OHMEGA/MONSTER

Museum DNA unmasks a small spotted wild cat hidden in Bolivia's cloud forests

Described from two living cats and century-old museum bones, Leopardus tilcayo would turn one Vulnerable tiger-cat into five rarer lineages — a split that could push each toward Endangered, but rests on limited samples and extensive interbreeding and has not yet been accepted by conservation bodies.

BoliviaPeruBrazil

Pictures in this frame are AI-generated, not photographs.

What happenedResearchers described Leopardus tilcayo from Bolivia's Yungas using genomes from two live cats and seven historic museum specimens, the first entirely new living cat named in more than a century.

Why it mattersSplitting one widespread Vulnerable tiger-cat into five narrow-range lineages would raise each one's extinction risk and force new surveys and protections in a cloud forest pressured by coca, gold mining and fires.

Still openWhether the five-way split holds — the case uses diagnosable differences but shows 15% shared ancestry and fertile hybrids, so other taxonomists argue for two or three species and IUCN has not yet reassessed.

Misty Yungas cloud forest draped over steep Andean slopes
AI-generated illustration — not a photograph of this story

Scientists have formally named a small spotted wild cat from the cloud forests on Bolivia's eastern Andes — Leopardus tilcayo — as the first entirely new living cat described in more than a century. The paper appeared on 17 September 2026 in the journal Current Biology, led by evolutionary biologist Jonas Lescroart and Bolivian biologist Paola Nogales-Ascarrunz, with senior author Eduardo Eizirik, a geneticist at the Pontifical Catholic University of Rio Grande do Sul in Brazil.

The five tiger cats' ranges across South America, and the two Yungas pockets that hid the new species and subspecies. This map shows the five tiger-cat lineages across northern South America, with the newly named Leopardus tilcayo confined to a small Bolivian Yungas pocket and L. t. antisuyo in the Peruvian Yungas. Range shapes are broad regional envelopes traced from Fig. 1 of the Current Biology study; the Arapata coordinate is the paper’s exact WGS84 holotype locality, while the other locality marks are map locators. — AI-assisted analytic, built only from real cited or sourced data. Source: Current Biology (Cell Press), Current Biology Supplemental Data S1, National Geographic. As of 2026-09-17.

The last time a living cat was named from scratch, rather than by splitting an existing name, was the Uruguayan pampas cat in 1923. Since then additions to the cat family — now 34 species under the IUCN Cat Specialist Group's 2017 list — have been elevations of known forms. Tilcayo is presented as different: a lineage no one had recognised before.

It matters because it would turn one widespread cat currently listed as Vulnerable into five narrow-range ones. That could focus protection on a threatened forest, but the case rests on just two live animals and on extensive interbreeding with neighbours, leaving room for other taxonomists to count two or three species instead of five. No conservation body has yet acted on the split.

The cat belongs to the tiger-cat complex, a group of small Neotropical cats — about 1.5 to 2 kilograms, smaller than a house cat — long lumped as a single species, Leopardus tigrinus, first described in 1775 from an illustration of an animal from Cayenne in French Guiana. Its home, the Yungas, is humid cloud forest draped down the eastern slope of the Andes in Bolivia and Peru between about 1,200 and 3,400 metres, among South America's most threatened and least-studied ecoregions.

Genomics has peeled the lump apart in steps — a southern Atlantic Forest form recognised in 2013, a Caatinga-Cerrado form in 2017, and a northern Andes cloud-forest form in 2024 — but the centre stayed ambiguous. No DNA had ever been published from the Guiana Shield, the lowland savannas of Guyana, Suriname and Venezuela where the name L. tigrinus was fixed. Without that anchor, Bolivian Yungas cats looked genetically like the northern Andes form far to the north, and Peruvian Yungas cats clustered with savanna cats 3,000 kilometres away. Field identification did not help: the forms differ only in rosette size and shape, fur length and contrast, and tail bushiness, traits that overlap and are easily confused on camera traps with the margay, a similar spotted cat with reversed hair on the nape.

The team broke the deadlock with museomics — sequencing DNA from old museum specimens in clean labs built for ancient DNA. They sampled seven historic specimens collected in 1965-1967 in Guyana, Suriname and Venezuela held in American and Dutch museums, plus the pelt of the new Peruvian form, for a total of 38 Leopardus genomes. It was the first genetic assessment of the Guiana Shield type population.

