3.2 ICZN.

Synonymies completed 1 April 2021

Nomenclatural statement.—A life science identifier (LSID) number was obtained for this publication: urn:lsid:zoobank.org:pub: 146B87F3-E6A7-4A49-8CFD-140C6DED3DE9