ON THE PERIPHERY OF THE KLONDIKE GOLO RUSH: CANYON CITY, AN ARCHAEOLOGICAL PERSPECTlVE
Thomas J. Hammer B.A. (Honours), Simon Fraser University, 1994
THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE OEGREE OF MASTER OF ARTS in the Department of Archaeology
@ Thomas J. Hammer 1999 SIMON FRASER UNIVERSITY June 1999
All rights resewed. This work may not be reproduced in whole or in part, by photocopy or other means, without permission of the author. National Library Bibliothèque nationale !*I of Canada du Canada Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395. rue Wdlington ûttawaON K1AW OnawaON KlAW Canada Canada
The author has granted a non- L'auteur a accordé une licence non exclusive Licence aiiowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, loan, distribute or sel1 reproduire, prêter, distribuer ou copies of this thesis in microfonn, vendre des copies de cette thèse sous paper or electronic formats. la forme de microfiche/nlm, de reproduction sur papier ou sur format électronique.
The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neitber the droit d'auteur qui protège cette thèse. thesis nor substantial extracts fkom it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. This thesis investigates an Upper Yukon River Klondike Gold Rush site known as Canyon City. Canyon City and other sites like it were integral. but peripheral to the Klondike Gold Rush. This study attempts to integrate the oral. archiva1 and archaeoiogical data to provide insights into the day-to-day life of the resident population at Canyon City during the Klondike Gold Rush,
1897-1900.
Over the course of four years, six localities within the former settlement were extensively tested. These include: the West Tent locality, the Canyon
Hotel and Saloon, the NWMP Barracks, the Canyon and White Horse Rapids
Tramway Co. building, the East Cabin and the Machine/Blacksmith Shop. A total of 17,395 artifacts and 156 faunal specimens were collected during the investigations.
lt is argued that the wrnpany structured the inhabitants' day-to-day lives within this single industry tom. This structuring is evident in the settlement characteristics such as the structured settlement layout, the lack of duplication of services, the predorninately male population and the dependence of the residents on the Company for sustenance. Although based on limited data, the settlement also appears ta have been organized socially-the workers and the ovmers. Furtherrnore, the cultural remains and architectural data suggest the general nature of the site was expedient and utilitarian, which not only reflects the logistical problems faced by Klondike era settlements but also the economic motivation of the townsite's owners. ... 111 The abandonment behavior present at Canyon City appears to have
largely been determined by company interests. Except for the East Cabin, wtiich burned with much of its contents still intact, the site undervuent planned.
permanent abandonment.
Since the investigations carried out at Canyon City wre the first of its
kind on the Upper Yukon River, it is unknown whether or not Canyon City can
be considered a typical settlement in this area. Work is needed at similar settlements along the Upper Yukon to better illuminate their role in the
Klondike Gold Rush and their significance in the development of the Yukon. This thesis was made possible by the generous support of the Yukon
Heritage Branch, Government of Yukon. The Canyon City Archaeology
Project was a joint endeavor of the Yukon Heritage Branch, Department of
Tourism, and the Kwanlin Dün First Nation with the support of the MacBrîde
Museum and Yukon Conservation Society. I would like to thank Jeff Hunston.
Director of Yukon Heritage Branch, for giving me the opportunity to work at the site over a period of four years. I greatly appreciate the continued support from Heritage Branch Archaeologists Ruth Gotthardt and Greg Hare. who provided me with insights, inspiration and edited parts of this thesis several times. I would also like to extend my gratitude to the rest of the staff of
Heritage Branch for their assistance. Thanks for the rnap Brent.
I am also grateful to my senior supervisor. Dr. David Burley, for his patience and rigorous editing. I would also like to thank Dr. Jack Nance and
Dr. Olga Klimko for their comments on my thesis project.
1 am grateful for the financial support for the Department of
Archaeology, SFU, for awarding me a Graduate Fellowship and a Teaching
Assistant position. Further financial support was provided by the Yukon
Heritage Branch, the Canada Employrnent Challenge Program. the Yukon
Government Student Training and Employrnent Program, the Yukon
Foundation, and DIAND Northern Scientific Training Program.
I extend thanks to John Hatch, John Scott and Lawrence Cyr for sharing their knowledge about the old days. Donna Hagen and Sweeney
v Scurvey, from the Kwanlin Dun First Nation provided the understanding of the traditional land use of the site through their oral history work. Valuable contributions wre made by elders Mrs. May Hume, Mrs. Julie Joe. Mrs. Lucy
Wren, Mrs. Mary James, Jimmy G. Smith, Edwin Scurvey, EMrdGordon.
Mrs. Virginia Vallevand, John Suits, Louie Smith, Rose Charlie and Ronald
Bill.
The project owes its success to the hard work of the field crew and site interpreters. These include: Ty Heffner, vvho also did the faunal work. Megan
Williams and Sarah Berquist (Heritage Branch STEP students); Azalea Joe,
Henry Taylor, Corey Pope, John Yaklin: Marilee Smarch, Charlie O'Brien. and
Michael Smith of Kwanlin Dün; Own Williams, Jacob Jirousek, Hillary
Walkley, Sara Neilsen, Sara Bryce, Greg Kubica. Georgina Nicioux.
Wilmonica VanBibber, Loic Markley and Hannah Hickling of the Yukon
Conservation Society; Andrea Hoyt, Rachel Pugh and Christie Colx of the
MacBride Museum.
I wouid also like to acknowiedge the Yukon Underwater Divers
-ociationJs efforts. Doug Davidge. and Perry Diamond conducted the underwater survey at Canyon City.
Thanks also go out to the support and assistance of the Yukon
Archives, MacBride Museum; Chief Lena Johns, Chef Joe Jack, Pat Joe.
Patty Ann Finlay and Jackie Shortie of the Kwanlin Dun First Nation. David
Neufeld of Parks Canada, Flo Whyard and Helene Dobrowolsky openiy shared their knowiedge and resources with me on Canyon City. Mrs. Whyard provided the NWMP log book for Canyon City dated 1899-1900.
An important and productive aspect of the project was the enthusiasm of the many volunteers who participated in the Canyon City excavations.
Thanks also to the thousands of visitors who took the time to visit Canyon
City.
Finally, I owe the wrnpletion of this thesis to my family and closest friends. My wife Myra's encouragement, support and patience over the years kept me going and sane. Afthough unknowingly my children, Melissa and
Nolan, provided me with much inspiration. And of course, thanks to Jim
Slater and Ruth Whitney.
vii TABLE OF CONTENTS
APPROVAL II .-. ABSTRACT III
ACKNOWLEDGEMENTS v
TABLE OF CONTENTS viii
LIST OF TABLES X
LlST OF FIGURES
CHAPTER ONE: INTRODUCTION
CHAPTER WO: CANYON CITY IN CONTEXT SITE LOCATIONAND GEOGRAPHICCONTEXT HISTORICAL OVERVIONOF CANYONCITY GOLDRUSH HISTORY- 1800-1 899/1900 Pre-Gold Rush Era Gold Rush Era Canyon City PoST GOLDRUSH ERA - 1900 - PRESENT
CHAPTER THREE: FIELDWORK, RESEARCH AND ANALYSIS METHODOLOGY PREVIOUS WORKAT CANYON CITY ARCHAEOL~GICAL FIELDMETHOOOLOGY DOCUMENTARYRESEARCH METHODOLOGY ARTIFACT ANALYSE MEHTODOLOGY
CHAPTER FOUR: HlSTORlC FEATURES AT CANYON ClTV CANYON CITY TOWNSITE WEST TENTLOCALITV: FEATURES CANYON HOTELAND SALOON: FEATURES NORTH-WESTMOUNTED POLICE BARRACKS: FEATURES CANYONAND WHITE HORSERAPIDS COMPANY TRAMWAYOFFICE: FEATURES THE EASTCABIN: FEATURES THE MACHINE/~LACKSMITHSHOP: FEATURES SUMMARY CHAPTER 5: CANYON CITY HlSTORlC ARTIFACT ASSEMBiAGE WEST TENTLOCALITY: ARTIFACTASSEMBLAGE Summary CANYONHOTEL AND SALOON:ARTIFACT ASSEMBLAGE Summary NORTH-WESTMOUNTEO POLICE BARRACKSARTIFACT ASSEMBLAGE Summary CANYONAND WHITE HORSE RAPIDS TRAMWAYOFFICE: ARTIFACT ASSEMBLAGE Summary EASTCABI N: ARTIFACTASSEMBLAGE Summary MACHINE/BLACKSMITHSHOP: ARTIFACTASSEMBLAGE Summary
CHAPTER 6: DISCUSSION AND CONCLUSIONS SITE CHARACTERISTICS SETLEMENT SUBSISTENCE ABANDONMENT CONCLUSION
REFERENCES ClTED Table 1. Table showing the historic features tested during the archaeological investigations at Canyon City including number of units, area excavated and approximate sampie size. Table 2. Functional groups used in the dassification of the Canyon City Historic Artifact assemblage and their associated artifact types. Table 3. Table of features identified at Canyon City and accompanying identifications. Table 4. Table showing artifact counts and each locality's relative contribution to the histonc artifact assemblage from Canyon City. Table 5. West Tent locality artifact functional groups and counts. Table 6. Faunal rernains coliected from the West Tent locality. Table 7. Canyon Hotel and Saloon artifact functional groups and counts. Table 8. NWMP Barracks artifact functional groups and counts. Table 9. Canyon and White Horse Rapids Tramway Office artifact functional groups and counts. Table 10. East Cabin artifad functional groups and counts. Table 11. East Cabin faunal remains. Table 12. MachineIBlacksrnith Shop artifact functional groups and counts. Figure 1. Map showing location of Canyon City in the Yukon. Figure 2. Map showing the major routes to the Klondike Gold Fields. Figure 3. Archival photograph of Miles Canyon, ca. 1899. Figure 4. Archival photograph of a tramcart on the tramway. Figure 5. Archival photograph of Canyon City Townsite. Figure 6. Archival photograph showing River steamers and freight on Canyon City Dock Figure 7. Map of the Canyon City townsite showing the location of the historic features documented during the investigations. Figure 8. Archival photograph of Canyon City showing the West Tent locality. Figure 9. Schematic diagram of the features documented at the West Tent Locality. Figure 10. Archival photograph of the Canyon Hotel and Saloon ca. 1899. Figure 11. Schematic diagram of the features documented at the Canyon Hotel and Saloon. Figure 12. Archival photograph of Canyon City townsite show-ng the NWMP Post and Storehouse in middle of the photograph. Figure 13. Schematic diagram of the features docurnented at the NWMP Barracks. Figure 14. Archival photograph of the remains of the NVVMP Barracks at Canyon City 1912. Figure 15. Archival photograph of original Tramway Office at Canyon City ca. 1897-1898. Figure 16. Archival photograph of completed Tramway Office at Canyon City ca. 1898. Figure 17. Schematic diagram of the features docurnented at the Canyon and White Horse Rapids Tramway Co. Building. Figure 18. Archival photograph of the Tramway Office, metal basin sits below the widow next to the second door to from the left. Figure 19. Schematic diagram of the features documented at the East Cabin. Figure 20. A) schematic diagram of the Machine/Blacksmith locality showing estimated perimeter based on depressions obsewed; 6)diagram of the composite feature located in the west portion of the MachineIBlacksmith shop; C) diagram of the composite feature docurnented in Unit 10 located in the east portion of the structure. Figure 21. Graph of the frequency of functional groups within the overall Canyon City historic assemblage. Figure 22. Graph of the frequency of functional groups wi-thin the West Tent Locality historic assemblage. Figure 23. Graph of the frequency of fundional groups within the Canyon Hotel and Saloon historic assemblage. 90 Figure 24. Graph of the frequency of functional groups within the North- West Mount Police Barracks historic assemblage. 97 Figure 25. Graph of the frequency of fundional groups within the Canyon and White Horse Rapids Tramway Office historic assemblage. 103 Figure 26. Graph of the frequency of fundional groups within the East Cabin historic assemblage. 110 Figure 27. Graph of the frequency of functional groups within the Machine/Blacksmith Shop historic assemblage. 118 . Figure 28. Archival photograph of tramway crew ca. 1899. 126 Figure 29. Archival photograph of tramway crew, of note is the wmen seated behind Norman Macaulay the man with the cane ca. 1899 (Yukon Archives 132
xii INTRODUCTION Pierre Berton has noted that the Klondike Gold Rush was one of the best documented adventures of the last 150 years. With the availability of cameras, film, newspapers, books, dianes and recorded interviews, the historian's problem really becornes one of where to focus and what to chmse (Davidson 1996). Archaeological studies conœming the Klondike Gold Rush have typically been centred on a few principal locations such as Skagway. the
Chilkoot Trail and Dawson City. In the Yukon, Dawson City and selected sites along the Chilkoot Trail have been the focus of the majority of historical and archaeological research so far camed out (Blee 1991 ; Bradford 1989; Burley
1985; Burley and Ross 1979a. 1979b; Cooper 1998; Hems and Nieuwhof
1994; Mini 1978; Murray and Hamilton 1986; Spude et al. 1993). Peripheral gold rush transportation settlements along the Yukon River, however, were essential to the success of the gold rush and these settlements ensured the relatively srnooth flow of the mass of stampeders who came from al1 over the world to the Klondike gold fields (Bennett 1978). The stories of these argonauts are familiar and well documented by historians; even romanticized by the likes of author Jack London and poet Robert Service. However, historians and anthropologists have largely ignored the peripheral settlements along the Yukon River and their resident populations.
With the centennial of the Klondike Gold Rush in 1998, Canyon City, located 10km upriver from Whitehorse was selected by the Department of
Tourism, Government of Yukon, to be developed as a histonc interpretive site.
In support of this program, archaeological investigations were conducted here over four field seasons between 1994 and 1997. The main objective of the
project was to gather architectural data for the possible reconstruction of
former buildings. Within this framework, the site was to be developed as an
interpretive museum displaying Canyon City and Klondike gold nish history.
As a result of the Canyon City study, I have had the opportunity to
examine this settlement and its role in the Klondike gold rush from an
archaeological perspective. The settlement was one of several peripheral
transportation oriented communities flourishing on the route to Dawson City
between 1897 and 1900. Aithough my overall goals are structured and limited
by the nature of the Yukon Tourism project. the data can also be used to
address other questions about the nature of life and economy at the site. This
study further contributes to our understanding of gold rush events on the
Upper Yukon River.
Canyon City and other settlements like it were important to the success
of the Klondike gold rush. Unfortunately, as prirnarily single purpose sites,
most were mentioned only in passing by the gold seekers. The typical
stampeder probably viewed towns Iike Canyon City much like today's highway
driver views a gas station-a necessity for continuing the journey but not worthy of detailed description. If Canyon City were studied solely through the documentary record, therefore, there would be little to no data on the day-to- day life of the occupants; at best, there would be simple lists of names, dates and critical events. As a general objective, my thesis attempts to gain further understanding of the nature of settlement within this peripheral gold nish site. Historical Archaeology has a unique position within the social sciences since it is "capable of gaining sirnultaneuus access to the past through multiple, independent categories of evidence" (Schuyler 1977 cf. Deagan
1988: 8). These categories. the database of Historical Archaeology, include
"the spoken word, the written word, observed behavior and preserved behavior" (Deagan 1982; Schuyler 1977 cited in Deagan 1988:8). To gain an accurate picture of what life was like at Canyon City, this thesis integrates the
'spoken word'-oral history, the 'written word'4ocumentary evidence including archival photographs, and the 'preserved behaviof-the archaeological record.
Central to this thesis is the premise that Canyon City was a single purpose company settlement, as indicated by the archival record. It is hypothesized that the company stnrctured the inhabitants' day-to-day lives within the single industry town. By both providing and regulating services. the
Canyon and White Horse Rapids Tramway Company dictated the construction and layout of facilities and workers' accommodation within the settlement and even the demographics and subsistence base of the resident population.
Finally. since the site was company owned, the abandonment behaviour should be uniform and consistent throughout.
It is argued here that the layout and buildings of Canyon City were specifically structured to meet the needs of the corporate owners in support of the tramway operation. Furthemore, it is expected that commercial senrices within the site were limited by the curporate structure. In addition. it is argued that the population was principally itinerate, non-local labourers hired by the Company as freight hustlen and for the operation of the tramway and related facilities. Women and children may have been present in small numbers, but their presence was peripheral to the organkation and functioning of the townsite. Therefore, few indicators of the presence of women and children such as toys, women's and children's clothing, toiletries and jewelry should be represented in the assemblage (Blee 1991; Spude and Scott 1993).
The social structure within the settlement is expected to be two tiered.
As with any company there are both owner/manager(s) and workers.
Differences between these groups should be seen archaeologically in several different areas. It is argued that owner/manager(s) likely lived in the log buildings and the workers in the tents. Because of the presurned difference in wealth between the two groups, the contents associated with domestic structures should refiect these differences. This should be evident in the archaeological record through the cornparison of utilitarian dominated assemblages versus assemblages with non-utilitarian artifacts such as ceramics. furthemore, social status should also be manifested in the faunal assemblage, more specifically the differences in meat cuts and types, between the two groups (Shulz and Gust 1983).
Archival data suggest the company supplied workers at the site with room and board. Therefore, it is expected that these occupants were heavily dependant on imported goods. Archaeological correlates for this include a high visibility of imported goods such as tinned goods and a low frequency of local faunal remains. One variable that must be taken into account is the possibility of the Company or individuals trading, likely with local First Nations for country foods.
Finally, the abandonment behavior present at Canyon City is predicted to have been consistent and unifom throughout the site. It is argued that the
Company's interests dictated abandonment behavior. There are no surface remains of the former buildings at the site suggesting al1 buildings were dismantled and building materials removed from the site in a systematic fashion. A low frequency of usable artifads remaining at the site would further indicate planned abandonment behavior (Schiffer 1989; Stevenson 1982a;
1982b).
In the following chapters an archaeological perspective of Canyon City is presented. Chapter Two reviews the historical setting of Canyon City, briefly discussing Yukon settlement types, Yukon history and the site itself in terrns of the historical record. Chapter Three provides the methodological frarnework and the research design that guided the four years of archaeological investigations. Chapters Four and Five provide the results of the excavations at the six Canyon City localities under study, detailing the site's layout, architectural data and recovered artifact assemblage. The final chapter synthesizes the results and addresses the questions posed in this introduction. -CHAPTER 2-
CANYON CITY IN CONTEXT
SITE LOCATIONAND GEOGRAPtIlC CONTUCT
The historic town site of Canyon City (Borden Number: JdUr-5) is
located in southwestern Yukon approximately 2km up river from Miles Canyon
and 1Okm south of Whitehorse (Latitude: 60°39'30"N; Longitude:
134O59'55'W; Elevation: 637 m a.s.1.) (Figure 1). The site is strategically situated on the east bank of the Yukon River beside the first large river eddy before entering Miles Canyon. At this point the river runs east ta west. The former settlement sits on an old north-south inclined river terrace approxirnately 1.Sm above today's river level. It is bordered to the north and to the east by a yet higher ancient river terrace, approximately 1Om above today's river level. Towards the western end of the town site the terrain rises, slightly marked with occasional outcrops of basalt.
Canyon City was originally founded as a tramway service around Miles
Canyon and the White Horse Rapids. The 15m vertical basalt walls of the
Canyon create a bottleneck of the river for about 1km. Approximately 800m down-river from the end of Miles Canyon the White Horse Rapids occur.
Canyon City falls within the Lake Laberge Ecoregion. Most of the terrain in this ecoregion lies between 600 and 1,500m a.s.1. Rolling hills dissected by plateaus characterise the topography of this area (Oswalt and
Senyk 1977). Vegetation consists of open white spnice forests in older stands on terraces and plateaus with lodgepole pine occupying burn areas ALASKA
Figure 1. Map showing location of Canyon City (JdUr-5) in the Yukon. (1crn=lOOkrn). (Oswalt and Senyk 1977). Black spruce dominates wetter sites and balsam poplar is more common on recent alluvium. Paper birch is scattered throughout the region wncentrating in cooler aspects. The understory vegetation is characterised by masses with a wide variety of shrubs and forbs
(Oswalt and Senyk 1977). The irnmediate area in and around Canyon City was heavily logged during the gold rush occupation of the site. The resulting vegetation that arose and characterises the site today is open stands of pine, spruce and trernbling aspen with an understory of grasses, fireweed, bearberry, soapberry , strawberry, roses, juniper and various herbs.
Canyon City lies within the traditional territory of the Kwanlin Dün First
Nation-Tagish Kwan. The Kwanlin Dun are Southem Tutchone and Tagish speakers and are grouped within the larger Athapaskan linguistic family
(McClelIan 1975).
The site is located on the upper Yukon River, which served as a core resource area for the First Nation people of southwestern Yukon. Both the
Tagish and the Southern Tutchone peoples used the river and the land in the immediate vicinity of Canyon City (McClellan 1975: 31). McClellan (1975: 34) reports that the traditional fish camps, one referred to as T' aqadji, were used by the Tagish Kwan in the general area of Miles Canyon. From her description of the locations, Canyon City was likely used as a fish camp.
Traditional native land use in the vicinity of Canyon City was documented by Donna Hagen, a mmmunity researcher hired by the Kwanlin
Dün First Nation in 1994. Aîthough there were very few personal recollections about the traditional use of the site, Kwanlin Dün First Nation elders did tell of
extensive traditional use in the general area during the 19th and 20th
centuries (tiare and Hammer 1995). Traditional activities included fishing,
hunting, trapping and recreation. In addition, some of the elders recalled their
rnothers and fathers talking of Canyon City, some of whom may have worked
for the owner of the tramway, Norman Macaulay, at Canyon City during the
gold rush. Elder May Hume talked of the trail, which is still visible and cuts
through Canyon City, as a traditional trail that begins at Manh Lake and
continues to Lake Laberge, extending over a distance of 100km in length.