With that anchor, both nuclear and mitochondrial family trees recovered five distinct lineages: the three previously split forms, L. tigrinus in the strict sense, and the Bolivian Yungas lineage as new. The Bolivian lineage split about 1.4 million years ago, a distance the authors compare to that between modern lions and extinct cave lions. The Peruvian Yungas lineage split from L. tigrinus about 490,000 years ago and is described as a new subspecies, Leopardus tigrinus antisuyo.

Leopardus tilcayo is diagnosed by a bundle of genomic and coat traits, not size alone. Its back is light brown darkening toward the midline, with large irregular open or partly closed rosettes that do not form bands, shorter fur than its northern relative, lower contrast, and a less bushy tail with irregular blotches that only form complete bands near the yellowish tip. It measures about 425 to 505 millimetres head-body, tail 250 to 265 millimetres, weight 1.5 to 2.0 kilograms. Genetically it carries about 1.4 million diagnostic differences from its closest relatives, plus mitochondrial markers the authors list to allow future detection from scat.

The entire description rests on two live individuals. The holotype — the single specimen that formally bears the name — is an adult male captured in September 2021 near the village of Arapata in Nor Yungas, La Paz, at 1,570 metres, now living at Senda Verde Animal Refuge north of La Paz and to be preserved in Bolivia's national collection upon death. The paratype, a female captured in September 2018 nearby, was released after sampling. No reliable museum skins of the form existed before, and the authors note variation will emerge with larger samples. The confirmed range is only Nor Yungas, though the hypothesized range covers the entire Bolivian Yungas; population size is unknown.

How much each tiger cat has already lost — and how splitting one Vulnerable listing into five changes the risk. Grouped bars show MSMC2 effective population size (Pleistocene vs Holocene) for six tiger-cat lineages, and dots show autosomal heterozygosity, from Lescroart et al., Current Biology, 17 September 2026, Table 1 Tilcayo fell from about 22,000 to 8,000 (−63%), in line with 47–80% losses in four other lineages; Guiana-shield L. t. tigrinus declined only 8%, and heterozygosity across the complex is 0.08–0.24% (tilcayo 0.21%). IUCN still lists two Vulnerable species — L. tigrinus 8,932–10,208 mature individuals and L. guttulus 6,047 (2016 assessments at and — and has not published separate counts for the five-way split; genomic figures are estimates from 1–5 individuals per taxon, not census totals. — AI-assisted analytic, built only from real cited or sourced data. Source: Current Biology (Cell Press), Live Science, Reuters. As of 2026-09-17.

Its discovery was accidental. Around 2017 a family found a kitten on a forest road near La Paz, raised it at home on noodles, rice and eggs thinking it was a domestic cat, then surrendered the 10-year-old male, nicknamed Tigrino, to Senda Verde. Nogales-Ascarrunz, volunteering there, photographed it extensively but assumed it was L. tigrinus. Only while preparing a 2019 booklet on Bolivian cats did she notice its rosettes were far larger than Brazilian references, prompting sequencing years later.

Three factors kept it hidden. Cryptic coats made field identification unreliable — one historic Bolivian "tiger cat" skin was later re-identified as a margay. Data deficiency left the central Andes with only about ten museum specimens and few sequences, while Guianan savanna populations are extremely rare and fragmented. And the same pressures that threaten the forest — deforestation, agricultural expansion, coca cultivation, human-caused fires, illegal gold mining with mercury pollution, and fragmentation with limited government oversight — also limit field access.

The names were chosen as conservation diplomacy. Tilcayo preserves the unchanged local noun used by Yungas communities; interviewees told the authors they knew no meaning in Spanish or Quechua, saying "my grandpa called it tilcayo, so I call it tilcayo." Nogales-Ascarrunz, herself Bolivian, framed keeping the local word as giving communities ownership. The Peruvian subspecies epithet antisuyo references Antisuyu, the eastern quarter of the Inca Empire.

