Coast Salish Woolly Dogs

  • By Russel Barsh
  • Posted 6/22/2016
  • Essay 11243
See Additional Media

Weaving with spun yarns was a defining characteristic of pre-Contact Coast Salish civilization in the Salish Sea (the marine waterways of what are now Washington and British Columbia), together with the cultivation of food plants such as camas (Camassia spp) and the construction of coastal clam gardens. Traditional Salish yarns utilized a variety of gathered materials such as duck down, mountain-goat wool, and fireweed cotton, but the primary material was the hair of a distinct dog breed selected for a long, soft, white undercoat. Woolly-dog flocks were a common part of Coast Salish village life and a substantial expense, and the production of woolens was a significant source of portable, storable wealth for Coast Salish families that was typically controlled by women. Introduction of cheap machine-spun yarns in the 1850s resulted in the neglect of dog flocks and loss of the woolly breed, although many Coast Salish women continued to earn their livelihood from weaving and knitting. Discovery of a single documented specimen of a woolly dog at the National Museum of Natural History in 2003 has made it possible to identify the nearest living relatives of this breed.

Colorful Woolens from Dogs Bred for Clothing

Accustomed to seeing Native peoples of the Pacific Coast clothed in furs and skins, Spanish sailors accompanying Juan Francisco de la Bodega y Quadra (1744-1794) on the first European exploration of the Salish Sea in 1791 were astonished to find Coast Salish villagers in colorfully decorated woven woolens, apparently of their own manufacture. Bodega y Quadra's officers debated what animal could possibly have provided the plush white wool for the yarn, concluding that it must have come from dogs, "partly because when the woven hair was compared with that of those animals no difference was found and partly from the great number of dogs they keep in their villages, most of which were shorn" (Spanish Explorations, 256). The following spring, exploring the vast natural harbor he named Port Orchard Bay in what is now Kitsap County, George Vancouver also saw Coast Salish dogs "all shorn as close to the skin as sheep are in England; and so compact were their fleeces, that large portions could be lifted up by a corner without causing any separation" (Vancouver, 266).

Twenty years later, in his 1808 exploration of the British Columbia river that now bears his name, Simon Fraser encountered finely woven "rugs" made of dog hair or mountain-goat wool (Howay, 88). James McMillan, chief factor at the Hudson's Bay Company establishment at Fort Langley near present-day Vancouver, B.C., recorded Coast Salish people wearing woolen capes and blankets, and saw canoes transporting flocks of shorn dogs in 1828.

"The blankets are many of them made by themselves of dogs wool," reported Augustus Case, an American sailor with the Wilkes expedition, in May 1841 (Charles Wilkes ..., 133). His shipmate George Sinclair wrote in his log for June 10, 1841, "They have no domestic animal except the dog, which are quite numerous" and a source of yarn for spinning, like the sheep of Europe (Charles Wilkes ..., 181).

The London artist Paul Kane (1810-1871), exploring and painting the Salish Sea in 1847, described the everyday winter clothing of Coast Salish men as a blanket "made either of dog's hair alone, or dog's hair and goosedown mixed, frayed cedar bark, or wild goose skin," and explained:

"They have a peculiar breed of small dogs with long hair of a brownish black and a clear white. These dogs are bred for clothing purposes. The hair is cut off with a knife and mixed with goosedown and a little white earth, with a view to curing the feathers" (Kane, 210).

He also described hand-spinning and the use of a loom, both of which also appear in his field sketches and in a large oil painting of the interior of a Coast Salish cedar-plank house, which he completed after his return to London in 1855. Later observers referred to yarn mixtures that included wild-goat wool from the Cascade and Olympic mountains as well as duck down and fireweed cotton.

British naturalist John Keast Lord (1818-1872), who traveled the Salish Sea in the 1850s with the British boundary survey team, reported in his memoirs that "Along the coast several tribes at one time kept dogs of a peculiar breed, having long white hair, that were annually shorn as we shear sheep, and the hair so obtained was woven into rugs, sometimes mixed with the wool of the mountain goat, at others duck feathers, or wild hemp, finely carded" and woven on a loom of their own invention (Lord, 215).

