Friday, January 6, 2017

How Many Furs, What Kinds, What Were They Worth?

The Hudson's Bay Company's Latin motto was "pro pelle cutem," which roughly translates as "skin for pelt," the skin alluding to the diligent efforts of Company employees in collecting the beaver depicted on its coat of arms. The focus was on beaver pelts, used mainly for the fur stripped from the skin and made into fine felt hats, but the HBC collected many other types of animal skins, often used for fur coats, capes, and trim on clothing. In the reconstructed fur store at Fort Vancouver visitors can see the pelts of otter, lynx, fox, wolf and other fur bearing animals.

Since the business was furs, the focus was on beaver, yet other furs were traded, several questions arise: What was a beaver pelt worth? What other pelts were collected and what were they worth, by themselves and in relation to beaver? How many pelts, beaver and others, were collected in a year? How much were they worth in total? Is there any way to estimate the values in today's dollars? Fortunately, the shipping manifests sent with the furs to London help provide answers to these and related questions.

Each fall, usually in October or November, a ship left Fort Vancouver for England, laden with furs and other products from the Columbia District and beyond, from Alaska to California, that had been accumulating at the District depot since the previous fall. In documenting each shipment, clerks listed the numbers of pelts by species and type, the value of each pelt, and the total value of each type of pelt. The actual income from these returns was eventually determined as much as a year later at public auctions in London, where furs were sold by the pound, not per pelt. We have no record of those auction results, but the assessed values in the shipping manifests provide a reasonably good proxy.

This "valuation" was done for two reasons: to provide a monetary value for insurance purposes during the sea voyage, and as the basis for a preliminary calculation of profit or loss. The Company would not want to pay for insurance over what they thought the shipment was going to yield when sold, and they wanted to get an accurate estimate of potential income to compare with expenses incurred. These figures are also good for showing the relative values of the various species and types of pelt.

This analysis starts with archival data on the annual shipment of furs and other products sent to England in November 1844, follows with a reworking of that information for clarity and additional calculations, and concludes with interpretive commentary on the data.

The image below is the summary sheet for the returns shipped from Fort Vancouver on November 13, 1844: "Valuation of Furs, Returns of Columbia [District], Outfit 1843 & 1844, shipped to England per Barque Columbia 1844." The columns are: number, species/type, value of each item (in shillings/pence), total value (in pounds/shillings/pence).

The spreadsheet below is a reworking of the same data, separating furs from non-fur sundries and sorting furs by the total value of each species or type (column F), in descending order.

Numbers presented this way can quickly cause one's eyes to glaze over, so let's try breaking the table down, picking up some interpretive points in the process.

Columns A and B are from the source document, with peltries totaling 57,354. Sundries include animal skins used for leather, not for fur, and items measured by the pound, such as castoreum, the contents of a mature beaver's musk glands, used in making perfume. Each adult beaver yielded only a few ounces of castoreum, so the 264 pounds in this shipment suggests that it was carefully retrieved from many of the 13,430 large beaver shown. Other sundries include isinglass, a substance found in the flotation bladders of sturgeon, used in clarifying beer and wine, and whalebone, used in corset stays. Feathers don't show up in every annual shipment, so those shown here may have been accumulated over several years. But we might wonder how many geese or swans had to be plucked to yield 168 pounds of feathers.

Column C shows the percentage contribution of each species to the total number of pelts. Thus we see that the 17,290 beaver (large + small) made up just over 30% of all pelts, while 21,236 muskrats made up 37%. Add the 7,900 martens, nearly 4,000 mink, and 1,428 raccoons to account for just over 90% of the pelts shipped--from just five species. The value per pelt in Column D is from the source document. Only sea otter(144s/4p), silver fox (98s), brown bear (71s), and grizzly bear (35s), were valued higher than a large beaver (32s). On the bottom of this measure is the lowly muskrat, at 6 pence (half a shilling) each.

Column E gives one large beaver (that is, the "Made Beaver" used as a unit of value for trading purposes) the value of 1, and the value of other species in relation to that unit. This shows at a glance that, for example, one sea otter was assessed at 4 1/2 Made Beaver, and a silver fox pelt was worth 3 Made Beaver. One muskrat, in contrast, was worth a tiny fraction of a Made Beaver. Put in other words, at 1/2 shilling each, it took 64 muskrat pelts to equal the value of one large beaver.

Column F, the total assessed value of each species or type within species, is from the source document, here sorted in descending order. Column G calculates that ranked list as a percentage of the value of all furs. This brings us back to focus on beaver, which accounted for 30% of all pelts by number, but made up nearly 70% of the assessed value of all pelts. Sea otters (worth 4 1/2 beaver pelts each) accounted for less than 2% of the value of all pelts, because there were just 89 sea otters in the Columbia's cargo. Similarly, each silver fox pelt was valued at three times the value of a large beaver, but there were only 48 silver fox pelts in this shipment, so they accounted for less than 1% by value. On the other end, the lowly muskrat made up 37% of all pelts by number, but accounted for just 1.5% by value.

Column G also shows that for all the range of species and types within species, the top five items together (beaver, marten, black bear, brown bear, and land otter) accounted for 90% of the value of the peltries in this shipment.

Columns H and I use the monetary conversion and inflation calculations from the website to give estimated conversions of 1844 British currency in today's US dollars. It turns out that the mid-range of those calculations puts £1 in 1844 at about $120 today, making each shilling $6 and each pence $.50 today--very approximately, but also conveniently (see the note below for more on this). These estimates put the assessed value of one large beaver pelt at about $190 today, and the dollar value of all beaver pelts in this shipment at something like $2.9 million today--again helping to explain the focus on beaver. And while "thirty-six thousand pounds in 1844" might not mean much, saying that the total value of this single annual shipment "would be something like 4.3 million dollars today" might make some ears to perk up in the back row. 