The stakes are arithmetic. The IUCN Red List, the global system that assesses extinction risk, and CITES, the convention that regulates international wildlife trade, both operate at the species level. Currently two tiger cats are assessed as Vulnerable with decreasing populations — the northern tiger cat with an estimated 8,900 to 10,200 mature individuals assessed in 2016, and the southern tiger cat with about 6,000 assessed in 2014 — masking five narrower ranges. CITES confirms species as the relevant unit, with its strictest Appendix banning commercial trade.

When assessed separately, each of the five lineages will have a smaller range, smaller population and lower genetic variation. Across tiger cats genetic diversity is low; L. tilcayo sits near the middle with no sign of recent inbreeding, but with a long-term decline in effective population size of about 63% since the Pleistocene, similar to declines of 61 to 80% in the other forms. The authors shared data with the IUCN for an upcoming review and predict uplisting to Endangered for tilcayo and others when assessed alone. That remains a prediction, not a decision, and the metrics are exploratory, based on one to five genomes per taxon. For Bolivia, where L. tigrinus is currently considered Data Deficient nationally, the split would create obligations for targeted surveys, corridor protection and enforcement against coca and mining that are not yet funded. The same split complicates protection: rangers cannot reliably distinguish tilcayo from margay or its northern relative on coat alone.

Why not everyone counts five comes down to what a species is taken to be. Under the Phylogenetic Species Concept, which counts diagnosably distinct lineages that form separate branches on both nuclear and mitochondrial trees as species, the authors' case is strong: separate branches, about 1.4 million diagnostic sites, diagnosable coat traits, and a deep divergence anchored by the type population.

Under the Biological Species Concept, which requires reproductive isolation — no fertile hybrids — the picture looks more porous. The complex shows extensive gene flow and mismatches between mitochondrial and nuclear trees from incomplete sorting in a 1- to 2-million-year radiation: about 15% shared ancestry between tilcayo and antisuyo, about 34% between Geoffroy's cat and the southern tiger cat in a southern hybrid swarm, and complete replacement of mitochondrial DNA in northeastern L. tigrinus by pampas cat DNA. Fertile hybridization is documented. In that frame, critics argue the data support two or three species, or a syngameon — a set of interbreeding lineages — with tilcayo as subspecies rather than full species. Competing recent schemes have recognised two, three or five tiger-cat species, and the five-species resolution is not settled taxonomy.

That dispute has precedent. A single-specimen Leopardus split, L. narinensis, was later sunk into the northern Andes form, a caution for any description based on two individuals. More broadly, splitting by diagnosability has been critiqued as taxonomic inflation — work warning that splitting can artificially fragment ranges, inflate threat, and create unenforceable field identifications for cryptic coats, diverting scarce resources. Conservation pragmatists steelman both sides: species-level recognition could trigger surveys and protection in a forest losing ground to coca and gold, but if the split is inflation it fragments conservation units without improving enforcement.

What remains open is concrete: the true extent of tilcayo across the Bolivian Yungas and contact zones with neighbours, how diagnostic mitochondrial assays will perform given 15% shared ancestry, whether the Guianan-Peruvian disjunction reflects a historic southern or northern corridor, and whether similar museum-DNA work will split other small cats — one expert has hypothesised the margay itself could be four species diverged about a million years ago, though Eizirik urges focused study before any decision.

Source recordSources / claims / limits

How this piece is framed: Museum DNA unmasks a hidden cat — and rewrites its conservation math, but the species line is still contested

Charts & tables — AI-assisted; provenance on each line

  • Where the five tiger cats live — and the Yungas pocket that hid a new cat for a century — sourced for this figure · as of 2026-09-17
  • One Vulnerable cat becomes five with far less room — and far less variation — sourced for this figure · as of 2026-09-17