What Did Woolly Dogs Look Like?

George Vancouver likened Coast Salish wool dogs to Pomeranians: a large, foxy-faced German Spitz-type breed with very long plush hair that would have been familiar to English gentlemen. In his oil painting now at the Royal Ontario Museum, Paul Kane depicted a small poodle-like dog sitting patiently beside a Coast Salish loom. The corresponding field sketch (reproduced in Eaton and Urbanek's Paul Kane's Great Nor-West, page 103), more resembles a small Shiba or Husky with a foxy face. Lord surmised that the native dogs actually originated in Japan, where there were dogs of similar appearance.

It is due to accident that the American naturalist C. B. R. Kennerly (1830-1861), while collecting specimens in the 1850s as a member of the American boundary survey team, supplied science with what is now the only known specimen of a woolly dog. On August 19, 1859, Kennerly wrote a letter to Spencer Baird, Assistant Secretary of the U.S. National Museum in Washington, D.C.:

"We got another splendid [mountain] goat skin which was sent to Camp Skagit where Mr. Gibbs happened to be & he took charge of it; but most unfortunately his famous Indian dog "Mutton" got at it and ate the head off. He sent it to me yesterday & when I opened the bag & saw the injury I could almost have cried. Mutton was sheared a short time ago, & as soon as his hair grows out we will make a specimen of him" (Barsh et al., 7).

The reference to shearing indicates that Mutton, who "Mr. Gibbs" -- pioneering ethnographer and linguist George Gibbs (1815-1873), another member of the survey team -- may have obtained from Salish villagers in southwest British Columbia, was a woolly dog. Indeed:

"[USNM specimen number 4762] has a pure white, very dense coat, with unusually long guard hairs and an exceptionally fine, dense undercoat of woolly or cottonlike appearance, perfectly matching the fine frizzy fibers found in 19th century Coast Salish dog hair blankets. The original Smithsonian catalog contains a note that this is 'The dog whose hair is used by the Indians in manufacturing of blankets'" (Barsh et al., 7).

Using DNA extracted from Kennerly's specimen, Robert Fleischer's laboratory at the National Zoo in Washington, D.C., found that Mutton shared Tibetan and/or Siberian ancestors with dog breeds such as the Shibas and Akitas of Japan and Siberian Huskies, thus corroborating John Keast Lord's surmise that woolly dogs were more closely related to Japanese than to European breeds.

More recently, Caroline Solazzo and her colleagues used proteomic techniques to confirm that Coast Salish blankets collected in the early- to mid-nineteenth century were indeed a mixture of dog hair with other fibers including mountain goat wool.

How Woolly Dogs Were Bred

Although many early explorers and settlers recognized the importance of dog-hair weaving to Coast Salish peoples, they recorded little about the dogs themselves, how they were bred, fed, or trained. By the time that anthropologists began raising these questions in the twentieth century, woolly dogs had long disappeared as a distinct, managed breed. Only a few clues survived in the memories of older people.

Canadian anthropologist Diamond Jenness (1886-1969) recorded some woolly-dog traditions in the 1930s:

"The dogs that supplied this hair belonged to a small breed now extinct, though very numerous before the introduction of trade blankets destroyed their utility. During the sockeye and humpbacked salmon season the Indians commonly abandoned them on islands with whatever dried fish remained over from the winter; then they recovered them in autumn and sheared them with mussel-shell knives" (Jenness, 46).

Similarly William Elmendorf learned that the "wool dog" of the Coast Salish people living along Hood Canal "was a special, separate breed with long hair shorn and used in woven textiles," and was known by a name that means "long-haired dog" (The Structure of Twana Culture, 95-97). He wrote:

"The wool dog was a distinct breed and was not used for hunting. Wool dogs lived in the house with their owners and were given special care and a different diet from the hunting breed. Owners tried to prevent their interbreeding with hunting dogs. When the 'wool' was long it was sheared with a knife" (The Structure of Twana Culture, 95-97).

Dogs were named, highly valued and often buried wrapped in a blanket. Dog remains and dog burials are not uncommon in Coast Salish archaeology. A dog buried with a young woman was unearthed on Lopez Island by a University of Washington field school in 1968. DNA extracted from archaeological dog remains provides physical evidence for Coast Salish villages maintaining locally distinct dog breeds.

