Parks Canada Banner
Parks Canada Home

Canadian Historic Sites: Occasional Papers in Archaeology and History No. 19

Yukon Transportation: A History

by Gordon Bennett

The Great Stampede


On 16 August 1896 gold was discovered on Bonanza Creek, a tiny tributary of the Klondike River. As news of the discovery spread throughout the territory, prospector after prospector, miner after miner abandoned his diggings and set off in frenzied pursuit of the new El Dorado. A year later Klondike hysteria enveloped the outside world.

In many respects the Klondike discovery and the stampede it fathered provided a fitting curtain to the 19th century. Debilitating depression with only brief and fitful interruptions had overhung the western world for a generation before 1898. The psychological effects of the Klondike discovery changed all this, precipitating an almost New Year's Eve type of celebration of purge and promise, as the worn-out 19th century indulged in one last grand binge. In an era that had been characterized by the Gilded Age, the Great Barbecue and the Robber Barons, the chain of events set in motion by the discovery served to democratize the maxim that "money making was the most prized career."1 No longer was wealth regarded as a private prerogative of a Rockefeller, a Morgan or a Carnegie: the New Year's resolution on everyone's mind was to strike it rich.

With unwitting foresight, a despatch datelined Forty Mile, Yukon Territory, 17 August 1896, predicted that "such is the lure of gold in depression ridden America that many are expected to come."2 Probably no other event in Yukon history has ever been anticipated with such understatement. From late July of 1897 when the Portland and the Excelsior, each laden with gold, landed in Seattle and San Francisco respectively, most of the English-speaking world and much of Europe found itself caught up in a maelstrom that knew only two words — "gold" and Klondike. Almost immediately a crush of humanity, doggedly determined to win its fair share of untold wealth, streamed north. Three factors made the Klondike gold rush possible: the existence of a vast amount of placer gold, the publicity given to the discovery in the press of the time, and the transportation system that had evolved since the era of the fur trade.3 Without a juxtaposition of the three in 1897-98 the gold rush would never have reached the proportions that it did.

If one accepts the definition of transportation in its broadest sense as the utilization of transportation routes and a variety of transportation forms to move men and supplies, the history of transportation during the gold rush can be said to have been practically synonymous with the history of the gold rush itself. Each interacted with the other; the one as cause, the other as effect as circumstances dictated. If the gold rush is conceived, moreover, not as an independent entity nor as a temporary aberration, but as one stage in the historical evolution of the Yukon, then an estimate of its effect on the transportation system can be attempted. With the exception of the railroad, the impact of which properly belongs to a future chapter, the gold rush did not substantively alter the nature or function of the gateways which had been established prior to 1896. No new routes of any consequence were discovered although several variations on the old routes were used with varying degrees of success. What the gold rush did do was to exaggerate the impracticality of the old Hudson's Bay Company fur-trade routes and to strain and emphasize the inadequacy of the coastal routes under gold-rush conditions. Where the effects of the gold rush were most clearly felt were on the modes of transportation. As might be expected, many of these effects were quantitative, but the building of roads and tramways, the introduction of sternwheelers on the upper river, and the construction of the White Pass and Yukon Route railway were major qualitative changes in the transportation system of the Yukon.


Another qualitative change which is not so immediately apparent also occurred. Before 1896, no government had attempted to influence or interfere with the transportation system. Adjustments in the transportation system had resulted from the action of natural forces and the response to them by individual men or trading companies. As we have seen, this process of "natural selection" had resulted in the practical extinction of the Hudson's Bay Company fur-trade routes, the domination of trade by the Saint Michael route and the rise of the Chilkoot route as the principal avenue of migration. With the gold rush, however, government for the first time became actively interested in influencing the flow of transportation into the Canadian North. The most conspicuous examples of this interference were the promotion of the old Hudson's Bay Company trade routes by the municipality of Edmonton; the promotion of the Stikine railway project by the dominion government, and the imposition of customs duties.

When news of the gold strike on Bonanza Creek reached the outside, Edmonton was only a small town of seven hundred people. Nonetheless, Edmonton qualified as a transfer point to the Klondike, being situated at the head of a known trail to the Yukon (the old Hudson's Bay Company Athabasca-Mackenzie-Porcupine-Yukon river route) and having the facilities to serve as a supply base, in this instance the northern terminus of the Canadian Pacific Railway.4 In an attempt to stimulate the economy, local politicians and merchants undertook an advertising campaign to attract stampeders. Billing the route through Edmonton as the "All Canadian Route,"5 they hoped to capitalize on the patriotic inclinations of the Canadian stampeders. On a more practical plane, the promoters emphasized the fact that "going via Edmonton" eliminated customs levies, an important consideration for those who had mortgaged themselves in order to take part in the rush. Had Edmonton advocates been content to stand on this one advantage, the Edmonton route would most likely have escaped the notoriety for which it was later known, but the city built its advertising campaign around the dubious assertion that the Edmonton trail was the fastest route to the gold fields. Even the most ardent optimist would have been hard-pressed to substantiate this claim for the truth of the matter was that the 90 days allowed for traversing the trail by the Edmonton promoters was patently unrealistic.6 That some people took this route is indicative of the delirium that overcame normally sane men during the gold rush. The Hudson's Bay Company presence in the Yukon had always been precarious, with transportation a basic problem. This obstacle, as we have seen, was never successfully overcome. Moreover, to exacerbate the plight of those who chose this route during the gold rush, the light canoes, the assistance of Indians and voyageurs, the series of supply bases, all of which had been available to the fur trader, were not available to the gold seekers. Nor was there any resemblance between traders who were capable of withstanding the rigours of the trail and the stampeders who were not.

22 The first known view of what later became Dawson City — 13 years before the discovery of gold on Bonanza Creek.
(Frederick Schwatka, Along Alaska's Great River [New York: Cassell & Company (1885), p. 243.)

23 The arrival of the Excelsior (below) in San Francisco and the Portland in Seattle in July 1897 signalled the beginning of the great stampede.
(Chicago Record, Klondike [Chicago: Chicago Record Co., 1897], p. 400.)

24 The Stikine route and main Edmonton trails to the Klondike.
(Map by S. Epps.) (click on image for a PDF version)

Of the one hundred thousand people who set out for the Klondike in 1897-98, only two thousand used the Edmonton route. Few of those who did ever reached their destination. "Not a single one," Pierre Berton has written, "as far as can be determined, found any gold at all."7 In almost every case it took two years to go from Edmonton to Dawson and by the time the lucky ones finally made it, the gold fields had been staked from end to end. This first example of an attempt by government, in this case municipal, to influence the direction of transportation flow into the Yukon would have been comic had it not had such tragic consequences. The overland routes running west of the Mackenzie River proved no more responsive to the wishes of Edmonton than they had been to those of the Hudson's Bay Company.

Edmonton was not the only would-be metropolis to vie for the Klondike trade. Operating on the valid assumption that the great majority of stampeders would opt for one of the coastal routes, the port cities of Victoria and Vancouver, with the active support of Montreal- and Toronto-based railroad interests, set out to supply the outfitting and transportation needs created by the rush. In this they were assisted by the Canadian government which imposed a schedule of customs duties on all foreign goods going into the Yukon. This schedule had the dual purpose of raising revenues and diverting business away from such American centres as Seattle and San Francisco which had early established a stranglehold on Klondike trade. As the Yukon was cut off from the coast by the Alaska Panhandle, however, all goods had to pass through American territory before reaching the Yukon. There, supplies that had been purchased in Canada were liable to retaliatory customs duties levied by American authorities, unless conveyed through Alaska in bond, in which case any prior advantage gained from outfitting in Canada was lost.

The trade issue was complicated by a longstanding dispute with the United States over the Alaska boundary.8 Despite Canada's claim that Dyea and Skagway, the coastal gateways to the Yukon interior, were on Canadian soil, the United States exercised sovereignty over both of them. Ever since the late 1880s Canada had pressed for a solution to the question only to meet with American indifference. The gold rush made a settlement all the more urgent from the Canadian viewpoint for reasons of state as well as for the potential effect that any settlement would have on the rivalry between Canadian and American ports. The United States, on the other hand, showed little interest in negotiating so long as it enjoyed de facto control over all supplies going into the Klondike via the coastal routes.

An alternative solution to the boundary imbroglio presented itself in the Stikine River route which originated at Fort Wrangell, Alaska, followed the Stikine River to Telegraph Creek, crossed overland to Teslin Lake on the British Columbia-Yukon border and thence down the Yukon River system to Dawson.9 Under the treaty of 1825 between Great Britain and Russia, Great Britain had secured free navigation rights on the Stikine which were ceded to Canada after 1867.10 The Stikine route, therefore, became a vital lever in Canada's attempt to influence transportation flow into the Yukon, especially in view of the failure to reach an accord with the United States on the boundary question. The surveyor general of Canada succinctly explained the need to develop the Stikine route. "We must have an independent road allowing free access to our country whatever complications may arise with the United States," he wrote, "and for this purpose it is imperative that a road be located from Telegraph Creek to Teslin Lake."11 With this in mind, the Canadian government signed a contract on 26 January 1898 with those ubiquitous Canadian railroad contractors, William Mackenzie and Donald Mann, to build a wagon road within six weeks and a narrow-gauge railway by 1 September 1898 from the head of navigation on the Stikine to Teslin Lake.12 To complete the transportation system, the government planned to connect each end of the rail line with a fleet of steamboats.13 In return, Mackenzie and Mann were to receive a land subsidy of 25,000 acres for each mile of line constructed. After a long and occasionally acrimonious debate in the House of Commons, a bill was passed approving the project.14 Twelve miles of track had been laid and tickets had actually been sold when the Conservative-dominated Senate rejected the terms of the government contract. This dealt the coup de grâce to the Stikine project and the railroad scheme was unceremoniously abandoned.15

Attempts to regulate artifically the flow of transportation into the Yukon during the gold rush failed. Only in the imposition of customs duties did the government achieve any partial success. This action may have persuaded some to use the Edmonton route and it seems evident that it was partly responsible for stimulating the trading and transportation companies operating out of Victoria and Vancouver. The failure of government to influence the selection of access routes to the Klondike, however, appeared in sharp contrast to what was taking place, without government intervention, on the traditional gateways of Saint Michael and the Lynn Canal.