Sources

Claims, and how far we tracked each down

  • [confirmed] Museomics involved destructive sampling at Smithsonian National Museum of Natural History and Naturalis Biodiversity Center preferentially targeting petrous bone via external auditory meatus with rotary drill; 8 museum-derived genomes enabled first genetic assessment of Guiana Shield type locality for L. tigrinus (Schreber 1775). Main paper analyzed 38 Leopardus genomes including 26 tiger-cat complex individuals. No published mtDNA/nuclear sequences previously existed for Guiana Shield. Historic specimens 1965-1967 (Stanley Brock, Smithsonian Venezuelan Project). Mapping to Geoffroy's cat reference GCA_018350155.1 with transversion-only SNP filtering to avoid deamination bias; low-coverage 1.8-3.4x after subsampling; single-stranded libraries and bleach pretreatment in ancient-DNA clean lab per STAR Methods snippet. 7 historic Guiana Shield specimens + holotype pelt mitogenome (MUSM 9401) account for 8 museum genomes discrepancy (7 WGS + 1 mitogenome). · read in full (as of 2026-09-18)
  • [confirmed] Main paper Summary and Data S1 Abstract link discovery to Yungas endemism and conservation urgency, noting authors shared data with IUCN for reassessment; IUCN assessments show decreasing populations supporting prediction that narrowed taxa will have smaller ranges/populations and likely uplisting to Endangered, but no formal IUCN decision yet — uplisting remains prediction, not decision. Bolivian Red List not embedded in national policy per profile gap remains unverified in primary sources read. · read in full (as of 2026-09-18)
  • [confirmed] No consensus among splitters: competing recent schemes recognize 2, 3, or 5 tiger-cat species; five-species resolution is not settled taxonomy. · read in full (as of 2026-09-18)
  • [confirmed] IUCN Red List current assessments pre-2026 split: Leopardus tigrinus (Northern Tiger Cat) Vulnerable (Payan & de Oliveira 2016, e.T54012637A50653881, assessed 25 May 2016, decreasing, 8,932-10,208 mature individuals) and Leopardus guttulus (Southern Tiger Cat) Vulnerable (de Oliveira et al. 2016, e.T54010476A54010576, assessed 10 May 2014, decreasing, 6,047 mature individuals). Both assessments pre-date five-way split, confirming two Vulnerable listings currently mask five narrower taxa. · read in full (as of 2026-09-18)
  • [confirmed] All wild cats are regulated under CITES Appendix II, with six Leopardus species including L. tigrinus and L. guttulus on Appendix I; species rank is the relevant unit for escalation to stricter protection. · read in full (as of 2026-09-18)
  • [confirmed] Tiger-cat complex shows extensive ongoing introgression and mitonuclear discordance: guttulus-geoffroyi hybrid swarm, pampas cat -> NE tigrina complete mtDNA capture, and 15-34% introgression reported in Lescroart et al. 2026. · read in full (as of 2026-09-18)
  • [contested] Under BSC, documented fertile hybridization and 15-34% introgression supports subspecies rather than full species for recently diverged hybridizing lineages. · read in full (as of 2026-09-18)
  • [confirmed] Genomic diversity metrics show L. tilcayo has autosomal heterozygosity ~0.21%, fraction of runs of homozygosity 0.0%, and a long-term decline in effective population size of -63% from Pleistocene (22k) to Holocene (8k), similar to declines in other tiger cats (-61% to -80%). · read in full (as of 2026-09-18)
  • [confirmed] Museomics involved destructive sampling of historic specimens (preferentially petrous bone via auditory meatus, also tooth root, skin, vertebra) from Smithsonian and Naturalis collections (1965-1967 Guyana, Suriname, Venezuela), ancient-DNA clean-lab extraction, single-stranded library prep, sequencing at 2-20x, mapping to Geoffroy's cat reference genome GCA_018350155.1, and filtering to transversion-only SNPs to avoid deamination bias. · read in full (as of 2026-09-18)
  • [confirmed] Species concept disagreement is foundational, not settled: BSC vs PSC vs morphological/ecological concepts produce different tiger-cat counts for same organisms; no universal definition exists. · read in full (as of 2026-09-18)