The source of this data is HBCA B/223/d/154, "Invoice of Furs and Peltries, Returns of the Columbia District Outfit 1844, Shipped per John McLoughlin Esquire on board the Barque Columbia, Alexander Duncan Master, for London . . ." From microfilm and digitized images in the ARC of Fort Vancouver NHS. In his letter to Governor & Committee of November 20, 1844, John McLoughlin reported: "The Barque Columbia under the command of Captain Duncan left this place on the 13th Instant . . . with the furs of the season." (The Letters of John McLoughlin, from Fort Vancouver to the Governor and Committee, third series, 1844-46, Toronto: Champlain Society, 1944, p. 52). Some information from this document is in John Hussey, Historic Structure Report, Fort Vancouver National Historic Site, Vol. 2, (1976), pp. 34-35. Hussey's purpose was to suggest what furs might be displayed in the reconstructed fur store, so he listed the number of furs by species, but not the assessed value of each type of pelt.

For historical monetary conversions is a best effort by professional economic historians to deal with the various problems of comparing values over time and across currencies. They also understand that whatever the methodology, no single answer can be considered "correct" nor "exact." Using the "Conversion ($ and £)" comparator, British pounds of 1844 to U.S. dollars of 2014, and the CPI result, gives an average value of £100=$12,300 (in a range from $8,040 to $16,500). For these estimates I have rounded that average down slightly to £1=$120. In pre-1971 British currency there were 20 shillings in a pound and 12 pence in a shilling, or 240 pence in a pound. These approximations thus make each 1844 shilling equal to $6.00 and each 1844 pence worth $.50, in today's U.S dollars. While these conversions must be considered only rough approximations, they are surely better than a wild guess or a shrug of the shoulders, when faced with the obvious question, "How much would that be in today's dollars?"

Friday, July 1, 2016

HBC Fur Returns for the Columbia Department, 1826-46

How many beaver and other furs did the Hudson's Bay Company take in? What were the trends over time? This posting supplies some answers to questions central to HBC's existence, and its presence at Fort Vancouver.

Background: The Hudson's Bay Company expanded into the Pacific Northwest when it merged with the Northwest Company in 1821, with Fort George (formerly Astoria) as regional headquarters. In 1824 HBC Governor George Simpson and Chief Factor John McLoughlin arrived from Canada to reorganize operations west of the Rocky Mountains. The regional headquarters was relocated to Fort Vancouver in 1825, and a plan was launched to expand trading operations into the coastal areas from Puget Sound to Russian-held Alaska, complementing the trading posts the NWC had established in its New Caledonia district (in the interior of modern-day British Columbia) and the upper Columbia River. At the same time, HBC began to send large annual trapping expeditions, called brigades, east into the southern Rockies and south into southern Oregon and northern California.

For the next two decades the HBC controlled the fur trade in what it called the Columbia Department--from Alaska to California and from the Rocky Mountains to the Pacific coast. The settlement of the international boundary along the 49th parallel in 1846 and the concurrent transfer of HBC's Department headquarters to Fort Victoria signaled the end of the era dominated by the fur trade. These graphs display the quantitative results in visual terms. The furs acquired, whether by trading (which predominated north of the Columbia River) or by trapping and hunting (which predominated to the east and south) were collected at Fort Vancouver for shipment to England. They were called "returns" followings the longstanding fur trade term for sending trade goods and supplies into the hinterland each fall, with the results of trading and trapping being returned to regional headquarters in the spring.

Each annual cycle was called an "outfit," and the trade goods distributed each fall were matched against the furs returned each spring, to calculate annual profit or loss. HBC accounting practice was to begin each outfit, what would now be called its fiscal year, on June 1, and end it on May 31. The annual data discussed here refer to the outfit years imbedded in HBC records. Thus 1826 starts on June 1 of that year, and ends on May 31, 1827--and so on.

Figure 1, Beaver Returns, by Region, 1826-1846: The main focus of trapping and trading was to acquire beaver pelts, primarily for fur used to make the felt hats that had been popular in Europe (and later in North America) since the 1600s. At first beaver had been acquired in northern Europe, then in eastern North America. By the early 1800s coastal trading ships, mainly from New England (thus called "Bostons" by Natives) were trading for furs on the Northwest coast, and the Montreal-based Northwest Company was trading and trapping in the northern Rockies and along the Columbia River. The Beaver Returns graph shows:
By 1831 HBC's successful trading operations on the Northwest Coast had forced out American competitors, increasing average beaver returns from the northern regions (yellow in the graph, roughly equivalent to modern British Columbia) from about 9,000 to about 13,000 per year.

For the next decade, 1831-41, beaver returns were notably steady about 21,000 per year for the Columbia Department as a whole. About 12,000 per year, nearly 60 per cent, came from the north, Through the early 1840s beaver returns trended downward to just over half what they were at the high point in the 1830s. This reflected both the decline of beaver populations, and a change in men's fashion from felt to hats made of silk, which reduced the market for beaver.

Over these 21 years, 1826-46, HBC took nearly 390,000 beavers from the Columbia Department. That accounts for most, but not all, of the beavers taken from the region--American coastal traders operated through the late 1820s, and there was occasional American competition inland. Of the 390,000 total, about 221,000 (57%) were from what's now British Columbia, 113,000 (29%) were from the middle Columbia River posts and the trapping brigades, and nearly 56,000 (14%) were from the area around Fort Vancouver and the Oregon coast.

Beaver returns for the middle Columbia River posts and the land-based trapping brigades (green on the graph) held fairly steady until about 1835, grew markedly from 1837 to 1840, then trended downward. Beaver returns for Fort Vancouver Indian Trade (the trade shop at Fort Vancouver plus Fort George after it was reopened in 1829 and Fort Umpqua after 1832, blue on the graph) trended slowly downward from the late 1820s.

Figure 2, Non-Beaver Returns, by Region, 1826-1846:

While beaver pelts were HBC's stock-in-trade, the Company dealt in the skins of many other fur-bearing animals. Although it receives little notice in fur trade history, the most numerous was the lowly muskrat. Fur from muskrats was also used for felt hats, but they are much smaller than beaver, and their fur is shorter and of lower quality. In trade, HBC valued 15 muskrat pelts as equal to one beaver. Next came marten, mink, and land otters, members of the weasel family whose more valuable pelts were used for coats, capes, and trim on winter clothing.

