At eight minutes past midnight on 17 December 2024, a camera strapped to a tree 3,112 metres up a Himalayan slope woke up, fired six frames and shut off again. Whatever walked past it had fangs.

None of that is supposed to happen in West Bengal. Musk deer, small hunched antlerless things whose males grow sabre-like upper canines, were quietly written off in the state generations ago. Their last occurrence report there dates from 1955, in Singhalila National Park, and every claim since has sat somewhere between ambiguous and unprovable.

Then a trap camera in Neora Valley National Park caught one walking past.

What the camera actually caught

Meghna Limboo and Govindan Veeraswami Gopi, both of the Wildlife Institute of India, reported the detection in the conservation journal Oryx, calling it the first photographic documentation of a musk deer anywhere in the state. Those images fell out of a camera-trapping survey begun in December 2023 under the Pan-India Assessment and Monitoring of Endangered Species programme, an effort set up to keep tabs on red pandas rather than deer. One camera, six consecutive frames, a single detection event in temperate forest.

Diagnostic features of the genus Moschus were all present: long hare-like ears, no antlers, upper canines plainly visible. There wasn’t enough detail to name a species, because the photographs were taken in the dark and grain does what grain does. India hosts four musk deer species, all of them classified as Endangered, and Limboo and Gopi decline to pick one.

One camera, one night, one short paper, and that is the entire evidence base.

Neora Valley covers 160 square kilometres inside the Kangchenjunga transboundary landscape and packs temperate, subtropical and tropical habitat into that space. Both researchers put the animal’s persistence down to that available habitat, which is a polite way of saying the park still holds ground steep enough and thick enough for a shy herbivore to vanish into for seventy years.

Which species it turns out to be matters more than it sounds. Enforcement priorities, habitat protection and Red List assessments all hang on knowing which of the four species this is, and a grainy frame shot at midnight settles none of it.

Why they have fangs

Why fangs instead of antlers? Musk deer never developed the second option. Males settle disputes during the rut with elongated upper canines instead, which grow down past the lower lip and stay visible even with the mouth shut. Very Hammer horror, considering the animal is roughly spaniel-sized and spends its life eating leaves, twigs and moss.

The gland that put a price on their heads

Adult males carry a musk pod between the navel and the genitals, and its contents have been prized in traditional medicine and perfumery for centuries. In its report On the Scent, the wildlife trade monitoring network TRAFFIC documented musk reaching roughly US$45,000 a kilogram on the international market by the end of the 1970s, two to three times the value of gold by weight.

What followed was predictable. TRAFFIC’s Russian figures record 300 to 380 kilograms of musk leaving the country’s east between 1989 and 1996, a volume equivalent to a total harvest of roughly 90,000 to 104,000 animals. Russia’s musk deer population is estimated to have fallen by 50 to 70 per cent between 1990 and 1993 alone.

Nobody knows how many are left

Counting musk deer is genuinely difficult. Solitary, mostly nocturnal, and confined to steep terrain, they resemble one another closely enough that species have been muddled for decades. Writing in Scientific Reports, a team modelling the Himalayan musk deer’s range described its distribution as historically misunderstood as a direct result.

So the numbers simply are not there. The IUCN Red List assessment by R.J. Timmins and J.W. Duckworth lists the Himalayan musk deer as Endangered with a decreasing population trend and no usable estimate of how many remain. Limboo and Gopi’s own paper asks for exactly that: a proper population assessment for Neora Valley, something nobody has attempted yet.

For a sense of how thin this kind of evidence usually gets, look at Afghanistan. A Wildlife Conservation Society team led by Stephane Ostrowski confirmed Kashmir musk deer in Nuristan province in 2014, the first scientific record since a Danish survey team passed through in 1948. Five sightings, none of them photographable. Bengal’s forest department managed better on its first go.

What happens to the cameras now

“Presence of musk deer in the forests of Neora Valley” is how North Bengal’s chief forest officer for wildlife, Bhaskar J.V., described the find to ETV Bharat. Around 50 trap cameras are now running across the hills, and work has started on identifying which species turned up. The paper’s authors want the same, plus stronger long-term protection across the wider landscape.

Confirming any of it will take a while. Scat genetics, repeat surveys and species identification all cost years and real money, and the deer have shown no particular interest in cooperating with either. They managed seventy years unobserved before anyone thought to look, so they can probably wait out the paperwork too.