Recently, Mark Lindsay, a researcher for the Kwanlin Dün First Nation,
uncovered an obituary from 1906 in the Whitehorse Star. The obituary was for a Mrs. John who resided at Canyon City and was predeceased by her
husband the year before.
HISTORICALOVERWEW OF CANYON CITY
GOLDRUSH HISTORY - 1800-1 8W'I900 Non-native settlement of the Yukon cm be divided into three periods.
The slow movement of Europeans into the region between the mid-1800s and
1896 characterises the first period. This early development of the Yukon produced much of the background geological and transportation related information on the region, which some researchers argue made the Klondike gold rush of 1898 possible (Bennett 1978; Newell 1987). The second period begins with the Klondike gold rush marking the start of rapid settlement of the region by non-natives and finishes at the end of the gold rush. The third period is the post-gold rush era (poçt-1900) marked by a stabilisation of the
population and the expansion of settlement out of the Yukon River corridor
(Coates and Momson 1988, 1989; Duerden 1980.).
Pre-Gold Rush Era
The first European incursion into the Yukon was for the quest for furs
(Bennett 1967; Webb 1993; Wright 1976). The Hudson's Bay Company
established itsetf in the interior region of the present day Yukon Territory
between 1842 and 1852. The presence of gold in the Yukon was not
unknown to the Hudson's Bay traders. Robert Campbell, a Hudson's Bay
Company explorer and trader. noticed traces of gold at Fort Selkirk but not in
enough quantities to merit exploitation (Friesen 1978). In the vicinity of Fort
Yukon, at the confluence of the Yukon and Porcupine Rivers, there were
rumours of vast quantities of gold discovered in the immediate region by an
early missionary narned Reverend MacDonald (Friesen 1978: 13). Concerned
only with increasing profits from the fur trade. clerks and traders at the several
posts established along the Yukon River dismissed these grandiose rumours
of gold. These rumours, however, did not fall on deaf ears. 80th successful and unsuccessful minen from the California and British Columbia gold fields started to filter into the region to prospect after the 1870s (Clark 1942; lnnis
1936; Powers 1974; Wright 1976). Between 1880 and 1895 the non-native
population within the Yukon consisted mainly of a small core of miners estimated to number approximately 1,000 (Gates 1994; Wright 1976) and it was not until the 1890s that govemmental agencies were established in the interior.
Transportation routes into the region. however, were already established by the time governments became involved in documenting the
North. Campbell, an Early Hudson's Bay Company trader, first travened the
Liard River route that was quickly considered too difficult to be feasible. Once the Yukon River was further explored by HBC traders and it was discovered to provide a continuous water route from Fort Yukon to Fort Selkirk, the HBC and the fint prospectors such as McQuesten. Mayo and Harper reached the north via the Mackenzie, Peel and Porcupine Rivers (Bennett 1978; Wright 1976).
This route became the al1 Canadian Route advertised by the growing city of
Edmonton in the 1898 rush. After 1867 the preferred route into the Yukon interior was the Yukon River via St. Michael's-an al1 water route (Bennett
1978). By 1869 the stemwheeler Yukon was making regular trips up the
Yukon River into the interior carrying supplies and news for the prospectors focated in the region (Bennett 1978: 17).
It was not until 1880 that the coastal Chilkat lndians granted entrance into the Yukon interior via the mountain passes from coastal Alaska to non- natives. All miners arriving in the Yukon over the steep mountain passes during this time had to confront Miles Canyon and White Horse Rapids.
Some travellers. who shot the Canyon and Rapids and lived to tell about it, made this stretch of the Yukon River legendary in their published accounts.
W.B. Haskell (1898: 121) wntes: the water was boiling through it [Miles Canyon] at such a terrific speed that it ridged up in the center, while along the perpendicular banks it wtiirled in huge eddies which had a very threatening look. The clouds of spray gave the water level a snowy appearance ...We pushed off, and in two minutes my heart failed me, and I would have given al1 the gold I ever expected to get in these regions had I staid out.
Most prospectors portaged around these bamers following the traditional native portage trails. Dawson (1889) in 1887 and Schwatka (1893) in 1883 document wind-lasses and log roll-ways set up along this section of the river by prospectors to aid in the portage of their boats and cargo.
Dawson (1889), who was travelling upriver, reported three portages between the start of the White Horse Rapids and the end of Miles Canyon.
The first portage was on the west bank of the Yukon River enabling travellers to avoid the White Horse Rapids. Approximately 1.2km above the Rapids was another short portage of 40m. The third portage, which avoids Miles Canyon, was located on the east bank of the river before entering the Canyon. A windlass was set up at this point to haul boats for portage up the steep basalt banks.
Gold Rush Era
The discovery of gold on August 16th, 1896 by Skookum Jim, Tagish
Charlie. Kate Cannack and George Carmack on Rabbit Creek (later named
Bonanza Creek), a small tributary of the Klondike River, triggered the Klondike gold rush. At Forty Mile, which was a log city located at a previous gold strike
40 miles up river from where Dawson City would soon flourish, George
Carmack's arriva1 to register his claim started an intemal gold rush within Alaska and the Yukon to the Dawson region. In the spring of 1897 with the docking of the steamer Excelsior in San Francisco with $750,000 worth of
Klondike gold, the Klondike goM rush was on (Hunt 1974; Newell 1987).
There were four main routes to the Klondike (Figure 2). First was the al1 Canadian Route that involvecl a 2,700 mile journey starting in Edmonton,
Alberta, up the Athabasca River to the Mackenzie River and then on to the
Peel and Rat Rivers until the Yukon River was reached. The second trait travelled to the gold fields was the al1 water route that took about a month.
Steamers were boarded in San Francisco, Seattle, and Vancouver- The steamers made there way up the Northwest Coast around the Aleutian Islands to Saint Michael's where the Yukon River drains into the Bearing Strait. Once at Saint Michael's, the stampeders boarded river steamers that plowed up river to Dawson City. A third and less popular route was the Ashcroft and
Stikine trails; crossing overland through British Columbia to the headwaters of the Yukon River.
The fourth and most popular route was the cheapest way to get to the gold fields (Neufeld and Norns 1996). A total of 35,000 to 40,000 gold seekers used this route during the gold rush (Neufeld and Noms 1996.). This is the Trail of 98. Stampeders boarded the many steamers headed to Dyea or
Skagway. settlements situated at the end of Alaska's Lynn Canal and at the foot of the Chilkoot and White Pass Trails respectively. After off loading, the would-be miners traversed the treacherous passes and made their way to
Bennett City on Bennett Lake. The geographically short joumey to Bennett Figure 2. Map showing the major routes to the Klondike gold fields. could be completed within three weeks, if one had enough money to hire packers, or it could take up to three months if one had to transport their goods themselves (Neufeld and Noms 1996). Once at Bennett, the gold seekers buitt boats or boarded steamers. The joumey continued through the system of southem lakes to Marsh Lake, drained by the Yukon River. Once on the
Yukon River the only bamers facing the Dawson City bound Stampeders were
Miles Canyon and the White Horse Rapids (Figure 3).
As a result of the mars migration to the Klondike, former settlements boomed and new towns were created. Prior to the gold rush, Dyea, which was once a traditional hunting and fishing camp, had one store owned by
Healy and Wilson (Neufeld and Noms 1996: 55). Shortly after news of a gold strike reached the rest of the world, Dyea boasted 200 businesses including
40 saloons, and a variety of theatres, concert halls and bordellos (Neufeld and
Norris 1996). The population expanded from approximately 1,000 during the summer of 1897 to 8,000 in the spring of 1898 (Neufeld and Norris 1996).
The same phenomenon occurred in Skagway.
There were two major types of settlement that developed along the
Klondike trails. The largest and most permanent were the service and distribution centres (Duerden 1980: 16). Skagway, Dyea, Saint Michael's and
Dawson City were settlements of this type. These towns were metropolitan, consisting of a variety of services for the stampeders and functioned as the hubs for distribution and transportation. For most stampeders. these types of settlements were jump stations rather than end destination spots. The second type and the most numerous along the trail and the Yukon River were settlements serving lines of communication (Duerden 1900). These settlements usually cansisted of a small resident population and had a single purpose. Lindeman and Bennett cities. the most varied of the lot, functioned mainly as short terni boat building camps. Both towns had a high population and provided a variety of services, minoring those of service and distribution centres. However, this mirroring was a resutt of the freezing up of the lakes and rives, therefore ceasing easy transportation and creating a back up of stampeders, who were waling for the spring thaw. More typical examples of the second type of settlement were situated along the Yukon River en-route to
Dawson City. North-West Mounted Police (NWMP) Posts were stationed approximately every 25 miles along the Southern Lakes system and Yukon
River and served not only to enforce Canadian law and collect customs from the stampeden but also to aid and infonn them dunng their joumey (Berton
1972; Dobrowolsky 1995). A number of wood cutting camps, which serviced the steamers on the Yukon, were also present. Other settlernents of this type offered places of rest and lodging as well as services directly related to transportation. Canyon City was one of the latter.
Canyon City
The historic settlement of Canyon City arose as a direct result of the
Klondike gold rush. f he community was built around the start of a wooden tramway operation that enabled the multitudes of Klondike bound stampeders to circumvent the treacherous Miles Canyon and White Home Rapids. The primary purpose of Canyon Cdy was to facilitate the tramway operation for the
waves of incoming gold seekers. In June of 1900 the White Pass and Yukon
Railway Company compkted its rail line linking the coastal town of Skagway.
Alaska. with the newly emerging town of Whitehorse. This rail link made trail
and transportation services between Skagway and Whitehorse nonessential.
Like most of the small transportation Settlements along the upper Yukon River
after the amval of the railway. Canyon City's importance to Yukon
transportation waned. It was quickly abandonad in 1900-1901.
In the fall 1897. in anticipation of the great flood of stampeders to come
and presumably with a farniliarity of the region, Norman Macaulay established
a roadhouse on the east bank of the Yukon River at the beginning of the
traditional portage trail around Miles Canyon and the White Horse Rapids
(Hare and Hammer 1995: 15). Macaulay constnicted a 8.8km long wooden
tramway that started in the vicinity of his roadhouse and ended at the foot of
the White Horse Rapids.
The tramway was a simple but effective means of transporting freight
(Figure 4). Archival photographs show the tramway fine consisted of peeled
logs placed parallel to each other approximately lm apart with cross ties every
1 -5-2m. The tramcars were constructed of wooden timbers with a slightly
concave wooden flat bed for freight. A running board extended from one side
of the tramcar for the operator where there was a vertical lever that probably functioned as a braking mechanism. The horsedrawn tramcars were pulled along the trarnline on concave, cast-iron wheels. Apparently the 8.8km tramway took 18 men and 23 horses 21 days to build (Horback 1975; Price
1898).
From al1 accounts it is ckar that Macaulay's tramway company-the
Canyon and White Horse Rapids Tramway Company-was in operation by the spring of 1898, ready for the mass of stampeders wintering over at Bennett
City. While Macaulay built and completed his tramline, John Hepbum struggled to finish construction of a second tramline located on the west bank
1km upriver from Macaulay's, which he began in 1897. It was not cornpleted until after November of 1898 (Wood 1898a: 38) and probably was not open for business until the spring of 1899. Hepbum's company-the Miles Canyon and Lewes River Tramway Inc.-was short-lived. In July of 1899 Macaulay bought out Hepbum for $60,000 (The Klondike Nugget 1899). It is uncertain whether Macaulay kept Hepbum's tramway open for any period of time after the pirrchase.
The freight levy for the tramway established by Macaulay was three to five cents per pound for goods, with a flat fee of $25 for boats. Otherwise. boats could be piloted through the Canyon and White Horse Rapids for a fee of between five and twenty dollars. At first people risked the trip through the canyon and rapids on their own. According to Rickard (1909) numerous people lost their outfits and 200 lost their lives. Although Rickard no doubt exaggerated the number of deaths, safety concerns led Superintendent Sam
Steefe of the NWMP to decree:
There are many of your countrymen who have said that the Mounted Police make the laws as they go along, and I am going to do so now for your own good. therefore the directions that I give shall be camed out strictly, and they are these :- Corporal Dixon. who thoroughly understands this work, will be in charge here and be responsible to me for the proper management of the passage of the canyon and White Horse Rapids. No women or children will be taken in the boats. If they are strong enough to come to the Klondyke they cm walk the 5 miles of grassy bank to the foot of the White Horse Rapids. and there is no danger for them here. No boat will be permitted to go through the canyon until the corporal is satisfied that it has sufficient free board to enabie it to ride the waves of safety. No boat will be allowed to pass with human beings in it unless it is steered by competent men, and of that the corporal will be judge. There will be a number of pilots sekcted, whose names will be on the roll in the Mounted Police barracks here, and when a crew needs a man to steer them through the canyon to the foot of the rapids, pilots will be taken in tum from that Iist. In the event of the men not king able to pay, the corporal will be permitted to arrange that the boats are run without charge. The rate now charged, 5 dollars, for each boat, seems reasonable (Steele 1915: 31 1- 312).
This declaration guaranteed the success of Macaulay's operation. All travellers were required to report to the NWMP stationed at Canyon City for approval to pass through Miles Canyon and the White Horse Rapids.
Violation of this regulation resulted in a fine of $100 (Steele 1898: 31). Those who were not deterred and attempted to bypass Canyon City without checking in were, at times. forced in by NWMP gun shots fired over the bows of their scows (Steele 1898: 31).
Early in 1898. a small Company settlement owned by Norman
Macaulay arose at the start of his tramway (Figure 5). The NWMP records and stampeder jounals refer to this settlernent and surrounding area by several ternis: Miles Canyon, White Horse, Miles Canyon and White Horse
Rapids and White HomRapids. At first, they used these names Figure 5. Archival photograph of Canyon City townsite. East cabin shown on the right margin of photograph; a woman stands in front (Parks Canada, Sinclair photo.) interchangeably for the area at the end of the White Horse Rapids, which is the present location of the city of Whitehorse, and the town site of Canyon
City. This adds much confusion for researchers in detennining exactly where events took place. Rickard (1909) appears to have been the first to cal1 the former settlement Canyon CQ.
Archival records for the development of this settlement are scarce, makîng it difficult to detemine when the buildings at Canyon City were established. We do know, however, that the settlement eventually consisted of at least 12 log and several tent structures (Figure 5). Archival photographs show that the first section of the Canyon and White Horse Rapids Tramway
Co. Office was constnicted by the spring of 1898, with the second section added shortty thereafter. A section of the Canyon Hotel and Saloon was also completed at this time. Construction of the NWMP Post was begun after the
!jthof July, 1898, and appears to have kencornpleted by the 30"
Septernber, 1898 (Steele 1898a: 21 ; Wood 1898b: 3). The NWMP storehouse was built in the winter of 1898ll89Q(Wood l898a). From archival photographs it appears the stable, machine/blacksmith shop, dock, a residential cabin and other miscellaneous buildings were al1 completed by the summer of 1899. By September of 1899 a telegraph line was connected and a telegraph office was established within the NWMP Post, with a second telegraph office in place at the foot of the White Horse Rapids (Wood 1899).
At the height of operation the tramline was processing between 70 and
100 tons of freight per day (Walley n.d.) (Figure 6). At maximum operation the tramway operated day and night with 23 horses (Prie 1898). The numbers of freight hustlers working for the Canyon and White Home Rapids Tramway
Co.. estimated from the archival photographs, was between 30 and 35. They were paid $4.50 a day plus board with fifty cents extra for overtime (Hitchcock
1899: 431).
The tramline transported goods from stampeders' handrnade scows and barges. as well as large, paddkwheel river steamers. Between its origin and Dawson City, the Yukon River was divided into two sections for steamer transportation. The section upriver from Miles Canyon was designated the
Upper Yukon and the section down river hmthe Whitehorse Rapids was referred to as the Lower Yukon. By the summer of 1898 there were full sized river steamers on both the Upper and Lower Yukon. The steamers on the
Upper Yukon rân between Bennett City and Canyon City and the steamers on the Lower Yukon ran between Dawson City and the terminus of Macaulay's tramline.
The success of Macaulay's operation prompted an official govemment survey of the tawn site. Macaulay also proposed to transform the tramline into a narrow gauge railway in 1899 (Dobrowolsky 1990: 1). It is uncertain whether these events indicate Macaulay's belief in the permanence of his tramway business or whether this was a means ta increase the value of his holdings. Macaulay had known that the completion of the Skagway-
Whitehorse railway was fast approaching and that its completion wouM adversely affect his business. Therefore, his possession of the east and west bank flght-of-ways along the Yukon River (Hepbum's trarnline and his own) and his daims for upcoming commercial improvements likely increased his property value.
In 1898 the White Pass and Yukon Corporation began construction of its railway linking Skagway to the foot of the White Horse Rapids opposite the terminus of Macaulay's tramway. In August of 1899 in a bid to gain a monopoly on the transportation business and a right-of-way for their rail line,
The White Pass and Yukon Corporation hired C.E. Peabody of the Alaska
Steamship Company to buy out Macaulay (Minter 1987: 320). The price paid to Macaulay is reported to have been $185,000 (Minter 1987). The Canadian
Development Company. a subsidiary of the Alaska Steamship Company, ran the trarnline for the next ten months, after which, it ceased to operate. This gave the railway a complete monopoly on the transportation of freight into the upper Yukon.
On the 8" of June of 1900, the rail link between Caribou Crossing and the new town site of Whitehorse was completed. Whitehorse quickfy grew as
Canyon City was abandoned over the next two years.
POSTGOLO RUSH €RA - 1QOO-PRESENT The post-gold rush period in the Yukon is marked by a significant drop and stabilization of the population as well as non-native settlement in areas outside of the Yukon River corridor (Bennett 1978; Burton 1972; Coates 1985;
Duerden 1980; Webb 1993). By 1899 the huge human migration into the
Yukon ceased and other gold rushes such as those in Nome, Alaska, and Atlin. British Columbia, drew a large part of the transient population out of the
Yukon. Other areas within the Yukon such as the Kluane-White River region and the Mayo-Keno area saw their own mini-rushes (Bennett 1978: Duerden
1980; Johnston n-d.; Stevenson 1980, 1989a, 1989b; Webb 1993).
Whitehorse located at the terminus of the rail-line became the distribution hub for the Yukon. Aithough Dawson City's population declined. it still played an important role politically and served the now commercialized placer gold industry. After 1900. roads began to be buiit further opening the Yukon to new settlernents (Bennett 1978; Durden 1980).
The establishment of the railway over the White Pass, a permanent link between the Yukon and the outside world, and large river steamers on the
Yukon River made many of the small settiements on the Upper Yukon non- essential. On the trails to the Klondike. towns that once thrived were being abandoned while others located along the rail-line were ensured of their permanency, such as Caribou Crossing. Neufeld and Noms (1996) state that by 1899 only those that could not afford to pay train fare hiked the Chilkoot.
Lindeman and various camps along the Chilkoot trail were abandoned by the fall of 1899 (Neufeld and Norris 1996). Bennett still survived, albeit in a smaller capacity, with other short lived settlements established during the first years of the rail transport.
Very little is known about Canyon City in the post-gold rush pend once the railway was established and the tramlines ceased to operate. Because
Canyon City was a one-industry town, most if not al1 of its occupants were linked to the Company. Tharefore, once the tramlines stopped operating and the new town of Whitehorse began to flourish, Canyon City was quickly abandoned.
At present there is no archival documentation recording the abandonment at Canyon City. Ercept for the lingering presence of the
NWMP, who stayed at the town site at ieast periodically until October of 1901 , the majority of the settlement was probably abandoned after completion of the rail link (Dobrowolsky 1990). Occupation of the site did occur after 1900, for a short period, by Mr. and Mrs. John as earlier described. At present, the extent and location of their occupation is unknown.
The resuits of the 1994 and 1995 Canyon City Project suggest that most of the log structures in the sefflement were systematically dismantled and removed. Accounts from both native elden and long-time Yukon residents report that no structures were present at Canyon City during the
1920s except for the wooden dock (Hare and Hammer 1995: 21).
Photographs of Canyon City dating to 1912 cleariy show the NWMP structure still standing. in a partial state of dismantlement. but no others.
Another photograph of the same date and photographer depicts a small log cabin.
The materials salvaged from the buildings during abandonment were more than likely reused in the emerging town of Whitehorse. This was so for other short-tived upper Yukon River settlements such as Bennett Crty (Ingram and Dobrowolsky 1994: 7). Aîthough not confirmed, it has been suggested that segments of the Canyon Hotel and Saloon may have ken used to '
construct part of the Closeleigh Hotel and Saloon (later narned the Pioneer
Hotel) located in Whitehorse (John Hatch personal communication 1996).
NWMP Corporal Dixon. who was dosely associated with Canyon City. buiît
the hotel on Front Street early in 1900 (Ingram and Dobrowolsky 1994: 54).
Horback (1976) reports that in 1900 Macaulay also brriît a hotel in Whitehorse.
It is possible that Macaulay and Dixon's venture were one and the same and that the logs from Macaulay's Canyon Hotel and Saloon were used in the
Closeleigh Hotel and Saloon's construction.
The 1994, 1995 and 1996 archaeological investigations at Canyon City revealed limited evidence of postgoid rush occupation of the site dating up to the present. A series of wire mesh and wooden pans may post-date the occupation at Canyon City and rnay be related to the John's occupation of the site. This type of structure is reminiscent of fox fam cages that were popular in the area during the early 1900s (Hare and Hammer 1995). As yet, no one recalls there ever king a fox fam located at Canyon City so the structure's use and time of use are still uncertain. More recent beer and soft drink cans, some with pull-tabs, are scattered intermittently throughout the town site.