To get an idea of the relative numbers of non-beaver pelts for the Columbia Department as a whole, outfit 1841 serves as a sample year. In that year, HBC collected 18,599 beavers in the Columbia Department. The non-beaver returns for the entire region ranked as follows, showing that muskrat, mink, marten, and land otter together accounted for 80 percent of non-beaver fur returns for 1841:

Species       pelts      percent
Muskrat    15,333       38.3
Mink           6,763       17.0
Marten        6,214       15.6
Land otter   3,515         8.8
Bear            2,367         6.0
Lynx           1,810         4.6
Raccoon     1,417         3.6
Fisher            667         1.7
Fox                513         1.3
Wolf              464         1.2
Badger           291        0.7
Wolverine      227        0.6
Sea otter        158        0.4

For the northern region (yellow on the graph), non-beaver fur returns: Nearly tripled from 1830 to 1833, Climbed to new highs in 1838 and 1839, After a dramatic fall in 1840 climbed again to new records in 1845 and 1846, Accounted for 64% of the Columbia Department total. For the middle Columbia and trapping brigades (green on the graph), non-beaver returns: Tripled from 1830 to 1834, Held fairly steady, with significant ups and downs through 1846, Accounted for 32% of the Columbia Department total. 

For the area around Fort Vancouver and southwest Oregon (blue on the graph) non-beaver pelts were much less numerous than beaver, held fairly steady with a significant bump in 1843 and 1844, and accounted for just 4% of the Columbia Department overall total. Here land otter was the dominant non-beaver species. From 1826 to 1846, HBC collected 55,748 beavers here, and in that 21-year period 30,365 non-beaver pelts were collected by Fort Vancouver Indian Trade, distributed as follows:

Species           pelts           percent
Land otter     17,016            56.0
Muskrat          8,557            28.2
Mink               1,049              3.5
Bear                   890              2.9
Raccoon             743              2.4
Fox                     534              1.8
Sea otter             480              1.6
Wolf                    363             1.2
Lynx                    304             1.0
Fisher                  201             0.7
Marten                169             0.6
Badger                  50             0.2
Wolverine               8             0.0
Seal                         1             0.0

Figure 3, Beaver and Non-Beaver Returns for the Columbia Department, 1826-1846:

This graph provides a visual comparison of the annual beaver returns (dark grey below) to non-beaver returns (light grey above), for the Columbia Department as a whole. It shows: When individual animals are tallied, from 1831 through 1846 non-beaver pelts always outnumbered beaver. For the 21 years shown, HBC collected just over one and a quarter million (1,269,320) furs in the Columbia Department. Of that total, about 390,000 (31%) were beaver, and nearly 880,000 (69%) were non-beaver species.

For the high point years of 1838 and 1839, more than three times as many non-beaver pelts as beaver were collected. After a sharp drop in 1840, non-beaver returns recovered significantly, while beaver returns continued to decline steadily. In 1846, the year of the Oregon boundary settlement, non-beaver returns reached 67,205, their highest to that date, accounting for 84% of the total 80,163 furs collected in that year.

Sources: For Fort Vancouver Indian Trade: Lester A. Ross, “Fort Vancouver, 1829-1860: A Historical Archaeological Investigation of the Goods Imported and Manufactured by the Hudson’s Bay Company,” (Typescript, Fort Vancouver NHS, 1976), Table 64, p. 1326. Data for 1835 are not available, and have been interpolated. For Columbia Department: James R. Gibson, Farming the Frontier: The Agricultural Opening of the Oregon Country, 1786-1846 (Vancouver, BC: UBC Press, 1985), Table 30, p. 201.

Sunday, September 6, 2015

Columbia Boats in the Fur Trade Era

Traveling overland in pre-railroad times was an arduous affair, moving freight was more so, and over mountainous terrain even worse. Compared to walking with a load on one's back or dealing with a string of semi-wild pack horses, transportation by water was usually more efficient both in energy expended and size of loads moved. Early explorers of western North America, including Lewis and Clark, McKenzie, Fraser, and Thompson, all looked for water routes, and used them wherever possible. They discovered that the many chains of ridges collectively known as the Rocky Mountains blocked river navigation from the east. The few rivers west of the Rockies where boats could travel were periodically punctured by impassible rapids and waterfalls. That meant that vessels moving passengers and freight had to be large enough for cargo but light enough to carry around obstacles. The result was the Columbia boat.

This painting by British Army officer Henry Warre shows a Columbia boat in use.
Hauling Up A Rapid (Les Dalles Des Morts) On The Columbia River.  Lt. Henry Warre, 1846
Image PDP00057 courtesy of the Royal BC Museum and Archives, Victoria, BC, Canada

Hudson's Bay Company carpenters, some of whom were specialized ship carpenters, built a variety of watercraft in the Columbia District. For example, this snippet from the Spring 1841 inventory of Fort Vancouver shows six Columbia boats (using that name), six batteaus, one ship's boat, and one scow:

(Source: HBC Archives, B.223/d/136, Fort Vancouver Inventory, 1841.)

In the Great Lakes and the extensive navigable river systems east of the Rockies, fur trade voyageurs famously used birch bark canoes in a range of sizes. Some bark canoes were used west of the Rocky Mountains in the early years of the fur trade, but they soon gave way to watercraft made entirely of wood, from canoes of wooden planks to larger batteaux, a French term sometimes used interchangeably with Columbia boat. (East of the Rockies, wooden craft called York boats came into general use in the 1800s. Named after York Factory, the HBC headquarters on Hudson Bay, York boats were larger than Columbia boats, and too heavy to be carried over the portage trails around the falls and rapids of the western rivers.)