These two gateways, which attracted the majority of stampeders, were linked to the West Coast outfitting ports by a thousand-mile stretch of water known as the Inside Passage. As a result of the unprecedented demand for coastal steamer facilities, the existing transportation companies were strained to capacity.16 To cope with the overflow, a number of new transportation companies sprang into existence, coastal steamer production was rapidly increased, and old, generally unseaworthy vessels that had been left to rot on the beaches were revived, Martha Black, a towering figure in the post-Klondike period of Yukon history, later described what for most of the stampeders had been a typical trip up the coast when the gold-rush hysteria was at its height.

The steamer was certainly a "has-been." She was dirty, and loaded to the gun wales with passengers, animals, and freight. Men slept on the floor of the saloon and in every corner. The captain was seldom, if ever, sober, and there were many wild parties. Poker, blackjack, and drinking went on night and day, and our safe arrival in Skagway was due probably to the Guiding Hand that looks after children, tools, and drunken men.17

The Saint Michael route responded to the gold-rush transportation challenge with a rapid expansion in facilities. By the summer of 1898 there were an estimated 110 steamers on the lower river, a tenfold increase over 1897 and a 14-fold increase over the pre-gold-rush number.18 Saint Michael itself became something of a boat-building or, more correctly, a boat-assembly centre.19 Yet despite the rapid increase in the number of boats on the lower river, the Saint Michael route never did regain control of population movement into the interior, a control it had lost to its Lynn Canal rival after 1882. Though potentially the fastest route to the gold fields, this advantage was never seized.

Several factors account for the lower route's failure to exploit the opportunity provided by the gold rush. The transfer of traffic from coastal steamer to river boat was not orderly. Passengers would be spilled off the decks of the ocean vessels at Saint Michael only to discover that there were no available sternwheelers to take them on the last leg of the voyage to Dawson. This problem was exacerbated by impediments to navigation on the lower river itself. On that section of the river below Circle City known as "the flats," the river spread out like a lake, cut through and through by innumerable bars and islands. As a result, the river was reduced to little more than a series of small shallow streams through which a pilot had to locate a channel large enough to accommodate large steamers. Every summer once the channel had been found, a pilot was stationed on the flats to take the boats safely through, but the annual spring floods made it necessary to locate a new channel each year.20 Initially the trading companies had used Eskimos and Indians as deck hands and pilots because of their knowledge of the river; however, their familiarity with the watercourse was limited to short stretches and as a consequence 20 or more native pilots were required for each trip. This system was abandoned once white pilots had gained enough experience to navigate the river without outside help, but the problem of shifting channels remained.21

Another difficult section of the river was encountered at the "Ramparts." At this point a sternwheeler running against the current was forced to tie up every 10 or 15 minutes so that an extra head of steam could be raised.22 At Fort Yukon, low water often made it impossible to take a sternwheeler above the post.23 Had the United States government acted upon the suggestion that a permanent channel be dredged through the flats and at Fort Yukon,24 the lower river might have attracted a great deal more of the Klondike traffic than it did. However, while the lower river route did not attract large numbers of stampeders, it did retain important function that it had performed since 1869 as the life line to the Yukon.25 Despite certain obstacles to navigation, it remained the only feasible route over which heavy freight could be brought in.26

25 "Built by the mile and cut apart in proper lengths" — thus did one observer describe the 12 virtually identical boats built by the Moran Brothers shipyard in Seattle for the Klondike trade. Included in this commercial armada were the J. P. Light, D. R. Campbell, F. K. Gustin, Mary F. Graff, Pilgrim, Victoria, and Oil City.
(Minnesota Historical Society.)

If most of the stampeders avoided the Edmonton trails and the Saint Michael route with remarkable if unconscious discretion, they made up for it by pouring onto the Alaska Panhandle with blitzkrieg force. It was here that the mad cascade that constituted one of the largest mass movements in peacetime history converged, only to be scattered upon one of the six trails which led into the Yukon interior. Of these trails, the most popular by far proved to be the trails over the Chilkoot and White passes on Lynn Canal.27

Before the gold rush, the density of traffic over the Chilkoot Trail had not been heavy enough to justify any major improvements. Consequently, when the first wave of stampeders hit the beaches at Dyea in August of 1897, the transportation facilities were wholly inadequate to cope with the assault: a roughed-out road between Dyea and Sheep Camp which required fording, a horse packing outfit that had been organized in 1894 by John J. Healy28 and the human packing service provided by the Indians. As a result, the stampeders were forced to transport themselves and their supplies under the most adverse circumstances. During late summer of 1897 many built canoes to move their supplies up the Dyea River to the head of canoe navigation at Canyon City, at which point everything was transferred to the trail which led over the summit. When the river froze over, some sledded their outfits over the ice to a point just below Sheep Camp.29 Most of the stampeders, however, moved their supplies along the Dyea-Sheep Camp trail that had been cut prior to the gold rush.

As the number of stampeders increased, some improvements became necessary. A toll bridge was built one-half mile out of Dyea and a good wagon road was constructed to Finnegan's Point, six miles outside the city.30 Horses were brought into Dyea in great quantity and packing operations were expanded. A number of communities sprang up along the trail where the stampeders could stop for food and shelter, only to disappear when the gold rush had spent itself and eliminated their function.31

That winter the trail was improved when 150 steps were cut out of the ice on the coastal side of the summit. Later more steps were added. A cord life line was strung up parallel to this somewhat incongruous stairway and shelves were hacked out at intervals so the stampeders could avail themselves of a few moments' respite. A toll was levied for the use of the steps which the operators collected without the arguments commonly associated with such an enterprise. As T.A. Rickard noted, "Everyone was in a hurry; and anything that facilitated progress was liberally compensated."32 In December a horse-powered tramway was constructed up the pass. It was superseded by a much more ambitious undertaking in the spring of 1898. The new tramway, which ferried goods from Canyon City to the summit and was later extended to Crater Lake, was built by the Chilkoot Railroad and Transportation Company. It consisted of a copper-steel cable supported by tripods anchored in concrete and was powered by steam generators. When it was finished, it was said to have "had the longest single span in the world, twenty-two hundred feet from one support to the next." Each car on the line had a 300-pound capacity. The tramway never stopped and in the spring of 1898 it was dropping freight on the summit at the rate of nine tons an hour.33

After the Chilkoot Railroad and Transportation Company tramway was completed, the Dyea trail was equipped with a coherent transportation system which eliminated most of the transportation problems that had previously impeded the movement of traffic along this route. Freight rates fell sharply, levelling off at 13 cents a pound for the through journey between Dyea and Bennett;34 however, the impact of this significant improvement was small. Ironically, by the time the tramway became operable, the stream of humanity that had set out for the Klondike the previous summer had practically dried up.

Since 1882 the Chilkoot Trail had been the funnel through which men in search of Yukon gold had flowed. The gold rush did not change this. Of the forty-odd thousand stampeders who are estimated to have made it into the Klondike, well over half went in by this route.35 But whereas before 1897-98 the Dyea trail had been regarded as the best route to the Yukon gold fields, the gold rush had the effect of exaggerating its deficiencies to the point where they overshadowed its utility. Martha Black spoke for most of those who took the Chilkoot Trail during the gold rush when she called it the "worst trail this side of hell." "Men talked of the Chilkoot as if it were a malevolent thing," Kathryn Winslow has written, "capable of wrath and punishment."36 Of all the photographs taken of the gold rush, the one that depicted a black line of lock-stepped humanity, stooped and bent as it inched its way through the pass, came to symbolize this adversity.

Why did the gold rush so alter the reputation of the Chilkoot Trail? Why did this transportation highway into the Yukon, heretofore regarded as the best route into the interior, come to be regarded as some kind of hell the stampeder had to survive before he was worthy of the Klondike treasure? Was it the trail itself or something else that made men look upon the Chilkoot as that "malevolent thing"? The evidence suggests that a combination of factors, the stampeders themselves, the size of the stampede, the season, as well as the inadequacies of the trail, wrought the change. Few of those who set out for the Klondike in the fall of 1897 were fit enough to cope with the rigours of the trail. Fewer still had any alpine experience. Improper diet and inappropriate clothing for both the task and the climate added to the hardship,37 but of all the burdens that each stampeder had to face before he reached the Klondike, none was more significant in altering the reputation of the trail than packing.