These recent artifacts are the remains left by the people who used and still use the site as a recreation area, Above the former town site on a flat portion of the eastem terrace is an area with scattered structural remains and artifacts
(rnetal containers). The artifacts (sanitary tin cans) definitely post-date the gold rush occupation of Canyon City, suggesting the occupation of this area does as well. XHAPTER 3-
FIELDWORK, RESEARCH AND ANALYSlS METHOOOLOGY
PREWOUSWORK AT CANYONCm
Archival research on Canyon City has kensporadically canied out
over several yean by area and local historians such as Dobrowolsky (1990),
Knutson (1990), Scuwey (1995), Sawatsky, Whyard and Horback (1976).
Over the last four years 1, in cosrdination with the Heritage Branch staff. have
been assernbling and reviewing these resources as well as pursuing further
archival sources conœming Canyon City.
Canyon City was firot recorded as an archaeological site by Hunston
during a reconnaissance of the area in the earfy 1980s (Gotthardt 1994). As
described, he documented not only an abundance of gold rush era cultural
material, but also recarded an eadier prehistoric camponent documented by
lithic flakes eroding out of a trail that passes through the terrace.
The next archaeologist to report on Canyon City was Easton (1987)
during an underwater suwey of the Upper Yukon River. His survey team
located the remains of the Canyon City dock, its foundations and part of the tramway track that were submerged as a result of the damming of the Yukon
River in 1958. Easton (1987) collected an oar rest and horseshoe.
In 1993, Gotthardt (1994) carried out preliminary archaeological testing at Canyon City. A total of two days was spent identifying and locating the site's main historic features as well as determining the extent of the prehistoric occupation. A baseline parallel to and 24m from the present-day Yukon River's edge was established. Small shovel tests were conducted along this line every 10m within the former town site and every 20m for 500m West of the site. The shovel test resuits suggested that the main concentration of prehistoric occupation was at the eastern end of the former town site. Nine additional shovel tests were placed within this area to further determine the extent of the occupation.
Artifacts recovered from Gotthardt's investigations were of both prehistoric and historic origin. Of the prehistoric assemblage. 14 lithic specimens were recovered both above and below the White River ash stratum, which is dated at 1,260 î 50 BP (Clague et al. 1995). The collection included a chert end scraper and a broken biface fragment. The historic assemblage cansisted of five nails, one screw, two flat ferrous strips and a metal buckle with a small fragment of leather stilt attached. The site's principal historic structures were located by companng gold rush era archival photographs of the settlement with foundation berm outlines. The 1993 investigations provided exploratory data from which a more indepth field investigation could be planned.
ARCHAEOLOGICALFIELD METHODOLOGY
The archaeological field methodology implemented at Canyon City refiected the goals, to a certain degree, of the sponsoring institution, Heritage
Branch, Government of Yukon. These goals were to recover as much detailed structural information as possible on the Canyon and Whitehorse Rapids
Tramway ûffice (year l), the NoraiWest Mounted Police Barracks (year 2), and the Canyon Hotel and Saloon (years 3 and 4). In addition, Heritage
Branch sought a sample of material culture rernains associated with these
buildings ta permit the interpretation of activities carried out at these localities.
The data gathered during the investigations and synthesized in the following
months were to be used for restoration, interpretation and display.
A second but important component of the project was its public orientation. High school and post secondary students were hired to assist in the excavations from the Kwanlin Dün First Nation, MacBride Museum and the
Yukon Conservation Society. Each year students assisted in excavations as well as provided on site interpretation services to the many visitors to Canyon
City, which averaged 2,000 per year. With such high public interest and because of the project's proximity to Whitehorse, several volunteers assisted in the excavations as well.
Third, although the project's applied objectives structured research, they also presented an opportunity to test additional localities in order to gain a broader understanding of the townsite. The additional localities were as follows: the West Tent LocalRy (years 2 and 4). the 'East Cabin" (years 2 and
3), the MachinelBlacksmith Shop (years 3 and 4) and other features as time allowed. Because the objectives of the Heritage Bnnch and the research were essentially in Iine with each other, testing of these lacalities was carried out under the same research design as discussed below and in Chapter One.
Based on initial site reconnaissance prior to the 1994 feld season. it was possible to locate al1 of the major structures shown in archival photographs. Reconnaissance. however, indicated that there were many . more archaeological features present than first anticipated and that other features were probably hidden by understory vegetation. Historic features refer to al1 of the anthropogenic landscape changes, architectural remnants or artifact clusters present at Canyon City.
To start the fieldwork, the townsite was cleared of vegetation after which a systematic surface survey was camed out. Crew mernbem walked parallel transects along the Yukon River bank approximately Sm apart.
Transects were then walked north to the terra- marking the end of the former townsite. Features encountered in each individual's path or field of vision were flagged. Transects were repeated until the entire site was traversed.
Field staff from the Historic Sites, Government of Yukon, subsequently prepared a site feature map (see Figure 7, p.40). Also at this time a 200m x
160m grid was established over the townsite with NOWO located on the southeastern terrace bordering the site. As well, an east-west running base line was staked every 10m on the N60 line.
Excavation units were tied into the east-west base Iine and the larger site grid. Units on the grid were assigned north and west identifier's, such as
N60W135, while others not on the gnd were assigned unit numbers particular to the structure or feature with which they were associated.
Excavation units were placed judgementally to identîfy and locate building features according to the characteristics of the feature or structure being tested. Units within each structure were placed in strategic positions based on the potential to yield structural data. For exampie. the corners of a
structure as well as fioor joist and orner depressions within it were targeted to
retrieve dimensional and structural data. No less than a five percent area was
excavated from each major structure tested; some localities such as the "East
Cabinn and the Machine1Blacksmith Shop had up to 50 percent their area excavated (Table 1). This latge variance regarding area sampled was a resuît of the size of the structure being tested, wlh the lowest excavated area fractions corning from the Canyon Hotel and Saloon and the Canyon and
Whitehorse Rapids Tramway Onice, which each enclosed an area of at least
All units were excavated by trowel with back-dirt screened through 3rnm
utility mesh. Excavation proceeded stratïgraphically with arbitrary 5crn levels excavated within each stratum until sterile deposits were encauntered. Three- dimensionai provenience (depth below surface. depth below White River Ash.
north coordinates and west mordinates) was recarded on ly for prehistoric
Feature Description No. of Un- Total AM Percent Excavafed Am Excavated Excavated CC71 Tramway Office 35 252mL 35mL 14% N.W.M.P. Barracks 22 Canyon Hotel and Saloon 25 West Tent Locality 23 MachinelBlacksmith Shop 16 'East Cabinw 25 N.W.M.P. Storehouse 1 Building outline? 1 Tent box frame ouUine 1 Tent box frame outline 3 'East Cabin" Privy 1 N.W.M.P. Privy 1 Privy? 2 Refuse dump 1 Table 1. Table showing the historic features tested during the archaeological investigations at Canyon City including number of units and area excavated. artifacts as per terrns of referenœ. For provenienœ recording of historic artifacts, 1xlm excavation units were divided into four quadrants based on cardinal directions. In situ historic artifacts and those found while screening were placed in level bags according to the stratum and quadrant from which t hey came. Significant artifacts, which were deterrnined according to their context, and al1 historic features were photographed in situ and recorded on floor plan sheets. On completion of excavation. unit wall profiles were drawn and photographed.
D~~UMENTARYRESEARCH M~OOOLOGY
Archival research was principatly carried out at the Yukon Archives located on the Yukon College Campus in Whitehorse, Yukon. The MacBride
Museum, Parks Canada as well as personal 'old timer" libraries were surveyed. Sources reviewed include newspaper accounts, starnpeder's journals, North-West Mounted Police annual and monthly reports and Gold
Rush era photographs of the former townsite. The photographie record of
Canyon City is quite extensive; however. primary documents regarding the fomer townsite are scarce. The North-West Mounted Police monthly and annual reports were the most valuable for the purposes of this thesis.
Published literature related to the Klondike Gold Rush and Canyon City was also surveyed. Sources used include older volumes such as Rickard (1909).
Ogilvie (191 3) and Steele (191 5) as well as more recent volumes such as
Burton (1972) and Gates (1994). These secondary sources provide a genenl overview history of the Gold Rush and only mentioned Canyon City in passing. Oral histories related to the former townsite were sought in each year of the excavations. Oldtimers and First Nation elders visited the site and were asked what they remembered. The results, unfortunately were disappointing.
The length of time since the occupation of Canyon City, two generations, coupled with its brief fluorescence are probably wntributing factors.
ART~FACTANALYSIS METHODOCOGY
A large histonc artifact collection (N=lï,396) was generated during the four years of feld investigations. The classificatory scheme used to type the artifacts is largely based on Sprague (1981). Hardesty (1988). Klimko and
Hodges (1993), and Blee (1991). The adapted classification attempts to place the artifact within the cultural context of use, based on its meaninghnction for the participants of that system (Sprague 1981: 252). Furthemore, the typology devised here attempted to be comparable to that of Bk's,(Blee
1991; Blee and Scott 1992), which is commonly used for assemblages in
Skagway.
In order to identify the function of several of the unknown structures and comment on the general activities camed out at Canyon City in different localities the material culture was organized into the following functional groups: household (domestic); personal rnanagementlclothing; leisure; transportation; communication; hunting/subsistence; construction/structural; workinglindustrial; and, miscellaneous. Al1 identifiable artifacts were typed according to their primary function and placed within these larger functional Functional Group Type of Artifacts Household Artifacts related to the household as a whole such as tableware, glassware, food storage. furniture. stovelheaten and lighting. Personal Management Artifacts related to the individual including ciothing, medicinal, personal hygiene, and adornment. Leisure Artifacts related to liquor and tobacca wnsumption. and gaming. Transpcrtation Artifacts such as bridle apparatus and wagon and cart parts. Communication Artifacts including glass insulators, telegraph equipment and telegraph wire Hunting/Subsistenœ Artifacts related to amis and ammunition and hunting such as fish hooks. Construction/Stmctural Artifacts related to the construction and structural components of buildings such as nails, screws, grommets, strapping, hooks. bolts, and window glass. Arîifacts such as industrial tools and those artifacts related to machining or blacksmithing activities. Miscellaneous Those artifacts that do not fall into the above categories or those where the function is not identifid. Table 2. Functional groups used in the classification of the Canyon Clty histonc artifact assemblage and their associated artifact types. categories (Table 2). Unidentad artifacts were classed acwrding to their material of manufacture.
The goal of this classification scheme is to identify, as best is possible, the different activities camed out at the site as well as within the different structures at Canyon City. Within a one Company owned senlement buildings would have had a primary purpose, although it is likely that other activities would have been carried out in them as well. Thus, by looking at the different functional groups and their frequency. 1 should be possibk to identify the primary purpose of each structure in addition to secondary activities carried out there.
It is recognized that not al1 artifacts recovered represent the function they were initially designed for (Bk1991: 84; Sprague 1981). even without artifact alteration. As well, some artifacts cross into two or more functional groups. In such instances the nature of artifacts and their relation to function is problematic; however, by using this type of classification one artifact is not the sole deteminant of an adMior function. Rather, the functional groups from a specific locality are viewed together and in relation to each other according to their relative frequency thus reducing the effect of anomalies such as the unaiteieâ. secondary reuse of artifacts.
Faunal remains were identifii and catalogued separately. Remains were identifîed to species where possible using the resources at the
Archaeology Branch, Government of Yukon. The number of identified specimens present (NISP) were tabulated for each tocalrty where faunal remains were present. 4HAPTER 6
HISTOR1C FUTURES AT CANYON CITY
A total of 98 surface historic features were identified and mapped during the 1994 transect suwey of the Canyon City townsite (Figure 7; Table
3). The historic features include 21 possible structures or tent frame outlines, eight large and two small metal container middens, two boffle rniddens, 21 individual or groups of depressions, remains of a wooden sled, several small artifact scatters, tramline earthen beds, and concentrations of bailing wire. Not al1 of the features shown on the site feature map (Figure 7) were tested. Table
3 summarizes the features identified and those that were tested during the four years of investigations at the townsite.
This chapter describes the townsite layout and surnmarizes structural remains for each of the localities. Six major lacalities were the focus of detailed excavations and include the West Tent Locality (CC2-CC7), The
Canyon Hotel and Saloon (CC1 l), the Machine/Blacksmith Shop (CC59), the
NWMP Barracks (CC39), the Canyon and Whitehorse Rapids Tramway Office
(CC71), and the 'East Cabinn (CC86).
CANYONCm TOWNSITE
The boundaries of the historic townsite of Canyon City run approximately 190m east to west by 120m north to south encornpassing an area of approximately 22,800m2 . The majority of features within the townsite are related to the Gold Rush occupation of the site. The pen enclosure (CC90) and CC8 and CC41, however, likely postdate the gold rush occupation.
Feature Description Festuli) ûescn'ption CC1 Sderof teather footwear Cc50 Small scatter of boffle glass Tent outiine' Scattered woode n feature Tent outline* Bailing wire Tent outiine' Borrow pit Scatter of footwear Large structural timbers ûepression' Benn outlirie and depression ûepression* Depressions' Benn outline' Two scatters of bailing wire Bonow pit Notch logs - corral corner Depression Machine shop outline* Canyon Hotel and Sabon' Scatter of femus objects BottWcan midden Stable outline Floor remains Baling wire Depression Depression Structure outline Stabk enüy? Small metal container scatter Brome? Notched building bg ûepression Rectangular depression' Scatter of cans and bottles (recent) Bottie midden Baling wire in depression Can midden (M-2)' Tramway spur line Can midden (M-3)' Borrow pits Can midden (M4)' CBWR Tramway Office' Can midden (M-5)' Depression Galvanized chicken wire Depression Can midden (M4)' ûepression Can midden (M-1)' Tent ouüine* Can midden (M-7)' Depression Circle of cobbles Depression Fencing? Depression (tent outline) Wooden sled Chicken wire with wood Can midden (M-8)' Gahranized chicken wire Depression Small metal container scatter Cluster of cobbles and FCR Chicken wire Cluster of cut logs and cans Chicken wire Small cluster of metal containers Structure outline Tent outline* Structure outline Fire pit, circle of cobbles 'East Cabin"' Teiegraph Wire Bonow pit NWMP Poste Can rnidden (M-IO)* NWMP Storehouse* 'East Cabin's" privy* Structure outline with burnt floor Pen enclosure NWMP Privy?' Square depression Depression Telegraph wire De pression Wooden feature De pression Plank Structure outline? Fenced enclosure assoc. CC86 Wooden feature - Tramcar? Deep square depression Depression Tramway bed CC49 ~kdenfeature - Tramcar? CC98 ~ramw&line spur 'fsmanrn Table 3. Table of features identified at Canyon City and accompanying identifications. The pen enclosure, located at the base of the eastem tenace, is one of the largest features observed at the site. It is rectanguiar in shape consisting of 10 pens and was fenced with chicken wire. The feature appears to be the remains of a fur fami which likely dates to around the First Woild War, 1914-
1920, when fur faning boomed al1 over the Yukon (McCandiess 1985).
Figure 5, page 22, is an archival photo of Canyon City and the only one so far encountered that provides a good view of the eastem tenace of the site. This photograph was likely taken in 1899 and shows the 'East Cabinnon its right margin. The two small tents nom of the cabin are likely CC84 and CC85.
Beyond these two wall tents, however, there is no indication of a significant feature such as the pen enclosure. The pens were likely constnicted much later than the gold rush occupation of Canyon City and may be associated with the post-Gold Rush materials atop of the eastem tenace.
CC8, a rectangular benoutline located behind the West Tent Localdy, and CC41, a buiiding outline, do not show up on any of the 24 historic photographs of Canyon Cdy. The photographs span the years 1897 to 1900.
Within CC41 several sanitary tin can fragments were observed along with metal flashing and other bumt building debns suggesting a post-ûold Rush date (at least post 1904) for the occupation of this structure. No artifacts were on the surface in and around CC8 and the one unit excavated within this structure yielded no time sensitive artifacts. Its absence from the photographs is the only evidence suggesting it is not contemporaneous with the Gold Rush. Those features that clearly are associated with the Gold Rush occupation appear to be quite structured in their placement. The townsite was organized around the tramline. The major structures dosest to the Yukon
River ail face south towards the river and the tramline and are oriented in two blocks running east to west. The first bîock consists of the East Cabin and the
Canyon and White HomRapids Tramway Co. onice, located in the eastem portion of the site. The second bW,about 20m north of block one, consists of the tents CC84, CC85, CC75 the NWMP Barracks and storehouse, the
Canyon Hotel and Saloon and the West Tent locality. The
MachinelBlacksmith Shop and the stables are located further north and set opposite frorn a spur of the tramline that runs south to north.
The largest middens of metal containers are wncentrated in the northwestern portion of the townsite. Pathways are numerous here and likely post date the gold rush occupation of the site. Their creation may have contributed to the further division of the larger middens sinœ there is no physical or photographic evidence of any structures in this vicinity. The large concentration of metal containers in this area suggests an intentional use of the area for refuse disposal.
The town layout into blocks pre-dates the official townsite survey carried out between September 26M and October 3d 1899 by Dominion Land
Surveyor Paul Dumais (Department of lnterior 9899). There are no natural features, such as manh or undulating topography, within boundaries of the site that would necessitate such a stnictured town plan. It is evident that buildings were placed according to some structureci plan and it follows that
Macaulay's Company imposeci this stniduring.
WEST TENT LOCALIW:FUTURES
The West Tent Locality was sampkd during the 1995, 1996 and 1997 archaeological investigations. It is located approximately 8m west of the
Canyon Hotel and Saloon. The locality wvers an area of approximately 17m east to west by 10m north to south. A number of Gold Rush era photographs depict three wall tents Iined up side by side from east to west (Figure 8).
Today, only the benn outlines of the three still exist (features CC2. CC3 and
CC4). lrnmediately behind the ben outlines are four shallow rectangular depressions, three of which uvere tested. The goals of the excavations at this locality were to first gather structural data about the tents including their dimensions, and second to infer what types of activities were associated with their use.
A total of 23 units were excavated at the West Tent Locality equaling
26m2. Placement of the unita was judgmental and rested solely on the presence of bems or depressions. As a resuît of the investigations several intact structural features were documented, three middens identified and 1,635 artifacts and 27 faunal specimens collected.
Intact structural features between 5 and 10cm below surface relate to the box frames buiît as bases for each of three tents (Figure 9). Two square bem outlines defined the location of the two western most tents and an L- year. They appear to be relying on the name association with the Canadian Standards Association (CSA) for credibility in the industry. Societe Generale de Surveillance (SGS) has been severely disrupted in the past year by the move to KPMG QRI of two of their senior staff in Western Canada. SGS stüi operates in a central Canadian rnindset. since its operationai activities are coordinated through their Toronto head oflice location. SGS will only issue about 5% of the new certificates in BC in 1999. and do not appear to have made sigdcant inroads into the local market in the past year. Approximately 20% of their clients switched to KPMG QRI with the recent move of their senior staff. Qualitv Control Bureau (QCB) will probable emerge as KPMG QRI's main competitor in the western Canadian market in 1999. They tend to target the srnail and emerging business community. and have developed excellent links to the BC Institute of Technology (BCIT). with whom they offer joint training courses in quality management systems. QCB wili issue approximately 25% of the new registrations in BC this year. Intertek Testina Services (ITS) is emerging as a serious competitor in the local market following the hire of a new manager for their ISO 9000 registration practice in late 1997. ITS tends to be on the bidder's list for the majority of contracts let competitively in BC. Many of the local QMS consultants are being eniisted by ITS as subcontractors. ITS also actively seeks to recruit current KPMG QRI ciients with proposais to transfer their registrations. However. they only gained about 5% of the new registrations in BC in 1998. Figure 9. Schematic diagram of the features docurnented at the West Tent Locality. shaped benn marked the east and norai walls of the eastem most tent. The bems' likely functioned to insulate Me Roor box frames from drafts. Sill legs were obsewed along the inner base of the benn outlines as were floor planks.
The westein most tent wntaineâ north-south running floor joists and the floor planks were placed on these joists. Tents CC3 and CC4 did not have floor joists. Rather, the planks were set directly on top of the surface. The two western most tents had a circle of rocks associated with a circular depression.
The presence of these features is problematic. They may relate to later recreational use of this locality or they might have functioned as areas for stove placement.
Three corners were exposed for the box frame of CC2 and from these corners the tent appears to have been 2.7~2.1m (8ftlOin.xGPtl Oin.) suggesting that the wall tent used for this dwelling was a standard 7x9ft wall tent. This type of tent could fit two comfortably, but might have had up to four individuals if it was intended to be a sleeping area only. The northeast and southwest corners of the middle tent (CC3) indicate 1 was a larger tent measuring
3.3x3.8rn (10ftiOin.xl2ft6in.). Two corners were excavated within the eastern most tent box frame. Its dimensions were 2.5x2.8m (8ft2in.x9ft2in.), which are very close to that of CC2.
The tent box frames were simply constnicted. Saddle notching is present on al1 of the corners observed and the sill logs or planks were set on the ground with littk to no preparation, such as digging or filling for leveling.
Each of the box frames had floor planks. However, if the two rock circles in CC2 and CC3 are directly associated, then 1 is likely that the entire fiwr area of each tent was not planked. Rather, a portion of the area inside the tent may have beeo left uncovered for the placement of a stove or heater. The flwr planks were placed directly on the ground in CC3 and CC4 but were raised with the aid of sleeper sills and floor joists in CC2. The archival photographs of this area fumish no further information of the box frames or platioms because the canvas wall tenh were placed on the outside thereby covenng the frames.
