The yellow-throated marten moves through Asian forests with a striking golden bib and a remarkably varied appetite. This wild mustelid hunts, scavenges and eats seasonal fruit. From snowy Russian woodland to tropical islands, its daily routines and feeding opportunities change with the landscape.
Natural-history reference: Kharza is another name for the yellow-throated marten. Its accepted range includes continental Asia and populations on Taiwan, Hainan and the major Sunda islands. Forest is its principal setting, with documented excursions into alpine meadow and modified vegetation. Its food includes live prey, carrion and fleshy fruit.
Classification: Animalia; Chordata; Mammalia; Carnivora; Mustelidae; Guloninae; Martes flavigula (Boddaert, 1785).
The marten genus sits within the mustelid family. In the Mammal Diversity Database, subgenus Charronia contains this species and the Nilgiri marten, Martes gwatkinsii. The latter is a distinct Western Ghats species. Taiwanese and Javan yellow-throated martens remain within M. flavigula under this treatment.
Small details with ecological meaning: Korean winter trails show martens inspecting empty bird nests as well as spaces beneath fallen wood. In central China, seed passage affected germination differently among plant species. A Thai carcass observation even captured an encounter with a hawk eagle competing for the same food.
Can yellow-throated martens live above the tree line?
Yes. Summer cameras in Nepal recorded them in alpine meadow well uphill from forest. Year-round occupation of those sites remains unconfirmed.
Are they strictly nocturnal?
No. Thai tracking and Russian cameras recorded predominantly daylight activity, with some twilight and nighttime movement.
Do they always hunt in pairs?
No. Solitary animals and larger groups are also documented. A witnessed paired macaque hunt demonstrates cooperation without showing that all hunting is social.
Is the yellow-throated marten endangered?
Its global category is Least Concern. Local declines, road traffic, fragmentation and hunting remain concerns in parts of its range.
Recognizing the yellow-throated marten
A golden neck and chest, dark head and legs, and a long dark tail give the yellow-throated marten an unusually bold pattern. The back can be yellowish or light brown, making the dark extremities especially conspicuous. Its elongated body is muscular and flexible. Males are generally larger than females.
The Martes Working Group gives approximate adult measurements of 45–60 centimeters for the head and body, with another 30–40 centimeters of bushy tail, and a mass of 2–4 kilograms. Yushan National Park’s Taiwanese account gives narrower measurements: 44-46 centimeters for the body and 35-37 centimeters for the tail. Both accounts are descriptive overviews, without population measurement samples accompanying the ranges.
In Taiwan, Yushan’s description includes a yellowish-brown back that darkens toward the tail, black feet, a bright neck and a white area around the jaw and throat. Animal Diversity Web also describes summer fur as darker and duller than winter fur.
One species across very different Asian landscapes
Martes flavigula, also called the kharza, belongs to the marten genus Martes. Its western range reaches Afghanistan and Pakistan and continues through Himalayan regions of India, Nepal and Bhutan, with Bangladesh farther east. Across China, its distribution extends toward the Korean Peninsula and the southern Russian Far East. These country records describe a broad geographic envelope, within which occupation depends on local habitat.
The species was first described as Mustela Flavigula; Nepal is its fixed type locality, the geographic reference attached to the name. Historical classifications have also placed it in Charronia and Lamprogale. Modern treatment retains Martes flavigula and recognizes Charronia as a subgenus shared with the Nilgiri marten. Names in older literature can consequently differ even when they concern this same species.
Farther south, yellow-throated martens occur through Myanmar, Thailand, Laos, Cambodia and Vietnam, and into Malaysia and Brunei. Island populations inhabit Taiwan, Hainan, Borneo, Sumatra and Java. The Mammal Diversity Database includes these populations within the same species account. Its name history includes chrysospila from Taiwan and robinsoni from Java, as well as lasiotis. The Nilgiri marten has its own species entry; those island populations remain within M. flavigula.
Forests occupied by yellow-throated martens span several climates. In the Russian Far East, the IUCN specialist account associates them particularly with mixed Manchurian coniferous and broad-leaved forests. They occur less frequently in the dark coniferous taiga of higher elevations and remain active through snowy northern winters. Taiwan’s Yushan account emphasizes mountains around and above 2,000 meters, including broad-leaved, coniferous and arrow-bamboo forest. In central China’s Houhe reserve, conifers, broad-leaved trees and bamboo occur in a subtropical landscape with a pronounced late-year fruit season.