26 The Chilkoot and White Pass trails.
(Map by S. Epps.) (click on image for a PDF version)

27-38 Few phenomena have been as well-documented in photographs as the rush to the Klondike. Figures 27 to 38 show only a few highlights. 27, A portion of the Skagway waterfront, 1898. (Public Archives of Canada.) 28, The Chilkoot Trail between Dyea and Canyon City. (Yukon Archives.) 29, The Chikloot Pass viewed from the scales. (Yukon Archives.) 30, The last 1,000 feet to the summit of the Chilkoot Pass. (Public Archives of Canada.) 31, The wagon "road" on the White Pass Trail, three miles from the summit. (Yukon Archives.) 32, A portion of the stampede settlement at Lindeman Lake. (Yukon Archives.) 33, Bennett, 1898. (Yukon Archives.) 34, North-West Mounted Police post at Tagish Lake where every boat and its occupants were registered. (Yukon Archives.) 35, The tramline between Miles Canyon and Whitehorse Rapids. (Public Archives of Canada.) 36, Shooting the Whitehorse Rapids. (Public Archives of Canada.) 37, The Klondike armada on Lake Laberge. (Yukon Archives.) 38, Stampeders arriving at Dawson. (Public Archives of Canada.)

The Klondike discovery occasioned a great mass movement to Dawson, the metropolis of the gold fields. In Dawson the population expanded so rapidly that the supply organizations were unable to keep pace with increasing demand. As winter descended upon the Klondike in 1897, the prospect of starvation loomed ominously. As a result, the government passed a regulation in January 1898 that no one was to be permitted to enter Canadian territory without having the means of survival on his person, or more specifically, as it turned out, on his back.38

Just how effective this regulation was in terms of forcing the typical stampeder to pack the legendary ton of supplies over the trail is open to question. While much has been written about the unreliability of many of the handbooks that purported to provide the stampeder with information about his trip, almost all were unanimous in recommending that the tenderfoot take enough supplies to get him over the trail and established in the gold fields. Because the regulation requiring the year's supply of provisions was proclaimed in January 1898 when most of the stampeders were already on the trail, it seems doubtful that it had much effect. There is, for example, no record of any bona fide stampeder being turned back at the summit because he lacked the requisite supplies. In fact, it appears that the North-West Mounted Police regarded the regulations as a means of denying entry to "undesirables." The fact remains, however, that packing was one of the most severe trials that the stampeder had to suffer in his quest for the Klondike.

Those who were fortunate enough to have money met the regulation without difficulty. They hired Indian packers, contracted the task out to one of the packing outfits that did business along the trail, or procured pack animals. For those who had mortgaged themselves to the limit in order to make the trip, however, and they constituted the majority, the import of the regulation was quite simple — they would have to transport their supplies on their backs. Few of them knew at the outset how to maximize space and distribute weight in manageable allotments. It was only through trial and error and after a great deal of hardship that the stampeders learned to pack and move their outfits in stout canvas bags 50 inches long. Arranging the pack so the centre of gravity rested on the shoulders, with a strap around the forehead to give extra support, the typical stampeder transported 50 to 60 pounds at a time. The outfit was moved in relays of about five miles and cached, the process being repeated until the entire outfit had moved from the coast to the lakes. The relay system had the effect of concentrating all the traffic on one part of trail, a fact that resulted in numerous blockages, loss of time and general deterioration of the trail. In all, about 30 trips were needed to freight an outfit from one cache to another and 90 days to move it over the trail from the coast to the head of navigation at Bennett Lake. It has been estimated that by the time a stampeder had completed this task he would have walked at least twenty-five hundred miles.39

The route leading over the White Pass, while less popular than its rival, the Chilkoot, attracted some five thousand of those who stampeded to the Klondike in 1897-98. This pass, named by William Ogilvie in 1887 after Sir Thomas White, then minister of the Interior, had been known for ten years before the gold rush.40 Captain William Moore, a former steamboat captain, is generally credited with its discovery. Between 1887 and 1897, Moore and the White Pass, which he virtually came to regard as his personal possession, waged competition with their rivals John J. Healy and the Chilkoot Pass in an attempt to siphon off the traffic going into the interior. Despite Moore's endeavours, however, the White Pass remained unused until the gold rush when it was suddenly thrust into prominence.

The White Pass route had two great advantages over the Chilkoot route: Skagway, at the foot of the White Pass, had a harbour whereas Dyea did not, and the trail through the White Pass was low enough to use pack animals over the summit. As a result, pack animals were shipped by the hundreds to Skagway. Of all the animals that were employed as beasts of burden, horses and mules proved to be the most adaptable to northern conditions. Burros were too small and oxen were too slow to traverse the boggy ground so characteristic of the White Pass route. At first some attempt was made to make the trail passable for the pack animals and some stretches of corduroy were laid, but, as one observer pointed out, "the moment a horse could by any means be got over the trail, all further improvement ceased and was never again resumed."41

The great demand for horses on the White Pass Trail during 1897-98 was reflected in the $300-price which horses commanded in Skagway. A few enterprising men recognized that here was a mine potentially richer than the gold fields themselves and horses that had been slated for the glue factory only days before found themselves relegated to the hell of the Skagway trail. As Robert Kirk noted, even excellently conditioned horses fell victim to the ignorance of their new masters, the weather, the bad trail, the poor food and the lack of rest that awaited them in Skagway.42

By late 1897 the White Pass route on the coastal side of the summit had become so littered with the carcasses of horses that the stampeders were referring to it as the Dead Horse Trail. In their insane urge to reach Dawson as quickly as possible, the stampeders had made no improvements. The boggy trail had deteriorated under the constant pounding of feet to the point where it became impassable. As a result, the trail was temporarily closed and a wagon road was built by George Brackett to White Pass City, ten miles up the trail from Skagway. Brackett charged a toll of $20 a ton for the use of his road, but, unlike the toll that was charged for the use of the stairway of ice on the summit of the Chilkoot, Brackett encountered one problem after another in his attempts to collect it.43

The trails leading over the Chilkoot and White passes converged on Bennett Lake, the former via Lindeman Lake. During the winter of 1897-98, a tent city, well in excess of ten thousand people, mushroomed at the head of the lake as stampeders stopped to build the motley armada that would carry them down the Yukon River system after the spring break-up.44 Although knock down boats constituted a standard item in most Klondike outfits and many stampeders packed the necessary construction materials over the passes, the demand for boats so exceeded the capacity of what few shipbuilders there were that priority was given to the construction of large freight scows. These freight scows, built by such contractors as King's Sawmill and Shipyard.

were made of two inch planking, forty-two feet long, and twelve feet wide, with straight sides. They were square at both ends, but sheered up like a barge, with pointed outriggers running about eight feet at the bow and stern, and a long heavy sweep at the end. They were decked fore and aft for eight feet, with the middle open, and a plank ran around the sides to walk on.

Each scow had a mast about twenty feet high, rigged with a square sail. The mast was set about eight feet back of the bow, so that a man could work the sweep in front of it. Sails were used only when crossing the lakes. Usually a tent was placed over the cockpit in the middle. After the cargo was loaded, this was where the crew lived, cooking on a little sheet iron stove. The scows were unpainted, were capable of carrying twenty tons, and drew from 24 to 26 inches.45

A similar problem of supply and demand beset sawmill operators who had established themselves at Bennett to satisfy those who could not afford the price of a ready-made boat ($300 to $500) or had failed to include the requisite materials in their outfits.46 A shortage of labour exacerbated the situation when few stampeders showed any inclination to work for wages and possibly defer their arrival in Dawson. The problem was partially alleviated when a compromise was worked out whereby those who wanted boat lumber supplied the mill with raw timber and took lumber that had already been cut into boards to build their vessels.47 For most of those who streamed north, however, cutting timber for their own use and assembling boats of their own manufacture was the rule.

After the lumber was whipsawed, a frame was constructed, the sides and bottom were nailed to the frame and the cracks were caulked and pitched. Then the oars or poles were cut, a mast erected and seats nailed down for the last leg of the race for Dawson.48 To increase the speed of these vessels — some fragile and unseaworthy, some barely adequate, and only a few built well enough to ensure safe passage down the watercourse — the stampeders used blanket sails and "current sails." The current sail was a device consisting of a piece of canvas weighted with rocks and dropped into the water. The undercurrent catching this submerged sail countered contrary winds and augmented downwinds where the water was deep enough so the current sail did not drag on the bottom.49

Having survived the perils of the trip up the coast, the adversity of the passes and the misery of the whipsaw pits at Bennett, the stampeder had yet two more serious obstacles to face — Miles Canyon and the Whitehorse Rapids. The canyon, a turbulent, dangerous stretch of water at the foot of which was a series of rapids, was to claim several lives before the North-West Mounted Police passed a regulation that only skilled pilots were to take boats through Miles Canyon and the Whitehorse Rapids.50 Fines of $100 were levied against all who violated this rule. The regulations also prohibited women and children from accompanying the pilot through the canyon to the foot of the rapids although this prohibition was occasionally evaded. The pilots, who were licensed by the North-West Mounted Police, charged $20 to $25 for each boat they took down. One of the most famous of them was Jack London, who was later to gain fame as a popular novelist. On a good day a pilot could make ten trips, returning to the head of the canyon from the foot of the rapids on horseback.51 One look at this violent stretch of water, however, persuaded many a stampeder against trusting the fate of his possessions to the skill of a pilot. As a result, a windlass was rigged up on the east side of the river in the spring of 1898 to haul the boats out of the water and a tramway was built by Norman Macaulay from the head of Miles Canyon to the foot of the Whitehorse Rapids, a distance of about five miles. The tramway consisted of peeled logs, eight inches in diameter, over which horses pulled wagons with cast-iron concave wheels. The freight rate on Macaulay's tramline, which he named "The Whitehorse Rapids Tramway Company," was three cents a pound and an additional $25 was charged for each boat. Shortly thereafter, another tramway, this one six and one-half miles long, was built on the west bank by a John Hepburn, but the competition affected both operations and Hepburn sold out to Macaulay for $60,000.52

Once through the Whitehorse Rapids, the way to Dawson was unimpeded except for the minor obstructions at Five Fingers and Rink rapids. These obstacles, however, were not difficult to negotiate with boats of the size used by the stampeders. By mid-June the flotilla began to stream into Dawson.53 The ordeal of the past few months meant nothing. The success or failure of the stampede was now to be determined in the Klondike gold fields.