Several descriptions of Columbia boats are scattered through accounts left by contemporary observers who lived in or visited the Pacific Northwest during the fur trade era. Details vary, but one common element was light weight to facilitate portaging. Boats ascending the Columbia from Fort Vancouver first had a four mile portage around the Cascades (present site of Cascade Locks, just above Bonneville Dam). Next there was a carry of 12 miles from the base of the Dalles to above Celillo falls (called the Chutes in early accounts). Some other rapids could be "run" coming down, and boats emptied of their cargo could be "lined" going up or down--left in the water and pulled by a crew walking on a path along the riverbank. But depending on water levels portaging was often necessary at other spots on the trip to Okanogan, terminus of the horse brigades bringing fur returns from Thompson River Post (modern Kamloops) and points north. Annual Express Brigades across the continent faced portages to pass Kettle Falls at Fort Colvile and other rapids on the way to Boat Encampment, the end of water transport at the base of the Rockies, nearly 900 miles upriver from Fort Vancouver.

Another detail frequently mentioned is the use of gum to seal the seams between sheathing planks. This sealant was made by heating the resin of evergreen trees and mixing it with tallow to make it soft enough to spread on the hull. A burning stick of firewood was then rubbed along the surface to melt the gum into the seams. Every boat carried a keg of gum, and journals of river trips often belabor the need to repair and replenish the gum on the hull seams, especially in warm weather.

Alexander Ross arrived with the Astorians in 1811, was posted to Okanogan that year, and worked for the Northwest Company then HBC over the next fifteen years. In his Fur Traders of the Far West, (p. 72) Ross noted that due to the "great partiality" of the Northwesters for bark canoes, "the country was ransacked for prime birchbark more frequently than for prime furs." Such preferences notwithstanding, they settled for plank boats:

"The description of craft used on the waters of Columbia by the Astor Company consisted of split or sawed cedar boats, strong, light and durable, and in every possible way better adapted to rough water than the birch-rind canoes in general use on the east side of the mountains. They carried a cargo or burden of about 3,000 lbs. weight, and yet, nimbly handled, were easily carried across the portages."

John Dunn worked was an HBC clerk and postmaster in the Columbia District for eight years in the 1830s. In his History of the Oregon Territory, pp. 60-61, Dunn wrote:

"There are no birch canoes used in the Oregon country by the company's servants. They use only the batteaux, which are made of quarter-inch pine boards, and are thirty-two feet long, and six and a half feet wide in midships, with both ends sharp, and without a keel—worked, according to the circumstances of the navigation, with paddles, or with oars. These boats are found to be better adapted to the lakes and rivers there, than the canoes of the north."

U.S. Navy Lieutenant Charles Wilkes, an experienced mariner who conducted detailed reconnaissance in this area in 1841, provided this description of Columbia boats (Narrative, IV, p. 371):

"The boat was somewhat of the model of our whale-boats, only much larger, and of the kind built expressly to accommodate the trade: they are provided yearly at Okonagan, and are constructed in a few days: they are clinker-built, and all the timbers are flat. These boats are so light that they are easily carried across the portages. They use the gum of the pine to cover them instead of pitch."
Later (P. 378) Wilkes gave more detail:
"The shape of these boats has been before described: they have great strength and buoyancy, carry three tons weight, and have a crew of eight men, besides a padroon. they are thirty feet long and five and a half feet beam, sharp at both ends, clinker-built, and have no knees. In building them, flat timbers of oak are bent to the requisite shape by steaming; they are bolted to a flat keel, at distances of a foot from each other: the planks are of cedar, and generally extend the whole length of the boat. The gunwale is of the same kind of wood, but the rowlocks are of birch. The peculiarity in the construction of these boats is, that they are only riveted at each end with a strong rivet, and being well gummed, they have no occasion for nailing. They answer, and indeed are admirably adapted to, all the purposes for which they are intended; are so light as to be easily transported over the portages by their crews, and in case of accident are easily repaired."

This detailed description does not mention a sail, but later in his Narrative, (p. 379), telling of his trip from Ft Vancouver to Cowlitz, Wilkes wrote: "They are provided only with a square sail, as the wind blows generally either directly up or down the river." Then describing a trip up the Columbia between the Cascades and the Dalles (p. 381): "On the 30th of June, they had a favourable wind, but it blew so hard that they were obliged to reef their sail, and afterwards found the waves and wind too heavy for them to run without great danger." So these boats could be fitted with a mast and square sail, as seen in some illustrations of the larger and heavier York boats used east of the Rockies. The mast would have been removable to facilitate portaging, and mounted equidistant from each end, to be used going in either direction. Steering was with a paddle or oar at either end (or both ends when needed) rather than a fixed rudder, as Wilkes described (p. 391-2):

"Each boat has also its bowman, who is considered the first officer and responsible man; the safety of the boat, in descending rapids particularly, depends upon him and the padroon, who steers the boat. They both use long and large blade-paddles; and it is surprising how much power the two can exert over the direction of the boat."

Peter Burnett, an American pioneer who arrived in this area via the Oregon Trail in 1843, described the Columbia boat that took him from Walla Walla to Fort Vancouver (Recollections, p. 128):

"We procured from Mr. McKinley, at Walla Walla, an old Hudson's Bay Company's boat, constructed expressly for the navigation of the Columbia and its tributaries. These boats are very light, yet strong. They are open, about forty feet long, five feet wide, and three feet deep, made of light, tough materials, and clinker-built. They are made in this manner so that they may be carried around the falls of the Columbia, and let down over the Cascades. When taken out of the water and carried over the portage, it requires the united exertions of forty or fifty Indians, who take the vessel on their shoulders, amid shouts and hurras, and thus carry it sometimes three fourths of a mile, without once letting it down. At the Cascades it is let down by means of ropes in the hands of the Canadian boatmen."

Burnett's recollection of the number of Indians required for portage seems excessive, unless much of the cargo was left in the boat while it was carried around the rapids. Also, if fifty men tried to crowd under a boat of this size, they would likely be tripping over one another. Other accounts say that the standard crew of 8 men could carry an empty Columbia boat when necessary, or that the crews from two boats might be combined to portage one boat. W.H. Grey, who came west with the Whitman-Spalding party and made the trip down the Columbia from Walla Walla to Fort Vancouver in 1836, provided this description of the portage at The Dalles (History, p. 146):

"The boats were let down with lines as near the fall as was considered safe, hauled out of the water, turned bottom up, and as many Indians as could get under them, say some twenty-five to each boat, lifted them upon their shoulders and carried them to the water below. For this service they each received two dried leaves of tobacco, which would make about six common pipefuls."