CANYONHOTEL AND SALOON:FEATURES
Approximately 8rn directly east of the West Tent Locality are the remains of the former Canyon Hotel and Saloon (Roadhouse). All that exist today of the former log structure is a rectangular berm outline surmunding north-south running floor joist depressions and a east-west running centre sill depression. This outline is well defined along segments of the north. east and south walls but fades into the natural terrain towards the east along the south and east walls. Gold Rush era photographs of the Roadhouse show that the log structure consisted of three cribs, or conjoined sections. with four entranceways along its south wall (Figure IO). The number of entrance ways likely reflect the intemal division of the different functions of the Roadhouse such as a restaurant or saloon and sleeping quarters. It is likely that the number of doomays do not exadly represent the nurnber of different fundions or interna1 divisions. In addition, a cold cellar existed 10m south of the
structure that was later used for refuse deposition. This latter feature was not
tested.
Investigations were camed out at the Roadhouse locality during the
1994, 1995, 1996 and 1997 field seasons. The goals were to gather structural
information and detemine and dinerentiate actnrity areas within the building.
A total of 25 units were excavated of which three were 50xSOcm, one
was 2xlm, another was 2x2m and the rernainder were Ixlm in size (Figure
11). The total surface area excavated was 26.75m2,which is a 12.5 percent
area sample of the entire structure. Like al1 the localifes investigated at
Canyon Cw the placement of excavation units was judgmental, guided by the
goals of the testing. Of the 25 units, 14 were placed along the peflmeter of the
building to provide a cross section of the berrn and determine the placement of
sill logs. The remaining eight units were excavated within the berm outline to test the inside of the former structure. Several intact structural features were documented allowing for ar! accurate estimate of the dimensions of the
building at 31.5x7m (103ft6in.x23ft).
Foundation and flaoring features of the Roadhouse were documented in 16 of 25 units excavated. All wood remains were in a severe state of decay. and none showed evidence of buming.
Wall sill logs were present within eight units of those placed on the building perimeter. The north sill remains in Unit 1 terrninated within the southeast quadrant of the unit and it is likely that the point of temination represents the northwestem corner. The southeast corner of the structure was documented within Unit 22 with the presence of the east and south sill log remains. These two units were wed for detennining the dimensions of the structure. A large basalt cobbîe and a complete evaporated mild can were observed to the norai and just abng the side of the north sill log in Unit 3. The cobble rnay have fundioned to stabilize the north sill Mile the metal container could have fundioned as a cap or a sleeve for a post.
A total of 11 units intersedecl Roor joist and centre sill depressions.
However. intact remains of floor joists in line with their respective depressions were observed in only four units. Two of the units had temains that were probable floor joists but appear to have been displaced. Their displacernent likely occurred during the dismantling of the structure. A stack of logs ninning north to south was docurnented in Units 10, 11 and 12 located in the southeast portion of the Roadhouse. The stack consists of nine logs in width and three to four logs deep. The logs, which may have been flwr joists. appear to have been stockpiled in this location after the Canyon Hotel and
Saloon was dismantled.
Remains of a doorstep or walkway were present in Units 19, 20. and
RHTP 1,2, and 3. These consist of four t 5x9cm (6ftx3ftll2in.) planks placed parallel to each other and secured by wire nails to three small poles, 4cm in diameter. Its presence indicates the placement of a doonuay at this location
(Figure 1 1). Oriented by archival photographs, Units 17, 18 and 21 were placed along the south perimeter whete other entranœways were believed present. Steps or walkways were not encountered within these units. The stratigraphie profiles of the units excavated mthin the Canyon Hotel and Saloon indicate that the construction of the building disturbed much of the surface deposits. Units Wh distu- stratigraphy are located along the north portion of the structure and the disturbance œases once north of the noRh sill log. It appears, therefore, that the surfa- was îeveled to prepare a building surface, a technique used for the construction of most other buildings at
Canyon City.
The relatively low frequency of wood remains suggests the building was systematically dismantled during or after the abandonment of Canyon City.
None of the features or depressions present illustrate intemal room divisions.
NORTH-WESTMOUNTED POLICE BARRACKS: FEATURES
Further west of the Canyon Hotel and Saloon are two former buildings that make up the North-West Mounted Police locality. CC39 has been identified through archival records as the Barracks and CC40 as the NWMP
Storehouse. Archival photographs only show the front of both structures
(Figure 12). 60th buildings were log and had sod roofs. The barracks has one entranceway located along the middle of the south wall, and two windows on each side of the doorway. The entianceway to the storehouse is located near the southeastem corner of the structure and a small window is evident towards the southwestern portion of the south wall. The investigations at this locality focused almost exclusively on the Barracks.
A total of 22 units were excavated in and around the NWMP Barracks.
Three of the 22 units were placed outside of the benoutline in order to test Figure 12. Archival photograph of Canyon City townsite showing the NWMP Post and Storehouse in middle of the photograph (Yukon Archives Beatty Coll.). the areas surrounding the former building. The berm outline encloses an area of 91m2 and the 19 units excavated within this outline constitute a 21 percent area sample. Units were placed in the corners of the berm outline to verify the actual location and dimensions of the building. Units were also placed along the sides of the bem in order to expose the four outside wall sill logs. Surfaœ depressions and artifact dusters largely detennined the placement of units within the structure.
Woaden remains in the NWPBarracks were in various states of decay, but al1 can be assessed as poor to very poor. Evidence of a previous fire was present in rnost of the features observed. Structural features in the north half of the structure displayed extensive charring whereas remains in the southern portion showed less intensive burning. It is clear that the building was dismantled before the fire occurred. That is, if the entire building bumed then it would be expected that deposits within the inner surface of the structure woufd have contained evidence of collapsed logs and extensive charcoal.
This was not the case, al1 of the structural rernains appear to be situated in their primary contexts.
The excavations revealed significant information on the constructiori of the former building foundations. Figure 13 is a plan of the foundations from the 1994 and 1995 investigations. Structural remains include floor joists. sill logs. a floor plank and two hewn centre sill logs running in the middle of the structure traverse to the north-south running floor joists. Chaned remains and the oxidation of the deposits where logs once sat enabled an estimation of the
diameter of the logs used. Flaor joists had an average diameter of
approximately 14cm and sill logs an average diameter of 30-35m.
The four inside corners of the building were documented with the
exposure of traces of the outer sill logs. The building was approximately 45ft
or 13.2111in length and 23R or 7m in width. These dimensions are relatively
consistent to those provided by Superintendent Wood (1898c: 3) in his
October 1898 report to the comptroller in Dawson City. Specifically, he states that the barracks construded at Canyon City were 4ûfWOft. Furthemore, he
mentions the building contained a men's quarters, mess room and kitchen.
Clear room divisions are not readily apparent from the archaeological remains.
The floor joists are generally spaced 1m apart frorn each other except to the
left and right of the doonivay. The close positioning of these joists may
represent intemal partitions. Archival photographs of the NWMP Barracks show two stovepipes. One pipe exits the roof in the east haH of the structure and the other on the west half. The presence of two stoves within a building of this size indicates there were at least two different rooms within the structure.
Post-hole depressions are distributed throughout the building. These depressions likely represent the location of former poles for interior roof support such as supports for ridgepoles and purlins. These posts do not appear to signify inner partitions of the structure. One post-hole depression in the northwest corner of the building probably represents an intemal log corner post. Why no post-hole depressions were apparent in the other corners is uncertain. One possible explanation may be that vertical posts were present but were set directly atop the ground. therefore, no depression would have been left The two post-hole depressions along the centre sill likeîy represent vertical logs to support the roofs ridgepoie.
Evidence of the doomay was observeci in units N62W111 and
N62WllO. Seven wire-ârawn 6in. spikes were recovered on the top of the south sill. nree of the spikes were parallel to each other with their shanks extending into the east wall and two spikes laid parallel to each other extended into the square from the west wall while another spike extended into the west wall. The placement of six of the spikes suggest they held in a door frame. The distance between the heads of the spikes was 94cm or approximately 3ft1 in. This measurement is close to standard door frame width. Although uncertain, the south sill log appears to be thinned by notching thereby making the sill part of the door frarne as well.
The stratigraphic profiles of the excavated units are informative. Before laying the foundation logs, the area was leveled through excavation. A trench running the length of the building was excavated for both the south and centre sill logs. Once this was cornplete the sill logs were put in place and the floor joists then placed directly on the ground. The rest of the building was constructed on this foundation.
A photograph dated to 1912 cleariy depicts the NWMP Barracks still standing with no evidence of Cire (Figure 14). This photograph reveals two important details. First, the Barracks were not dismantled and moved during the abandonment of Canyon City and second the fire that chaned the logs of Figure 14. Archivai photograph of the remains of the NWMP Barracks at Canyon City 1912 (Mac Bride Museum Chadwell Coll.) the barracks must have occuned after f 912. The structure, however, must
have been salvaged kfore its buming since extensive charcoal was not
observed within the structure itseîf.
CANYONAND WHITE HORSERAPIDS COMPANYTRAMWAY OFFICE: FEATURES
The remains of the former Canyon and White Home Rapids Company
Tramway OfTice building are located 1Om east and 20m north of the NWMP
Storehouse. It is prominently located in the eastem portion of the site and
would have been one of the first buildings seen by the oncoming starnpeders.
This structure and the MachineBlacksmith Shap are the only two structures
d irectly related to Macaulay's tramway operation.
Archival photographs show that the building was consttucted in at least
two phases. The western most section (Figure 15). consisting of three cribs
and two entranœways, was buitt first, probably in late 1897. The eastern half
of the structure, consisting of two cribs and twa entranceways, was then
completed by the end of 1898 (Figure 16). The east and noRh berms and
several floor joist depressions are the only evidence today of the building. A
well-used footpath that runs along the southem perimeter of the site has
erased any traces of a south berni that may have existed.
The Tramway Office was the focus of archaeological investigations
during the 1994 field season. A total of 35 units (1xl m) were excavated. Few
intact wooden feature were present and those present were in an extreme
state of decay making a functional identification difficult. Features ranged in depth from 5-20cm below surface and some of the remains encountered did
show signs of previous buming. The majority of features consist of scattered
building debris such as small timben and fioor planks. Portions of the south.
north and west sill logs wen present (Figure 17). but only the northwestem
corner was confidently located. Using these data. the building is
approximately 36x7m (118~23ft) in sue, which makes it slightly larger than the
Canyon Hotel and Saloon.
Floor joist depressions, running noRh to south and spaced
approximately lm apart were observed on the surface only in the eastem half
of the structure. The lad< of joist depressions in the western haif suggests this
part of the building had a dirt floor. This may reflect the initial rush to ered the western section of the structure in order to get tramline business up and
running. The termination of Roor joists approximately in the middle of the structure is the only indication of an interna1 division from the archaeological remains observed. The stratigraphie profiles of the excavated units suggest that the ground was leveled in a northward direction in order to create a building surface. The deepest ground disturbance occurs along the north wall.
A unique feature was uncovered within units S6E4 and S7W24 located approximately 8m west of the southeastem corner. It appears to have been a wooden troughldrainage pipe that extended south from S6E24 into S7E24 with a southward incline towards the river. It is a hollow rectangular wood plank pipe 1Ox8cm that had kenplaœd in an excavated ditch. A circular depression along the north wall within SE24 started at the surface and continued down until connecting with the pipe's north terminus. This feature suggests that the Tramway office may have had some sort of rudimentary plumbing. One archival photograph of the Tramway office shows a large wooden or metal container, possibly a water storage tank, mounted on the building wall below a window (Figure 18). It appears to be situated almost directly above the location of this buried feature. From the location of the drain, it is likely that some type of basin. sink or bath tub was located immediately inside the south wall of the building. The pipe K cleafiy located in the eastern half of the structure, which was the latest addition to the Tramway Office.
The low frequency of intact in situ structural remains again indicates that the Canyon and White Horse Rapids Tramway Company office was systematically dismantled during or shortly after the abandonment of Canyon
City. The archival photographs indicate that the office was constructed in two phases and the archaeological record demonstrates an important differenœ between the two parts of the building. The later addition. although expedient, appears to have more planning and thought in its construction including a wood floor and a drainage pipe.
THE EASTCABIN: FEATURES
Approximately 30m further east of the Canyon and Whitehone Rapids
Tramway Office and perched on a platfonn along the base of the upper east terrace in the southeastern corner of the townsite are the subsurface remains of the East Cabin. Archaeological investigations at this locality were camed during the 1995 and 1996 feld seasons. At the start of the t 995 season it was decided to place volunteers in this area since, at that time, no structures were known to have been located there. Excavations soon revealed the charred remains of a former cabin. During the 1995 season a photograph of the Canyon City townsite was acquired from Parks Canada that included the cabin (see Figure 5). This photograph confimied that these remains were from a log cabin with a canvas tent extension, and that the cabin was contemporaneous with the gold rush occupation of the site.
The photograph was taken fram across the river on a steep bank overlooking Canyon City. This is a considerable distance and inhibits a detailed observation of the cabin. The photograph, however, shows the southem and western elevation of the structure and provides a general view of the cabin features that would have been impossible to gain solely from the archaeological record. The building was constructed of logs 4th a gabled roof. The entrance into the structure was from the south through the tent extension. The cabin had one window midway along its western wall and a stove pipe extended from the northwestern corner of the cabin's sod roof.
During the 1995 and 1996 investigations a total of 25 units (lxlm) were excavated at this locality providing a 75 percent area sample. The main motivation for such an extensive excavation was not related to the historic archaeology project. Rather, a prehistoric site was located beneath the cabin.
An Agate Basin-like point dated to 2,600 I60 B.P. (Beta 099458) was subsequently wllected (Hammer 1997: 40). All of the features documenteci within the East Cabin locality were ' heavily charred, presumably a result of the building king destroyed by fire.
All of the features encountered above the intact floor joists were fragmented and in extreme disanay, also indicating wall and roof collapse.
Logs placed paralkl to each other in the southern and northem excavations units are identified as floor joists andlor foundations remnants
(Figure 19). Planks were most likely secured to these joists in order to create a smooth floor surface. The absence of floor joists in two of the uni& located near the middle of the building suggests the presence of a shallow cellar. The outer dimensions for the cabin are infened by the outer limits of the logs.
Thus, the building can be estimated to be 5.2x4.9m (17x1 6ft).
During removal of floor joists in the northeastem units of the locale a door latch was recovered. In addition. the logs, some of which have been displaced within units N25W22 to W24, suggest the presence of some sort of door step or porch area. This entranceway would have kenlocated within the southern tent extension.
The stratigraphy of the cabin reveals that a dirt platfom was created on a 3 degree incline to create a level building surface. The overburden above the floor joists Iikely came from the sod roof of the cabin and tent extension.
This layer was riddled with fragmented burned wood and was rich in historic material culture. Displacea juiifii~grernainci
Figure 19. Schematic diagram of the features documented at the East Cabin. THE MACHINUBLACKSMITHSHOP: FEATURES
Test excavations in 1994 tentatively identified the Machine/Blacksrnith
Shop, located at the back of the townsite across frorn the stables. None of the
archival documents or oral data piovide accounts of such an activity occurring
at Canyon City. Further excavations in 1995,1996 and 7997 confimed the
building's function largely through the types of artifacts recovered. Although
neither documentary nor oral history data mention the presence of such a
building, gold rush era photographs do depict a log structure in the northeast
portion of the town (Figure 5). It is assumed this building is the one king reported upon here.
A total of 18m2 was excavated within the 6x6m perimeter of the
MachineIBlacksmith Shop. These excavations recovered a large sample of artifacts, second in abundance only to the East Cabin locality. As well, several features were documented.
The perimeter of the MachinefBlacksmith Shop is rnarked by four linear depressions that fom a square 6x6m (19ft6in.xl9ft6in.) in size. These depressions, probably resulting from sill logs, are taken to represent the exterior walls of the structure. No in situ sill logs were encountered during the excavations. Floor joists and floor joist depressions were also absent. A heavy concentration of artifacts indicates the structure likely had a dirt floor.
Like most other buildings at the site, it appears to have been dismantled during the abandonment of Canyon City. =-a~card h.noc - -.- -
Figure 20. A.) schematic diagrarn of the Machine/Blacksmith locality showing estimated perimeter based on depressions observed. B.) diagram of the composite feature located in the west portion of the Machine/Blacksmith shop. C.) diagram of the composite feature documented in Unit 10 located in the east portion of the structure. During the 1996 and 1997 excavations a large composite feature was exposed and dowmented. This is tentatively identifid as a forge. It was confined to an area of 1x3m located in the northwestern corner of the building
(Figure 20b). Consisting of wood and sheet rnetal, it is associated with a high frequency of bumed coal and slag. Intemiixed with the sheet metal were vanous artifacts including a solder bar, a horseshoe and various small reworked iron fragments. One of the unburned pieces is a milled fragment with a peg hinge. Charred and unbumed fragments were present as well.
A plank feature, also in the northwest comer runs north to south and is
170crn long and 55crn wide (5ft6in.xlftlOin.). It consists of two sections that are joined together by the western most plank. The planks that make up the feature range between 15-20cm (6-7.8in.)in with and 6Ocm (2ft) in length. At their northern Iimit al1 five planks are attached with nails to an underlying cross piece. The western most plank, wtiich has a latch at its northern end is 97cm
(3ftîin.) in length and is attached to two short cut logs before its southern terminus. Underlying these remains is a relatively thick dark stained floor deposit (34cm) littered with reworked metal pieces and nails. This suggests that the planks represent a collapsed intemal feature of the Machine Shop. most likely a workbench for working metal near the forge.
A second composite feature was located in the eastern half of the structure (Figure 20c). Three southwest to northeast ninning 5x15.2cm
(2x6in.) planks with three underlying cross pieces were directly associated with numerous nibber strapping fragments. This feature is reminiscent of a flat bed for a tramcar. Furthemore an east-west running 5xlOair (2x4in.) piece of lurnber that extends out of the west wall of Unit 10 with a large ferrous handle at its eastem terminus is similar to those oôsenred on the tramcars; possibly a break handle or hitch. These remains were recovered with a vafiety of unrelated artifacts, that taken together, probably represent a junk or parts pile.
Except for a post documented in Unit 1, the remaining features observed during the excavations appear ta be displaced structural remains of the Machine Shop. The post, which is 20cm in diameter and continues for
46cm below the dirt floor of the Shop, likeiy represents a support for the ridgepole of the structure. It appears that like the majority of the buildings at
Canyon City, the Machinel Blacksmith Shop was dismantled during or shortly after the abandonment of Canyon City. Summmv
Architectural data were document4 for each of the six localities investigated at Canyon City. These data allowed the penmeter of each structure to be detemiined. Furthemore, construction details, particularly building surface preparations, were documented.
Except for the West Tent Locality, al1 of the buildings investigated were of simple log cabin construction. Surface preparation. when present. was minimal for the Canyon Hotd and Saloon, the Canyon and White Horse
Rapids Tramway Co. Ontce and the Machine/Blacksmith Shop. The East
Cabin and the NWMP Barracks surfaœs appear to have undergone more effort. Trenches were excavated for the N\/VMP Barracks and an earthen platforni was constnicted for the East Cabin. Nevertheless, al1 of the architectural data collected suggests that the structures were expediently buitt without signifiant improvements added over time. -CHAPTER S
CANYON CITY HlSTORlC ARTIFACT ASSEMBLAGE
A total of 17,395 historic artifacts was recovered during the four years of excavations camed out at Canyon City. The overall goal of the artifact analysis has been to interpret activity patterns carried at the different localities.
Thus. artifact functional groups were developed. as discussed in Chapter 3, in order to achieve this purpose. The artifact assemblages from each of the main localities are the focus of this chapter, with each assemblage discussed separately under the assumption that the assemblage reflects, to a certain degree, the activities canied out at each respective locale.
A total of 14 localities as well as surface atiiacts and those recovered undenvater wntribute to the historic arofact assemblage. Table 4 lists the frequency of artifacts recovered from each locality, as well as their proportional contribution to the overall Canyon City assemblage. The six major locales, as examtned in the previous chapter, contribute most to the assemblage. Of particular note is the richness of the East Cabin assemblage as compared to the other five locales. This is explained by it being burned, as opposed to some other abandonment process. The comparable proportions represented by the other five locales Iikely reflect planned abandonment, an issue discussed in the final chapter.
Overall, the vast majority of artifacts collected at Canyon City appear to relate to the Gold Rush occupation of the site. Their relation to this occupation is determined , wherever possible, by their respective terminus post quem Locality Altlf.ct P~contofA- R.ktivo Conîffbuüon Rank Count Exc.v.t.d lo Assunblm East Cabin 61S8 57% 24.4% 1 NWMP Barracks 2413 21% 17.6% 2 MachineIBlacksmith Shop 3372 50% 15.6% 3 Canyon Hotel and Saloon 1846 12.5% 14.9% 4 Tramway Office 1661 14% 13.2% 5 West Tent Locality 1635 15% 12.8% 6 Surface and Undenivater 121 - 0.3% 7.5 CC89 1O1 7W/a 0.3% 7.5 CC75 22 8% 0.2% 8 CC36 93 11% 0.1% 9.5 NWMP Storehouse 12 >lOh O.l0h 9.5 Cc42 22 40% 0.1 % 9.5 CC56 16 20% 0.1% 9.5 CC8 2 >I% 0.02% 10 CC18 1 25% 0.01% 11 Total 17395 - lm - Table 4. Table showing artifact counts and each locality's relative contribution to the historic artifact assemblage ffom Canyon City.
(date after which) of manufacture. Of course this is problematic, since the teminus post quem of an artifact refers to the earliest date it was introduced to consumen, and artifacts may be used long after achial production. Wnh this in mind, however, very few artifacts have a teminus post quem dating after
1900, or the period in which site abandonment took place.