The gold rush gave birth to many elaborate transportation schemes for facilitating the movement of men and supplies into the Yukon. This was a natural response to the news coming out of the Klondike that gold was there for the taking. A myth was in the process of being created — that nuggets of gold covered the Klondike landscape in such profusion that they only awaited some lucky soul to gather them up. Nor was this myth challenged by those who had struck it rich in 1896-97 and who now found themselves deluged by gold seekers eager for information about the Yukon, "You might just as well believe what you hear," the former confided, "because you just can't tell lies about Klondike. It's all true."54 In the popular imagination, therefore, it was neither the finding nor the mining of the gold that caused people to hesitate about leaving for the Klondike; the obstacle to be overcome was that of getting to the gold fields themselves.

It is hardly surprising that two of these transportation schemes which promised to convey the prospective stampeder to the Yukon without the exertion required by more conventional forms of transportation were based on inventions that had captured the popular imagination of the time: the bicycle55 and the balloon. In Kalamazoo, Michigan, a man by the name of Frank Corey built an airship to carry him to the Klondike when news of the phenomenal discovery broke in the American press. Corey's plan was to take two men on the inaugural flight so each of them could stake claims and thereafter to run fortnightly flights to the Yukon. His scheme reached its climax when requests for tickets started to pour in, but, despite the enthusiasm it had generated, neither the plan nor the airship ever left the ground. In Seattle, Washington, meanwhile, the Jacobs Transportation Company invested $150,000 in another balloon scheme, this one conceived by a Don Carlos Stevens. It was Stevens's plan to operate regular flights from Tesklo Bay near Juneau to Dawson and so enthusiastic did he become about the whole thing that he let it be known that when he got to Tesklo Bay he would hang out a sign — "All Aboard for Klondike" — "and when I've got my passengers I'll cut the rope and away we'll go." In this instance, Tesklo Bay's proximity to Dawson, in comparison to that of Kalamazoo's, was of little consequence. The promoters, it turned out, had more hot air than the balloon.56

Other transportation schemes that were talked about but never made it off the drawing boards were legion. The Pullman Palace Car Company completed a prototype for an electric sleigh which the Great Northern Mining and Transportation Company intended to use on the Klondike creeks. Complete with steam heat, electric lights and elegant interior appointments, the prototype was designed to travel at the then breath-taking speed of 60 miles an hour. The degree of disappointment felt by Yukon plutocrats and would-be plutocrats must have been great when news was received that the scheme had failed, but unfortunately their disappointment was never recorded. Another solution to the northern transportation problem was conceived by the Klondike Combined Sledge and Boat Company, which designed a steel sledge and barge which it intended to market in the summer of 1898. This contraption had sails and oars, air chambers for buoyancy and, as an added inducement, burglar-proof compartments for gold. Like its predecessor, this scheme suffered an unheralded end and one can only speculate whether its failure was the result of impracticality or the end of the gold rush.57

The experience of a stampeder setting out from Edmonton would tend to suggest the former. He also built a "sleigh-boat," the overland motive power for which was to be supplied by a horse until a river was reached, at which point the device would be capsized to make a boat. The one problem never resolved by the inventor, however, proved to be the contraption's failing. "When the snow and ice were on the ground," an eyewitness remarked incredulously, "the rivers were also frozen." Edmonton was privy to another scheme which did make some sense. This device was a steam sleigh which had a cogwheel for traction. Nevertheless, the progress it made failed to create a favourable impression — only 18 inches. The steam sleigh only bored itself deeper and deeper into the ground as a result of the action of the cogwheel. Other ideas which attracted attention during the gold rush included carts mounted on buggy wheels and large unicycles around which platforms had been built.58

Two other schemes to expedite the movement of supplies, particularly food, had Scandinavian origins. The first was made by a captain of the Royal Norwegian Army, Nils Muller, who suggested to Clifford Sifton that a series of stations be built between the coast and Dawson, and manned by a corps of Norwegian skiers. Muller wrote that

A skiloper with a full load of provisions is able to cover 15 miles per day — with one or two days provisions about 30 or 35. But with a sufficient number for relieving at each station the distance between Dawson City and Dyea may easily be covered in 8 days — counting 20 working hours per day.59

How this mode of human transport was to adequately supply the needs of some thirty thousand people was never answered and for its part the Department of the Interior showed no interest in finding out. In the meantime, the Alaska missionary, Sheldon Jackson, had persuaded the United States government to relieve the food crisis by driving a large herd of reindeer overland from the coast, the reindeer to supply not only transportation, but ultimately the fare of hungry Dawsonites. Like its predecessors, this scheme failed when most of the reindeer died of starvation before reaching their destination.60

Of the many transportation schemes that were stillborn during the gold rush, none was more interesting nor perhaps more deserving of attention than the proposal for a monorail. The brain child of a David Jones from San Francisco, the monorail appeared to be an effective answer to the northern transportation problem. Its originator claimed that "it can be carried above the snow and the line can always be available for service, it can be run over steep grades either by friction rollers or cog and back gearing and could be built rapidly."61 Unlike so many others who were out for some fast money, James did not want to sell his plan; rather he sought employment in the civil service to develop his scheme. However, the Canadian government gave short shrift to his proposal and both his idea and his desire for employment were refused.62

All of these transportation schemes had one thing in common: all failed to find application for one reason or another. Many of them were products of the hysteria that was typical of the gold-rush period, poorly thought out, and perhaps in no way dissimilar to the contraptions which amateur inventors try to pass off as perpetual motion machines, but others, if not successful, were forerunners of things to come. In the invention of the steam sleigh propelled by a cogwheel, for example, we can see a precursor to the snowmobile, a device well-suited to the needs of northern transportation. For the immediate future as men conceived it in 1898, however, the solution to the northern transportation problem was not to be found in snowmobiles. For that, men would turn to the 19th-century panacea for all transportation problems — the railroad.


The Klondike gold rush caused a stampede of railroad incorporations which in its own peculiar fashion was as great as its human counterpart over the passes. In 1897, 32 railroad companies applied for federal charters to build lines into the Yukon. In the same year the Province of British Columbia incorporated 10, while between 1897 and 1899 another 12 filed articles of incorporation in the United States.63 In 1897 the Canadian government commissioned a series of surveys to determine the most feasible route for a railway and in 1898 it sponsored its own scheme to build a line between the head of navigation on the Stikine River and Teslin Lake.64

The one railroad scheme that finally materialized was not wholly inspired by the gold rush. In 1895 a group of English capitalists who had formed the British Columbia Development Association sent one of their number, Charles Herbert Wilkinson, to Canada's westernmost province to investigate investment prospects. From Ernest Billinghurst, the brother of another member of the syndicate, Wilkinson learned of Captain William Moore. Moore, discoverer of the White Pass and its foremost promoter, had previously sought money from Billinghurst to develop a route over this pass. Consequently Billinghurst introduced Moore to Wilkinson. Wilkinson showed enough interest in Moore's scheme to send Billinghurst to Skagway. Billinghurst reported favourably and in 1896 a decision was made by the syndicate to proceed with a transportation project of an undetermined nature. A small sum of money was placed at Moore's disposal and in 1896 Moore used this money to cut a rough trail a few miles out of Skagway.65 This initial development work was followed by the incorporation of two Canadian companies in May and June of 1897, the British Columbia-Yukon Railway Company and the British Yukon Mining, Trading and Transportation Company, to build a railroad from the summit of White Pass to the trading post of Selkirk on the Yukon River.66 The absence of enabling legislation providing for railroads in Alaska prevented the syndicate from obtaining a right of way between Skagway and the summit.67

The incorporation of these two companies predated the beginnings of the Klondike stampede by some three weeks. Thus the decision to build a railroad into the Yukon, if not actually proceed with its construction, was not a result of the gold rush. In fact, considering the short duration of most gold rushes, the Yukon's remoteness and the cost of building a railroad over such difficult terrain as the Alaska Panhandle, the gold rush was not of itself an altogether attractive proposition for investing in a transportation system as permanent as a railroad. This is not to suggest, however, that the promoters were unaware of the Bonanza Creek discovery nor that the discovery had no effect in their application for the railroad charters. As Tappan Adney has pointed out, the Klondike discovery was "common property outside six months before" the Portland and Excelsior caused the "acute attack of insanity" that precipitated the stampede.68 What should be noted is that other factors not directly related to the gold rush played an important role in the syndicate's decision to build the railway. Information culled from official and unofficial sources suggested that gold was not the only resource which could be profitably exploited. Reports concerning grazing land, timber and other metals indicated that the Yukon had a potentially rich and diversified economic base, a crucial consideration in light of past gold rushes and the exhaustible and nonrenewable nature of gold mining.69

