In situ Caching of a Large Mammal Carcass by a Fisher, Martes pennanti

1. What they did

Many animals cache food: they handle it so it keeps for later and is harder for competitors to get. Fishers usually carry prey or loose food away to a sheltered place, such as a den. When they find something too big to move, such as a large mammal carcass, they have been known to stay nearby and return to feed. The authors knew of no published account of a fisher, or any other mustelid, caching a large, immovable item.

On 20 October 2008 they began a study of carrion feeders at a carcass depot in Algonquin Provincial Park, Ontario. The carcass was an adult male American black bear that had been shot and left there fresh on 18 October. They fixed a motion-triggered trail camera to a tree 2.5 m behind it. Each trigger took three photographs 30 seconds apart, followed by a one-minute pause. Ethan Hüner visited every two to five days (seven visits in all) to check the camera and its batteries.

When they saw that a fisher was covering the carcass, they ran a small, informal test. On 1 November, Hüner left a pile of dead bracken fern and a pile of fresh white spruce boughs beside the carcass, to see whether the fisher would use them. On 5 November and again on 7 November, he removed cache material from the bear’s rump to see whether the fisher would cover it again. The study ended on 17 November 2008.

2. What they found

Signs at the carcass

• On the second visit (26 October), Hüner found scrapes in the gravel forming a semicircle round the rear of the carcass, about 0.9 to 1.2 m out, with the scraping directed towards the carcass.
• The hind legs were completely covered with dead and green grasses, pine needles, moose bones and twigs. The material was packed between the legs, over the tail and rump, and round the edges of the carcass.
• The photos showed that a male fisher had placed the material, starting late in the afternoon of 22 October.

Caching and feeding by the fisher

• The fisher placed material with their mouth and forelimbs. They used all the bracken fern on 1 November, carrying mouthfuls onto the carcass, and dragged a spruce bough onto the bear’s shoulder, where the bullet holes were.
• After each removal of cache material, the fisher covered the rear of the carcass again: on the night of 5 November, and on 9 November.
• The fisher was not recorded feeding before the first caching. They fed mainly through the anus, apart from two brief feeds (under a minute each) at the bullet holes. They fed under the cover either by uncovering a patch or by pushing their muzzle through the material.
• By 17 November the camera had recorded 36 feeding episodes (mean 14.9 minutes, range under 1 to 65 minutes) and 21 caching episodes (mean 7 minutes, range under 1 to 13 minutes).
• Visits were almost all at night or at dawn and dusk. The fisher covered up to about a third of the carcass and apparently scent-marked at least once.

Other scavengers

• Red foxes: 12 episodes between 29 October and 11 November, always at dawn, dusk or night, twice with two foxes together. A fox pulled the cache material off the rear on 29 October to feed; the area stayed uncovered until the fisher returned after a second fox feed. No direct meetings between fox and fisher were recorded, but foxes seemed preoccupied by something just before or after the fisher’s visits.
• Common ravens: present by day throughout, up to seven at once. They fed only at the bullet holes from 23 to 28 October. They first fed at the rear on 29 October, after the fox had disturbed the cache, and then used exposed areas, picked through the cache material or fed at the widening feeding hole. One raven may have removed the spruce bough. Once a raven landed on the carcass while the fisher fed and edged towards them.
• American marten: fed briefly at the rear on 31 October.
• Eastern wolf: one investigated the carcass on 23 October but did not feed.

3. What they concluded

• This is apparently the first documented case of a fisher caching a large mammal carcass where it lay.
• The partial covering did not keep other scavengers away: foxes, ravens and a marten all fed from it, including where the fisher had concentrated the cover. It did delay ravens feeding at the rear, which may have kept the food available to the fisher for longer, though the data cannot confirm this.
• Covering could matter most against ravens, which feed by day while the fisher is mainly active at night. Ravens cannot break through hide, so they rely on other animals to open a carcass. Concealment may deter ravens that have not already seen the food, which might explain why it stopped working when it did.
• Caching may also slow decay and reduce odour, as experiments with cougar caches have suggested. The authors did not measure temperature or odour, had no uncovered carcass for comparison, and their own removal of cache material may have reduced any benefit.
• Such caching would be unlikely to stop wolves, which feed on black bear carcasses in Algonquin and will dig through deep snow to reach a carcass.
• An intact large carcass is an unusual and large food supplement for a fisher, especially here, where there is no deer harvest and snowshoe hares were in a cyclic decline. Being able to make it last could give a fisher a significant fitness advantage.

4. How tracking was or was not used

The caching was first noticed from signs at the carcass: scrapes in the gravel in a semicircle directed towards the carcass, and grass, needles, bones and twigs packed over the hindquarters. The camera photos then showed which animal had made them. The paper also records how each scavenger fed (fisher through the anus and under the cache, ravens at the bullet holes and through the cache material). No tracks or trailing were reported.

5. Tracker or observer certification

Not reported. Hüner checked the site and the photos; no tracking experience or qualification is described.

6. Other technologies, skills or innovations

• A motion-triggered trail camera (Moultrie Game Camera) that recorded the ambient temperature with each photo.
• A simple field test: adding new cache materials (bracken fern, spruce boughs) and removing cache material, to see whether the fisher’s caching was a real tendency.