The passage that follows is from the report of British Army Lt. Mervin Vavasour, who with Lt. Henry Warre made a reconnaissance mission to this region in 1845 ("Secret Mission," p. 145). The crew of 5 oarsmen, presumably plus a steersman and bowman, is similar to other accounts, and the description of the boat matches the more detailed version by American Charles Wilkes from 1841:

"I left Fort Colville on the 19th August, embarked below the [Kettle] Falls in a boat belonging to the H. B. Company expressly adapted to this dangerous river navigation, and descended the rapids. These boats are built of cedar after the model of a bark canoe, the planks being rivetted to the ribs, having no knees,and the seams filled with pitch and gum. They are propelled with oars by 5 men and steered with a paddle. . . . having passed several rapids, at one of which we found it necessary to carry the baggage, and the boat being let down by a line, we reached Okanogan, a small post on the right bank of the river 138 miles from Colville. This post is used as a provision station for the Brigade crossing the Mountains in the spring."

Other contemporary accounts, including those by HBC Governor George Simpson, use the term batteau (plural batteaux, which is simply the French word for boat, standard spelling bateau/bateaux) or the English word boat, with little further description and without making a distinction for Columbia boat.

The secondary literature also includes relevant commentary on Columbia boats. Whitman College professor William Denison Lyman wrote a comprehensive overview of The Columbia River. On p. 134 he gives this description:

"The bateaux were boats of peculiar shape, being built very high and broad so that in an unloaded condition they seemed to rest on the water almost like a paper shell. Both ends were high and pointed as prows. They were propelled with oars and steered with paddles. One of the usual size was about thirty feet long and five feet wide. Being of light-draft, double-enders, capable of holding large loads and yet easily conveyed around portages, more steady and roomy than canoes, these bateaux were the typical Columbia River medium of commerce during the era of the fur-traders."

Dennis F. Johnson has written a book on York boats, the larger and heavier river craft used east of the Rocky Mountains, and a summary article, in Wooden Boat magazine, In the article (P. 56) Johnson makes this brief mention of boats west of the mountains, different from York boats:

"West of the Rockies, the construction of craft on the Fraser and Columbia rivers was different from that of York boats. One type used on the Columbia was described by company employee John Work on his journey from York Factory to the Columbia District in 1825: 'These boats were about 30 feet long with a 5 foot beam, clinker-built and pointed at both ends . . .  Planks of cedar formed their outer skin, since nails were scarce they were only used to secure the planks to the stem and stern piece. The overlapping seams were gummed with pitch to render the craft watertight.'"

Geographer James R. Gibson's comprehensive study, The Lifeline of the Oregon Country includes this description of Columbia boats (p. 102):

"The vessels were made at Fort Colville (previously at Spokane House) from quarter-inch yellow pine or red cedar boards, 30-32 feet long and 5 1/2-6 1/2 feet at the beam, pointed at both ends, large enough to transport up to 50 pieces [of 90 lbs. each] (or up to 500 bushels of wheat), and light enough to be navigated and shouldered by a crew of 8-10 men. In the fall of 1825 at Spokane House, sawyers in fifteen days finished cutting enough boards for three boats: 73 boards 6 inches wide and 40 feet long for sideboards, 3 pieces 14 inches wide and 40 feet long for keels, and 6 pieces 2 inches wide and 40 feet long  for gunwales, so each vessel comprised twenty-four 40-foot boards, 12 per side, making each about 3 1/2 feet high (allowing 1 1/2 inches for overlap and 1 inch for curvature) and about 35 feet long (allowing 5 feet for curvature). Batteaux varied somewhat in their dimensions, depending on when and where they were made, perhaps as well as by whom."

Building Columbia boats:

The following account of on-the-spot construction of five boats for use on the rivers on the western side of the Rockies at Fort St. James, on the east shore of Stuart Lake in the HBC District of New Caledonia, now north-central British Columbia. It covers a period of 42 days from March 9 to April 19, 1843, and is from the journal of HBC clerk Alexander Caulfield Anderson (Fort St. James Post Journal, 1840-46, HBCA, B.188/a/19), transcribed by Anderson's descendant Nancy Marguerite Anderson and posted in her blog,  <>

These journal notes show that boats could be built by a few men, several of whom were on the sick list, working often in the open rather than in a fully equipped shop. The timbers or cross ribs of the boat's frame (called by the French term "verangues" in this account) were bent by heating in a steam box, and the thin planks for the hull were sawn. The sheathing planks, or strakes, overlapped at the edge, a technique known as clinker-built or lapstrake.