When news of the Klondike discovery broke in the English press in August of 1897, the British Columbia Development Association found itself in an excellent position to exploit the transportation potential of the ensuing stampede. The syndicate, however, had fallen on bad times and a lien was taken on it by the English financial house of Close Brothers. When the syndicate failed in March 1898, Close Brothers appropriated its assets, including the two railroad charters that had been secured the previous spring.70

On 29 March 1898 Close Brothers obtained a West Virginia charter to build a railroad between Skagway and the summit of White Pass. With the passage of a bill "Extending the Homestead Laws and Providing for Right of Way for Railroads in the District of Alaska" by the United States Congress on 14 May 1898, and the subsequent approval by the secretary of the Interior of the company's application for a right of way from Skagway to the summit, the last legislative obstacles to a Skagway-Selkirk railroad were overcome.71

In the meantime, a representative of Close Brothers, Sir Thomas Tancred, had travelled to Skagway to determine the feasibility of building the railroad. With two United States representatives of the firm, Samuel H. Graves and E.C. Hawkins, in tow, Tancred arrived in Skagway in April.72

Once in Skagway, Graves and Hawkins made a preliminary survey of possible routes leading over the White Pass. After a cursory examination of the area, they met with Tancred in the bar of the St. James Hotel in Skagway. There they discussed the feasibility of building the railroad and debated what recommendations Tancred should relay to the investors on his return to England. On the basis of their survey, Graves and Hawkins were agreed that the rail line could not be built. At this point the discussion was on the verge of breaking up and with it the railroad scheme when something occurred that was to be of major consequence for the railroad scheme, the subsequent course of transportation history in the Yukon and indeed the history of the Yukon Territory itself. Michael J. Heney, an independent railroad contractor, had overheard their conversation. Heney, who had been lured to the north by the vision of building a railroad into the Yukon, had conducted an extensive survey of the area leading out of Skagway and was convinced that a railroad could be built. The only thing he lacked was capital. With the conviction of a man obsessed and the persuasiveness of one who knew that his dream hung in the balance, Heney set out to convince Tancred, Graves and Hawkins that he was right. His determination and his experience won the Close Brothers' representatives over. On 27 May 1898, men and supplies were landed at Skagway. The following day construction commenced.73

The construction phase was beset by many problems. Skagway was over a thousand miles from Vancouver, Victoria or Seattle, the nearest supply bases, and the only connecting transportation service was by water. The capacity of the coastal steamer fleet that served Skagway and other points on the Panhandle, moreover, was taxed to the limit by the stampeders. With the end of the gold rush a momentary respite ensued, only to disappear with the outbreak of hostilities in Cuba and war with Spain. The war brought the United States government into direct competition with the railroad for transportation services on the coast and the American government forced its advantage by pressing practically every available vessel on the Pacific coast into government service. In Alaska the terrain over which the roadbed was to be laid was devoid of material adequate for ballast and as a result, ballast had to be hauled from the bed of the Skagway River at one end and from Fraser at the other. These difficulties were exacerbated by an insufficient labour supply. "It was obviously out of the question to engage men in the ordinary way and convey them in hundreds at our cost to Skagway, the first president of the railroad later recalled, "because while the gold fever was at its height, the moment they set foot ashore in Skagway would be our last glimpse of them."74 Nevertheless, the harsh reality of the gold rush partially offset this problem. There were literally hundreds of vagrants in Skagway awaiting the arrival of friends, money or both before hitting the trail for the gold fields. Here was a ready-made labour market if, albeit, an unskilled one. But these vagrants were also a "captive market" and the merchants of Skagway who dominated the town council realized that once the railroad began operating, their captive market would disappear. As a result, the town council made it known that it expected the railroad to pay for the privilege of operating out of Skagway and if this failed, the council threatened to seek an injunction preventing the railroad from laying track on that section of the roadbed that ran through town.

39 White Pass and Yukon Route railway.
(Map by S. Epps.) (click on image for a PDF version)

When the controversy was at its height and while the town council was engaged in a volatile meeting, the railroad crews worked through the night. By morning the track was laid. Presented with a fait accompli, the town council capitulated.75 Despite this small success, however, the company's labour problems were not fully overcome. Turnover continued at a rapid rate as men, heretofore forced to remain in Skagway for a variety of reasons, found these reasons obviated with the arrival of their friends or money. No longer restrained, they would drop their tools and leave for the Klondike.76

Like so many other transportation projects of the period, the White Pass and Yukon Route77 railway was built with the intention of returning an immediate profit to its investors. But while other transportation companies, cognizant of the boom and bust nature of the gold-rush phenomenon, operated on the "high-grading" principle with planned obsolescence in mind, the railroad was designed as a permanent fixture in the Yukon transportation system. It would be built, Samuel Graves noted,

in the belief that the line that would pay best was a well located one, with the lowest possible gradients and a very solid roadbed over which heavy engines could haul heavy loads up the hill in summertime, and which would admit of modern appliances for snow fighting in the winter.78

The intentions of those on the site, however, could and often did exceed the expectations of observers in Skagway and indeed the expectations of some of the investors themselves. In Skagway local wags referred to the project derisively as the "Jackass and Yukon Railway," while in England some of the less enthusiastic shareholders tried to jettison the scheme with arguments in favour of a tramline.79 But the contractors and the Close Brothers representatives in the field persisted and as the weeks slipped by the line finally began to take shape.

By 21 July four miles of track had been completed.80 This brought the roadbed to the foot of the pass where the most difficult section in terms of construction would begin. Here the White Pass rose to an altitude of 2,885 feet in 14 miles. In order to reduce the gradient, 21 miles of track were required. The problems involved in finding a satisfactory roadbed for the additional 7 miles of track were complicated by Brackett's wagon road. The right of way that Brackett had reserved for his road was the most direct and feasible line to the summit and while the roadbed for the railway could cross this line, it could not run parallel to it. Finally, after several attempts to skirt Brackett's road, which were made even more costly by the need to remove debris that constantly fell on the wagon road as the railroad crews carved out the roadbed, the White Pass and Yukon Route bought Brackett out.81 As the railroad slowly inched its way up the pass to Heney Station during August of 1898, word of the gold strike at Atlin, British Columbia, reached the construction camps. Its impact was immediate as an estimated 65 per cent of the work force abandoned their jobs and stampeded to Atlin. It was not until October that the company was able to replenish the crew to their pre-Atlin levels and by that time two valuable summer months of construction time had elapsed.82

With the onslaught of Alaskan winter, construction entered a new stage. Winter work required an almost superhuman effort. "The strong winds and severe cold made the men torpid, and benumbed not merely their bodies but their minds," Graves wrote, "so that after an hour's work, it was necessary to relieve them by fresh men."83 At mile 15 the crews reached the most difficult stretch on the line. Here a perpendicular wall of granite rose nearly two thousand feet from the canyon floor. Polished by the action of long-extinct glaciers and worn sooth by the winds that ripped through the pass, this stretch presented a problem of the severest order. Working on platforms supported by crowbars drilled into the granite below and suspended by ropes secured from above, the labourers blasted and hacked the face of the rock until a tenuous horizontal ribbon had been cut for the roadbed. A mile further on, a tunnel 250 feet long was bored through the mountain. At mile 19, the Dead Horse Gulch Viaduct, a switchback 215 feet above a rock-strewn stream was built. Finally on 18 February 1899 the summit was reached. Two days later freight and passenger trains were placed in service.84

Once over the summit, construction proceeded apace to the head of Bennett Lake. To facilitate supply, Heney organized a teamster operation which doubled as a passenger- and freight-carrying service between the summit and the lake. When the railroad reached Bennett on 6 July 1899, the horses and wagons were withdrawn and made ready for use over the frozen surface of the lake that winter.85

40 Cutting the grade for the White Pass and Yukon Route railway on Tunnel Mountain.
(Yukon Archives.)

41 A crew pauses for a noon meal at the entrance to the tunnel.
(Yukon Archives.)

42 The switchback at mile 19.
(Yukon Archives.)

43 A railroad crew clears the grade and lays track. Occasionally the rails dipped and swayed with the configuration of the ground.
(Yukon Archives.)

44 Wheelbarrow crews remove rock near the summit of White Pass.
(Yukon Archives.)

45 A cut through glacial frost and what appears to be a causeway in the distance.
(Yukon Archives.)

46 The first passenger train to the summit of White Pass.
(Yukon Archives.)

From Bennett the survey called for the line to skirt the eastern shore of the lake to Caribou Crossing (now Carcross). Because heavy rock blasting was required on this section, the company transferred most of its operations to the Caribou Crossing-Whitehorse section with the exception of the rock crews which were left to cut out the road beside the lake. The decision to complete the first and last sections of the railway before tackling the middle made it necessary to move rails, engines, rolling stock and other material over the middle section to Caribou Crossing. For this purpose the lake was used as a connecting link and a power barge was built. As with Heney's horse-drawn wagons that had operated between the summit and Bennett, the barge was also used to move people and supplies going on to Dawson.86 When the lake froze over that fall, the White Pass and Yukon Route organized the "Red Line Transportation Company." This cartage service moved "a large number of passengers and hundreds of tons of freight, including material, engines and boilers etc., for a number of steamers built at the head of Lake Bennett that spring." On 8 June 1900 the Caribou Crossing-Whitehorse section was finished. In the meantime the company had bought out Macaulay's tramlines at Miles Canyon and Whitehorse in order to secure a right of way. With the completion of the line between Caribou Crossing and Whitehorse, a through transportation service was inaugurated between Skagway and Whitehorse, utilising the power barge on Bennett Lake. Finally, on 29 July 1900 the Bennett-Caribou Crossing section was completed. The following day through trains between Skagway and Whitehorse began operating.87