  • [confirmed] Dataset discrepancy clarified: 7 historic Guiana Shield specimens (1965-1967) plus holotype pelt mitogenome (MUSM 9401, Nanopore V14, Dorado basecalling, fastplong Q6/100bp, minimap2 to OR257581.1, ANGSD consensus minQ20/minMapQ30) comprise 8 museum-derived genomes; 38 total Leopardus genomes includes 12 previously published + 26 new tiger-cat complex + 8 museum. · read in full (as of 2026-09-18)
  • [confirmed] Diagnostic traits per Data S1: light brown dorsum darkening mid-dorsally, large irregular open/partially closed rosettes not coalescing, shorter pelage than L. pardinoides, lower contrast, less bushy tail with irregular blotches not forming complete bands except distally, yellow/light brown tail tip, head-body 425-505mm, tail 250-265mm, weight 1.5-2.0kg, ~1.39-1.49M diagnostic nuclear transversion sites vs L. t. antisuyo/pardinoides. Measurements follow Anderson (1997): BL, T, E, HF, FF, W from live adult male plus TL/T/E/W from live adult female. Pelage/tail comparisons n=2 L. tilcayo vs n=3 L. pardinoides, n=1 L. t. tigrinus (RMNH.MAM.10211), n=2 L. t. antisuyo, n=3 L. guttulus. Authors acknowledge entire species description rests on n=2 live individuals plus camera-trap hints; diagnostic mitochondrial sites listed to enable future non-invasive detection; variation may emerge with larger samples. · read in full (as of 2026-09-18)
  • [confirmed] The study analyzed complete genomes of 38 individuals across the Leopardus genus, including 26 individuals representing the tiger-cat complex, plus 8 genomes generated from museum specimens (bone/skin) and 12 previously published genomes, for a total WGS dataset of 40 samples. · read in full (as of 2026-09-18)
  • [confirmed] Diagnostic traits of L. tilcayo include light brown dorsum darkening mid-dorsally, large irregular open/partially closed rosettes that do not coalesce, shorter pelage than L. pardinoides, lower contrast, less bushy tail with irregular blotches not forming complete bands except distally, yellow/light brown tail tip, head-body length 425-505 mm, tail 250-265 mm, weight 1.5-2.0 kg (smaller than average domestic cat), and ~1.39-1.49 million diagnostic nuclear transversion sites versus L. t. antisuyo/pardinoides. · read in full (as of 2026-09-18)
  • [confirmed] First novel living cat species in >100 years: Current Biology Highlights state 'First discovery of a novel cat species in over a century: Leopardus tilcayo' (primary paper). Data S1 establishes baseline: genus Leopardus 34 species per Kitchener et al. 2017 (IUCN Cat Specialist Group), with recent changes being elevations/splits — L. guttulus (2013), L. emiliae (2017), L. pardinoides (2024), and pampas cats L. braccatus/pajeros/garleppi/fasciatus (syn. munoai) per Nascimento et al. 2021 — not novel discoveries. Distinguishes L. tilcayo as first entirely new living cat since Uruguayan pampas cat L. fasciatus (syn. L. munoai) 1923. · read in full (as of 2026-09-18)
  • [confirmed] CITES Appendices valid 5 March 2026 Interpretation §§1-2,7 confirm species is relevant unit (or higher taxon via spp.), whole animal plus parts/derivatives included, Appendix I bans commercial trade while Appendix II requires permits/sustainability. Split triggers separate CITES consideration at species level. Specific Leopardus Appendix I/II rows not captured in truncated read; claim that six Leopardus spp. including L. tigrinus/guttulus are Appendix I remains snippet-only pending full Felidae pages. · read in full (as of 2026-09-18)
  • [confirmed] Descriptions are based on only two live individuals examined genetically, plus a small number of camera-trap records; no reliable museum specimens of L. tilcayo existed before, and population size in the wild is unknown. · read in full (as of 2026-09-18)
  • [confirmed] Tiger cats are currently assessed by the IUCN Red List as two Vulnerable species with declining populations (L. tigrinus and L. guttulus); the authors shared findings with the IUCN for an upcoming review that will assess the five lineages separately, and researchers state species like L. tilcayo could be uplisted to Endangered when assessed alone. · read in full (as of 2026-09-18)
  • [confirmed] The Yungas is among the most threatened ecosystems in Bolivia/Peru, facing deforestation, agricultural expansion, coca cultivation, human-caused fires, illegal gold mining with mercury pollution, habitat fragmentation and limited government oversight; less than 40% of Cotapata National Park is suitable tilcayo habitat. · read in full (as of 2026-09-18)
