Open Collections

BC Historical Books

BC Historical Books

BC Historical Books

Camping in the Canadian Rockies : an account of camp life in the wilder parts of the Canadian Rocky mountains,… Wilcox, Walter Dwight, 1869-1949 1897

Item Metadata

Download

Media
bcbooks-1.0340553.pdf
Metadata
JSON: bcbooks-1.0340553.json
JSON-LD: bcbooks-1.0340553-ld.json
RDF/XML (Pretty): bcbooks-1.0340553-rdf.xml
RDF/JSON: bcbooks-1.0340553-rdf.json
Turtle: bcbooks-1.0340553-turtle.txt
N-Triples: bcbooks-1.0340553-rdf-ntriples.txt
Original Record: bcbooks-1.0340553-source.json
Full Text
bcbooks-1.0340553-fulltext.txt
Citation
bcbooks-1.0340553.ris

Full Text

  
THE LIBRARY
THE UNIVERSITY OF
BRITISH COLUMBIA
Gift of
University of Toronto
à        Mount Assiniboine. CAMPING IN THE
CANADIAN ROCKIES
AN ACCOUNT OF CAMP LIFE IN THE WILDER PARTS OF
THE  CANADIAN   ROCKY  MOUNTAINS, TOGETHER
WITH A DESCRIPTION OF THE REGION ABOUT
BANFF,  LAKE  LOUISE, AND GLACIER,
AND A SKETCH OF THE EARLY
EXPLORATIONS
BY
WALTER DWIGHT WILCOX
SECOND EDITION, WITH MAP
WITH TWENTY-FIVE FULL-PAGE  PHOTOGRAVURES, AND MANY TEXT
ILLUSTRATIONS FROM PHOTOGRAPHS  BY THE AUTHOR
G. P. PUTNAM'S SONS
NEW YORK LONDON
27 West Twenty-third Street 24 Bedford Street, Strand
&§t JLnitkubocker |)rcss
1897
*m Copyright, i8q6
by
G. P. PUTNAM'S SONS
Entered at Stationers' Hall, London
Ube Knickerbocker press, flew Borft
i PREFACE.
THE Canadian Rocky Mountains offer exceptional
attractions to those who enjoy natural scenery»
sport, and camp life. Few regions of the world
combining mountain, lake, and forest scenery possess the
additional advantage of a delightful summer climate, such
as obtains in the Canadian Rockies.
The extremely wild character of this part of the Rocky
Mountains, and the very short time since it was opened
up to travellers, are probably, in great part, the reasons
for the lack of literature and the absence of any thoroughly illustrated publication concerning this region.
During a period of four years, the author has made
camping excursions into many of the wilder parts of the
mountains and effected a considerable number of ascents.
An excellent camera has been an almost inseparable companion in every excursion, so that photographs of the
typical scenery have been obtained from every possible
point of view. Moreover, throughout all the processes of
photographing, no expense of time or labor has been
spared in order to obtain true and artistic representations
of nature.    Nor have these results been obtained without iv Preface.
considerable sacrifice, for in many cases the proper light
effects on lakes and forests required hours of delay, and
frequently, on lofty mountain summits, high winds made
it necessary to anchor the camera with stones ; while the
cold and exposure of those high altitudes made the circumstances unfavorable for successful work.
The author makes grateful acknowledgment of the
assistance received from many friends in the preparation
of this book. Special thanks are due to Prof. J. H. Gore,
of Columbian University, and to the Hon. Chas. D. Wal-
cott, Director of the United States Geological Survey, for
the valuable aid and information given by them ; to M.
Guillaume La Mothe for an interesting letter concerning
the first exploration of the Fraser River ; and to Sir
William Van Home for the many courtesies extended.
Washington, D.C., July, 18
W. D. W. CONTENTS.
CHAPTER I.
Banff—Its Location—The Village—Tourists—Hotels—Topography
of the Region—Rundle and Cascade Mountains—The Devil's Lake—Sir
George Simpson's journey to this Region—Peechee the Indian Guide—
An Indian Legend—The Missionary Rundle—Dr. Hector—The Climate
of Banff—A Summer Snow-Storm—The Mountains in Winter        I-I5
CHAPTER II.
Lake Louise—First Impressions—An Abode of Perpetual Winter—
The Chalet—Visitors—Stirring Tales of Adventure—Primeval Forests—
Forest Fires—Mosquitoes and Bull-Dog Flies—Mortal Combats between
Wasps and Bull-dogs—The Old Chalet—Morning on the Lake—Approach
of a Storm—Sublimity of a Mountain Thunder- Storm—Cloud Effects—
The Lake in October—A Magnificent Avalanche from Mount Lefroy—
A Warning of Approaching Winter ......      16—35
CHAPTER III.
Surroundings of the Lake—Position of Mountains and Valleys— The
Spruce and Balsam Firs—The Lyall's Larch—Alpine Flowers—The