The species also ventures well above forest. A 2012 camera survey in Nepal’s Kanchenjunga Conservation Area recorded it between 3,252 and 4,510 meters. Three detections at the highest station occurred in late July and August. Some alpine-meadow records lay roughly 1,400–1,700 meters from the nearest forest patch downhill. These summer encounters show substantial use of open mountain terrain, without establishing year-round residence at those elevations.
Telling it apart from other martens
The Nilgiri marten is the closest comparison in overall appearance, with a lemon-yellow bib and a dark brown coat that becomes rufous around the shoulders and toward mid-body. Its range is confined to the Western Ghats. Published observations there include pairs, trios and a pair killing an Indian chevrotain. Joint hunting consequently cannot identify a yellow-throated marten by itself; locality and the extent of pale body coloration are more useful distinctions.
Beech martens range through Europe and central Asia into the Himalayas, where geographic proximity can make comparison relevant. They are generally brown to gray-brown, with a white or buffy throat-to-chest marking, although that marking varies regionally. They also often use buildings. The broad golden body coloration of a yellow-throated marten is a stronger distinction than a pale bib alone.
The sable offers a northern comparison. It inhabits northern Asian forests and taiga, and is predominantly brown with a variable pale throat patch. Its shorter tail relative to the body produces a different silhouette.
Searching the forest from roots to tree hollows
Korean winter tracking reveals how the marten investigates a forest rather than simply passing through it. Trails led beneath fallen logs and branches, around uprooted tree bases and living roots, and into small ground holes. The animals also climbed to inspect tree hollows and empty bird nests.
Across thirteen tracked routes totaling 49.8 kilometers during three winter seasons between January 2011 and February 2013, travel and marking traces were concentrated along ridges. Searching occurred in valleys, on slopes and at forest edges. A ridge record can consequently represent movement between feeding places, while searching traces elsewhere identify places investigated for food. Rock crevices, burrows and hollow trees are reported resting places.
Yellow-throated martens also enter altered vegetation. Wildlife Conservation Society Indonesia records visits to plantations and gardens alongside use of tall and secondary forest. The IUCN account includes degraded forest and some oil-palm settings in Southeast Asia. Such visits demonstrate flexibility within a forested landscape; whether a plantation can sustain a population independently of native forest is a separate question.
Food choices follow local opportunities
Yellow-throated martens eat both animal and plant foods, but their menu changes with place and season. At Houhe in Hubei, researchers identified 53 food types and 603 food items in 276 scats. Mammals, birds, invertebrates, snakes, frogs and fruit all appeared. Rodents supplied the principal mammalian component, including Leopoldamys edwardsi, Berylmys bowersi and species of Niviventer.
Small mammals were eaten throughout the sampled year and led the diet during much of the January–August sampling. Fruit became prominent from September into January. The important detail is that rodents were also abundant when fruit consumption peaked. This seasonal switch therefore suggests an attraction to readily available ripe fruit, rather than a forced response to a disappearance of mammalian food.
Many Houhe plants produced fleshy fruit between August and November. The martens consumed Chinese kiwifruit, Actinidia chinensis, along with Sorbus hemsleyi, Diospyros lotus and Hovenia dulcis, among others. Different ripening schedules spread opportunities across the season. Bird remains appeared in later winter, spring and summer samples, while invertebrates and snakes were frequent in samples collected from April to October.
The apparent importance of a food depends partly on how it is measured. At Houhe, invertebrates occurred in 51.8 percent of scats but accounted for only 1.3 percent of reconstructed fresh food biomass. A small insect meal can leave recognizable remains without contributing much mass.
Wasps, wax and small vertebrates
In South Korea’s Jirisan National Park, analysis of 952 scats collected from 2009 to 2011 identified twelve insect species from eight families and three orders. Social wasps, including hornets and yellowjackets, were prominent. Wasps dominated the autumn insect component, while beetles and orthopterans also appeared in summer and autumn. Winter and spring insect remains in this collection belonged to the wasp family Vespidae.
Bee wax occurred in all four seasons, with its strongest representation from spring into early summer. This places nest-associated material alongside the identifiable insect foods, although wax in feces does not establish its nutritional contribution. These seasonal patterns describe the insect portion of the Jirisan diet.