Eight functional groups represent the Canyon City historic artifact collection as well as a miscellaneous group that was created for unidentified items. The construction/ structural and household functional groups dominate the collection with the rest of the groups, except for the miscellaneous group. having frequencies lower than 10% (Figure 21). ------Figure 21. Graph of the frequency of functional groups within the overall Canyon City historic assemblage.
WEST TENTLOCALITV: ART IFACT ASSEMBLAGE
A primary research question regarding the West Tent locality was to detennine the function of these tents. Two main possibilities exist: 1) the tents were used as storage, possibly for the Canyon Hotel and Saloon due to their proximity to this establishment; or, 2) the tents were used as dwelling quarters for sorne of the residents of Canyon City. Excavations at this locality recovered a total of 1,635 historic artifacts and 27 faunal specimens resulting in a total of nine functional groups represented within the collection (Table 5,
Figure 22). The groups' frequencies and divenity seern indicative of domestic and personal FUNCTIONAL GROUP TOTAL FUNCnONAL GROUP TOTAL Houuthold Ceamics Botlklglassware Complete bottk Bottk fragments Medicinal fragments Jar fragments Lead foi1 Co* stopper Metal container Misœllaneous container Total
Personri MI~lOthi~ Buttons "Mother of Pearl" Glas Fenous Building hardware Clathing Strawing Unidentifieci textiles Fntcnings Wool coat Naiis Fwtwear Sciaiv Leather and rubber foohveor Grommet Heel tao Nut Eyeiet window gbss Personal hygiene rot.[ Mirror Shave crearn cap Noiungiindultrial Eye dropper Axe Fik Toîal r0t.l
Leisure Miscoilanoou8 Tobacco Btvekd g lass Toîal Unidentified fenous fragments r-I ARTIFACT T0 Table 5. West Tent locality artifact functional groups and counts. activities, thus suggesting that the tents were used as dwellings. The total assemblage. as discussed below. suggest the more permanent residents of
Canyon City used the dwellings rather than hotel guests.
Household related artifacts comprise 28 percent of the West Tent locality collection. A total of 15 ceramic sherds were recovered from Midden
1, a srnall concentration located north of the tents. All of the cerarnic sherds West lent AtW.ct Grmup. and Frequsncks
Figure 22. Graph of the frequency of functional groups within the West Tent Locality historic assemblage. are white glazed from tableware vessels. Two conjoinable base fragments provides a partial base mark of '. .. E Chinan probably representing Fine China or Made in China. Three partial container fragments made out of an unknown material, sirnilar to compressed paper, also represent tableware. One of the fragments is a base, body and rim fragment, and suggest the vesse1 was rectangular and deep. most likely representing a type of serving dish.
Bottle fragments are the most abundant artifact type within the household functional group. Bottle fragments were rewvered from al1 features tested at the West Tent locality. Colou~sindude arnber, dear. clear with green tint. clear with aqua-blue tint, ciear with manganese tint and green.
These fragments likely represent various activities including food and
80 condiment storage, liquor consurnption and medicinal. Contents of most of the fragments can only be speculated upon since glass colour is only loosely
associated with contents (Hardesty 1988). To type them as liquor or
condiment bottles solely on the basis of colour would be tentative at best. The
majority of bottle fragments collected do appear to be related to liquor
consumption however.
Several base fragments were recovered but only one has a maker's
mark. This specimen is green and has partial body and base segments intact.
Ernbossed on the base is "CS. 8... 165.. .". This is the manufacturing mark of
C.S. 8 Co. of St. Helens, Lancaster. England. The Cannington, Shaw and
Company produced bottles with C.S. 8 Co. between ca. 1875 and 1913
(Toulouse 1971: 147). This wmpany produced narraw neck and wide mouth pale and dark green bottles with the narrow neck bottles for wines and spirits
(Toulouse 1971).
Other alcohol related bottles include one complete specimen, a green
12 oz dip-mould bottle with a down-tooled finish with cork wire still attached
(Jones and Sullivan 1989). Two clear bottle fragments with paper label still attached are definitely associated with liquor consurnption. The label indicates that the fragments are from a Jarneson's Blend bottle, a brand of scotch whiskey. A total of 104 fragments from a green gin case bottle were recovered. In addition, a Hiram Walker and Sons lead foi1 fragment was collected. The Hirarn Walker Company produced and still produœs several brands of liquor. Fm-two fragments of a plate mould medicinal boffle were collected
from Midden 1. The bottle is clear with manganese tint. Embossed on its
base is "Pat May 2 1899", in addition to a possible maker's mark consisting of
the letter M with a star outline. No manufacturer has been identified.
A number of other bottle fragments have embossing. Once ciear
specimen has '...AUCE". the word probably king Sauce and another has the
ietter ". . . EA & P.. .". All of these fragments were recovered from Midden f and
represent at least one Lea and Pemns bottle. One clear wide mouth jar finish
fragment, representing food storage and consumption. and 11 white glass jar
fragments with the embossing 'K[?]lNG CO ...An were collected. Neither the
manufacturer nor the contents for the white jar have been identified.
Most of the metal container fragments likely represent food
consumption at the West Tent locality. Three relatively complete hole-in-top
condensed milk cans were collected, one that was identified though its embossing as an Eagle Brand milk cm. Two sardine-like fish or meat cans were also present; one a pressed drawn tin and one a hole-in-top, lapped side seam tin. Although commonly believed to be modem. the pressed drawn
"sardine" can has been in existence since the 1880s (Fontana and Greenieaf
1962: 72). As well, one lever lid. rectangular metal container was collected and would have stored non-perishable foodstuffs such as coffee or baking powder (Murray and Hamilton 1986).
All but one of the of 29 faunal specimens collected during the three years of excavations came from the middens behind the tent benn outlines- Taxon Element ~ang insguler Vembreé, kktatarsmt/ Ulm Ribs libi8 Secrum NISP Bwn, Bons Mammal Goat? 1 1 Cow 1 2 1 4 Moose 1 1 Caribou 1 t Law 3 3 5 1 6 1 18 Medium 1 1 2 Unidentified 1 1 Total 4 S 8 1 1 8 1 1 29 1 Table 6. Faunal remains collected from the West Tent locality.
CC6 and Midden 2 (Table 6). The exception is a sawed large mammal rib excavated from the middle tent. At least four species of mammal are rep resented in the assemblage induding goat or sheep (domestic?)
( Ovis;/Capra),cow (Bos taurus), rnoose (Alces alces), and caribou (Rangifer tarandus). Cow is represented by one butchered vertebrae, two sawed ribs and a sawn sacrum. Two butchered vertebrae fragments indicate the presence of moose and caribou. Large mammals, which would most likely be cow, moose or caribou, are represented by 18 specimens, all butchered by saw. One long bone fragment and one irregular bone fragment represent medium mammals. One metatamal of what appears to be a domestic goat or sheep was collected, this has evidence of butcher marks on its proximal posterior face.
The fauna recovered, as well as other food related artifacts seems low considering the tents were likely occupied seasonally for three Yeats.
Hitchcock (1899: 431 ). however, states that the freight hustlen who worked for
Macaulay on the tramline received room and board. Therefore, it is likely that food consumption at the West Tents was minimal since rnost of the meals 83 would have been provided by the Company and would have been taken within the Canyon Hotel and Saloon (Wood 1898d: 36). The faunal assemblage does illustrate that the occupants of the tents did not entirely rely on imported foods. Local species are represented (Maase and Caribou) as are impofted species (Cow and domestic goat or sheep). Furthemore. the presence of local fauna suggests that the residents of the tents were not used by people passing through. who would have either obtained their meal from the
Roadhouse or used supplies from their outfit.
The personal rnanagement/clothing fundional group comprises 31 percent of the West Tent assemblage. Buttons, clothing, footwear and personal hygiene items are al1 represented. The buttons were manufactured from metal, shell and glass. The majonty of the metal buttons are rivets for trousers with a few snap buttons most likely for shirts and jackets. One ferrous snap has "PAT 1116 89 MADE IN FRANCEnstamped on its face. The remaining buttons, based on size, appear to be fasteners from shirts and undergannents.
A sleeve of a wool coat was documented within the confines of the eastern most tent. As well, 55 textile fragments were recovered from behind the tents. These clothing fragments were small and in a heavy state of decay and could not be identified as to origins.
Footwear fragments were wllected from al1 features tested. The refuse area behind the tents contributeci the most footwear fragments (n=362).
Footwear specimens include leather and rubber boot fragments. two sole fragments as well as heel plates and heel tacs. Many of the leather fragments have boot. brass lace clasps andlor eyelets. These items likely represent the types of footwear used by the occupants and appear to be work-rubber boot related. Footwear is second onty to nails in its frequency at this locality.
Personal hygiene items used by the tent's occupants include mirrors, shaving cream and an eyedropper. What has been identified as the top of a shaving cream tube manufactured by the Crown Perfumery in London,
England, was collecteci from the western-most tent. The mirror fragments came from the middle tent as did the eyedropper. The eyedropper was recovered in two fragments, a pipette fragment and a pipette with pipettecap fragment. Both fragments ff together and the pipette-cap appears to be manufactured from rubber.
A Lambert and Butler tobacco metal container slip-lid was collected and represents smoking as a leisure activity. The poor representation of the leisure activities group (0.1%) is somewhat surprising . However. it is likely that the majority of the bottle fragments recovered at the West Tent locality and classified in the househoM functional group represent alcohol consumption. The low representation of smoking paraphernalia is perplexing and cannot be explained.
Two buckles. a bearing for a tramcar. and a bailing wire fragment make up the transportation functional group. This group is under represented at 0.2 percent, which is not surprising given the domestic use of the tents. Hunting and subsistenœ is also poorly represented within the West
Tent locality. A total of five cartridge casings were collected. Four of these
casings come from the tents and one was recovered the midden. All of the
casings were spent. two king -22 calibre Dominion cartridges, one a
Winchester Rifle Company shotgun primer, one a Dominion -44 calibre and a
Winchester 45-75. The latter pair are revoiver cartridges. As well. a partial
rifle butt-plate, a lead shot bal1 and a .22 calibre slug were also collected.
Base stamps for these cartridges (after Bames 1981) indicate they were
available during the gold rush. ft is suggested. therefore. that these artifacts
belonged to the occupants of the tents and are not a result of post
abandonment activities at the site.
At least two glass insulaton were recovered from Maden 1. Several of
the pieces have embossing that matches complete insulators collected from
the MachineBlacksmith Shop and by the divers in search of the submerged
dock in the Yukon River. W. Brookfield New York manufactured the insulators
and their prÏmary fundion was likely related to the telegraph line established at
Canyon City in 1899.
The construction1structural functional group represents the largest
proportion of the overall West Tent collection. It is dominated by nails
(n=382), 95 percent of which are wire-drawn and five percent machine-cut.
The vast majority of nails (95%) were recovered from within the three tent
outlines. These were likely used to secure the box frames of the tents as well as other interior wood features, Thin strapping fragments with nail holes are also included within the group. The original fundion of the strapping was most likely for binding crates, but due to their fragmentary condition. they probably functioned as interior fastening devices. Flat glass. which has been tentatively identified as window glass, was recovered from al1 three tents. This is sornewhat problematic sine no windows are evident in the archival photographs of the tents. It is just as likely that the flat glass represents mimr fragments or picture frame glass. Rounding out the construction/structural functional group for this localrty are three grommets and one screw.
The worMndustrial group has a very low representation within the collection, less than one percent. The lone artifact is an axe file recovered within the middle tent.
Finally, the miscellaneous group includes one beveled glass fragment and a large number of unidentified ferrous fragments. The latter were collected from two of the middens and al1 three tents. They are thin and flat and likely came from metal containers and crate strapping but were too small to categorize confidently. The glass fragment is flat and rectangular in shape with three factory beveled edges. Its function is unknown.
Summary
The assemblage from the West Tent locality suggests that the tents were used as dwellings or sleeping quarters rather than as warehouse or other purposes. The high frequency of househoîd and personal management/clothing related artifacts within the collection supports this conclusion. The occupants were likely the freight hustlers who worked for Macaulay. The low frequency of food remains suggests that most meals were
taken within the Roadhouse. an interpretation supported by archival data
(Wood 1898a). The only leisure activities document& at this locale are the
consumption of liquor and tobacca use; activities that are far from extravagant.
Such a high frequency of footwear may be related to the nature of work that the occupants engaged in such as freight hustling which would have involved
high Wear and tear on garments especially footwear.
CANYONHOTEL AND SALOON:ARTIFACT ASSEMBLAGE
The goals of the excavations at the Canyon Hotel and Saloon other than in the structural realm were to attempt to define and differentiate adivities within the building. A total of 1,846 artifacts are included in the artifact assemblage for the Roadhouse (Table 7, Figure 23). Functional groups and their proportionate representation include household (46%), personal managementklothing (2.5%), leisure (3.5%), transportation (0.2%).hunting and subsistence (0.3%), construction/structural(31.5%), and miscellaneous
(16%).
Five main categories exist within the household functional group, which is the most abundant group. Bottle body fragments are the most frequent and most likely represent the consurnption of liquor. Various colours are represented with amber and dark green kingthe most numerous comprising
90 percent of the group. A total of 16 base fragments are present. Based on sire they were likely standardized 26 oz and 12 oz bottles. The dark green bases include what appear to be two tum paste-mould and two cup-bottom specimens. The amber bases include nine cupbottom mouM and two tum- paste mould bottles (Jones and Sullivan 1989). One ciear base fragment recovered from the surface in the northem part of the Roadhouse has a 'P" with a circle border representing the Pierce Glass Company of Pennsylvania who made bottles from 1905-191 7 (Toulouse 1971: 412). Given this date range, the specimen represents a post-
FUNCTlONAL GROUP TOTA FUNCTiONAL GROUP TOTAL Hou~hold L.hum Tableware Toôacco plug 2 Ceramic Pipe dsm 1 Ferrous spoon and bawi Polrer chip fragments 60 Ferrous handk lugs rml 03 Boltleiglauwan Tabkware Bottk Medianal Bot!k/glasswan? Lead foi1 fragments Cork Wire carie clamp Metal container Total C~e~~tnittiMJStnrcurd Personaf ManrgernenUClothi~ Building hardware Button Stripping "Mother of Pearl" Fastenings Glass Ferrous Clothing Textiles Footwear Personal hygiene Mirror Sewing and Safty pins Pend Money Personal adomrnent Total 1 ARTIFACT TOTAL 1846 Table 7. Canyon Hotel and Saloon artifact functional groups and counts. Gold Rush use of the site. A 'U" is embosseci on one of the amber cup-bottom mould bases and appears to be associated with the Upland Glass Company,
Upland Indiana. This company produced bottles mostly by mould blowing from 1890 to 1909 (Toulouse 1971: 508). A second arnber base collected from Unit 3 had W.T." embossed on its base. Its diameter is 1 112in. and is associated with a threaded amber finish recovered from the same unit. The bottle makef s mark has not kenconfdently identified but the Glasstown
Plant in Millville, New Jersey, is a possibility. This company manufadurad bottles between 1809 and
Canyon Hotel and Saloon ArtWatt Graups and Fmquencks
Figure 23. Graph of the frequency of functional groups within the Canyon Hotel and Saloon h istoric assemblage. 1935 (Toulouse 1971: 544). The small base diameter of the bottle coupled with the threaded finish suggests the bottle was either for medicinal or condiment purposes. Three wide mouth jar finishes, with wrk-like closures, were collected and appear to be associated with food storage. As well, four plate-mould medicinal boffle fragments are present.
A total of eight clear and amber bottie, stopper-finishes were recovered as well as two down-tooled wrk-finishes. The finishes are likely associated with the wnsumption of alcohol since one amber finish had a "Canadian Club
Whisky" lead cork foi1 still adhering. In addition. a similar clear finish was recovered on the surface near the river with a "Glenlivet" lead cork foi1 still adhering to it. Although it cannot be definitively stated, it is also more than likely that the vast majonty of other bottle fragments are associated with the consumption of alcohol. Other artifacts related to the consumption of alcohol include two corks. seven fragments of wire cork clamps and 53 fragments of lead cork foil. Several types of liquor and wmpanies manufacturing the liquor are represented. lncluded within the collection are Imperia1 Brand Whiskey,
Canadian Club Rye Whisky, Glenlivet Scotch, Whyte and Mackay of Glasgow
Scotch Liqueur and Compass Brand liquor. The foi1 fragments were recovered throughout the Roadhouse; however, they did tend to cluster within Units 19,
20 and 3. The quantity of liquor related artifacts reflect the building's use. at least in part as a licensed saloon (Wood 1898d). Bottle fragments were recovered from each une at the Roadhouse locality except for Units 11 and 14. with no real clustering to demonstrate a particular sedion of the structure designated for drinking.
Tableware is represented by 16 artifacts within the collection. Three are plain white-glazed cerarnic fragments and one is a blue rimmed fragment from a plate. One fenous bowl and a ferrous spoon were recovered within the same unit as were two ferrous handle lugs that appear to be associated with a kettle or coffeepot. The remaining eight tableware artifacts are fragments of crazed, thin gfass that appear ta be from a tumbîer. Eight other thin clear fragments were recovered but were too small to identify as to function; however, they are considered to be glassware fragments.
A total of 232 metal container fragments were collected within the
Roadhouse. Most are too fragmented to detemine contents and were classified as metal containers bas& on their thickness and cuwature. Six lever-lid, one lapped side seam and one single-end seam fragment were recovered from Unit 13. These Iikely represent one tin cm. The tin would have Deen used for dry storage such as tea, coffee or coca. Unit 3 produœd one condensed milk can that functioned in a secondary context as earlier described. Unit 8 contained a hole-in-top lid fragment with 'MILKn embossed on it.
A diverse but proportionately small number of artifacts are present within the personal rnanagementfclothing fundional group. Artifacts related to clothing include glass, shell and femus buttons, textile fragments, and footwear items. Loosely associated with clothing are one sewing pin and one safety pin. Personaf adomment items within the collection include two ferrous plates from badges and one badge pin. Sixteen mirror fragments are included in the personal hygiene category. One 1891 copper United States five-cent piece and two pencil fragments complete the assemblage. Items were recovered from a number of units from both the western and eastem portions of the structure. The baleen fragment may be associated with female clothing.
Gambling and smoking are the leisure activities represented with the
Canyon Hotel and Saloon artifact assemblage. A total of 60 clay (ceramic)). blue and white poker chips were recovered from Units 3,19 and 20. The 60 fragments represent at least 12 individual poker chips and were recovered from near the north and south sill logs. lnciseâ on the chips is a scene of dogs playing poker with the words 'NO MONKEYING" bordering their circumference. The high frequency of chips suggests gaming activities were taking place within the saloon portion of the Roadhouse, an activity not reported in archival documentation.
Smoking related artifacts include two tobacco plug clasps. one circular- shaped and one heart-shaped, and a pipe stem. The manufacturer for the tobacco has not been determined.
The hunting and subsistence and transportation functional groups make up less than two percent of the collection. Five spent brass cartridge casings, al! of which were being produced during gold rush times, were recovered.
One bridle ring and a horseshoe nail fragment are included within the transportation functional group. Nails and window glass dominate the constniction/structural functional group. The majority of the naik recovered were wire-drawn with only a few machine-cut nails present All were less than 4 1/4in. in length suggesting their use in interior construction. Units 2, 17, 19, 20 and 21 wntained the most window glass (n=428). Unit 2 is located along the north wall of the structure near the eastern corner with the remaining units situated along the south wall moving east to west respectively. The recovery of Rat glass, including large fragments (>5cm2) within these units suggest the presenœ of windows within their immediate vicinrty. Unit 13 produced the only brick recovered from Canyon City.
Five wire fragments, probably bailing wire, and 289 unidentified ferrous fragments are included within the miscellaneous group. As well two brass plates and one brass tube make up the remaining items in the Miscellaneous group. The latter fragments may be components of a gas lantem.
Summrry
The Canyon Hotel and Saloon assemblage appean to be representative of the activities one would expect to occur within a Roadhouse.
Bottle glass. of which the vast majonty is most likely a result of liquor consumption. contribute to over 60 percent of the household functional group.
As well a large number of metal container fragments coupled with ceramic and ferrous tableware represents food consumption. No faunal remains were recovered during the excavations within the Roadhouse. This is likely a function of refuse rernoval, which would have likely been more intense because of the business nature of the Saloon. The kisure functional group
has the highest frequency within the Roadhouse assemblage in comparison to
other locale's assemblages. Poker chip fragments are the most notable
leisure item recovered from the Roadhouse. Their concentrations in the
centre of the structure along with lead foi1 caps suggest that the Saloon portion
of the Roadhouse was located in this area. The distribution of bottle glass,
although not really conœntrated, tends to be more numerous in the west ha#
of the structure. Looking at the archival photographs (Figure 5), it appears
that the western portion of the Roadhouse was one large room as suggested
by the lack of entranœ ways along this section. The three entrance ways to
the east may be to smaller rmms related to a hotel operation.
It is difficult to determine exactly how the artifacts hmthe Roadhouse
came to rest where they did since the structure appears to have had a floor.
The smaller artifads. which are the majority of the bottle glass collection. rnay
have passed through the cracks between the flwrboards. Larger pieces may
have been deposited either prior to the fiwr being put in place or dunng the
structure's dismantlement. Except for the green bottle base recovered on the
surface with the postgold rush maker's mark. al1 the artifacts are contemporaneous with the gold rush.