When completed, the White Pass and Yukon Route railroad was 110.7 miles long. Of this, 20.4 miles of track were in Alaska, 32.2 miles in British Columbia and 58.1 miles in the Yukon Territory. The railway was of the narrow-gauge (three feet) type, 45-pound rails having been used. From Skagway to the summit of White Pass, the most difficult stretch in terms of construction and maintenance, the line ascended 2,885 feet. The maximum gradient on this section was 3.9 per cent, the average being 2.6 per cent. The highest elevation was not at the summit of the pass, however, but at Log Cabin, British Columbia, mile 33, where the railroad reached an altitude of 2,916 feet. The average track curvature on the Alaskan side ranged from 16 to 20 degrees. The total construction cost was approximately ten million dollars and an additional 2.5 million dollars were spent on rolling stock and equipment.88

All told, some thirty-five hundred men were employed at one time or another while the railroad was being built. Of this number, 35 died from all causes including illness as well as accident.89 Writing a few years after the railroad had been in operation, Graves noted that probably "no other railway in the world was built by such highly educated men as worked on our First Section." This phenomenon, a demographic product of the Klondike stampede, was only one of the remarkable factors associated with the building of the railroad. Another was the "White Pass spirit" that prevailed during the construction phase.90 Except for a short strike in March 1899, which was precipitated by the company's decision to reduce wages and extend the working day, no other serious labour-management problems were encountered.91

As the first train ever to ride on the new rails confidently made its way from Whitehorse to Skagway, pulling the empty cars which had accumulated in Whitehorse since June of 1900, an Irish crewman remarked, "Be Jakers — the first thrain into this country was a thrain OUT."92 Whether he realized it at the time or not, this Irishman perceived both the challenge and the continuing dilemmas that were to confront the transportation system of the Yukon, and indeed the Yukon itself, for the next half century.

47 The Red Line Transportation Company's last run into Bennett from the summit, 6 July 1899.
(Yukon Archives.)


The pressure of the gold rush on the river transportation system had an immediate impact that was reflected in the tremendous expansion of navigation facilities. By August of 1898, 30 additional transportation companies had joined the original firms, the Alaska Commercial Company and the North American Transportation and Trading Company, in the competition for traffic on the Yukon River. Taken together, these companies operated some 60 sternwheelers, 20 barges and 8 tugboats.93 Despite the unprecedented pressure on the navigation system, however, the gold rush did not result in the introduction of any new forms of water transportation. The sternwheeler retained its role as the basic form of river communication during the period. While the gold rush was not accompanied by any qualitative changes in sternwheeler technology, it was marked, nonetheless, by a significant modification in water routes to the Yukon interior. Before 1898 the sternwheeler had worked the lower Yukon exclusively. Sternwheeler traffic originating at Saint Michael had never been extended above Robert Campbell's old fur-trading post at Fort Selkirk. It was for this reason that Fort Selkirk had been designated the original terminus of the White Pass and Yukon Route railway.94

The gold-rush popularity of the upper river system was to have a significant effect on the transport function of the upper and lower routes. While thousands of people busied themselves in the spring of 1898 building the wind- and pole-propelled vessels that would convey them from Bennett Lake to Dawson when the ice broke up, a few began the historic experiment of assembling the steam-powered craft they had sledded over the passes the previous winter. Though there is some dispute as to which sternwheeler was the first to successfully navigate the upper river between Bennett and Dawson that spring,95 that controversy is less important than the success of the experiment itself. An historic breakthrough had been accomplished — steam had been introduced on the upper route. The effect of this achievement was to be of profound consequence for the future of the Yukon transportation system.

Having demonstrated that the sternwheeler could be successfully used on the upper river route, a number of sternwheelers were built at Bennett Lake. Because of the obstacles to through navigation posed by Miles Canyon and the Whitehorse Rapids, the fleet was separated into two divisions — one serving the Bennett-Miles Canyon section, and the other the Whitehorse-Dawson run.96 In 1898 the Bennett Lake and Klondike Navigation Company operated the Flora and Nora below Whitehorse to connect with the Dra which plied between Bennett and Miles Canyon. These vessels were approximately 75 feet long. The trip between Bennett and Dawson took four and one-half days. Sleeping accommodation consisted of wooden bunks ranged in tiers of three and the passengers supplied their own bedding. The cost for the through trip was $75. Meals were an additional dollar each. In the meantime, the Canadian Development Company had placed the Willie Irving, the Goddard and the Anglian between Whitehorse and Dawson. In the summer of 1898 steam service was expanded to include Lindeman Lake with the operation of a steam ferry to connect with Bennett.97 The upper route proved to be so successful that it was able to challenge the historic pattern of river transportation dominated by the lower river via Saint Michael. As a result, various sternwheelers like the Victoria were withdrawn from the lower run and transferred to the upper Yukon.98

With the introduction of sternwheelers on the upper river, the desirability of extending the railroad to Fort Selkirk, a need which had previously existed, was obviated. Whitehorse, situated at the foot of the major impasse to through navigation on the upper river, Miles Canyon and the rapids, was selected as the terminus for the railroad. When the railway was finally completed in July of 1900, the upper river route, served by the railroad and a complementary fleet of sternwheelers, was in a position to supplant the lower route as the main life line to Dawson.


Recalling his trip to the Klondike, Robert Kirk wrote that "the present advanced state of development of the northern mines is due largely to... dog-teams."99 While Kirk may well have exaggerated the importance of dogs as a transportation factor in the Yukon to the detriment of other forms of transportation, it must be admitted that the dog and the dog team played a crucial role in northern life during this period.

No one knows when dogs were first used to fulfill a transportation need in the Yukon. The Indians used them to the extent that their technologically primitive society demanded and the Hudson's Bay Company had utilized dogs in the prosecution of the fur trade. With the coming of the prospector in the 1880s, the use of dogs had declined in proportion to the falling off of the trade in furs. This occurred in part because the mines were worked during the summer months only. Many of the miners left for the outside before the annual fall freeze-up, while the rest passed the winter months inside their cabins. The evolution in mining technology occasioned by the introduction of winter digging in the early 1890s and the discovery of the rich gold fields in the Klondike in 1896 renewed the demand for dogs and dog teams.

Dogs were used for a variety of purposes during the Klondike period: to haul poles, logs and lumber for sluice boxes, drift burning and cabins; to deliver mail; to reach the small outlying settlements in the territory; to commute from creek communities to Dawson; to freight supplies, food and equipment; to carry gold from the mines to Dawson; to labour in the mines themselves, and to deliver water door to door before Dawson acquired its water system. During the summer dogs were used as pack animals; during the winter to draw sleds.100

It has been estimated that by 1899 there were some four thousand dogs regularly employed in the town of Dawson. While most of these dogs were privately owned, many of the transportation companies used dogs as well. Men often made $100 a day freighting supplies to the mines and in the spring of 1898 a return of $150 a day was not uncommon. During the summer months gold was brought into Dawson from the mines by dog trains consisting of 15 to 20 dogs. Each dog in the train carried a pack which weighed between 20 and 30 pounds. An indication of the wealth conveyed in this manner is apparent from the fact that a train of 15 dogs, each dog carrying a 30-pound pack, transported gold to the value of $122,400 (gold valued at $17 an ounce). For two and one-half months during the summer, the dog trains operated 24 hours a day, six days a week. During the winter, dog punchers worked eight hours a day, averaging 20 miles with a load of twelve hundred pounds. Occasionally, the dog teams were relieved from freighting supplies to the creeks and used for trips to the coast on Lynn Canal, 500 miles away. During these trips the packers carried private mail and light express. The charge for carrying a letter was one dollar, and the driver often increased his revenue by taking a miner from Dawson to the coast. The passenger generally paid some $500 for the privilege of accompanying the sled, not for riding in it, and he was expected to assist in the making of camp and the cutting of firewood and to furnish his own blankets and robe.101

48 The Clifford Sifton, the Bailey and an unidentified sternwheeler at Bennett. Until the railway was completed to Whitehorse, Bennett was the head of navigation on the upper Yukon.
(Yukon Archives.)

49 A steamboat race. The vessel on the right is probably the Canadian since it and the Bailey were involved in one of the most famous, and wisely one of the few, steamboat races on the Yukon. According to a report in the Victoria Daily Colonist of 29 September 1900, "the fight was a draw. A battle royal had been fought (the Bailey rammed its opponent twice). The Canadian, although having somewhat the best of the struggle, was so badly damaged that no victory was claimed."
(Yukon Archives.)

50-56 Dogs performed just about every transport function conceivable. They packed in summer, made long-distance trips in winter, provided Dawson and Whitehorse with primitive "municipal waterworks," and drew loads which belied their small size.
(Yukon Archives.)