Stable isotope analysis of a fragment of a woven blanket from the lower Fraser River by Rick Schulting revealed that the hair came from an animal that was primarily eating fish. Indeed, analysis of archaeological remains from the lower Columbia River (an area that was not Coast Salish but part of the Salish Sea trade network) by Kenneth Ames and colleagues found that dogs ate proportionately more fish than their human companions. Dried salmon would have been a perfect year-round dog food, made possible only by the large scale on which the Coast Salish produced dried salmon from reef-net fishing and river weirs.

The Antiquity of Dog Woolens

It has long been recognized that the Coast Salish loom and weaving technique are unique and not derivative of any European, Asian, or South American weaving tradition. The antiquity of this textile tradition is unknown since neither the soft products, nor the wooden implements used in spinning and weaving, have been preserved in archaeological sites thus far investigated inside the Salish Sea.

An indirect clue comes from the archaeology of Coast Salish fisheries. Feeding an adequate number of dogs for a textile industry required a large surplus of dried salmon. A simple hook-and-line fishery could not possibly feed both humans and dog flocks. Hence the rise of large-scale fishing technologies was a pre-condition for weaving. Julie Stein has suggested that rapid population growth in the Salish Sea roughly 2,500 years ago was a result of an intensification of salmon fisheries. Marco Hatch was able to demonstrate, using the genetic identification of salmonid remains in a Lopez Island archaeological site, that reef-net fishing replaced hook-and-line technology at least 1,800 years ago.

The woolly dog itself offers a clue. It was necessarily bred for a consistent fluffy white coat with long guard hairs some time after its ancestors traveled by land or by sea from northeast Asia to northwestern North America. As the study of archaeological dog DNA improves, the dating of changes in genes governing pelage will provide the marker for the rise of a distinctly Salish textile industry.

Why Woolly Dogs Disappeared

The decline of the woolly dog began with the establishment of fur-trading posts at the Fraser River and Nisqually deltas in the 1820s, and their introduction of inexpensive machine-made blankets from mills in the British Isles and New England. Bernard Seeman, who visited Fort Victoria on Vancouver Island in 1846, three years after it opened, observed that the supply of Hudson's Bay Company blankets had lowered the value of indigenous dog-hair blankets. John Keast Lord commented that since the introduction of cheap machine-loomed sheep's-wool blankets by the Hudson's Bay Company, the native woolly dog had become "almost extinct" (Lord, 216).

A traditional blanket required a flock of salmon-eating dogs, whereas a machine-made blanket could be purchased for a few beaver skins. As non-Native settlers streamed into the Salish Sea in the 1850s, bobbins of machine-spun, pre-dyed sheep's-wool yarns also became widely available to Coast Salish weavers and knitters. Coast Salish women continued to make their living from textiles well into the early twentieth century, including as dressmakers and milliners, but without their dogs, which no longer had practical value.

A glass-plate negative taken by James Orville Booen in the mid-1890s (now in the collection of the Chilliwack Museum and Archives) shows a dog in the arms of two young Native women that has been identified as a woolly dog. However, while the pelage is long and white, the wide face does not resemble Mutton's. It would be surprising if examples of woolly dogs survived forty years after the breed's pelage characteristics were no longer valued or maintained. On the other hand, longhaired pure-white dogs were probably rare in Coast Salish communities after the 1860s, and sports with these characteristics may have been called "woolly dogs" without actually sharing the original genotype.

This may explain stories that Elmendorf heard about two men on Hood Canal that purportedly owned woolly dogs between 1920 and 1940. Elmendorf himself concluded that the original woolly breed had "probably been long extinct" (The Structure of Twana Culture, 97). At about the same time, Diamond Jenness wrote that he had "noticed an old, creamy-white dog on the east Saanich reserve that seemed to carry some of the old strain," and learned that it was being sheared for knitting mittens (Jenness, 46). These were isolated cases of individual dogs, with no evidence that selective breeding continued to be practiced anywhere in the Salish Sea.