NORTH-WESTMOUNTED POLICE BARRACKS ART~FACTASSEMBLAGE
One of the goals of archaeological investigations at the North-West Mounted
Police Barracks was to determine, if at al1 possible, the full range of activities carried out within the structure. Wood (1898a) states that the Barracks contained sleeping quarters, an office and an officer's mess. A total of 2,413 artifacts were collecteci during the investigations (Table 8, Figure 24). Artifacts related to construction of the building (n=1729) dorninate the assemblage, the vast majority being window glass fragments however.
Construction matenal contributes to 75 percent of the collection. The househoM functional group is the next largest within the WMP assemblage, representing 18 percent of the collection. Wiihin this group bottle glass is the most abundant. The colour of bottle glass and types of finishes collected (two stopper finishes) imply the presence of alcohol. as do 12 lead foi1 fragments.
Unfortunately, no embossing was observed on these fragments so their identification as liquor cork foils cannot be confirmed. It might also be the
FUNCTIONAL GROUP TOTAL WNCTïONAL GROUP TOTAL Household 4unting and Subristuuo Tableware 2 Ammunition 28 Ceramics 1 Sluq 3 Fork 1 CamCasing 25 Bottle/glassware 490 FOh hook 1 Cork 1 rami 29 Lead foi1 12 Metal container fragments 43 Candie hoMer 1 Building Hardware Furniture related artifacts 3 Fiashing Total 552 Stmpping Fastenen Peraonal Manag.m«itlClothing Grommet Button Backer Naiis Button Wmdow glas Ferrous Gbzier point Bone toggle Chinking Rivet 9ot.~ FootuRar Bu&k Textile Penal Tobl i 1 Artn.ct Total 2413 Table 8. NWMP Barracks artifact functianal groups and counts. case that the foils and alcohol nlated bottle glass represent wnfiscated liquor
since wnsumption of alcohol by NWMP officers and non-commissioned
ofken was looked down upon by the upper echelons of the NWP(Steele
1898a). Thin clear and manganese tint bottk glass were recovered as well.
indicating the presence of medicinal aids as well as one clear wide mouth,
cork-finish jar fragment. One metal three-pronged fork and a white-glazed
ceramic sherd represent tableware. An omate metal plate and two C-clasps
relate to fumiture.
North-West Maunted Police Brrncks Artifoct Groups and Fmquencies
Figure 24. Graph of the frequency of functional groups within the Norîh-West Mount Police Barracks historic assemblage. Several tin can lids and fragments were wllected. The lids recovered
were of the hole-in-top variety and represent the wnsumption of food within
the NWMP Barracks. One can Iid was recovered in the northeast section of
the building along with 22 unidentifiabk mammal long bone fragments. These
remains suggest that the officer's mess section of the building was located in
this area. The fork and wide mouth jar finish fragments were collected near
the doorway of the structure.
A variety of items are represented within the personaVmanagement
clothing fundional group. This group contains the one artifact directly
associated with the NWP. With 'NWMP" stamped on its face, it is a metal
button backer for polishing buttons. It looks like a fiattened tuning fork and
was slipped in between the button and garment enabling the officers to polish
their brass buttons without soiling the cloth of their unifonn. Four buttons and
three textile fragments were also collected. Buttons include two trouser rivets,
a ferrous button and a bone toggle button fragment likely for outerwear. Two
buckles were collected and appear to be associated with footwear. Rounding
out the group are six lead fragments belonging to pencils.
A total of 25 brass cartridge cases were recovered from within the
NWMP Barracks as well as one lead slug. All are grouped with the hunting
and subsistence group although their fundion more obviously relates to
policing activaies. The Winchester Repeating Amis Company (WRA Co.). the
Dominion Anns Company (D), and Enfield are the identifia manufacturen of the cartridges. All were supplien to the NWMP dunng the lgthCentury
98 (Phillips and Klancher 1982). The Enfeld specimens (n=4) are -45 calibre revolver cartridges. Other cartridges include .22 calibre rim-fire (n=lQ), .30 calibre (n=l) and 7mm (n=l) centre fire. A single fishhook included in this functional group is a large jigging hook and relates to subsistence activities.
Telegraph wire was observecl along the westem wall of the NWMP barracks but not collected. This suggests that the location of the telegraph equipment within the building was in the vicinty of the western portion of the structure. The telegraph was established at Canyon City within the banacks during September of 1899 (Wood 1899).
As earlier noted, window gfass is the majonty (65%) of the construction/structural functional group. A total of 960 window fragments was recovered within N69W108 locatd just outside of the north wall of the building. This quantity suggests a window was present in the immediate vicinity. Both wire-drawn and machine-cut nails were collected and make up
32 percent of this functional group. Other construction1structural items collected include flashing, strapping, chinking fragments and a glazier point.
The miscellaneous group contains wire fragments and unidentified ferrous fragments. It contributes little to the interpretation of the activities at the NWMP Barracks.
Summary
Although the majority of the assemblage was collected outside of the structure, in the northwestern corner, the types and distribution of the artifacts within the structure suggest at least two room divisions-the westem portion of the structure as an office and the eastem portion as a living area. This is
based on the presence of telegraph wire outside of the western portion of the
building. which suggests that its entranœ into the structure wouM be the from the West wall, and food related items in the eastem portion. This interpretation
is consistent with Wood's (1898a) description of the MNMP barracks having at
least two sections.
The archival information suggests that the NWMP stationed at Canyon
City were there year round and twk their meals within the Roadhouse, which
may account for the relative low frequency of food related items. No leisure activities were identified from the assemblage recovered. However, it may be that the liquor was consumed by the MNMP rather than confiscated. As well. the fishhook, although placed within the hunting and subsistence group, may
also represent leisure activities. Since the NWPsupplied the officers with sustenance they would have had no real reason to fish except for leisure. One
benefit of fishing would be to add variety to their rationed diet. The officers kept a log of rations received and the lists indicate their diet was restricted to staple food items (monthly Canyon City NWMP reports in author's possession). The Yukon River would have provided a source of grayling and pike.
CANYONAND WHITE HORSERAPIDS TRAMWAYOFFICE: ARTIFACT ASSEMBLAGE
The Canyon and White Horse Tramway Oniœ localrty was the first structure tested at Canyon City in 1994. The major focus during the excavation of this building was to retrieve as much structural information as possible since it was slated by Yukon Heritage Branch for reconstruction. A
14 percent area sample of the Tramway Onice was cornpleted resolting in the
recovery of 1,661 historic artifacts (Table 9. Figure 25). Six functional groups
are represented including household (18%). personal management clothing
(6%). transportation (1%), hunting and subsistenœ (1%),
~nstruction/structural(66%) and Miscellaneous (8%).
The tramway office was identfied through archival photographs. A
white banner with 'CANYON & WHITE HORSE RAPIDS TRAMWAY
COMPANY CO" in several of the photographs identified the log structure as
the location for the wrnpany office. The structure is 30m in length and it is
likely that only a srnall portion of the building was allocated for the
administration of the Tramway business. So far, no records on how business
was carried out have been found. It is assumed that arrangements and
payments for freighting goods and boats around the Canyon and Rapids were
carried out within the building as well as other business related duties such as
paper work. storage and money collection. The assemblage. however, suggests other uses for the building. Although the constructionlstnictural functional group dominates the assemblage the variety of household and personal managementldothing artifacts indicate some use as a dwelling.
A total of 296 historic artifacts are represented within the household functional group. A variety of types are present including tableware, bottle glassware, metal containers. lighting related items and stove pipe rernains. FUNCTlONAL GROUP TOTALI 1 FUNCITONAL GROUP TOTAL
Tableware Gold nugget Ceramic 1d.l Ferrous spoon Knife handle TmnspOrmiori Ferrous knife Made Bridle apparat- Bottle/glassware Bracket Bottle glass Tau Medicmal Lead foi1 Hunting md Suimidence Cork Ammunition Metal Containers Cartrige casmg Larnp Slug Stove pipe Fish weight Total Fish hook TdJ Persocrat Marugcmari(lCleng Button Coridruct~budlN8I Moiher of Pearl Building Hardware Glass (white) Flashing Fenous Strapping Ferrous rivet Fasteners Bone Washer Leather Grommet NWMP Bras Nail Clothing eolt Textile Nut FeR hat Wiridow glas Baleen rai Buckle Footwear Miscellaneoui Adomment W ire Monetary Unidentified Coin * ral Artifact rotai 1661 Table 9. Canyon and White Home Rapids Tramway Office artifact functional groups and
Twenty-six ceramic fragments wre collected. All are small white glazed fragments that, being recovered from a 3m2 area, are probably from same vessel. Their shape and curvature suggest a aip. Other tableware fragments include three ferrous teaspoon fragments, two bone handles and one ferrous knife blade.
Bottlelglassware contribute to 68 percent of the household assemblage.
Three relatively complete medicinal bottles wre collected from the east half of Canyon and White Home Rapids Tnmumy Ollke Met Goum and Fmquencier
Figure 25. Graph of the frequency of funcüonal groups within the Canyon and White Home Rapids Tramway Office historic assemblage. the structure. All were plate-mould with a flanged or patent lip (Jones and
Sullivan 1989). One clear medicinal type bottle with some of the contents still inside, had the mark "PC O..Om. This rnay be a maker's mark but could also represent a lot number. Based on the contents aroma, the contents within the bottle appear to be a clove extract of some type. A second camplete medicinal bottle is a stout green tinted bottle with "Ayers", 'Lowel Massn.and
"Pills" embossed on the sides and the maker's mark on its bottorn is an incised
"cn. The pill bottle provides the manufacturer of the pills and its origin, however, the maker's mark is unidentifiecl. The third bottle is a wmplete clear
specimen with no embossing or makers ma*. A complete glass via1 was also
recovered and is included within the medicinal bottle sample. No distinguishing features were observed on this artifact. The remaining nine fragments are al1 clear and appear to originate from plate mould bottles.
Several bottle base fragments were recovered from the Tramway
Offîce. Nine of these bottles were cup-bottom moulds and seven were post- bottom moulds. All except two appear ta be from liquor bottles. Two of the bases had maker's marks present. The two campanies represented are the Cannington . Shaw & Company from England which produced bottles carrying "CS & Co" mark from 1875 to 1913, and the Nuttall8 Company from England that produced bottles bearing 'N & Confrom 1872-191 3 (Toulouse 1971: 147;
380). One white glass cup-bottom mould jar fragment was callected with
". .. O.. .The O.. .OMA.. ." embossed on its base. A square, clear, post-bottom mould bottle was also collected. Its dimensions of 3x3cm suggest it was a condiment or medicinal type bottle. Only two finish fragments were collected and are likely frorn 2602 liquor bottles. The rernaining bottle pieœs are body fragments. The colours represented include clear, manganese tint, green tint, aqua tint. green amôer and white. Three cork fragments as well as seven foi1 fragments were recovered from within the structure. Four of the fragments are from the same Robert Brown Ltd of Glasgow Whiskey lead foi1 cork cover.
Except for the medicinal bottles, which were recovered from the east portion of the office, there is no apparent clustering of bottle glass. Despite this lack of clustering. the bottles do represent a variety of activities which appear to have been carried out within the building including liquor consumption, food consurnption and medicinal related.
Five complete metal containers were collected from the Tramway office.
A single end seam, with a capped-on-threaded Iid can and a smaller condensed milk can were excavated from within the wooden pipe feature previously described. A hole-in-top lapped seam container with a filler diameter of 1 4116in.. a diameter of 3in. and a height of 4 4116in. was collected from SIE2. This container approximates Buriey and Ross's (1979: 63) type
44 No.2 (21b) which is pre-1907 and held either meats, fruits, or vegetables.
Another tin recovered matches Burîey and Ross's (1979: 68) type 47, a pre-
1904 "1 Lb. Net" can that likely held pork and beans. soup or stew. A unique tin with a height of 10f16in. and a diameter of 2 8J16in. was collected and may be tobacco related (Murray and Hamilton 1986). The top edge on this container is curled and a lid fragment accompanied the container. The remaining metal container artifacts are key and tin fragments. The latter were classified as metal containers based on their thickness and curvature.
Rounding out the household functional group are two brass lamp hardware fragments and ten stove pipe fragments collected from the northwestern corner of the building.
The personal managementklothing artifact functional group has a diverse array of artifact types. A total of 34 buttons are included in the collection from the Tramway Onice. The buttons were manufadured from various types of materiab including rnetal. glass. shell, kather. bakelite and bone. Most are of simple design with either two or four holes. lncludeâ in the button sarnple is one NWMP brass button manufactured by 'Smith 8 Wright
Birminghamn. All of the buttons except for the latter are associated with trouser, shirt and undergarment wear.
Other clothing related artifacts include small textile fragments, two baleen fragments, a buckle and eight foot Wear fragments including a ieather sole. A felt hat was block fifted and sent to Ottawa for conservation. It was recovered from the north western corner of the building. Adomment related artifacts include broach pin fragments as well as a badge. The baleen fragments, which may have originated from a corset, crinoline or other female undergannents, and the broach are indicative of the presence of the female gender.
Three coins were excavated from the Tramway Office. The oldest is a copper 1854, Bank of Upper Canada haif-penny bank token, recovered from
S3E9. The other two, a 1894 United States of America Hatf dollar and a 1891
Canadian 10 cent piece were collected from SSE18. One small gold nugget (1 gm) was also collected from the eastern haif of the Tramway Office.
Transportation related artifacts include bndle rings and rivets and a bracket that appears to be tramcar related. A Csh weight. which may be intrusive, a jigging hook, four lead slugs and 17 brass cartridge casings represent the hunting and subsistence fundional group. Several different cartridges are present. Ten of the 17 cartridge casings and three of the four slugs are .22 calibre, six of which were made by the Dominion Ams
Company. Revolver cartriiges indude a 45mm. two .44 calibre (one Smith and Wesson the other a Winchester) and a -38 calibre Winchester casing.
Two 30-30 United Metallic Cartridge Company casings were also present.
One 12 gauge shot gun primer was colkcîed. According to Bames (1985) al1 of these cartridge types were produced during gold rush times.
The constniction/stfuctural group dominates the Tramway Oflice assemblage with window glass cantributing 73 percent. Window glass clusters suggest six general areas of concentration (S6E8; S6E19;S6E16-
S8El8- SQE16, S2E20; S6E20; and. S5E26; see figure 20). These concentrations reflect the location of windows both on the outside and possible windows piaced inside the structure. lnterior windows may reflect the office nature of the building. The abundanœ of window glass suggests that windowpanes were not salvaged when the structure was dismantled. A total of 26 1 nails are present in the Tramway Onice collection. Most (82%) were wire drawn, however, some machine-cut nails are present as well. The rnajority of the nails came from the eastern part of the building, the portion having a floor. Other eanstruction/structural related artifacts include washers, grommets flashing and strapping .
The Miscellaneous group wntains 130 unidentifiable fragments and 10 wire fragments. Summary
Except for window glass there is no apparent clustering of artifact types between and among functional groups at the Canyon and White Horse Rapids
Tramway Company Office. When looked at as a whole. the presenœ of househoid and personal managementlclothing related artifacts suggest that the structure was not only used as a company office but also as a dwelling.
The number of individuals or location of domestic activities within the building are unknown. It is. however, one of only two major features tested at Canyon
City that suggest the presence of women. Except for a company sign identifying the building as an office within the archival photographs, such a use could not be discemed frorn the artifacts.
EAST CABIN: ARTlFACT ASSEMBLAGE
The presence of the East Cabin at Canyon City was detemiined during the 1995 investigations, and excavations continued in the 1996 field season.
The goals of the investigations at this locality were to gather structural information about the cabin and to detemine its function. A total of 6,158 historic artifacts was recovered (Table 10, Figure 26). The assemblage is abundant when campareci to the other localities at the site. With the household related items accounting for 53 percent of the Cabin's assemblage, there can be Iittle doubt that it was used as a dwelling.
Wihin household functional group, ceramics are represented by three porcelain bowlkup fragments, 15 white-glazed body fragments and three white-glazed rim fragments. The ceramic sherds were rewvered frorn seven FUNCï'ïONAL GROUP TOTAI FUNC77ONAL GROUP TOTAL Housohdd Ta blewan Ammunition Cenmic c.rtribgs cging Spoon Sluo Utensil fragments Fih Hook Ferrous frying pan robl Ferrcus coffee pot Fenous pot lid (6û frognents) Fenous plate/bow( (6 fragments) Fenous plate (16 fragments) Tin white enamel plate Unidentifiecl Codtmm Pail Pot lugs BotWglass wan Bottks Medicinal Lead foi1 =orwtiudioci/StruaunI Metal containers Buiiding hardware Stovdheater FMing Fenous body fragments swJm9 Cast iron stove legs Oaor hardware Fumiture Pad Lodr Lantem Fastenen Total w8sher smw ho& Personal Managemenffclothhg Gromma Buttons Naib Glass Stapks Sheil Nut Clay Boit Ferrous Wmdow Gkss Clothing Chinking Textile Fencing (Chidten wim) Footwear rot84 Penonal Hygiene Laundry bluing lisdlaiiiour Hair dye bonle Win Mirror fragments Femus fragments Penonal Adornment: Hair pin 'a4al Ink Botüe Tohl 11 Table 10. East Cabin artifact functional groups and counts.
units and show no spatial concentrations. Two ferrous spoon fragments were collected along with what appear to be ferrous fragments from utensil handles.
Also recovered were a sample of metal cooking/tableware artifacts including a East Cabin Artif.ct Groupi and Frequencies
Figure 26. Graph of the frequency of funcüonal groups within the East Cabin historic assemblage. complete frying pan, a complete coffee pot. a fragmented pot or fry pan lid , a fragmented bowl or deep plate, a fragmented plate, and a fragmented pail with lugs. It appean that these items wre mole prior to the destruction of the cabin. These items indicate both food preparation and consumption occurred within the cabin.
Bottle fragments rewvered are of four colour types: arnber, clear, clear with green tint and clear with manganese tint. One clear base fragment is 110 from a post-bottorn mould botk and two dear with green tint fragments originated from a cupbottorn mould bottle. Four of the clear with green tint fragments are rim shards from wide-mouth jars. Two complete vials were also collected; one with a cork and contents still intact. The content's aroma suggests cinnamon extract. One clear body fragment was collectecl that still had portions of its paper label adhering. The letters identifii are 'Crosse.. .
B.. .". This likely represents a bottle of which Crosse and Blackwell. Ltd, of
London England manufactured the contents. This Company bottled food and condiments ôeginning in 1830 with continued production to the present
(Toulouse 1971: 113). The consumption of liquor is represented by the bottle fragments recovered as well. One amber stopper finish with the lead foi1 still adhering from a Canadian Club Whiskey bottle was collected as well as two clear stopper-like finishes which are similar to the Glenlivet Scotch finish recovered from the surface of the site. A total of 17 lead cork foi1 fragments are present, one of which is from an Imperia1 Whiskey bottle. One complete but melted medicinal bottle and 16 fragments from plate mould bottles complete the collection.
Metal containers are represented by 1000 fragments. Several different closure types are evident including slip lidlcap. threaded lidkap, lever lid and hole-in-top. Except for the latter, al1 are commonly used for non-perishable goods such as spices, tea and coma (Burley and Ross 1979; Murray and
Hamilton 1986; Rock 1981 ). Both single- and doubleend seams are represented. The double end-seam for perishable goods was not in cornmon use until after the 19Ws but was used eadier (post 1890) for non-perishabie
goods (Rock 1983: 106). It is uncertain whether the double-end seams
recovered represent open-top sanitary tin cans or just bases for non-
perishable dry storage. Two relatively complete condensed/evaporated milk
cans are the only fragments where contents were confidently deterrnined. The
main concentration of metal container fragments was in the southeast area of
the building, suggesting tinned goods were stored here.
A total of 105 faunal remains was recovered within the East Cabin
locality. They are summarized in Table 11. The charred and fragmented state
of the collection makes identification to genus or species difficult. Mammal
remains are the most frequent and include large. medium and small mammals.
Represented within the collection are ww, rabbit and arctic ground squirrel.
Aves (birds) are well represented within the collection as well. At present it is
unknown if they are wild or domesticated in origin. F inally, two fish vertebrae were recovered. Except for the ground squirrel. the fauna probably represent the types of fresh meat consumed by the occupants of the cabin. Butchering marks were identified
Taxon Ekriwnt PnalanQas UnrdenM Cnnial Vemiwm Sapula ri Femur Humerus k.rpiIs Mammais S. panyii 1 Bovidae 1 Lagomorph 1 Large 45 1 Medium 1 7 Small 9 Aves 26 1 3 4 2 Pisces I 2 2 Total 71 1 4 1 4 1 4 19 105 Table 11. East Cabin faunal remains. on one of the large mammal specimens and five of the medium mammal . phalanges. The latter probably represents the skinning of the animal.
Four cast iron stoveiheater legs and 1896 fenous stove body fragments are included within the household group. The stove fragments. which were significantly thicker than the usual tin can fragments and not CU-, were recovered within units N2M8,N28W27 and N28W28. This indicates that the stovelheater was located in the western portion of the structure. One ornate upholstery button and 24 C-clasps represent fumiture, and four brass tube fragments represent the presenœ of a lantern.
The personal managernenüclothing functional group is relatively small but diverse. Twenty-three buttons were collected including glass, shell, clay and ferrous. The ferrous buttons are most numerous with 11 from Levi
Strauss and Company clothing. A femus Carhart glove button was also recovered. Also associateci with clothing are 12 textile fragments and two footwear fragments. The textile fragments are small and burnt hampering further identification. A chaned leather fragment and sole are identfied as footwear.
Personal hygiene activities are represented by three mirror fragments a small cluster of laundry bluing, and a fragmented hair dye bottle. The bottle was manufactured in a plate mould and ernbossed on it is 'C DAM SCHINSK
LlQUlD HAlR DYE NEW YORK". Caution should be used in autornatically assuming gender since it may have kenused for mustaches as well. The recovery of a hair pin, however, appears to reflect the presence of a female. This is not surprising since Figure 5 is a photograph of the East Cabin with a woman standing in front of it. A square moulded bottle with a threaded finish with the embossing '. .. RWOODS INKS" rounds out this functional group. The ink bottle is metted and in four fragments.
