Souris à la Crême

Souris à la Crême, or Creamed Mouse, is a recipe developed during the narrator’s attempt to test whether small rodents could sustain a large carnivore. Its culinary details emerge from a larger reversal: the Wolves of Wolf House Bay, apparently well fed despite bringing no visible prey home, were living largely on mice. The recipe makes that discovery bodily and practical. Rather than treating rodents as insignificant scraps, it presents them as food substantial enough to study, prepare, and eat.

Origin in the Mouse-Diet Experiment

The recipe arose when the narrator needed a controlled way to assess the nutritional value of mice. Unable to conduct a controlled trial on the Wolves, and with Mike away, the narrator devised a human trial using two equal timed periods: one on a mouse diet and another on canned meat and fresh fish. Physiological tests at the end of each interval were intended to show whether the mouse regimen left bodily functions unimpaired. The narrator recognized that such a single-subject experiment could not prove the Wolves’ case conclusively, but considered it potentially strong evidence that they could survive and function normally on the same food.

The first meal came from a basin of small-animal specimens left after a morning of skinning. The narrator cleaned the carcasses, put them in a pot over a primus stove, and boiled them into a stew. Their odor seemed appetizing, and the narrator was hungry by the time the cooking was finished. Eating them required an adjustment: the mice’s many tiny bones initially seemed troublesome, but could be chewed and swallowed without much difficulty. The meat tasted pleasant but bland. As that flavor became monotonous, boredom and diminished appetite prompted the narrator to experiment with different preparations; Creamed Mouse was judged the best of the resulting recipes.

Preparation and Ingredients

The recipe begins with one dozen fat mice. They are skinned and gutted, but their heads are left on, and the carcasses are washed before being covered with ethyl alcohol for a marinade of about two hours. Meanwhile, sowbelly is cut into small cubes and fried slowly until most of its fat has rendered. The recipe notes that sowbelly is ordinarily available in the Arctic and that ordinary salt pork can substitute for it.

After marinating, the mice are removed from the alcohol and rolled in salt, pepper, and flour. They are sautéed for about five minutes, with care taken not to let the pan become too hot: excessive heat, the instructions warn, would dry the delicate meat and make it tough and stringy. A cup of alcohol and six or eight cloves are then added. The pan is covered and the mice simmer slowly for fifteen minutes. A cream sauce is prepared by a standard method, poured over the carcasses, and left to settle with them, covered, in a warm place for ten minutes before serving. The procedure is unusually exact about time and heat, treating the small animals as delicate food rather than a novelty to be cooked carelessly.

What the Experiment Revealed

During the first week of the mouse diet, the narrator’s vigor remained unimpaired and no apparent ill effects occurred. A new craving for fats, however, exposed a flaw in the experiment. Wolves ate entire mice, whereas the narrator’s dissections had shown that much of a rodent’s fat was stored in the abdominal cavity, attached to the intestinal mesenteries, rather than beneath the skin or in the muscle. By eating only the prepared portions, the narrator had omitted a nutritionally important part of the prey. The diet was corrected: for the rest of the experimental period, the narrator ate whole mice, though not the skin, and the craving for fat eased considerably.

This correction gives the recipe an important place in the book’s reasoning. The narrator’s eating trial was not simply a stunt or a tasting exercise; its usefulness depended on attending to what the Wolves actually consumed. The first version of the diet had separated edible meat from the rest of the animal and therefore failed to reproduce the wolves’ feeding practice. Eating the whole mouse brought the experiment closer to the ecological fact under investigation, while also showing why a food’s nutritional value cannot be judged from its muscle alone. The evidence supplied here records the narrator’s initial physical response and the adjustment, not the complete results of the later comparison with canned meat and fresh fish.

The Ecological Argument Behind the Dish

The recipe’s premise took time for the narrator to accept. In the mossy sphagnum bogs around Wolf House Bay, small rodents could burrow and nest in abundance. As June waned into July, lemmings, red-backed mice, and meadow mice became conspicuous; meadow mice even invaded Mike’s cabin and nested in the narrator’s sleeping-bag pillow, where one gave birth to eleven young. Yet the Wolves seemed healthy, and the two males returned from nightly hunts without any visible catch. The narrator’s expectations about the mythical wolf had made a rodent-based diet seem too ludicrous to consider, despite the evidence accumulating nearby.

Angeline’s hunting supplied a direct clue. The narrator watched her pounce repeatedly in grassy muskeg and eventually saw a mouse in her jaws. At first the performance seemed like playful snacking; after she had eaten about twenty-three mice, its significance became harder to dismiss. The narrator concluded that the Wolves at Wolf House Bay, and probably Barren Land wolves raising families beyond the summer Caribou range, lived largely or almost entirely on mice. Because mice could not be carried externally in sufficient numbers, the males’ habit of returning from hunts and going straight into the den also acquired an explanation: they were regurgitating partly digested food for the pups.

The evidence for a rodent-rich diet was broader than the narrator’s observation of Angeline. Ootek confirmed the importance of mice and described Wolves eating large numbers of ground squirrels, sometimes apparently preferring them to Caribou. He also reported Wolves fishing for Northern pike in spring channels and catching spawning suckers in shallow tundra streams. Later, an analysis of scats found rodent remains, chiefly teeth and fur, in about 48 percent of the samples; other identifiable material included Caribou bones and hair, bird feathers, and a corroded brass button. In this wider account of opportunistic feeding, Creamed Mouse is not offered as a complete description of wolf nutrition. It is the vivid, experimentally motivated human counterpart to one significant food source—and a reminder that the Wolves’ apparent dependence on large game had been an assumption, not an observation.