Other Native Dog Breeds

By no means were all Coast Salish dogs woolly dogs. Early settlers and explorers referred to dogs kept for other purposes including hunting. Indeed, Kennerly collected a second specimen for the Smithsonian, an unsupervised dog that got into Kennerly's skins and paid with its life. This animal (USNM 3512) is similar in size and overall physique to Mutton, but with short light brown hair that would have been useless for spinning yarn.

"Dogs were valuable and were treated accordingly," ethnographer Homer Barnett concluded, adding that those not bred for their hair "were put through a course of training involving magical applications along with a more realistic regimen. They were taught to drive deer and elk into deep water and to raise mountain goats" (Barnett, 123). On Hood Canal, according to Elmendorf's sources, some dogs were specifically trained for hunting. The hunting breed was reputedly "small and thin, in coat and build like a wire-haired terrier, with a ruff around the neck," and disappeared not long after contact with European dogs (The Structure of Twana Culture, 97). Kennerly encountered feral packs of mongrel dogs during the first decade of European settlement.

An extraordinary photograph taken in 1873 near the mouth of the Fraser River shows a number of Coast Salish people standing in front of cedar-plank houses, with at least four dogs visible (Waterman, Plate 8). The dogs are terrier-like but shorthaired, with black markings, and do not resemble Mutton in any way. They appear more closely related to the so-called Tahltan "bear dog," documented in a single historical photograph of the same period.

The Tahltan breed is a useful comparison with the woolly dog. Tahltan people live in the far northwest of British Columbia and are linguistically and culturally unrelated to Coast Salish. The hunting dogs kept by the Tahltan were recognized as a distinct breed by the Canadian Kennel Club in 1942 but had disappeared by the 1990s. Genetically, they formed a distinct group from the woolly dog. Small, shorthaired, black and white, they resembled Fox or Rat terriers, which also were originally bred for hunting, albeit in Bronze Age Europe. Many Native American peoples may have selected their hunting dogs for distinctive shapes and color patterns, but none other than Coast Salish appear to have selected for spinnable hair.

Cultural Significance

Archaeologist Dale Croes argues that weaving technologies represent a "defining characteristic" of the Coast Salish socioeconomic system ("The Salish Sea"). According to Wayne Suttles, women were usually the owners of flocks and looms, as well as cedar-bark-collecting areas for basketry ("Central Coast Salish," 461). The status and power of pre-Contact Coast Salish women was secured by the value of textiles as a medium for storing, trading, and displaying wealth. Coast Salish women continued to earn a livelihood from textiles in the half-century following contact. As elder Mary Hansen told this author 40 years ago, "nice Victorian white ladies had their outfits made for them by nice Indian ladies."

Although looms have also been identified in the remains of seventeenth century cedar-plank houses at the Makah village of Ozette on the Olympic Peninsula's ocean shore, as described by Richard Daugherty, weaving was not observed on the Pacific Coast by early explorers, and may have been learned by the people of Ozette from Coast Salish neighbors, kinfolk, or captives. Indeed, when James Cook (1728-1779) and his Royal Navy expedition stopped at Nootka Sound on Vancouver Island, where Nuu-Chah-Nulth people linguistically related to the Makah lived, he saw a few woolen blankets but the Nootkans could not tell him what animal provided the wool (Howay, 84-85). A generation later, when John Jewett (1783-1821) spent two years as a captive of the Nuu-Chah-Nulth at Nootka Sound, he was told that woolen blankets were "procured from tribes to the south" (Howay, 88).

Similarly, Franz Boas's collaborator George Hunt (1854-1933) recalled seeing a Kwakiutl dog with hair "long like wool" whose owner had spun the hair into yarn for trimming baskets and blankets (Boas, 1317-1318). This would have been in the 1860s, about the time that woolly dogs were going out of fashion among Coast Salish people. It should be borne in mind that both the Makah at Ozette and Neah Bay, and the Kwakiutl of the northern Strait of Georgia, lived on the Coast Salish frontier and could have learned weaving from neighboring Coast Salish villages.