Trail among the Cliffs—The Beehive, a Monument of the Past—Lake
Agnes, a Lake of Solitude—Summit of the Beehive—Lake Louise in the
Distant Future .........      36-46 vi Contents.
CHAPTER IV.
Organizing a Party for the Mountains—Our Plans for the Summer—
William Twin and Tom Chiniquy—Nature, Habits, and Dress of the
Stoney Indians—An Excursion on the Glacier— The Surface Debris and
its Origin—Snow Line—Ascent of the Couloir—A Terrible Accident—
Getting Down—An Exhausting Return for Aid—Hasty Organization of
a Rescue Party—Cold and Miserable Wait on the Glacier— Unpleasant
Surmises—"I Think You Die"—A Fortunate Termination    .      47-64
CHAPTER  V.
Castle Crags—Early Morning on the Mountain Side— View from the
Summit—Ascent of the Aiguille—An Avalanche of Rocks—A Glorious
Glissade—St. Piran—Its Alpine Flowers and Butterflies—Expedition to
a?i Unexplored Valley—A Thirsty Walk through the Forest—Discovery
of a Mountain Torrent—A Lake in the Forest—A Mountain Amphitheatre— The Saddle—Impressive View of Mount Temple—Summit of
Great Mountain—An Ascent in Vain—A Sudden Storm in the High Mountains—Phenomenal Fall of Temperature—Grand Cloud Effects,      65-83
CHAPTER VI.
Paradise Valley—The Mitre Glacier—Air Castles—Climbing to the
Col—Dark Ice Caverns—Mountain Sickness—Grandeur of the Rock-
Precipices on Mount Lefroy—Summit of the Col at Last—A Glorious
Vision of a New and Beautiful Valley—A Temple of Nature—Sudden
Change of Weather—Temptation to Explore the New Valley—A Precipitate Descent—Sudden Transition from Arctic to Temperate Conditions—
Delightful Surroundings—Weary Followers—Overtaken by Night—A
Bivouac in the Forest—Fire in the Forest—Indian Sarcasm,    84—100
CHAPTER VII.
The Wild Character of Paradise Valley—Difficulties with Pack-
Horses—A Remarkable Accident—Our Camp and Surroundings—Animal Contents.
Friends—Midsummer Flowers—Desolation Valley—Ascent of Hazel Peak
—An Alpine Lake in a Basin of Ice—First Attempt to Scale Mount Temple
—Our Camp by a Small Lake—A Wild and Stormy Night—An Impassable
Barrier—A Scene of Utter Desolation—All Nature Sleeps—Difficulties
of Ascent— The Highest Point yet Reached in Canada—Paradise Valley in
Winter—Farewell to Lake Louise  ......    101—118
CHAPTER VIII.
The Selkirks—Geographical Position of the Range—Good Cheer of
the Glacier House—Charming Situation—Comparison between the Selkirks
and Rockies—Early Mountain Ascents—Density of the Forest—Ascent of
Eagle Peak—A Magnificent Panorama—A Descent in Hie Darkness—
Account of a Terrible Experience on Eagle Peak— Trails through the
Poorest—Future Popularity of the Selkirks—The Forest Primeval—
An Epitome of Human Life—Age of Trees—Forests Dependent on
Humidity        . .......    119-136
CHAPTER IX. -
Mount Assiniboine—Preparations for Visiting it—Camp at Heely's
Creek—Crossing the Simpson Pass—Shoot a Pack-Horse—A Delightful
Camp—A Difficult Snow Pass—Burnt Timber—Nature Sounds—Discovery of a Beautiful Lake—Inspiring View of Mount Assiniboine—
Our Camp at the Base of the Mountain—Summer Snow-Storms—Inaccessibility of Mount Assiniboine            137—157
CHAPTER X.
Evidence of Game—Discovery of a Mountain Goat—A Long Hunt—
A Critical Moment—A Terrible Fall—An Unpleasant Experience—
Habitat of the Mountain Goat—A Change of Weather—A Magnificent
Panorama—Set out to Explore the Mountain—Intense Heat of a Forest
Fire—Struggling with Burnt Timber—A Mountain Bivouac—Hope and
Despair—Success at Last—Short Rations— Topography of Mount Assiniboine—The Vermilion River—A Wonderful Canyon—Fording the Bow
River 158-182 Contents.
CHAPTER XL
The Waputehk Range—Height of the Mountains— Vast Snow Fields
and Glaciers—journey up the Bow—Home of a Prospector—Causes and
Frequency of Forest Fires—A Visit to the Lower Bow Lake—Muskegs —
A Mountain Flooded with Ice—Delightful Scenes at the Upper Bow
Lake — Beauty of the Shores — Lake Trout — The Great Bow
Glacier ...........    183-204
CHAPTER XII,
Sources of the Bow— The Little Fork Pass—Magnificence of the
Scenery—Mount Murchison—Camp on the Divide—A High Mountain
Ascent—Future of the Bow Lakes—Return down the Bow—Search for a
Pass—Remarkable Agility of Pack-H or ses—The u Bay" and the "Pinto "
—Mountain Solitudes—Mount Hector—Difficult Nature of Johnston