Various breeds of dog were used for freighting in the Yukon. By far the most sought-after were the native breeds; the husky, the malemute and the "Siwash" or Indian dog. These dogs were aptly suited to the rigours of the northern environment, being "well-boned, deep chested, and strong in the back, fore and hind quarters," as well as having thick outer and inner coats of hair and paws that were well-furred between the pads and the toes. Arthur Treadwell Walden, a noted Yukon dog puncher, preferred the husky to the malemute because it was larger and stronger, and observed that the "Siwash" or Indian dog was generally less preferable than either because of its supposed indolence. It is interesting to note that Walden's ascription of greater size and strength to the husky instead of the malemute reversed the conventional distinction drawn by professional breeders between the two types of dogs, thereby suggesting that Walden mistook one for the other. In addition to their superior physical characteristics, the native dogs showed a marked propensity for scavenging, to which the elevated food caches that dotted the northern landscape bore mute testimony. Accounts of their bad temper were legion as they were quick to attack one another when confused or frightened and no gold-rush story was considered complete with out a graphic description of a particularly vicious dog fight.102

During the gold rush a brisk business was done importing outside dogs for sale. In fact, contemporary photographs give every indication that the number of outside dogs far exceeded the number of native breeds in use at the time. Though not as valuable as the native breeds, these dogs proved adaptable to northern conditions, nature furnishing them with a thick coat of hair once the cold weather had set in. They had neither the strength nor the endurance of the native dogs and their dietary requirements were also greater, but they were particularly well-suited to short-distance hauling. Tappan Adney, an acute observer, noted that for this purpose the St. Bernard and the mastiff were unsurpassed.103

On the trail, dogs were fed dried salmon, each dog being given approximately two pounds of fish each day. Not only was dried salmon relatively cheap, but experience showed that it was also more nourishing than other foods, "one pound of dried King salmon being equal to five pounds of fresh meat." The Indians did a brisk business selling fish to the dog punchers. When fish was not available, a concoction of bacon and rice was prepared that was frequently adulterated with cornmeal or oatmeal, each dog being given three to four pounds. The dogs were fed once a day and always at night. This was done because it encouraged them "to make better time on the trail." Another reason was that the dogs tended to become lazy and indolent after feeding "and practically unfit for work."104

Of all the dogs in a team the most important was the leader. During the winter of 1897-98 when dogs were selling for $200 in Dawson, lead dogs were commanding $300 apiece. A well-trained lead dog relieved the driver of half the work involved in driving the team. The qualities required in a good lead dog were those of intelligence, discipline and responsiveness. The leader was not expected to augment the pulling power of the other dogs. It was his job to keep to the trail and to hold the harness taut. Leaders were generally of the native breed although this was a matter of individual preference. Jeremiah Lynch, a former United States senator, favoured an outside dog, believing that the Scottish collie made the best leader. The professional dog punchers generally had two lead dogs, one to relieve the other when breaking trail.105

The two most common types of sleds used during this period were the Yukon sled and the basket sleigh. The Yukon sled, or standard freight sled, originated in the Cassiar district of northern British Columbia and was brought into the Yukon during the early 1890s. Seven feet in length, the Yukon sled was 16 inches wide on the runners and set 4 inches off the ground. It was narrow in comparison to other contemporary styles, a characteristic which gave it more mobility. To compensate for this width limitation, the sled box was designed to overhang the runners by two inches on each side, thereby increasing the carrying capacity by four inches. The frame of the sled was light but strong, consisting of four pine slats laid lengthwise over four crosspieces of ash. The runners were also made of ash, with a brass plate attached to the bottom. Wooden runners were not used alone, as the ice had the effect of cutting them to pieces, and steel runners were apparently not popular because they had a tendency to split in low temperatures and produced too much friction under the same conditions. There is no evidence to suggest that the Indian custom of shoeing the sled runners with a thin veneer of bone was ever adopted by the white dog drivers.106 This was unfortunate as the Indian type of runner proved particularly well-suited to northern conditions.

A "gee-pole" was lashed to the side of the Yukon sled.107 Approximately six feet long and three inches thick at the butt end, the gee-pole extended upward from the front of the sled at an angle of 45 degrees. The gee-pole facilitated steering and was used to keep the sled upright. It was also used to break the sled loose when the runners froze to the surface and as a braking device when going down small hills.108

The second style of sled, the basket sleigh, originated among the Indians of the lower river, It was larger than the Yukon sled, varying in length from 8 to 12 feet and in width from 20 to 22 inches, and the body was 6 inches above the runners. Unlike the Yukon sled, the carrying surface did not overhang the runners. The basket sleigh had a light, flexible frame that was made from hickory, oak or white birch. The sides of the sleigh consisted of narrow wooden uprights set at two-foot intervals, a foot high in the front and two feet high in the rear. Each side was covered with a rail and the interstices of the frame were filled with a netting of cord or rawhide. Instead of a gee-pole, the basket sleigh was fitted with a plough-handle arrangement that extended from the middle of the sleigh to the back.109

The load that could be drawn by a dog team depended on two factors: the dogs themselves and the condition of the trail. Under optimum conditions a team of five native dogs could pull a sled with a thousand-pound load a distance of 15 to 25 miles in a single day. Generally speaking, the sled was loaded in the ratio of 160 pounds to each dog.110 In loading a sled, weight was distributed so that the heaviest part was in the front. The load was placed on a light canvas sheet approximately eight feet by ten feet which covered the bottom of the sled. Once loaded, the canvas sheet was drawn up and wrapped around the load. Two long ropes at the end of the sled were then woven back and forth through the side ropes to secure the sled. For extra security, these ropes were woven through V-shaped ropes at each end. After this was done, water might be splashed over the rope joints, to freeze them together.111

Only after the sled had been loaded were the dogs harnessed and the traces hitched. During the gold rush the standard dog harness consisted of a leather collar with a back band and a belly band to which the traces were attached. In harnessing the team, a dog puncher started with the lead dog and worked back to the sled dog. The dogs were generally arranged in tandem fashion rather than in pairs. Once harnessed, the dogs were hitched together, starting with the lead dog, until the dog in front of the sled dog was reached. Then the dog in front of the sled dog was harnessed to a whiffletree at the front of the sled. The sled dog was hitched separately to the whiffletree with a short pair of traces so as to enable him to jump out of the line and pull at right angles when the sled went around a corner.112

When the sleds were loaded, the dogs were coupled closely together to increase their pulling power. Another set of rings at the junction of the back band and the traces was used when the sleds were empty. This gave the dogs an additional 18 inches between them and made it easier for the dogs to run. Three miles an hour, or the speed at which a man could walk, was usually as fast as the dogs could go when the sled was loaded; when the sled was empty a dog team travelled at the rate of six miles an hour.113 To reduce friction the runners were always iced while on long trips a sail was sometimes used to take advantage of fair winds.114

While it was not uncommon for a thousand pounds to be loaded on a single sled, the experienced dog punchers generally used three sleds when hauling heavy freight. This was done in order to distribute the weight over 21 feet, thereby allowing a greater degree of mobility and speed. The sleds were connected by cross-chains. Hitched together, the three worked as one as the mode of connection forced each sled to follow in the track of the one in front. The sleds in these "sleigh trains" were reinforced with iron braces to reduce stress. As with the single sleds, weight was distributed so the heaviest part was in the front, the usual distribution being six hundred pounds in the first sled, four hundred in the second and two hundred in the third.

If the sleds tipped over, they were righted one by one. On long hills the sleds were unhitched and hauled up separately. On steep descents the dogs were unsnapped and the sleds allowed to go down on their own power. To reduce the momentum of a sled on a downhill run the driver rode on the gee-pole by leaning back on it and using his feet as a brake. This was considered the most dangerous part of sledding. Several fatalities occured and in one instance a driver was impaled, but riding the gee-pole was a necessary evil. Mechanical braking systems were tried, proved unsatisfactory, and abandoned.115

Over-sweating was a constant problem when the dogs drew heavy loads. To offset this condition, the driver frequently stopped and unhitched the team to enable the dogs to cool off by rolling around in the snow. This sudden chilling had no apparent ill effects on the dogs. The most difficult surface for the dogs to work on was ice. Their paws, warmed by the strain of pulling, would melt the ice, allowing frost balls to form under their nails and between the pads. These frost balls cut the paws like broken glass. As soon as the driver saw a dog limp he would stop the team and thaw out the pads by putting the dog's foot in his mouth, after which he dried the foot with his shirt. Frequently an outside dog, ignorant of the consequences, would stand still on the ice when the team was stopped, causing his warm feet to freeze solid to the trail.116

On long trips, especially between Dawson and Skagway, a dog puncher had to carry sufficient provisions to last him throughout the trip. These included fur robes, blankets, a small tent, a small sheet-iron stove, food for the party and for the dogs, cooking utensils, snow goggles and snow shoes. Within these limits, the weight of the outfit was reduced to the bare minimum as comfort was sacrificed to speed. On the trail, men wore drill parkas, fur caps which covered the ears, and heavy wool-lined, moose-skin mittens. Several pairs of moccasins were carried as it was necessary to keep feet dry at all times. Drill parkas were favoured over fur ones because the latter were generally found to be too warm. Fur parkas were worn at night and in the early morning when the temperatures were lower.117

On the trail the drivers and the passengers drank tea exclusively as it was considered to be more nourishing than either coffee or cocoa. Liquor was never touched as it dulled the senses.118

Travel on the Yukon River was difficult. The water did not freeze evenly and the surface was rough. Wind and snow exacerbated the surface conditions. Wherever it was necessary to break trail, the driver would precede the dogs and pack the snow with his snowshoes. The trip to the coast took 30 days. If light freight or express were carried, two sleds might be taken, one of which could be dropped along the way once the driver and team had consumed enough food to permit the balance of the supplies to be transferred from one sled to the other. During the late fall and early spring when the ice was not frozen during the day, it was customary to travel at night.119