[Begin transcription:]
Thurs. 9th [March 1843] Cold weather. Nothing new occurring. Commenced preparations for making the boats, … serres, &c. The cold weather & Bourgeau’s sore leg, have caused us to delay this necessary labour.
Mon. 13th. Clear & cold. The weather being too severe to bend the varangues, the three men who were occupied about the boat wood are preparing the keels &c … [Pierre] Gouin preparing rivets for boats, assisted by the lathe. Others at fire wood.
Tues. 14th. .. Still too cold for bending the varangues, but always preparing for the boats.
Wed. 15th. Weather gradually moderating. Commenced today bending varangues. [Joseph] Bourgeau, [Supplie] Larance & [Pierre] Roi, bending and preparing. [Jacques] Coutureau [Couturier] who is still lame, attending the steam box.
Thurs. 16th. Fine mild weather. Continued as yest[erda]y & bent the varangues of 1 1/2 boats — say 2 1/2 now bent.
Sat. 18th. The snow vanishing fast. Finished the varangues of 6 boats, wood cutters employed cleaning the fort.
Mon. 20th. Men arranged as below. Bourgeau, Larance, Roi, [Edouard] Crete, [Charles] Touin, Ignace Calument — Boats…. Bourgeau laid the keel of a boat.
Wed. 22nd. .. Turned a boat, ready for putting on the boards in the evening. Courturier for the last two days making wedges for the [fur] press.
Mon. 27th. Touin laid up with boils. Laid down another boat. Nothing new occurring.
Wed. 29th. Couturier working at boats in place of Touin, who is still laid up.
Saty. 1 April. .. Bourgeau turned his third boat ready for planking.
Tues. 4th. Afternoon finished another [the third] boat… I omitted to mention that on Sunday Roi [drove] a long splinter into his wrist, which incapacitates him from working. The men employed as under: Bourgeau, Larance, Crete, Ignace, [Amable] Lacourse, Courturier at boats
Wed. 5th. We began pressing packs in the morning, but after several interruptions at length gave it up & the men wrought in salmon store…  Began 4th boat. Two men are obliged to begin sawing as far from being sufficient for six boats as Mr. [Peter Skene] Ogden had anticipated, there is found not to be sufficient for the completion of five, which I am very sorry for, as it is now too late to think of attempting to remedy the evil.
Mon. 10th. Fine, but chilly, with Westerly winds. Bourgeau finished 4th boat & laid the keel of another.
Thurs. 13th. .. Bourgeau finished his fifth boat today. Mr. Ogden directed me to get six made, and were the wood ready, there would yet be time to finish another. But there are no materials, one of the keels having been found useless, and there being no more boards. I am thus reluctantly constrained to leave the sixth unattempted.
Fri. 14th. Fine. Bourgeau arranging apichimons &c for two boats. Larance working at sheels[?]. Crete, Leonard, Couturier, Louis Thibeault, & rest cutting wood, except Brunel & Roi who are laid up.
Saty. 15th. Larance as before. The other men employed till noon cleaning out boat shed, storing boats for the summer, piling the timber lately dragged by Gouin and other necessary jobs, preparatory to beginning to gum on Monday.
Monday 17th. Rained at intervals yesterday — today fine. Named crews of boats & all hands are preparing oars, caulking boats &c.
Wed. [April] 19th. Snowed a little in the morning. Afterwards fine. Finished gumming. Preparing for a start tomorrow. This is rather earlier than was intended; but the season is very forward, and everything being ready, I have decided on moving. It was my intention to go off the day after tomorrow; but the Canadians have a superstitious reluctance to starting on a Friday, which it is customary and prudent to humour, where there is no probability of eradicating it.
[End of transcription.]

Joseph Bourgeau, the boatbuilder in charge of this project, was born in L'Assomption/St. Paul District of Montreal, Lower Canada about 1807. He started his career with HBC as a middleman (HBC's term for boat paddlers, also used for common laborers employed in a variety of tasks) at Fort Colvile, where he worked from 1829 to 1835. Colvile was a center for the construction of Columbia boats, as it was close to a good supply of the necessary timber and just upstream from Fort Okanogan, where boats were needed for the brigade route linking the Thompson River and points north (via Kamloops) to the Columbia and points downstream. Bourgeau probably learned the boatbuilding trade during his years at Colvile. By 1839 he was listed in Company personnel rolls as "Boatbuilder," and held that position until 1844, when he retired to farm in the Willamette Valley. He died in California in 1849.

Supplie Larance, also mentioned above, was probably born in Lavaltrie, Lower Canada around 1808. He worked for HBC as a middleman in the New Caledonia (HBC's term for what's now northern British Columbia) from 1831 to 1845, as a boatbuilder in New Caledonia from 1845 to 1850, and in the same capacity at Fort Colvile in 1850-51. (Bio details from Watson, Lives Lived, pp. 221, 568.)

References cited:

Burnett, Peter H. Recollections and Opinions of an Old Pioneer. New York: Appleton, 1880. (Available for free download from Google Books.)

Dunn, John. History of the Oregon Territory and British North-American Fur Trade. London: Edwards and Hughes, 1844. (Available for free download from Google Books.)

Gibson, James R. The Lifeline of the Oregon Country: The Fraser-Columbia Brigade System, 1811-47. Vancouver: UBC Press, 1997.

Gray, W.H. A History of Oregon, 1792-1849, Drawn from Personal Observation and Authentic Information. Portland, OR: Harris and Holman, 1870. (Available for free download from Google Books.)

Johnson, Dennis F. York Boats of the Hudson's Bay Company: Canada's Inland Armada. Markham, ON: Fifth House, 2006.

Johnson, Dennis F. "The York Boats of the Hudson's Bay Company" Wooden Boat magazine, #144, (Oct.-Dec. 1998), pp. 50-58.

Ross, Alexander. Fur Traders of the Far West: a Narrative of Adventures in the Oregon and Rocky Mountains. (Vol. 1) London: Smith & Elder, 1855. (Available for free download from Google Books.)

"Secret Mission of Warre and Vavasour" The Washington Historical Quarterly, Vol. 3, No. 2 (April, 1912), pp. 131-153.

Watson, Bruce M. Lives Lived West of the Divide: A Biographical Dictionary of Fur Traders Working West of the Rockies, 1793-1858. Okanagen, B.C.: The Centre for Social, Spatial and Economic Justice, 2010. (Available for inexpensive download from

Wilkes, Charles. Narrative of the United States Exploring Expedition During the Years 1838-1842 (Vol. IV). London: Wiley and Putnam, 1845. (Available for free download from Google Books.)

Comments welcome (email: tomholloway62(at)gmail(dot)com).

Thursday, February 28, 2013

Tobacco at Fort Vancouver, Tons Of It

How many puffs on a pipe would it take to consume 96,000 pounds of tobacco? That’s forty-eight tons, equal to the maximum cargo weight allowed on two of the largest semi-trailers pulled by today’s 18-wheeler trucks. If you divided that much tobacco into 200-pound loads for pack horses, it would take a string of 480 animals to carry it all. If you separated that much tobacco into one-pound piles, each of the 96,000 piles would look something like this:

(Photo by Tom Holloway, weighing tobacco twists from the shelves in the reconstructed Fort Vancouver Indian Trade Shop. For size, the plate of the scale is 6” square. Thanks to the fine folks at Portland's Culinary Workshop for the loan of the scale.)