Dogs eat flesh; sheep eat herbage. Feeding a herd of dogs with salmon required diverting a substantial and expensive part of the supply of food for people. All other things being equal, dog woolens required far more labor than sheep woolens. Coast Salish people of the mid-nineteenth century made the practical decision to exchange their salmon for machine-made sheep-hair yarns and cloth that they could cut, stitch, and sell back to European settlers.

Note: This article is part of Cultivating Washington, The History of Our State’s Food, Land, and People, which includes more agriculture-related content, vidoes, and curriculum. 


Russel Barsh, Wayne Suttles, and J. Megan Jones, "History, Ethnography, and Archaeology of the Coast Salish Woolly-dog," in People and Dogs in Social, Working, Economic or Symbolic Interaction ed. by Lynne Snyder and Elizabeth Moore (Oxford: Oxbow Books, 2006), 1-11; Kenneth M. Ames, Michael P. Richards, Camulla F. Speller, Dongya Y. Yang, R. Lee Lyman, and Virginia L. Butler, "Stable Isotope and Ancient DNA Analysis of Dog Remains from Cathlapotle (45CL1), a Contact-era Site on the Lower Columbia River," Journal of Archaeological Science, Vol. 57 (2015), 268-282; Spencer F. Baird Collection, Record Unit 7002, Box 26, Smithsonian Institution Archives, Washington, D.C.; Homer G. Barnett, The Coast Salish of Canada, American Anthropologist, Vol. 40, No. 1 (1938), 118-141; Russel L. Barsh, "Backfire from Boldt: The Judicial Transformation of Coast Salish Proprietary Fisheries into a Commons," Western Legal History, Vol. 4, No. 1 (1991), 85-102; Russel L. Barsh, "Puget Sound Indian Demography 1900-1920: Migration and Economic Development," Ethnohistory, Vol. 43, No. 1 (1996), 65-97; Charles Wilkes and the Exploration of Inland Washington Waters: Journals from the Expedition of 1841 ed. by Richard W. Blumenthal (Jefferson, NC: McFarland & Company, 2009), 133, 136, 152-53, 181; Franz Boas, Ethnology of the Kwakiutl, Based on Data Collected by George Hunt, 35th Annual Report of the Bureau of American Ethnology (Washington, D.C.: Government Printing Office, 1921); "Proof That the Sto:lo Wooley Dog Did Exist," Canada's History website accessed May 15, 2016 (; Susan J. Crockford and Cameron J. Pye, "Forensic Reconstruction of Prehistoric Dogs from the Northwest Coast," Canadian Journal of Archaeology, Vol. 21, No. 2 (1997), 149-153; Dale R. Croes, "The Salish Sea: Using Wet and Dry Site Archaeology to Explore the Defining Characteristics of this Inland Sea, Journal of Wetland Archaeology, Vol. 15, No. 1 (2016), 72-108; Richard D. Daugherty, "Problems and Responsibilities in the Excavation of Wet Sites," in Wet Site Archaeology ed. by Barbara A. Purdy (Boca Raton, FL: CRC Press, 1990), 15-28; Diane F. Eaton and Sheila Urbanek, Paul Kane's Great Nor-West (Vancouver: UBC Press, 1995); William W. Elmendorf and A. L. Kroeber, The Structure of Twana Culture: With Comparative Notes on the Structure of Yurok Culture (Pullman: Washington State University Press, 1960), 94-99; Alexandra Harmon, Indians in the Making: Ethnic Relations and Indian Identities Around Puget Sound (Berkeley: University of California Press, 1998); Paula Gustafson, Salish Weaving (Vancouver: Douglas & McIntyre, 1980); Marco B. A. Hatch, "Identification of Archeological Sockeye Salmon Remains as a Proxy for Reef-net Fishing" (undergraduate honors thesis, University of Washington School of Fisheries and Aquatic Sciences, 2005); Frederic W. Howay, "The Dog's Hair Blankets of the Coast Salish," Washington Historical Quarterly, Vol. 9, No. 2 (1918), 83-92; Diamond Jenness, "The Saanich Indians of Vancouver Island," 1934-1935, Manuscript 3 VII-G-8M, Canadian Museum of History, Gatineau, Quebec; Paul Kane, Wanderings of an Artist Among the Indians of North America (London: Longman, Brown, Green, Longmans, & Roberts, 