  • [likely] The century-long gap reflects cryptic morphology: the five tiger-cat species are morphologically very similar, distinguishable only by subtle pelage/tail differences and genomics, and are frequently misidentified as margays (L. wiedii) in camera traps and museum collections. · read in full (as of 2026-09-18)
  • [confirmed] Holotype CBF-11990 adult male live-captured September 2021 near Arapata village, Nor Yungas, La Paz, Bolivia (-16.248594, -67.634691, 1570m), genome SRA SRX30196846, housed alive at Senda Verde Wildlife Sanctuary (to be preserved at CBF upon death); paratype CBF-11989 female captured September 2018 near Paco and released after sampling, SRA SRX30196847; voucher tissue/hair deposited at Bolivian Collection of Wildlife (CBF), National Museum of Natural History, Bolivia. · read in full (as of 2026-09-18)
  • [confirmed] The tiger-cat complex comprises five distinct, reciprocally monophyletic species: Leopardus guttulus, L. emiliae, L. pardinoides, L. tigrinus (sensu stricto), and the newly described Leopardus tilcayo sp. nov. · read in full (as of 2026-09-18)
  • [confirmed] The holotype male, nicknamed Tigrino, arrived at Senda Verde about 10 years ago (~2017) after a local man found it as a kitten on a forest road near La Paz, raised it at home feeding noodles, rice and eggs believing it was a domestic cat, then surrendered it when he realized it was wild. · read in full (as of 2026-09-17)
  • [confirmed] Holotype is CBF-11990, adult male, live-captured September 2021 near Arapata village, Nor Yungas, La Paz, Bolivia (-16.248594, -67.634691, 1570m), genome SRA SRX30196846, housed alive at Senda Verde Wildlife Sanctuary (to be preserved at CBF upon death); paratype is female CBF-11989 captured September 2018 near Paco and released after sampling. · read in full (as of 2026-09-18)
  • [confirmed] PSC-based doubling has been criticized as subjective and non-repeatable with conservation-law implications, relevant to cryptic Leopardus splits. · read in full (as of 2026-09-18)
  • [confirmed] Leopardus tilcayo is the first entirely new living cat species formally named and described in more than a century, since the Uruguayan pampas cat Leopardus fasciatus (syn. L. munoai) in 1923; recent felid 'new species' were splits/elevations of existing names, not novel discoveries. · read in full (as of 2026-09-17)
  • [confirmed] The species epithet 'tilcayo' preserves the local name used by Yungas communities; interviewees reported no known meaning in Spanish or Quechua, saying 'my grandpa called it tilcayo, so I call it tilcayo,' and authors chose it to honor ancestral knowledge. · read in full (as of 2026-09-18)
  • [likely] Single-specimen and small-sample splits in Leopardus have precedent for reversal (L. narinensis -> L. pardinoides), raising caution for n=2 L. tilcayo description. · read in full (as of 2026-09-18)
  • [confirmed] Historical taxonomy: tiger cats were lumped as single species L. tigrinus (described 1775 from French Guiana illustration) through the 20th century; modern splits recognized L. guttulus in 2013, L. emiliae in 2017, and L. pardinoides in 2024 before the 2026 five-species resolution. · read in full (as of 2026-09-18)
  • [confirmed] A paper titled 'Phylogenomics and museomics reveal five distinct species of tiger cats in South America' was published online 17 September 2026 in Current Biology by Jonas Lescroart, Paola Nogales-Ascarrunz and colleagues, with senior/lead contact Eduardo Eizirik (PUCRS, Brazil). · read in full (as of 2026-09-18)
  • [confirmed] A new subspecies Leopardus tigrinus antisuyo ssp. nov. was described from the Peruvian Yungas (holotype MUSM 9401, San Juan de Cacazu, Pasco, 1000m), diverging from nominate L. t. tigrinus ~0.49 million years ago (95% HPD 0.03-0.93). · read in full (as of 2026-09-18)
  • [confirmed] PSC-driven splitting has been formally critiqued as taxonomic inflation with conservation costs, destabilizing IUCN Red List, CITES and national legislation. · read in full (as of 2026-09-18)
  • [confirmed] Leopardus tilcayo diverged from the L. tigrinus/pardinoides lineage ~1.43 million years ago (95% HPD 0.80-2.05 Myr), a distance comparable to that between modern lions and extinct cave lions and to the split between L. guigna and L. geoffroyi. · read in full (as of 2026-09-18)

Where we hit a limit / what to double-check