Type.—Female, adult, skin and skull; no. 642 Bangs Coll. in Mus. Comp. Zoöl.; Osler, Saskatchewan; July 15, 1893; obtained by W. C. Colt; original no. 79 according to describer.
The skull lacks the basioccipital, basisphenoid, and left zygomatic arch. The "crowns" of the lower canines are missing; otherwise the teeth are present and entire. The skin is fairly well made, with soles of hind feet up, in good condition and in summer pelage.
Range.—From northern British Columbia and Great Slave Lake south on the west side of the Rocky Mountains to Ootsa Lake, British Columbia, and on the east side of the Rocky Mountains, south to central Montana, North Dakota and Minnesota; eastward in Canada, entirely north of St. Lawrence River, to Atlantic Ocean. See figure [28] on page [180].
Characters for ready recognition.—Differs from M. r. eskimo in longer tail averaging 19 rather than 16 per cent of length of head and body and extending beyond outstretched hind feet in study skins, rather than to a point short of tips of toes; boundary between brown upper parts and white underparts extending straight across cheeks from upper lip to side of body well below eye and ear, rather than with reëntrant angle from upper lip carrying white upward to point behind eye, and with breadth of rostrum less, instead of more, than 85.5 per cent of orbitonasal length; from M. r. campestris by smaller size: hind foot less than 25 in males and ordinarily less than 22 in females; in males total length less than 216 and tail averaging less than 34, and in females total length averaging less than 182 and tail averaging less than 29; color said to average darker; from M. r. allegheniensis by three average differences, namely lighter color, longer tympanic bullae and larger size of males; from M. frenata and M. erminea of same region by basilar length of skull less than 32; tail less than 50, and lacking black pencil.
Description.—Size.—Male: Six adults and subadults from Shaunavon, Saskatchewan, yield average and extreme measurements as follows: Total length, 202 (188-208); length of tail, 32.5 (31.5-34.0); length of hind foot, 22.8 (21-24).
Female: One adult and 3 subadults from the same area yield average and extreme measurements as follows: Total length, 172 (162-190.5); length of tail, 27.4 (24-34); length of hind foot, 19.6 (17.5-22).
Color.—Winter pelage all white, rarely brown; as described in M. r. eskimo except that line of demarcation on side of head between upper parts and underparts passes almost straight back without the dorsally directed reëntrant area of white behind the eye and ear; least width of color of underparts averaging 52 per cent of greatest width of color of upper parts.
Skull (Based on those from Shaunavon, Sask.)—See measurements and plates [14] and [15]; weight, 0.88 (0.70-0.98) grams in males and 0.55 (0.54-0.56) in females; basilar length, 29.5 (28.4-30.4) in males and 26.1 (24.7-27.0) in females; otherwise as described in M. e. richardsonii.
Remarks.—As early as 1858 (p. 159) Baird recognized an individual of this race from Pembina, Minnesota, as pertaining to a distinct species. Although he used for it the specific name pusillus originally proposed by DeKay for a small weasel from the state of New York, Baird wisely noted that the specimen he described "may be different from the New York species. . . ." After preparing this account, Baird included a second specimen, from Fort Steilacoom, Washington Territory, which he thought might be the same, but the differences that he was careful to point out, in the light of later knowledge, show it to be of the species Mustela erminea. Only a few other naturalists followed Baird in distinguishing the least weasel as a separate species until Bangs in 1896 (p. 21) clearly differentiated it and proposed for it the name Putorius rixosus, which continues in use today and applies to the species.
The accumulation at the National Museum of Canada, through the energy of Dr. R. M. Anderson, of a good series of specimens from Saskatchewan in the general vicinity of the type locality allows for the first time an adequate conception of the amount of secondary sexual variation and individual variation and permits recognition of subspecific characters to differentiate between M. r. rixosa and the subspecies eskimo and campestris. In comparison with the subspecies allegheniensis the basis for segregation is less clear and will remain somewhat in doubt until additional adults of allegheniensis from, say, Pennsylvania, become available with accurate external measurements taken in the flesh and especially with complete skulls.