It was no coincidence that both the gold rush and the golden age of dog transport occurred at the same time. The gold rush created the need for a winter transportation system, a need that had been practically nonexistent before 1896, and the demand created thereby could be met by the only available resource — dogs. As a result, dogs and dog teams became synonymous with winter transportation between 1897 and 1900. But the gold rush also created pressures that were to limit the supremacy of the dog to a brief interval. Although the dog was well-suited to northern conditions, the gold rush underlined the fact that dogs could not satisfy northern transportation needs adequately because of their size and pulling capacity. Once the horse was introduced, it was only a matter of time before it would supersede the dog. By 1900-01 the transition was pretty well complete. Nevertheless, the dog retained certain transportation functions which extended well beyond 1900. Of these, the most important were communication with outlying settlements not served by roads and in Mounted Police winter patrols.120


Although horses had been used in the Yukon prior to the gold rush, it was not until 1900, after it had been demonstrated that the horse could successfully survive the Yukon winter, that horses were to supplant dogs as the principal medium of overland transportation. Before 1896 it was generally believed that no horse could survive the northern winter on a diet of native hay and the demand for horses had not been great enough to warrant the expense of importing hay for the winter season. In Circle City, where horses had been used to pack supplies to the mines, the entire horse population had been killed each autumn in order to prevent mass starvation. Between 1896 and 1898 when the demand for horses increased enormously, the exorbitant price of hay, which fluctuated between $400 and $1,200 a ton, discouraged large importations of horses to meet the increased demand. Yet despite inflated prices and the limited seasonal value of horses, the teamsters who operated during the gold rush managed to realize substantial profits. Freight charges for horse cartage were high, ranging from a flat rate of 45 cents a pound on gold, to 25 cents a pound from Dawson to anywhere on Bonanza Creek, 35 cents a pound to Eldorado, and 50 cents a pound to Hunker Creek for general merchandise. One observer estimated that each horse employed in Dawson during the gold rush earned $4,500 a season for its owner.121

In spite of the horse's physical superiority — a horse could draw or pack at least five times the weight that a dog could122 — the dog retained its status as the Yukon's primary work animal until 1899. Until the question of winter survival could be answered affirmatively, the use of horses would always be limited. Although William Ogilvie had reported in June 1896 that horses "have been in use here, packing to the mines in the summer and hauling wood in the winter for several years... notwithstanding that they live only on the coarse grasses of the country," it was not until the winter of 1897-98, when 12 horses were brought in from Circle City and Forty Mile and kept working on a diet of native hay and sheltered in stove-warmed tents, that it was popularly believed that the horse could survive the cold.123 The successful wintering of a number of horses in 1898-99 by the North-West Mounted Police confirmed the year-round practicality of horse transport and was followed by the importation of large numbers of horses the following spring.124 In the summer of 1899 the cutting and drying of local red-topped grass assumed the proportions of a small industry in preparation for the winter ahead and that winter over one thousand horses were worked.125 The primacy of the dog had been broken.

57-59 Some typical examples of turn-of-the-century teamster operations. Figure 57 shows a team hauling a heavy boiler to the creeks.
(Yukon Archives.)

60 The Klondike gold fields. (click on image for a PDF version)


Prior to the gold rush, no internal transportation system of any consequence existed in the Yukon. To be sure, the basic outlines of a domestic transport system based on water had been established as early as the 1870s by such men as McQuesten, Harper and Mayo, but the overriding consideration of the pre-gold-rush era had been to develop an efficient system of communication with the outside. Communities that were little more than tiny trading posts, a small, constantly floating and ever fluctuating population, and the seasonal nature of the mining industry — each characteristic of the Yukon before 1896 — were not in themselves sufficient to create an internal transportation system. Under gold-rush conditions, however, the pressures for an internal transportation system mounted, only to be increased by the problems posed by the shallow, unnavigable creeks that linked Dawson and the gold fields, and the introduction of horses.126 Taken together, these factors generated a need for a system of roads and trails with subsidiary bridges and ferries to connect the Klondike mines with Dawson on a year-round basis.

Initially, road construction was undertaken by the individual miners and the teamsters who operated between Dawson and the mines. Between 1897 and 1898 a footbridge was built between Dawson and Klondike City, a scow ferry was placed in operation on the Klondike River connecting the north shore with a point just west of the mouth of Bonanza Creek, and a road was cut up Bonanza valley to Grand Forks, located at the confluence of Bonanza and Eldorado creeks.127 Nevertheless, it soon became apparent that the private sector was neither willing nor able to deal adequately with the road requirements of the district.

Given the avarice which underlined the stampede, this is hardly surprising. Men had not rushed to the Klondike to establish a permanent community replete with a system of roads: they had come in search of gold. Once successful, and all thought they would be, they planned to leave as quickly as possible. As a result there was no inducement to improve the trails which connected Dawson with the mines apart from the desire of each individual to facilitate his own movement. The only other alternative within the private sector, a toll road, was tried and proved a failure. As the newspaper campaign against the Pioneer Tramway Company demonstrated, neither the miners, the merchants nor the teamsters were interested in a solution of this type.128

In the meantime the one road between Dawson and Grand Forks had deteriorated to the point where "mud and water were up to the knees in many places." Along with the inflationary pressures set off by the stampede, the condition of this road was responsible for the generally exorbitant freight charges characteristic of the gold-rush period. Exasperated by inadequate transportation facilities, the community began to agitate the issue with the demand that government assume responsibility for road construction and maintenance. Citing the British Columbia gold camps as a precedent, many petitioned the gold commissioner for redress. Finally, the much-vilified gold commissioner, Thomas Fawcett, advised his superiors that there "ought to be some assistance in building roads."129

Fawcett's recommendation was acted upon in 1899. In that year the first government-built road in the territory was completed between Bonanza and Hunker creeks, with branches to Gold Bottom and Caribou. This was followed in 1900 by the construction of a new road to join Dawson and Grand Forks. Seventy-seven miles of wagon road and 170 miles of sled trails were built between 1899 and 1900.130

Certain soil characteristics made the construction of roads a complicated undertaking. The roads that were built in the area contiguous to Bonanza Creek, an area that was densely populated and hence in need of good roads, required a great deal of drainage, much corduroy and substantial quantities of fill. As a result, these roads were always better in the winter after the first snowfall than in the summer.131 The hard-packed snow offered a smooth, regular surface which permitted an ease of movement not to be found at other times of the year and eliminated the dust problem that was ever present during the summer months. During the spring, summer and autumn months, travel, whenever possible (which admittedly was not often), was done at night to take advantage of the stabilizing effects which lower temperatures had on the road surface.132 But climate and soil characteristics were not the only drawbacks to road construction during this period. Unusually high labour costs reinforced the tendency to temporize with the road problem in the belief that the problem would disappear once the gold rush had spent itself.133

Despite these obstacles, the foundations of an internal overland transportation network were well on the way to being realized by 1900. This new development, marking a significant stage in the evolution of the Yukon transportation system, was accompanied by direct government involvement through the appropriation of public money to pay for road construction and maintenance. The assumption of this responsibility was also attended by a change in attitude on the part of the dominion government, which had previously tended to view the Yukon and the gold rush as passing phenomena. Nevertheless, the government's admission of its responsibility did not settle the roads question to the satisfaction of many Yukoners. Rather, it transferred the issue to the political arena — an arena already characterized by a substantial amount of anti-government feeling.134 Further, this dependence on the government was to have long-term implications which could not before seen in 1900 and which were to be exacerbated by the Yukon's territorial status. So long as the central government could regard the Yukon as a hinterland to be profitably exploited, the essential road requirements of the territory would be satisfied, but once the Yukon became a millstone around the neck of the imperial power, as it did after 1914, the domestic needs of the territory would be sacrificed to other priorities as defined by Ottawa.135

61 A suspension bridge over the Klondike River and freight scows on both banks, Dawson, 1899.
(Yukon Archives.)


The Klondike gold rush altered irrevocably a substantial portion of the pre-gold-rush transportation system even though that system had been largely responsible for the scope and nature of the great stampede. Dyea and the Chilkoot Trail, which had served almost a generation of pre-1896 prospectors as well as the majority of those who had rushed to the gold fields in 1897-98, virtually disappeared in the wake of events that transpired during these years. Their demise was offset, however, by the rise of Skagway and the White Pass route, thereby vindicating Captain William Moore's ten-year act of faith that Skagway would one day replace its Lynn Canal rival as the gateway to the Yukon. For this the railroad was totally responsible.

The railroad survived as the most lasting legacy of the gold rush. That life in the Yukon did not return to its pre-1896 state is attributable in full to its existence. Without the railroad it is difficult to conceive of the mining industry adapting as successfully as it did to the conditions that prevailed after 1900 and without this adaptation it is inconceivable how the Yukon could have sustained the population that remained. It is inconceivable, moreover, that thousands would have chosen to remain had they been compelled to accept eight months of complete isolation — a fate that was theirs had the railroad not been built.

The railroad was also responsible for the subsequent shift in trade away from Saint Michael to the upper river and the emergence of Whitehorse as the transportation hub of the territory. Whitehorse's evolution to its present status was born of the decision to locate the railroad terminus there.

Less spectacular but no less important was the need for winter transport which the gold rush created. Before the stampede, transportation and communication had been limited to the summer months. This proved adequate so long as the traditional prospector and his values prevailed, but both were overwhelmed by the great influx of 1897-98. The introduction of overland transport to satisfy this seasonal requirement was an event of long-term significance and added another dimension to the Yukon transportation system.

previous Next

Last Updated: 2006-10-24 To the top
To the top