1859), 210; Mary Lois Kissell, "Organized Salish Blanket Pattern," American Anthropologist, Vol. 31 (1929), 85-88; Ben F. Koop, Maryann Burbridge, Ashley Byun, and Ute Rink, "Ancient DNA Evidence of a Separate Origin for North American Indigenous Dogs" in Dogs Through Time: An Archaeological Perspective, Proceedings of the 1st ICAZ Symposium on the History of the Domestic Dog, August 23-29, 1998, Victoria, B.C., Canada, ed. by Susan Janet Crockford (Oxford: British Archaeological Reports International Series 889, 2000), 271-285; John Keast Lord, The Naturalist in Vancouver Island and British Columbia, vol. 2 (London: R. Bentley, 1866), 215-216; Jennifer A. Leonard, Robert K. Wayne, Jane Wheeler, Raúl Valadez, Sonia Guillén, and Carles Vilà, "Ancient DNA Evidence for an Old World Origin of New World Dogs," Science, November 2002, pp. 1613-1616; The Fort Langley Journals, 1827-30 ed. by Morag MacLachlan (Vancouver: UBC Press, 2005), 90; Rick Schulting, "The Hair of the Dog: The Identification of a Coast Salish Dog-hair Blanket from Yale, British Columbia," Canadian Journal of Archaeology, Vol. 18 (1994), 57-76; Robert Fleischer to Russel Barsh, April 11, 2006, copy in possession of Russel Barsh, Lopez Island, Washington; Caroline Solazzo, Susan Heald, Mary W. Ballard, David A. Ashford, Paula T. DePriest, Robert J. Koestler, and Matthew J. Collins, "Proteomics and Coast Salish Blankets: A Tale of Shaggy Dogs?" Antiquity, Vol. 85, No. 330 (2011), 1418-1432; Julie K. Stein, Exploring Coast Salish Prehistory: The Archaeology of San Juan Island (Seattle: University of Washington Press, 2000); Wayne Suttles, "Central Coast Salish," in Handbook of North American Indians, Vol. 7: Northwest Coast ed. by Wayne Suttles (Washington, D.C.: Smithsonian Institution, 1990), 453-475; Wayne Suttles, "Prehistoric and Early Historic Fisheries in the San Juan Archipelago," unpublished report prepared for National Park Service, San Juan Island National Historical Park, Friday Harbor, Washington; George Vancouver, A Voyage of Discovery to the North Pacific Ocean, and Round the World, in Which the Coast of North-West America Has Been Carefully Examined and Accurately Surveyed, Vol. 1 (London: Robinson and Edwards, 1798), 266, 284, copy available at Internet Archive website accessed April 20, 2016 (; Spanish Explorations in the Strait of Juan de Fuca ed. by Henry R. Wagner (Santa Ana, CA: Fine Arts Press, 1933), 256; T. T. Waterman, North American Indian Dwellings, Annual Report of the Board of Regents of the Smithsonian Institution, 1924 (Washington, D.C.: Government Printing Office, 1925), 461-485, Plate 8; Accession 45SJ280/1996-121, Burial 1 (1968 University of Washington Field School, David A. Munsell), Accession Records, Washington State (Burke) Museum, Seattle, Washington; Kasia Anne Zimmerman, "Changing Ways, Constant Companions: The Ancient DNA and Local Knowledge of Tla'amin Dogs" (master's thesis, Simon Fraser University, 2014).

Licensing: This essay is licensed under a Creative Commons license that encourages reproduction with attribution. Credit should be given to both and to the author, and sources must be included with any reproduction. Click the icon for more info. Please note that this Creative Commons license applies to text only, and not to images. For more information regarding individual photos or images, please contact the source noted in the image credit.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License
Major Support for Provided By: The State of Washington | Patsy Bullitt Collins | Paul G. Allen Family Foundation | Museum Of History & Industry | 4Culture (King County Lodging Tax Revenue) | City of Seattle | City of Bellevue | City of Tacoma | King County | The Peach Foundation | Microsoft Corporation, Other Public and Private Sponsors and Visitors Like You