A different method revealed vertebrate foods in northeastern China. In a 2021 survey in Northeast Tiger and Leopard National Park, DNA analysis assigned seven scats to yellow-throated martens and recovered ten dietary vertebrate taxa. Siberian flying squirrel and Korean field mouse each appeared in three samples. The small scat sample provides locally specific evidence of consumption, without measuring the mass eaten.
The same northeastern Chinese analysis found Dybowski’s frog and Korean field mouse in samples from Eurasian otters and Siberian weasels as well as yellow-throated martens. Their different representation among consumers suggests that sharing prey species need not mean using them in identical proportions. Because the analysis targeted vertebrate DNA, fruit and invertebrate foods fall outside that particular view of the food web.
Joint hunts and meals from carcasses
A witnessed hunt in India’s Corbett National Park gives a close view of cooperation. At about eight in the morning on 5 December 2017, two martens stalked and attacked an adult rhesus macaque. The macaque initially resisted while standing upright. Alternating attacks developed into simultaneous assaults, with the martens biting different body regions and renewing their attack when it rose again.
The sequence ended with the macaque’s death and removal into scrub. It directly documents cooperative capture by that pair, but supplies no estimate of how often macaques enter the population’s diet. The IUCN assessment also reports group hunting of Siberian musk deer in the Russian Far East, drawing on northern field literature.
Feeding on a large animal can have another origin. On 5 December 2012, a marten fed on a recently dead red muntjac in western Thailand’s Huai Kha Khaeng Wildlife Sanctuary. The deer’s cause of death remained unresolved. A Changeable Hawk Eagle also used the carcass and was photographed fending off the marten.
Fruit feeding carries seeds through the forest
At Houhe, the seed contents of scats exposed an ecological consequence of fruit eating. Investigators recovered 38,643 seeds from thirteen identified plant species in 120 of the 276 scats, plus four unidentified seeds. Nearly all were intact. Seed-bearing scats occurred from August through January, and 92 percent of recovered seeds were deposited during September–December.
Seeds must survive digestion to retain a chance of growing after deposition. In germination experiments, six of eight tested plant species germinated. Passage through martens increased germination for Diospyros lotus, Celtis biondii and Hovenia dulcis, but reduced it for Sorbus hemsleyi and Kadsura longipedunculata.
Fruit handling also influences which seeds enter this pathway. The Houhe study recorded selection of multi-seeded fruits; among the largest-seeded food species, Diospyros lotus, smaller seeds were recovered after ingestion. The results establish transport and continued viability for several plants, while the distance traveled by individual seeds and their eventual establishment as forest seedlings remain unmeasured. The seed and diet analyses use the same Houhe scat collection.
Following individual martens through Thailand
Radio tracking in Phu Khieo Wildlife Sanctuary followed four males and one female for 4–16 months each. They occupied a north-central Thai plateau around 800–1,100 meters high, with closed forest, open forest-grassland and abandoned orchard. The local wet season ran from April through October, followed by a November-March dry season, so seasonal comparisons there concern a monsoonal tropical setting. Across 297 locations, their estimated home ranges varied from 1.7 to 11.8 square kilometers, averaging 7.2 square kilometers under the study’s 95 percent minimum convex polygon method.
A home range describes an animal’s recorded area of use; it need not be an exclusively defended territory. The Phu Khieo estimates overlapped by an average of 34 percent, with considerable differences among pairings. The sole female’s estimate was 8.8 square kilometers. Ranges generally expanded in the wet season among animals tracked in both seasons, although one male showed the opposite pattern.
Consecutive-day locations averaged 966 meters apart across 95 paired fixes. These straight lines connect two positions; they omit all the turns, searches and climbs made between them. They therefore describe displacement rather than daily walking distance.
Daylight activity changes with weather and season
The Phu Khieo collars registered activity in 656 of 1,127 readings. Activity occurred mainly in daylight, with morning and late-afternoon peaks and some activity at night. Active readings comprised 63 percent of wet-season readings and 55 percent of dry-season readings. Night activity increased under moonlight.
Across four analytical sites in Russia’s Primorsky Krai, camera research from 2010–2021 recorded about 80 percent of detection events by day, about 18 percent at twilight and about 2 percent at night. Twilight was defined as the hour before and after sunrise and sunset. Winter detections peaked around midday; autumn showed late-morning and late-afternoon peaks, and summer encounters were stronger in the afternoon.