Hunting and subsistenœ is represented within the East Cabin assemblage by two fishhooks. 16 cartridge casings and five lead slugs.
Thirteen of the cartridges are -22 calibre and two are 7mm calibre casings. A
'KYNOCK 7.9 MAG casing and one revolver casing with 'W.R.A. Co. 41 LDA" was also collected. All of the casings within the collection were in use by the
1890s and continue to be used today (Barnes 1985).
The transportation functional group represents a small portion of the overall East Cabin assemblage. Chain and buckles from bridle gear, a horseshoe nail, and an oar Io& from a boat are included.
Three files are the only items within the workinglindustrial functional group. Fasteners of which nails are the most numerous dorninate the wnstruction1structural group. Over 95 percent of the nails are between 14 inches in length and likely were used for flooring and smaller interna1 structural features. Nails are both wire-drawn and machine-cut. Unique to the cabin is the presence of 1 V2in. fencing staples, concentrated within unit N25W26.
This suggests storage in this area. The staples may be associated with the L- shaped depressions and chicken wire located just to the north of the East
Cabin documented in 1994. However, since many of the staples were adhering to fragments of wood, they may also have been used a fasteners within the cabin. ûther items included within the construction structural group are window glass. chinking, chicken wire and building hardware. A small key- padlock is included with the building hardware but is more likely a lock for a strong box.
Miscellaneous items includes 19 wire fragments and 807 small unidentified ferrous fragments.
Summary
The assemblage recovered from the East Cabin cleady supports the assumption that it functioned as a dwelling. Activities camed out included the preparation and consumption of food as well as personal management type activities. The predorninance of metal tablewares over ceramic tablewares is informative since it explains the relative absence of ceramics in other buildings. The evidence suggests that metal tableware was chosen over ceramic, likely because it is more durable.
The matnx in which the assemblage was recovered consisted of bumed wood and fire burned clay. There is little doubt that the cabin burned down with much of its material contents still inside. In this respect. it is a case of catastrophic abandonment representative of a single moment in time (Kent
1990). The timing of the fire is of concern. None of the archival documents mention a fire during the gold rush occupation of Canyon City. One archival photograph (Figure 3) shows evidence of what appean to be a recent fire atong the tramway during the time the tramway was in operation. The only other structure showing this amount of fire damage is the North-West Mounted Police Barracks. However. at the NWMP the fire appears to have been
confined to just the sill logs and floor joists suggesting that burning occurred
after the structure's disrnantiement.
None of the artifacts within the East Cabin assemblage have a post- gold rush teminus post quem. Yet the presence of staples within the cabin and chicken wire appear to associate its use with the L-shaped depressions to the north of the cabin. These depressions may have kenassociated with fur faming, an activity that would post date the tramway operation. The East
Cabin may thus represent a mixed assemblage including Gold Rush and Post-
Gold Rush activities.
MACHINE/BLACKSMITH SHOP:ARTIFACT ASSEMBLAGE
A total of 3,372 artifacts from several different functional groups was recovered at the Machinehlacksmith Shop locality (Table 12, Figure 27).
However, most artifacts refiect industrial activities and day to day equipment repairs during Canyon City's existence. The settlement's isolation and the logistics involved in the Klondike Gold Rush are also reflected through an emphasis on recycling within this assemblage.
The household, personal management/clothing, leisure, hunting and subsistence and communication functional groups cumulatively represent but
3 percent of the Machine Shop assemblage. The low frequency of these groups suggests an absence of domestic activities in the building.
The household functional group is represented by ceramics. bottlelglassware and metal container fragments. The seven ceramic sherds FUNCTIONAL GROUP TOTA FUNCnONAL GROUP TOTAL Pipe colbr Ceramics BoltWglasswlre Bottk Jar Metal container fragments Total
Personal MarUgmi.ritlCiathing Button Fenous Footwear Buckle Heel piece Rubber boot Total bisum Tobacco plug Total
Tmrmporbtiori BrÎdle apparatus Oxen sha Horsesha nail Bailing wire Bearing -bucbrril Total Building hardware Strapping Hunting and SubsbC.IIC. Latch Cartridge casing Hinge Lead pellet Braœ Gun deaning apparatus Glvier points Butt plate Fastenings Tobl Washer Grammet Communication Scnw Glass insuiator Stapk Total Nutr and ûotts Haoks Workinghdustrial Nails Tools Wmdow glau Hardware tool box ocil Handles Axe file Ihcdlrnoow Cleaver (homemade) Simet metal File Stripping Metal punches Rubber Chisel Leatbr Lid lifter Texti k Clamp Win Ferroris adze head? C~Y Chain Unidentifml Femws fragments
Repair links ml - - Ferrous pipes RnFACf TOTAL 3372 Total weight of amount of coke and slmg fragments r ,rd is 3.4kg or 7.51br. Table 12. MachineJBlacksmith Shop artifact functional groups and counts.
117 Figure 27. Graph of the frequency of functional groups within the Machine/Blacksmith Shop historic assemblage. appear to be fine stoneware and al1 are from the same narrow, cylindrical cream coloureâ vessel. A clear glass with green tint widtmouthed jar finish fragment and seven body fragments of the same wlour were collected. One of the fragments has a '. .. WHY.. . GLA.. ."embossed on it. which has not ken identified to the manufacturer. The wide-rnouthed jar likely first functioned as a vessel for food storage (Jones and Sullivan 1989). but may have been recycled at the Machine Shop-possibly to store nuts, bob, nails etc. Nine clear glass fragments were rewvered from Unit 10 and are of the same thickness and most likely represent one bottle. One has a portion of the label still ad hering with a woman's head and bordering letters '. .. oria.. .^. The size of the remaining label has made it difficuît to determine origin or contents.
Judging from the curvature of the glass it appears to have been a relatively small bottle that may have cantained condiments. One white coloured glass shard is included within the househoid group but its function has not been identified. All of the bottle glass was collected within the east half of the structure.
Metal containers are represented by four body fragments. They are identified as tin can fragments due to their curvature and thickness. Too Iittle remains of them to associate them with a particular food group or function.
They rnay also have served a secondary function such as containers for storing fasteners.
One button and ten footwear fragments make up the personal managementlclothing functional group. The button is a ferrous cast. two-hole button most likely from a jacket or trousen. Five heel plates. four rubber boot fragments and a boot buckle constitute the footwear assemblage. The footwear fragments were distributecl throughout the locality without clustering.
One tobacco plug clasp was present. This is the only indication of a leisure activity occurring within the confines of the Machine Shop.
Hunting and subsistenœ also have a low frequency within the total assemblage. This is somewhat unexpected sinœ a blacksmith shop is the most Iikely place within the former townsite where reloading ammunition would have taken place. The lad< of this activrty may suggest that the occupants of the site were not concemed with anns, which may be indicative of the
Tramway Company's suppîying and importing their sustenance needs.
One mplete glass insulator, manufactured by W. BrookfieM of New
York, represents the communication functional group. This type of glass insulator has also been recovered from the West Tent locality and from the
Yukon River near the now submerged dock. The insulator is associated with the telegraph Iine and office established within the North-West Mounted Police
Barracks in 1899 (Wood 1899).
The transportation, working/industrial and consûuction/structural functional groups are the most frequent and informative type of artifacts regarding the activities occumng at thb locality. In addition, they also provide insights about the nature of the settlement in general.
The transportation functional group makes up 27 percent of the
Machine Shop's artifact assemblage. Horseshoe nails have the highest frequency here. with most king either head or shank fragments. This is consistent with the shoeing of horses. A horseshoe. which was stolen and not included within the assemblages counts, and an oxen shoe were also recovered within the Machine Shop. The latter is the only evidenœ suggesting that an ox or oxen were used to pull tram cars. The remaining artifacts within the transportation functional group include bailing wire fragmsnts, three bearings for the wheels of a tramcar and artifacts related to bridle apparatus. The latter category include bridie hamess fragments. various sizes of bridie rings, brass rivets and an end-femle for a neck yoke.
Approximately 34 percent of the Machine Shop assemblage consists of artifacts related to machining and blacksmithing. During excavation 3 -4kg of what appears to be coke and slag were collected from the northwestern units of the locality. Two solder bars were present, including one with the embossed label 'ES. and Stetsonn. Several melted solder fragments and what has kententatively identified as a graphite contact rod for weiding were also recovered. Tools within this group number 18. They include a small hardware rnetal box, triangular file, an axe fik and a ferrous adze head. Toots that appear to be directly related to bbcksmithing include a machined cleaver. six chisels, a hot-plate lid lifter and two metal punches. The metal punches and machined cleaver show evidence of direct pressure from hammer blows.
Two hundred and ninety ferrous items have clear evidence of king reworked. This consists of non-factory reshaping and cutting. The homemade cleaver and punch, which are bath included in the tool types, are examples of finished products from blacksrnithing and machining. Merreworked items include nuts and bolts, hooks, strapping, banel hoops, iron bars and strapping.
These may be the byproduds of reworking other items and not actual functional items.
SmithingmAachining detritus is the most abundant artifact type within the workinglindustrial functional group comprïsing 67 percent of this group.
Angular ferrous fragments and circula?ferrous pieœs make up the majonty of this type. The latter do not exceed 20-25mm in diameter and range from 25-
10mm in thickness. They appear to be the byproducts of punching hoks in metal, most likely metal bars and banel hoops. The detritus' distribution was concentrated in the northwestem portion of the structure around and under the forge feature discussed in Chapter 4.
The construction/structural functional group is weli represented within this locality's assemblage. Building hardware and fastenings make up the majority of this type. The washers, nuts and bolts are most likely associated with the construction and maintenance of tramcars. Glazier points, which are small triangular metal pieces used to secure single panes of window glass within a frame, appear to have been stoied items and are directly related to window repairs. This type of maintenance was probably done on location with the glazier points stored within the Machine Shop. Nails present within the collection range in size fmm finishing nails (less than 1 112in.) to spikes
(greater than 5in.) and were most likely stored as well as used in the construction of the building. Aithough the nails are predominantly wiredrawn, machine-cut nails are present.
Summary
The Machine/blacksmith shop is clearly associated with industrial activities. The assemblage rewvered from this locale is significantly difFerent than any other excavated at the site. It was a specialized locale with activities dedicated to the upkeep of the settiement and the day ta day running of the tramline. Furthemore, it is evident that recycling metal for new parts or the re- adaptation of parts into other items was an essential activity.
The low proportion of domestic, personal management/clothing and leisure groups indicate that the structure was not used as a dwelling. The activity of blacksmithing is not well documented for the Klondike Gold Rush. It is Iikely that rnost settlements, even peripheral settlements like Canyon City, would have had a blacksmith within the community. Although beyond the scope of this thesis, the MachineBlacksmith Shop provides a unique look into the work of the Blacksmith in the Klondike. DISCUSSION AND CONCLUSIONS
The historic settlement of Canyon City arose as a direct result of the
Klondike gold rush. In 1897 a small company tom developed around the start of a wooden tramway enabling the multitudes of Kiondike bound stampeders to circumvent the treactierous Miles Canyon and White Horse Rapids. The primary purpose of Canyon City was to support the Miles Canyon and White
Horse Rapids Tramway Co.'s operation for the waves of gold seekers and tonnes of freight headed for Dawson. In June of 1900, the White Pass and
Yukon Railway Company wmpleted its rail line linking the coastal town of
Skagway, Alaska, with the newly emerging town of Whitehorse. This rail link rendered trail and water transportation services between Skagway and
Whitehorse non-essential. Like most of the small one industry-related settlements along the Upper Yukon River after the arriva1 of the railway,
Canyon City's importance to Yukon transportation waned, operations ceased. and the townsite was subsequently abandoned.
Archival records dealing with Canyon City are relatively scarce. It is clear, however, that the community of Canyon City was a company ovmed settlement devoted to the operation of the tramline. This was a seasonal endeavour, since the tramline and its docking facilities did not operate in the winter, Norman Macaulay established and owed this operation until 1899.
During this time, he initiated the construction of approximately 12 log structures and several tent structures including a roadhouse, a tramway office, a NWMP post and associated storehouse, stable, machine/blacksmith's shop.
a wooden wharf as well as a number of tents for storage and dwellings. The
resident population, wtiich is estimated from historic photographs, was
approximately 30 to 35 individuals and was predominantly male (Figure 28).
In August 1899, Macaulay sold his interests in the tramway to the White
Pass Railway, for $185,000. The tramline continued to operate for the next ten months, after wtiich operations ceased, allowing the railway a complete monopoly on the transportation of freight in the Upper Yukon.
Beginning in 1994 four field seasons of archaeology were conducted at
Canyon City that resulted in the documentation and mapping of 98 historic features, and the collection of 17,395 historic artifacts and 156 faunal specimens. Excavations concentrated on six major features or adivity areas at Canyon City that indude the West Tent locality, the Canyon Hotel and
Saloon, the North-West Mounted Police Barracks, the Canyon and White
Horse Rapids Tramway Company building, the "East Cabinn. and the
MachineIBlacksrnith shop. During the excavations at these localities architectural data was collectecl with particular reference to construction strategies and materials. Artifacts recovered from each locality provide information on activities carried out within each of the buildings tested.
The history and archaeology of Canyon City are unusual in the wntext of the Klondike gold rush in that Canyon City was primarily, if not exclusively, a single industry towsite that was not mining based. Furthemore, the almost
complete absence of postqold rush occupation of the site provides for a
material culture record that is largely lirnited to the gold rush period.
The archaeology of Canyon City offers an opportunity to describe a
peripheral Klondike gold rush era settlement. Specifically, the structural data
and the material culture rewvered frorn Canyon City provide insights into the
processes of a single industry settlement within the northern frontier. The
inhabitant's day to day lives appear to have been stnictured by the artificial
environment created by the wmpany. This structurïng is apparent in the
settlement characteristics of the site and in the nature of the subsistence base
of the occupants. The dominant type of abandonment apparent at Canyon
City does appear to have been as predideâ-planned permanent
abandonment. Within these areas data is lacking in the historidarchival
records of the Klondike gold rush. The excavation of Canyon City and the
analysis of the rnaterial culture appear to demonstrate that the single industry
focus of the townsite dictated the layout and composition of the tom's
infrastructure and activities; that the profitability of this gold rush period
transportation enterprise was sufficient to support a seasonal settlement
almost entirely supplied through external and rernote sources; and, that the
processes of abandonment at Canyon City were primarily detemined by
company interests. As a conclusion each of these issues is examined.
SmCHARACTERIST~CS
The gold rush occupation of Canyon City was but three hectic years.
To refer to Canyon City as a tom let alone a city is an overstatement. Canyon City was a wrk camp. Although having unique qualities, it was Iike many of the camps that arose along the Yukon River with the onset of the gold rush- That is, it had a dedicated transpoftation purpose. Within the period of the gold rush, however, it did not expand from a camp to a town, like Bennett or Dyea. The stampeders did not pause at this locale save for a few hours.
Therefore, without long-terni residence on the part of the stampeders, the population of Canyon City remained the same as did its level of services.
The work camp nature of Canyon City is readily apparent in the layout of the site. The map of Canyon City (Figure 7) and the archival photographs show that the gold rush occupation was highly structured and cannot be explained by the orientation of the River alone. Canyon City was wmprised of two east-wst ninning Iines of buildings and one north-south line. The first east-west line includes the "East Cabinn and Tramway office and the second comprises the tents CC85, CC84, CC75, the North-West Mounted Police
Storehouse and Barracks, the Canyon Hotel and Saloon and the three wall tents that make up the West Tent locality. The north-south line of structures are the Machine/Blacksmith Shop and the Stables that sit opposite each other separated by a spur of the tramline.
It is not unusual for a Company to organize its structures in such away.
What is somewhat unusual is the spatially strudured placement of the dwellings at Canyon City such as the West Tents. Photographs of Lindeman and Bennett (see Neufeld and Noms 1996) show that the business end of the townsites, such as the rail yards, were highly stnictured. The wall tents at these sites, wtiich are assumed to be dwellings, were scattered about in a haphazard fashion. This haphazardness did not occur at Canyon City during the gold rush occupation of the site even though, as it appears within the photographs, there were suitable, undeveloped areas for occupation. Another indication of the st~cturingof the camp is the concentration of metal container middens in the northwestern portion of the site. There are no indications of structures in this area, which suggests that the townsite had a designated area for refuse disposal. The placement of structures and other features at Canyon
City appears to have been largely detemined by Macaulay's plan of the site and considerations of continued use by him.
The archaeology identified the function of several of the features that had been previously unknown induding the West Tents as dwiellings, the East
Cabin as a dwelling, a portion of the Tramway Office as a dwelling and the
CC59 as a MachineIBlacksrnith shop. The Canyon Hotel and Saloon and
NWMP Barracks and the Miles Canyon and mite Horse Tramway building were identified through photographs. Besides identifying the activities that were carried out within each of these structures, the archaeological data indicate no duplication of commercial services. With thousands of gold seekers passing through the site on their way to the gold fields more than one restaurant or saloon likely could have been supported. lt is apparent, therefore. that Macaulay exercised monopolistic control over al1 economic activities camed out within the settlement. This interpretation is supported further by Macaulay's efforts to obtain a complete rnonopoly on freight transportation along this stretai of the Yukon River. In this respect he bought out his cornpetitor, John Hepbum, who had built a rival tramline operation on the opposite side of the river.
Building techniques at Canyon City are characteristic of a temporary camp without intention of long terrn use. All of the major structures were log and site preparation, if present at all, consisted of leveling the ground through excavation. The NWMP Barracks is the one exception. Prior to its construction trenches wre dug for the sill Iogs and a hem centre sill was used to stabilize the floor. More preparation was camed out on the NWMP barracks even though Macaulay wnstructed the structure. No doubt he had to follow NWMP guidelines as part of his contractual obligation.
The initial log structures built by Macaulay were expedient and hastily erected in preparation for the stampede to corne. For example, the initial section of the Tramway Office did not even have a floor. Over the next two years Macaulay built additions to these structures, but, except for the installation of floors, there was no real change in building techniques. That is, construction remained basic and utilitarian and balloon-framed, milled-timber buildings were absent. The expedient nature of this site is also emphasized by the fact that some of the wrkers at Canyon City were housed in tents. If the camp were intended to be even semi-permanent, it is likely bunkhouses or more substantial structures wuld have been built for the workers.
Canyon City artifacts were used not only to determine the function of buildings but to provide ches as to the general nature of settlement. It is notable that ceramics are nearly absent making up less than 0.1 percent of the collection. This is surprising in tM, respects. First, with the presence of a hotel and saloon, one might expect ceramic wares to have played an important role, especially in the restaurant fundion. Second, the near absence of ceramics is not duplicated in other household related artifacts, which account for 30 percent of the total assemblage. Two wntributing factors likely account for the low frequency of ceramics observed.
First, virtual lack of œramics may be explained by togistics, the utilitarian nature of the settlement and Macaulay's wncerns with profits. Tin or metal sewing dishes wuld likely have been the choiœ of the wmpany due to their durability, thereby avoiding the potential for broken dishes needing replacement in a region where logistics make the replacement of items an expensive exercise. The second factor is that ceramic consumption is greatly influenced by women (Klein 1991). Through examining different lsmcentury ceramic assemblages, Klien suggests that logistics is not the sole determinate influencing the frequency within an assemblage. Klien (1991 ) argues that the increased role of women in the dornestic sphere is also a major contributing factor. The Canyon City assemblage supports this conclusion. A single wman is present in but tvvo archival photographs of Canyon City (Figures 5 and 29) and it is difficult to canfirm whether or not these are the same women.
The assemblage contains four artifacts that can be confidently associated with the presence of women at the site; hm baleen fragments and a broach pin from the Tramway Office and one hair pin from the East Cabin. Coupled with Figure 23. Archival photograph of tramway crew, of note is the women seated behind Norman Macaulay the man with the cane Ca. 1899 (Yukon Archives H.C.Barley Coll.) the archival photographs, the assemblage suggests that -men were not - major contributors to the archaeological record at Canyon City; thefefore. supporting the site's demographics as male dominated.