          There are approximately 16 ‘twists’ of tobacco in this photo. I’m not a smoker, but I would guess that each twist would be enough tobacco, shaved off into a pipe, for at least 10 pipe-fulls. Let’s allow 40 puffs per pipe-full. Let’s see now: that’s 400 puffs per twist, times 16 gives 6,400 puffs per pound, times 96,000 equals 614,400,000 puffs.
          These last calculations are just idle speculation, of course, meant to push the point: Hudson’s Bay Company did not just deal in tobacco. It dealt in a LOT of tobacco!
          Ninety-six thousand pounds is also the amount of tobacco the HBC had in stock at Fort Vancouver and its subsidiary trading posts in the Columbia District, in the summer of 1844.
          Let’s back up a little and provide some context and detail. NPS Park guide Alysha Bronson has written an informative "Did You Know" overview of tobacco as an HBC trade item. As she points out, the most important point of origin of HBC’s trade tobacco was Brazil.(1) This was in part due to the long-standing commercial ties between British and Portuguese merchants. The Portuguese, in turn, produced large amounts of tobacco in their Brazilian colony. Brazil became independent in 1822, but the commercial chain persisted connecting its tobacco plantations with Portugal, Great Britain, and British trading companies. This means that the annual HBC supply ship sailing around South America on its way to the Pacific Northwest passed right by Brazil, the original source of the tobacco it had loaded on board at the London docks.
          In June 1844 the HBC supply ship Brothers arrived at Fort Vancouver loaded with goods intended to be used in Outfit 1845 (that is, what would now be called the fiscal year running from June 1, 1845 to May 31, 1846). This was consistent with standard Company policy that sought to ensure that a two-year supply of goods would be on hand or in transit, as a precaution in case something happened to the annual supply ship.(2) The shipping manifest of the Brothers included this group of items:

This is the transcription of the tobacco listed:
Quantity          Item                      cost per lb.      Total cost
18,925 lbs.      leaf tobacco                 5d              £394/5/5
11,465 lbs.      Canada roll tobacco   10d             £477/14/2
579 lbs.           Irish roll tobacco        10d             £24/2/6

(followed by 10 “assortments carpenters tools” like those discussed HERE, followed by
700 beaver traps without springs and 200 pairs of trap springs, discussed HERE. Most intriguing are the 41¾ dozen “assorted children’s toys” at an average cost of 2 shillings for each assortment, on the line between the carpenter's tools and the beaver traps.)

          Thus we see that the Brothers carried a total of 30,969 pounds of tobacco—more than 15 tons—to be stored at the Fort Vancouver depot for eventual distribution to the various trading posts, including the Indian Trade Shop and the Sale Shop also located at Fort Vancouver, for Outfit 1845.
         In the spring of each year, in anticipation of submitting an order to the London headquarters requesting supplies for two years in the future, HBC clerks conducted an inventory of "sundry goods remaining on hand." This is the listing of the tobacco in stock in the Fort Vancouver Depot storehouse at the time of spring 1844 inventory, just before the Brothers arrived from London: (3)

This is the transcription of the tobacco listed:
Quantity          Item                        cost per lb.      Total cost
16,200 lbs.      Canada roll  tobacco    10d              £645/0/0
961 lbs.           carrot tobacco                8d              £32/0/8
938 lbs.           carrot tobacco                9d              £35/3/6
467 lbs.           Irish roll tobacco           10d             £19/9/2
1,647 lbs.        Irish roll tobacco           11d             £75/9/9
31,807½ lbs.   leaf tobacco                   6½d           £860/9/1

         This comes to 52,020½ lbs. of tobacco, more than 26 tons. In addition there were 1,592½ lbs. of tobacco in stock in the Fort Vancouver Sale Shop, and 172 lbs. in stock in the Fort Vancouver Indian Trade (which included the Indian Trade Shop at Fort Vancouver itself, as well as the subsidiary posts at Fort George [formerly Astoria] and on the Umpqua River). The same 1844 inventory reported 11,493 lbs. of tobacco in stock at the 13 other HBC trading units in the Columbia District.(4) So the total 65,278 lbs. reported in the inventory, added to the 30,969 that arrived on the Brothers, comes to 96,247 pounds of tobacco in the Columbia District in mid-1844.
          The aggregate total discussed here was intended as a two-year supply, but any way you puff it, it was a lot of smoke.
          The total book value of this tobacco supply, the Company’s wholesale cost, was £2312 pounds, 13 shillings, and tuppence. My favorite online inflation calculator,, suggests that the equivalent value in today’s US dollars would be something on the order of $271,000. Let’s say “about a quarter of a million dollars worth of tobacco in today’s money” and call it good. It would then have been sold at cost plus 50% to employees in the servant class, or cost plus 100% to outsiders--missionaries, settlers, and free trappers.(5)
          As for how much a pound of tobacco was worth in "made beaver" trade equivalents, for simplicity's sake let's take "leaf tobacco," the most common form. In the inventory leaf tobacco was valued at 6½  pence per pound. According to the Standard of Trade for the Fort Vancouver area in 1844, a native trapper bringing in a prime beaver pelt could get 3 pounds of 'twist' tobacco in trade.  (Twists like those pictured above were made from leaf tobacco to prepare it for retail trade or sale.) The 3 pounds of tobacco (three piles like the one on the scale above) cost the Company 19½ pence, or 1 shilling 7½ pence. To put that figure in comparative context I refer the reader to an earlier blog posting, "What Was a 'Made Beaver' Worth?" where tobacco is included in the calculations. 
          Most, if not all, of this tobacco was smoked in pipes. This is the listing of the pipes in stock the Fort Vancouver Depot storehouse reported in the spring 1844 inventory:

The transcription of the lines listing pipes:
Quantity          Item                     cost per gross      Total cost
150 gross      hunters clay pipes     2s/3d              £16/17/6
27 gross        long clay pipes          5s/0d              £6/15/0
2 gross          dutch pipes                2s/9d              £0/5/6