Intergradation with the subspecies eskimo is suggested by the short tail of the specimen from fifteen miles east of Atlin, British Columbia; in other particulars that specimen, a skin-alone, agrees with the subspecies rixosa. Intergradation with campestris is indicated by increased size of some specimens from North Dakota, and is suggested with allegheniensis by the color of specimens from Wisconsin and Illinois. Three specimens from Winona County, in southeastern Minnesota, unfortunately are skulls-alone without external measurements. Also, two of these skulls are of young animals. The one adult, unsexed, is from Crystal Springs. Selected cranial measurements are: basilar length, 28.5; length of tympanic bulla, 10.9. These measurements accord with those of males of the subspecies rixosa to which the specimens from Winona County, therefore, are here assigned. The possibilities have not been excluded, however, that the adult is an unusually large female of the subspecies campestris or a male of allegheniensis that has tympanic bullae longer than average for that subspecies.
Some hesitation is felt in assigning the specimens, 8 in all, from eastern Canada to the subspecies rixosa. The skin-alone from Eagle River and the skin, with part of the skull, from St. Michael Bay, are in transitional pelage and are of no help in appraising subspecific characters. The one adult specimen which does have a complete skull is from an island south of the Comb Hills. This animal in all respects agrees with selected individuals of M. r. rixosa from Saskatchewan, but each of the five other skins in summer pelage has spots of dark brown color on the breast. Only about one specimen in three of rixosa from Saskatchewan is similarly marked. Furthermore, on some of the specimens from eastern Canada the spots are larger than on any of the animals from farther west. The greater frequency of brown spots on the breast, the larger average size of these spots, and the darker average coloration of the upper parts are suggestive of geographic variation, the existence of which has to be proved by additional and more complete specimens from eastern Canada. For the time being, specimens from there are tentatively assigned to the race rixosa.
Of 56 subadult and adult skulls only 3 (1 North Dakota; 1 Calgary, Alberta; and 1 Island S Comb Hills, Queb.) display lesions resulting from infestation of the frontal sinuses by nematode parasites. None of the young skulls shows such infestation.
Specimens examined.—Total number, 87 as follows. Arranged alphabetically by provinces and states and within each from north to south. Unless otherwise indicated, specimens are in the United States National Museum.
Alberta. Miette River, 1[77]; 5 mi. NW Camrose, 1[77]; Camrose, 2 (1[77], 1[31]); "near Camrose," 2[77]; Forks Blindman and Red Deer rivers, 1[60]; Innisfail, 1[86]; Veteran, 1[93]; Diddsbury [= Didsbury], 1; Calgary, 2 (1[93], 1[2]); Shepard, 1[86].
British Columbia. Clarks Ranch, Halfway River, Peace River Dist., 1[85]; 15 mi. E Atlin, 1[8]; Wistaria, P. O., 3 (2[77], 1[85]); Ootsa Lake, 1[85].
Labrador. Davis Inlet, 1[60]; 30 mi. upriver and 20 mi. toward Groswater Mts., Eagle River, 1; St. Michael Bay, 1.
Mackenzie. Old Fort Reliance, 1[2]; Fort Resolution, 2; Fort Smith, 1.
Manitoba. Gypsumville, 1[86]; Lake St. Martin Reserve, 1[86].
Minnesota. Roseau County: Cedarbend, 2[14]; Grimstad, 1[14]; America, 2 (1[14], 1[74]); Malung, 1[74]; Norland, 1[41]; Falun, 3 (1[14], 1[74], 1[41]); Palmville, 1[41]; Spruce, 1[74]; Stokes, 1[74]. No locality more definite than Marshall County, 1[14]. Clay County: Moorhead, 1[36]. Winona County: "near" Whitman, 1[34]; Altura, 1[98]; Crystal Springs, 1[98].
Montana. Sun River Valley, 1; Wibaux in Wibaux County, 1.
North Dakota. Walsh County: Grafton, 15 (3[60], 1[93], 5[2], 2[14], 1[74], 1[1], 1[76]). McHenry County: 4 and 4-1/2 mi. N Upham, 2. Wells County: 1[36]. Morton County: Mandan, 1[60].
Ontario. Algoma Dist.: Tatnall, near Oba, 1[86]. Moose Factory, 1[75].
Quebec. Island S of Comb Hills, James Bay, 1[9]. Saguenay County: Natashkwan, 1.
Saskatchewan. Osler, 1[75]; "near Regina," 1[77]; Dollard, 2[31]; Shaunavon (and "near" and 1 mi. NE), 9[77]; Klintowel P. O. (about 15 mi. N of Eastend), 1[77]; Eastend and "near" Eastend, 2[77].
Mustela rixosa allegheniensis (Rhoads)
Least Weasel