The isolation and utilitarian nature of the site is best illustrated by the assemblage from the machine/blad Canyon City's isolation demanded that repairs be conducted on the spot with the materials on hand. Archibald (1981), for example, indicates that incoming supplies to the Klondike muid be received within a rnonth at the very earliest (Archibald 1981: 46-53). Another factor affecting the intensity of recycling would have been expense. For example, a simple repair chain link, an item that would have been utilized by the Tramway Company frequentiy, had cost $0.20 per dozen in 1910 (Hudson's Bay Company 1977) and each repair link weighs approximately 0.6 of a pound. Based on freight rates provided by Steele (1898), Bennett (1978) and Archibald (1981) it would have cost $6.04 to ship four dozen repair links to Canyon City via the Chilkoot trail or 7.5 times the original cost of $0.80. To ship the same units via the St. Michael's route was less expensive at $1.82 or 2.3 times the original cost but there would have been a time cost involved since this is a longer route than the Chilkoot. Despite the preceding statement, recycling and reuse of household goods was not evident ta any considerable extent at the site. A total of 740 tin cans (25 percent sarnple) wre documented within the tin cmmiddens in the northwstem portion of the site and less than 1 percent had reuse. Metal containers documented dunng excavations also did not show alterations suggesting reuse, except for a possible post-sleeve recovered from the Roadhouse. Metal container reuse appears to be fairly typical of Yukon gold rush era sites (Gates 1985). Gates (1985) doaiments tin can reuse in the gold fields and suggests it was a regular adivity of individual miners. Items reused include five-gallon tins cut open for roofing materials as well as pails and fruit and vegetable tins for sieves. Like the gold fields, it seems most likely that tin can reuse at Canyon City wuld have been camed out by individuals on an as- needed basis. Its near absence suggests that basic material needs were being met by the Company. A final consideration of Canyon City settlement patterns is its possible indication of status differentiation. The artifact assemblage itself does not support this statement in so far as the domestic assemblages from each of the localities are similar. However, if the type of domestic dwellings at Canyon City are looked at, some individuals Iived in tents Mile others including the NWMP were housed in log structures. The NWMP would have held a certain status within the community on the basis of their authority. Regarding the others, it appears that the only extravagance present at the site is the East Cabin. It is perched along the base of the eastern terrace surrounded by a picket fence with its own privy. It is an anomaly within the townsite. The cabin's location and amenities suggest that the individual(s) that lived there held some sort of distinct status within the camp. Although hypothetical. this could have been Norman Macaulay's cabin. The wman in Figure 5 standing outside the entrance of the East Cabins entrance may have been a resident of the cabin as well. The faunal rernains support a hypothesis that individuals residing within the East Cabin had a higher status within the community. The two assemblages with the highest faunal remains recovered include the West Tents (n=27) and the East Cabin (n=105). While the frequency of cuts within each assemblage are too small to accurately compare, the NlSP difference between the two assemblages may be an indicator of status. As discussed below, the occupants of the site wre dependent on imported goods; therefore, fresh meat muid Iikely have been a luxury. SETTEMENT SUBSISTENCE Archival data indicate that the occupants of Canyon City relied heavily on imported fwds. Macaulay supplied his wrkers with room and board and the NWMP received monthly rations. These rations wre merged with the Company's, and the NWPtook their meals within the Roadhouse with the rest of the workers. Dominated by tin cans, the artifac! assemblage recovered from Canyon City reflects this dependence. Faunal remains (n=156) are relatively scarce throughout the site. Only 20 identifiable bones are present and most of these are rabbit recovered from the East Cabin. Moose and Caribou are represented by one bone each. If the settlement was even moderately dependent on local resources for sustenance or if the Company routinely traded or brought in wild game, greater numbers should occur. The low frequency of faunal rernains does not imply that meat was not regularly consumed. Canned and preserved meat, such as bacon and wmed beef, Iikely wnstituted an important dietary base for the occupants and it is clear that canned fruits and vegetables were important staples as weII. The presence of fish hooks and firearms related artifacts indicate some srnall scale hunting and fishing. Nevertheless, this was likely for the occupants' enjoyment rather than for survival. ABANDONMENT Except for the NWMP, it can be expected that al1 occupants at the site were employees of the Miles Canyon and White Horse Tramway Company. Therefore, once the tramlines stopped operating and the new town of Whitehorse began to develop, Canyon City wuld have been quickly abandoned. NWMP logbooks indicate the Post at the townsite remained in operation, at least seasonally, until 1902. Occupation within the area did occur after 1900 but the extent of the occupation has yet to be confidently established. None of the artifacts recavered in buried deposits provide a terminus post quem after 1900. Today, berm outlines provide the only surface evidence of the buildings that once stood a Canyon City during the gold rush. Intact subsurface features, 5-20cm deep primarily wnsist of sill logs, floor joists and the occasional floor plank; materials that wre presumably unworthy of salvaging. The lack of surface or subsurface remains suggests that buildings were systematically dismantled and moved. The most likely place for reuse of these materials would have been in the nearby tom of Whitehorse. During the winter of 1900 both buildings and businesses were moved to Whitehorse from other short-lived Yukon River settlements such as Bennett City, British Columbia (ingram and Dobrowolsky 1994: 7). Although not confirmed, John Hatch, a Yukon old-timer, has indicated that he now resides in hm sections from the Canyon Hotel and Saloon. Presumably these had been used to constnid part of the Closeleigh Hotel and Saloon, later named the Pioneer Hotel, located in Whitehorse (John Hatch personal communication, 1995). Not al1 of the structures were dismantled at the end of the gold rush. Photographs dated to 1912 show two buildings at Canyon City (Figure 14). Two of the photographs were taken from the dock and dearly depict the NWMP Post with its roof missing. The angle of the photograph shows the rest of the townsite except for the eastem corner. No other structures are standing, demonstrating the Tramway OfFice, Storehouse, Roadhouse, West Tents, Machine/Blacksmith Shop and the stable had been dismantled prior to 1912. Two additional photographs show the front face of an abandoned Cabin, similar to the East Cabin. This structure, however, appears to be located on the top of the eastem terrace of the site and likely represents the post-gold rush occupation of the site. It appears that tM, types of abandonment behaviors are displayed at Canyon City. The first is planned permanent abandonment (Mills 1995; Schiffer 1989; Stevenson 1982). This irnplies that the occupants or users of an area have previously made the decision to leave and not retum. lnherent in this type of abandonment is a decision over what items to take and what items to leave behind. Usable goods are usually rernoved during planned abandonment. One of the most important factors infiuencing this decision is logistics, or the distance to the next habitation and the available modes of transportation to get there (Mills 1995). Whitehorse's growth, proximity and accessibility-the river and the tramline-suggests there wrefew logistical obstacles to overcome when abandoning Canyon City. Therefore, in locations where planned, permanent abandonment processes are at work there should be a low frequency if not a total lack of usable items. Two other important factors infiuencing this type of abandonment, which are more site specific. must also be taken into consideration. There were two groups or decision-makers present at Canyon City during its abandonment: 1) the workers and 2) the company (this includes the Tramway company and the NVVMP administration). It is likely that both wuld have had different wncems regarding abandonment decisions. Because of the camp- like nature of this settlement, the wrkers probably were most concerned with personal items including clothing, personal hygiene, adornment and possibly some household items. The wmpany, on the other hand, would have been concerned about its assets ranging from buildings and machinery to hammers and axe files. Five of the six structures-the West Tent locality, the Canyon Hotel and Saloon, the NWMP Post. the Canyon and White Horse Rapids Tramway Office. and the MachinelBlacksmith shop-undemnt planned, permanent abandonment. The NWMP barracks appears to have been the only one of these structures not to be dismantled during abandonment. A detachment had already been established at Whitehorse by 1902, the year the Canyon City force was removed. This wuld have reduced, if not eliminated the need to move the structure. The four dismantled structures most likely represent the liquidation of the tramway operation's Canyon City assets. However, with a ready market for building materials but five miles down river in the growing tom of Whitehorse. the company liquidated their remaining assets at the site through their removal. In so far as usable artifads were scarce. planned abandonment is also reflected in the excavation assemblage. Another factor in the systematic dismantlement of the settlement is large scale economics. White Pass and Yukon Railway were the indirect owners of the settlement in its post 1899 days. The railway company on several occasions bought out potential competition only to shut it dom thereby further establishing its monopoly. Bennett (1978: 64) states that by 1907 the Railway's aggressive corporate strategy of eliminating competition was successful. It is likely that the operations at Canyon City were one of the several victims of the large corporation. The East Cabin seems to be the only structure at the site that entered the archaeological record ~Îthoutbeing removed or scavenged. Rather, it bumed with its contents still inside. Consequently, the East Cabin has a rich artifact assemblage induding the highest frequency of usable artifacts. Artifacts recovered from the charred floor indude, a tin caffee pot, a metal fry pan, a white enamel tin plate and a number of files. Other items that wre most likely usable but were recovered in a highly fragmented state include a ferrous pot, a ferrous bowi and plate and at least five cutlery pieces. As well, unlike the other structures that at one time had stoves, four stove legs were recovered CONCLUS~~N The results from the archaeological investigations of Canyon City provide a glimpse into the nature of settlement integral to but on the periphery of one of the largest gold rushes the world has seen. Canyon City was a transportation camp on the Klondike frontier. Wth the end of the Klondike gold rush there was no further requirement for a hastily wnstructed horse drawn traml ine. Much of the tom's infrastructure was systematically dismantled and removed but traces of this settlement are still preserved in the archaeological record. Four seasons of excavations have provided limited new insights into the history of Canyon City but added to our knowledge of a settlement that is usually considered a way point on the Trail of 98. There are many questions still to be answered such as the nature and extent of the post gold rush occupation of the site or how typical the cultural processes at Canyon City were as compared to other sites along the Yukon River during the gold rush era. Work is needed at similar settlements to better illuminate their role in the Great Starnpede of '98 and their significanœ in the development of the Yukon. The use of the archival and oral history data coupled the archaeological study of a settlement integral to the Klondike gold rush provides a fresh perspective on, as Pierre Burton States, "one of the best documented adventures of the last 150 years." REFERENCES ClTED Barnes. F. 1985 Cartridges of the Wdd, Sm Edition. K Wamer (ad.). DBI Books Inc., USA. Bennett, G. 1978 Yukon Transpottation: A Histoty. Canadian Historic Sites Occasional Papers in Archaeology and History No. 19. Parks Canada, Ottawa. Blee, C. 199 1 Sorting functionally-mixed artifad assemblages with multiple regression: A comparative study in historical archaeology. Unpublished Ph. D. dissertation, University of Colorado, Boulder. Burley, O. 1985 Historical Archaeology and the Cognitive Paradigm: Developing a Research Frarnework for the Yukon gold Rush Frontier. The Western Canadian Anthmpologist 2: 15-25 Burfey, D. and B. Ross 1979a The North Dawson Hillside Prow: A Final Report- Park Canada Manuscript Report No. 384, Parks Canada. 1979b The Dawson City A~chaeollogicalPmgram: Structural Repod for 1978 Operations. Parks Canada Manusmpt Repori No. 384' Parks Canada. Burton, P. 1972 Klondike: The Last Great Gold Rush 1896-1899. McClelland and Stewart Limited, Toronto. Clague, J., S. Evans, V; Rampton, and G. Woodsmrth 1995 lmproved age estimates for the White River and Bridge River tephras, western Canada. Canadian Journal of Earth Scr'ences 32L 1172-1 179. Clark, S. 1942 Mining Society in British Columbia and the Yukon. In The Social Development of Canada, S.D. Clark (author), pp. 308-379. University of Toronto Press, Toronto. Coates, K. 1985 Canada's Colonies: A History of the Yukon and Northwlest Temtones. James Lorimer 8 Company, Publishers, Toronto. Coates, K. and W. Monison 1988 The Gold Rush, 1896-1900. In Land of the Midnight Sun: A History of the Yukon, Coates, US.& W.R. Momson (authon), pp. 77-1 15. Hurtig Publishers, Edmonton. 1989 Transiency in the Far Northwest After the Gold Rush: The Case of the Pnncess Sophia. In Interpmting Canada's North: Selected Readings, K-S. Coates & W.R. Morrison (eds.), pp. 185-198. Copp Clark Pitman Ltd., Toronto. Cooper, D. 1998 Archeological Investigations in Skagway, Volume 6: Residential Life on Block 39. U.S. Deparment of the Interior, Klondike Gold Rush National Historical Park, Skagway Alaska. Dawson, G. 1889 Report on an Exploration in the Yukon District, N. W. T- and Adjacent Notthem Portion of British Columbia 7887. Geological and Natural History Survey of Canada Annual Report Volume III, Part 1. William Foster Brown 8 Co., Montreal- Deagan, K. 1982 Avenues of lnquiry in Historical Archaeology. In Advances in Archaeological Method and Theory, vol. 5., pp. 151 -1 77 , Academic Press, USA. 1988 Neither History Nor Prehistory: the Questions that Count in Historical Archaeology. Historïcal Archaeology 22(1): 7-1 2. Dobrowolsky, H. 1990 Miles Canyon: Macauley and Hepbum Tramways. In Study Tour of the Yukon and Alaska. Society for Industrial Archaeology, Ottawa. 1995 Law of the Yukon: A Pictonal History of the Mounted Police in the Yukon. Lost Moose Publishers, VVhitehorse. Durden, F. 1980 The Development of the Non-Native Settlement Pattern of the Yukon Territory. Canadian Issues Z(2): 11-31. Easton, N. 1987 Hetitage Resources lnventory of the Yukon Underwater Diving Association. Prepared for the Yukon Heritage Branch, Y.T.G.. Report on file at the Yukon Heritage Branch, Whitehorse. Fontana, B. and J. Greenleaf 1965 Johnny Ward's Ranch: A Study in Historic Archaeology. Kiva 28 (1-2): 1-1 15. Friesen, R. 1978 The Chilkoot Pass and the Great Gold Rush of 7898. Manuscript Report Series No. 236. Parks Canada, Ottawa. Gates, M. 1985 Modification and Re-Use of Arîifacts of the Klondike Region, Yukon Temtory. Unpublished Paper. 1994 Gold at Fortymile Creek Early Days in the Yukon. University of British Columbia Press, Vancouver. Gotthardt, R. 1994 Archaeological investigations at Canyon City in 1993. Report on File with Heritage Branch, Govemment of Yukon, Whitehorse. Hammer, T. 1997 The Canyon City Archaeology Projd, 1996. Ms on file at Heritageeranch Government of Yukon, Whitehorse. Hardesty, O. 1988 The Archaeology of Mining and Miners A Vïew from the Silver State. Special Publication Series, No. 6. Society for Historia1 Archaeology, U.S.A. Hare, P. and T. Hammer 1 995 The Canyon City Archaeology Project, 1 994. Ms on file at Heritage Branch Government of Yukon, Whitehorse. Haskell, W. 1898 Twyears in the Klondike and Alaskan Gold -Fields: 1896 1 898. Hartford Publishing Company, Connecticut. Hems, D. and P. Nieuwhof 1994 Bennet City: a Gold Rush Phenornena. Research Bulletin No. 305. Parks Canada. Hitchock, M. E. 1899 Two Women in the Klondike. GB Putnam's 8 Son, U.S.A. Horback, H. 1976 Yukon News (Silver Anniversary Edition). Yukon Archives 8011 13 MSS 145. Hudson's Bay Company 1 977 Autumn and Mnter Catalogue 19 10- 19 11. Original1y published by Watson & Dwyer Publishing Ltd., 1910. Hunt, W.R. 1974 Part II: The Klondike Era. In Noeh of 530, W.R. Hunt (author), pp. 29-94. MacMillan Publishing Co. Inc., New York. Ingram, R. and H. Dobro~~~lsky 1994 Edge of the River, Heart of the City. Yukon Historical and Museums Association, Whitehorse. Innis, H. 1936 Settlement and the Mining Frontier. In Canadian Frontiers of Settlement, W.A. Mackintosh and W.L.G. Joerg (eds. ), pp. 178-269. The MacMillan Company of Canada Ltd., Toronto. Jackson, L. and T. MacKay 1990 Glacial tirnits and ice-Flow Directions of the Last Cordilieran /ce Sheet in Yukon Tenitory between 60 and 63 Degrees North. Map O.F. 2329, Geological Survey of Canada. Johnston, D. n.d. Canying the Tenitory: Silver-Lead Mining in the Mayo District, 1919-1942. In Essays on the Political Ewnomy of the Yukon Territory, R. Stuart (ed.). Unpublished. Jones, 0. and C. Sullivan 1989 The Parks Canada Glass Glossary. Studies in Archaeology Architecture and History. National Historic Parks and Sites. Canadian Parks Service, Environment Canada, Ottawa. Kent, S. 1990 A Question of Inference: lnterpretations of Site Abandonment Behavior from Survey Data. Historical Archaeology 20: 7081. Klein, T. 1991 Nineteenth-Century Ceramics and Modeis of Consumer Behavior. HistoricalArchaeology 25(2):62-76. Klirnko, 0.and J. Hodges 1993 Last Mountain House. Western Heritage Services Inc., Saskatoon. Knuston, A. 1991 "Can We Do 1T?" "Hell, We Gotta!". Knutson Enterprises, Washington. McCandless, R. 1985 Yukon Wldlfe-A Social History. University of Alberta Press, Edmonton. McClellan, C. 1975 My Old People Say, Part 1. National Museum of Man Publications in Ethnology 6(1). National Museums of Canada. Ottawa. Mills, R. 1995 Historic Archaedogy and Abandonment: A Study in Site Formation Processes. Paper Presented at the 22nd Annual Meetings of the Alaska Anthropological Association, Anchorage. Mini, S. 1978 Archaeological Eiploration of the Klondike Histonc Sites 19 76 and 1977. Manuscript Report Number 309, Parks Canada. Ottawa. Minter, R. 1 987 The White Pass: Gateway to the Klondike. McClelland and Stewart, Toronto. Morrison, W 1974 The North-West Mounted Police and the Klondike Gold Rush. Journal of Contemporary History 9(2): 93-1 06. 1985 Showing the Flag. University of British Columbia Press, Vancouver. 1989 Eagle Over the Arctic: Americans in the Canadian North, 1867- 1985. In lnterprating Canada's North: Selected Readings, K.S. Coates 8 W.R. Morrison (eds.), pp. 169-184. Copp Clark Pitman Ltd., Toronto. 1990 Policing the Boomtown: The Mounted Police as a Social Force in the Klondike. Norfhern Review 6: 83-97. Murray, J. and 3. Hamilton 1986 Cultural Resource Inventory, Proposed Chilkoot Trail National Historie Park: Summary of 1984 Archaeological Survey. Microfiche Report Series 349, Environment Canada and Canadian Parks Service, Ottawa. Neufeld, D. and F. Norris 1996 Chilkoot Trail: Heritage route to the Klondike. Lost Moose Publishers, Mitehorse. Newell, Diane 1986 Importance of Information and Misinformation in the Making of the Klondike Gold Rush. Journalof Canadian Studies21(4):95-7 11. Ogilvie, W. 191 3 Early Days on the Yukon and The Story of b Gold Finds. Thorbum & Abbott, Ottawa. Oswalt, E. and J. Senyk 1977 Ecorepions of Yukon Temtory. Canadian Forestry Service Pacific Research Centre, Victoria. Phiiiips, R. and D. Klancher 1982 The first One Hundred Years: Arrns and Accoutrements of the Mounted Police 1873-1973- Museum Restoration Service, Ontario. Powers, M. 1974 Modem Man to Hunting Band: A Readaptation. M.A. Thesis, Department of Sociology and Anthropology, Simon Fraser University. Price, J. 1898 From Euston to Klondike. Sampson, Lowe, Marston and Co., London. Rickard, T.A. 1909 Through the Yukon and Alaska. Mining and Scientific Press, San Franciso. Rock, T. 1983 Cans in the Countryside. Histoncal A~nhaeology18: 97-11 1. Schwatka. F. 1893 A Summer in Alaska. J.W. Henry, St. Louis. Scuwey, S. 1 995 Canyon City Oral History: Intervie w with Louie Smith. August 2. 1995. Unpublished. Shulz, P. and S. Gust 1983 Faunal Remains and Social Status in 1gth Century Sacramento. Histoncal Archaeology 1 7(1): 44-53. Schiffer, M. 1989 Formation Processes of the Archaeological Record. University of New Mexico Press, Albuquerque. Sprague, R. 1981 A Functional Classification for Artifads from 1gm and 2om Century Historical Sites. North American Archaeologist 2(3): 25 1-26 1. Spude, C. and D. Scott 1993 Father Tumell's Trash Pit. U.S. Department of the lnterior Archeological Investigations in Skagway, Alaska, Volume 4. Klondike Gold Rush National Historical Park, Skagway Alaska. Steeie, SB. 1898 North- West Mounted Police, Upper Yukon District Lake Bennett, B.C., July 9, 1898. Yukon Archives. 19 15 Forty Years in Canada. Dodd, Mead and Company, New York. Stevenson, MG. 1980 Looking for Gold: Historic Sites Survey of Kluane National Park, Southwest Yukon. Unpulbishes M.A. Thesis, Simon Fraser University, Bumaby. 1982 Toward an Understanding of Site Abandonment Behavior: Evidence frorn Historic Mining Camps in the Southwest Yukon. Journal of Anthmpological Archaedogy 1 : 237-265. 1989a Sourdoughs and Cheechakbos: The Formation of Identity- Signaling Social Groups. Journal of Anthropological Archaeology 8: 270-312. 1989b Sourdoughs and Cheechakaos: The Formation of Identity- S ignaling Social Groups. Journal of Anthroplogical Archaeology 8: 270-31 2. The Klondike Nugget 1899 Saturday July 8. Dawson City, Yukon. Toulouse, J. 1971 Bottle Makers and Their Marks. Thomas Nelson Inc., New York. WaIley n.d. Notes on file at the MacBnde Museum, VVhitehorse, Yukon. Webb, M. 1993 Yukon: The Last Fmntier. University of British Columbia Press, Vancouver. Wood, Z.T. 1898a Annual Report of Superintendent Z.T. Wood. November. Yukon Archives 351 740 62. 1898b North- West Mounted Police, Tagish, Upper Yukon. November 1d", 7898, to the Comptroller N.W. M. P.. Ottawa. Yukon Archives. 1898c North-West Mounted Police, Tagish, Upper Yukon, October 1, 1898. Report for petiod ending Seplember 30, 1898. Yukon Archives. 18984 North-West Mounted Police Annual Report of Superintendent A.T. Wood, In Appendix A of Annual Report of Superintendent Steele. Yukon Archives. 1899 Appendix A. In A.B. Peny's Annual Report of the North- West Mounted Police. Yukon Archives. Wright, A. 1 976 Prelude to Bonanza. Gray's Publishing Ltd., Sidney. British Columbian. Zaslow, M. 1989 The Yukon: Northem Development in a Canadian-American Context. In lnterpreting Canada 's North: Selecîed Readings. K. S. Coates 8 W.R. Morrison (eds.), pp. 733-148. Copp Clark Pitman Ltd.. Toronto.