            A gross being 144, there were 25,776 pipes on hand, not counting the stocks already distributed to the various trading units at the time of the inventory. That may have been taken to make up a two-year supply, because no smoking pipes are recorded in the cargo of the Brothers.(6)
          Note also that the clay pipes were very cheap. At 2 shillings 3 pence for 144 pipes, the company's cost for the "hunters clay pipes" was 0.1875 pence per pipe, a tiny fraction of the 10 pence per pound for Canada roll tobacco.
          Unlike tobacco, the pipes used to smoke it, being made of non-perishable materials, have a large presence in the archaeological record. NPS Museum Technician Heidi Pierson has compiled a must-read profusely illustrated overview of Tobacco Pipes, based on material in the Fort Vancouver museum collection, available HERE to read or download. She reports (p. iii) that “of the almost 7,000 tobacco pipe fragments found near the former site of the Sale Shop building, most were concentrated outside the east wall.” This suggests that there was considerable standing around and smoking of pipes near the Sale Shop (the large double building that was located along the west wall of the stockade, where Company employees, settlers from around the region, and Oregon Trail emigrants acquired supplies), perhaps as customers waited for their turn at the counter or visited with one another. A similar concentration of pipe fragments was recovered from around the doorway of the Indian Trade Shop, the building now reconstructed just west of the blacksmith shop inside the Fort Vancouver stockade.
          For more from the archaeological record go HERE for an image and information on a Stepney tobacco pipe bowl fragment recovered in the 2012 Public Archaeology Field School conducted by the Northwest Cultural Resources Institute, Directed by NPS Archaeologist Doug Wilson.

         Several readers have asked in separate emails about the practice of putting leaves of tobacco in bundles of furs as they were packed for shipment to England to guard against insect damage. The inventories and ship manifests used in this posting make no reference to such practice, which is commonly mentioned in interpreting the Fort Vancouver Fur Store for park visitors. This is what I have been able to track down after a quick search:
          In the chapter on the Fur Store in his 1976 Historic Structures Report, NPS historian John Hussey notes that one of the most common measures to reduce damage from insects on the long voyage from Fort Vancouver to London "was to place tobacco between the layers of skins."
          Hussey cites two sources (in his footnote 68). One is Henry Martin Robinson, The Great Fur Land (London: Sampson Lowe et al., 1879--available for download at GoogleBooks). On p. 340 Robinson, discussing HBC practices east of the Rocky mountains in Rupert's Land, notes that “when the brigades depart, the furs are all sorted and repacked, and pressed into bales by an enormous lever—rum and tobacco being placed between the layers of skins to keep out the insects and moths." 
          Hussey's other source is a letter sent by Fort Vancouver Chief Factor John McLoughlin to the Governor and Committee of the HBC on August 5, 1829. On p. 80 of the collection of his letters published by the Champlain Society, McLoughlin reported that The Furs have been repeatedly Beaten and Tobacco put in with the small Furs in the Casks and Cases, the latter had tobacco in them coming here. . ." (Clicking on the colored text should cause a page from the 1941 published version of McLoughlin's letters to appear on your screen.) Note that this does not refer to bundles wrapped in hides, but to furs packed in small barrels (casks) and boxes (cases). The casks and cases in question had been used to ship tobacco from England, and were being recycled. If any readers have information on other sources documenting the use of tobacco as an insect repellent in bundles of furs, or where that tobacco came from (especially if it was grown at Fort Vancouver), please let me know at tomholloway62(at)

          1. For more background on tobacco in HBC’s trading system over time see Marius Barbeau and Clifford Wilson, "Tobacco for the Fur Trade," The Beaver, March 1944, pp. 36-39. I thank Fort Vancouver Park Guide Mike Twist for pointing me to this important source, which explains and illustrates the various types and packing methods that appear in the inventories and shipping manifests discussed here.
          (Readers may be interested to know that much of HBC’s historical magazine, The Beaver, from its first issue in 1922 through the 1970s at least, has been digitized and is available online.)

          2. The full list of the goods shipped on the Brothers in June 1844 is in “Fort Vancouver, Columbia District Account Books, 1843-1844,” Hudson’s Bay Company Archives (HBCA), B.223/d/158, Pp. 6-32. It is transcribed in full in Lester A. Ross, “Fort Vancouver, 1829-1860: A Historical Archaeological Investigation of the Goods Imported and Manufactured b the Hudson’s Bay Company” (Typescript, Fort Vancouver NHS, 1976), Appendix II, pp. 1,384-1,418.

          3.  The spring 1844 inventory is in “Fort Vancouver, Columbia District Account Books, 1843-1844,” HBCA, B.223/d/155. I thank Curator Theresa Langford and Museum Technician Meagan Huff for facilitating access to these archival sources.
          A common way of packaging tobacco was to wrap a bundle of cured leaves in cloth and then bind the bundle with twine or small-diameter rope. These were called "carrots" from their resemblance to the vegetable. Visitors to the Indian Trade Shop at Fort Vancouver NHS can see reproductions of such bundles hanging from the ceiling beams. Photos of modern reproductions of roll and carrot tobacco can also be seen on the website of the Museum of the Fur Trade in Chadron, Nebraska.

          4. The other units in HBC’s Columbia District trading network were as follows, with the total number of pounds of tobacco each had in stock when the 1844 inventory was taken (in parentheses): Fort Colville (530), New Caledonia (319), Thompson’s River (68), Snake Party (1,241), Fort Nisqually (457), Fort Victoria (1,116), Fort Langley (381), California Establishment (1,254), Steamer Beaver (932), Fort Stikine (650), Fort Simpson (4,545) Fort Nez Perces (0), Schooner Cadboro (0).

          5. These pricing policies are discussed in John Hussey, “Historic Structures Report” Vol. 1 (1972), p. 190-1, note 17, reproduced in Lester A. Ross, “Fort Vancouver, 1829-1860: A Historical Archaeological Investigation of the Goods Imported and Manufactured by the Hudson’s Bay Company,” (Typescript, Fort Vancouver NHS, 1976), Table 12, p. 150.

          6. There were reportedly 960 pipes in stock at other trading units in 1844, including 552 at the Fort Vancouver Sale Shop, 336 at Fort Simpson, 48 in the supplies of the Snake party, and 24 on the Steamer Beaver; source same as in Note 3, above.