Colder inland sites had more pronounced midday peaks than milder coastal sites. Detections declined toward extreme winter cold and, at several sites, the highest summer temperatures. The pattern is consistent with movement shifting toward more favorable thermal conditions. Camera encounters were also seasonally strongest in April–May, with another peak in October; December-January and August produced lower detection levels. Encounter frequency measures passage at camera positions, so it cannot be read directly as a change in population size.
Solitary animals, companions and groups
Yellow-throated martens appear alone and with companions. The Kanchenjunga survey’s eleven independent daytime detections included seven singles and four duos. In Phu Khieo, researchers made five daytime sightings of adult pairs on roads and trails. Neither set of encounters identified the animals’ family relationships. Korean winter tracking followed solitary animals and groups of up to six, with an average group size of 2.9 across the routes.
Other observations show companions feeding together: three martens foraged on a river sandbar and adjacent elephant-grass in Myanmar’s Hukaung Valley in December 2005, while a Thai pair chased squirrels through forest mid-storey. The duration of these associations and the animals’ kinship remain poorly known; traveling together alone does not establish a lifelong mating partnership.
Breeding and lifespan
Wild reproductive timing and juvenile development remain poorly documented. A specialist review summarizes two zoo females with distinct histories. One copulated on 18–21 August and gave birth to two kits on 18 April the following year. The other copulated on 25–29 October and delivered one kit on 27 March. The review also records captive mating behavior between November and February.
These observations are valuable captive records, but do not establish a single wild breeding calendar or species-wide gestation duration. Resting places such as hollow trees and rock crevices are known, while reproductive den selection is less securely documented. Reliable wild ages for weaning, independence and maturity remain unresolved. A reported sixteen-year captive longevity record likewise describes survival under human care, rather than an average lifespan in the forest.
People, forest connections and conservation
The Mammal Diversity Database lists the yellow-throated marten as Least Concern. The accessible IUCN assessment, published in 2016 from a 2015 evaluation, gives a decreasing population trend. Its broad range, use of varied habitats and occurrence in many protected areas support the global category. A 2025 specialist review also reports Least Concern, while identifying regional pressures that require attention within that extensive distribution.
The global assessment also flags potential concern for particular island populations. A broad mainland distribution cannot guarantee the condition of geographically separated populations on every island. Likewise, records in degraded forest show that some altered places are used, while maintaining links among occupied forests remains a distinct conservation issue.
Forest fragmentation and road traffic are conservation concerns in South Korea. A study examining modeled habitat connections alongside marten roadkill linked occurrence with vegetation cover, elevation and greater distance from roads and settlements. In Russia, incidental capture during hunting for other animals is a concern. A 2022 regional review described contraction particularly toward the Amur River, with southern Primorsky Krai assessed more favorably. Its population totals were expert estimates, rather than a dedicated census. The authors estimated a reduction from about 3,000 to 1,500 animals over roughly forty years; those are historical regional opinions, not a current global population total.
Past treatment of the species also shaped its Russian conservation history. Martens were classified as harmful to game management and systematically shot, including in protected areas, before declines became apparent from the 1960s. The 2022 review estimated a roughly one-quarter contraction in regional range over about forty years.
Hunting exposure varies geographically. The IUCN assessment found little evidence of targeted harvest in tropical parts of the range, but recorded incidental capture in non-selective traps. Poaching is reported in Nepal and India, and the IUCN assessment records fur hunting in Pakistan and Afghanistan. At a Khao Yai guard post in Thailand, three martens repeatedly ate boiled rice supplied for a wild bear, carried portions into trees and tolerated nearby human activity. On two observation dates in winter 2004–2005, they approached the station despite loud conversation and movement nearby; January observations continued beyond dusk.
Animal Product Reviews
Dive into Animal Streets’ Animal Product Reviews – your guide to the best gear, gifts, books, and gadgets inspired by the animal kingdom. From land-roaming lions to ocean-gliding whales, from household pets to prehistoric giants, we explore top-rated products for enthusiasts, educators, and animal lovers alike. Whether you’re shopping for toys, decor, field gear, or fun collectibles, we’ve sniffed out the best so you can discover it all in one wild place!