Creek — A Blinding Snow-Storm — Forty-Mile Creek—Mount Edith
Pass 205-219
CHAPTER XIII.
HISTORICAL.
Origin and Rise of the Fur Trade— I he Coureurs des Bois and the
Voyageurs—Perils of the Canoe Voyages—The Hudson Bay Company and
the Northwest Company—Intense Rivalry—Downfall of the Northwest
Company—Sir Alexander Mackenzie—His Character and Physical Endowments — Cook's Explorations — Mackenzie Starts to Penetrate the
Rockies—The Peace River—A Marvellous Escape—The Pacific Reached
by Land—Perils of the Sea and of the Wilderness .        .        .    220-236
CHAPTER XIV.
HISTORICAL.
Captain Cook's Explorations—The American Fur Company—First
Exploration of the Fraser River—Expedition of Ross Cox—Cannibalism
—Simplicity of a Voyageur—Sir George Simpson's Journey—Discovery Contents.
i
of Gold in 1858—The Palliser Expedition—Dr. Hector's Adventures-
Milton and Clieadle—Growth of the Dominion—Railroad Surveys—
Construction of the Railroad—Historical Periods—Future Popularity
of the Canadian Rockies       .        .        .        .        .        . 237~257
CHAPTER XV.
The Pleasures of the Natural Sciences—Interior of the Earth—
Thickness of the Crust—Origin and Cause of Mountains— Their Age atid
Slow Growth—System in Mountain Arrangement—The Cordilleran System—The Canadian Rockies—Comparison with Other Mountain Regions—
Climate—Cause of Chinook Winds—Effect of High Latitude on Sun and
Moon—Principal Game Animals—Nature of the Forests—Mountain
Lakes—Camp Experiences—Effect on the Character        .        .    258-275
Index
277-283
^J  FULL-PAGE PHOTOGRAVURES.
Mount Assiniboine	
Banff Springs Hotei	
Bow River and Cascade Mountain
Lake Louise	
Mount Lefroy and Mirror Lake
LAKE   AGNES   (In early July, 1895)
TOM   CHINIQU Y   (By courtesy of Mr. S. B. Thompson, New We
Frontispiec
Mount Temple, from the Saddle
Discovery of Paradise Valley    .
Camp in Paradise Valley
Mount Sir Donald, from Eagle Peak
Head of Rocky Mountain Sheep
North Lake      .       .       .
Summit Lake, near Mount Assiniboine
Head of Rocky Mountain Goat (shot juiy rs, ^5)
THE   WAPUTEHK   RANGE   (Looking across the range from near I
4
IO
l8
38
42
50
78
92
IO8
126
132
152
154
I64
184 mmm
Full-Page Photogravures.
Mount Daly  192
Upper Bow Lake (Looking east)  196
Upper Bow Lake (Looking west)  200
Source of the Little Fork of the Saskatchewan
River  206
Storm in Little Fork Valley     ..... 208
Mount Hector and Slate Mountains
(From summit of a mountain near Little Fork Pass, 10,125 feet in altitude)      . 2IO
Camp at Little Fork Pass  212
Upper Bow Lake (Looking south)  270
Emerald Lake and Mount Field       .... 272 ILLUSTRATIONS IN THE TEXT.
Rundle Mountain and Bow River
LAKE   LOUISE   Crooking toward  chalet)    .
Anemones   . ■    .
A Cool Retreat in the Forest
Summit of Mount Temple
Glacier House
Peyto ....
Packing the Buckskin
Calypso
Approaching the Pass
NORTH   LAKE   (Looking northwes
Haunt of the Mountain Goat
Mount Assiniboine (From northwest)
Lake on Vermilion Pass
Ready to March
Camp at Upper Bow Lake
The " Bay "...
Falls of Leanchoil
15
31
40
75
115
120
140
142
143
149
157
165
167
181
186
202
214
249  CAMPING IN
THE CANADIAN ROCKIES.
CHAPTER I.
Banff-—Its Location—The Village—Tourists—Hotels—Topography
of the Region—Rundle and Cascade Mountains—The Devil's Lake—Sir
George Simpson's Journey to this Region—Peechee the Indian Guide—
An Indian Legend—The Missionary Rundle—Dr. Hector—The Climate
of Banff—A Summer Snow-Storm—The Mountains in Winter.
THE principal resort of tourists and sportsmen in
the Rocky Mountains of Canada is Banff. The
location of the town or village of Banff might be
briefly described as being just within the eastern-most
range of the Rocky Mountains, about one hundred and
fifty miles north of the International boundary, or where
the Canadian Pacific Railway begins to pierce the complex system of mountains which continue from this point
westward to the Pacific coast.
Banff is likewise the central or focal point of the
Canadian National Park. There is so much of scenic
interest and natural beauty in the surrounding mountains
and valleys, that an area of some two hundred and sixty îîiCTfmfff]
The Canadian Rockies.
square miles has been reserved in this region by the government and laid out with fine roads and bridle-paths
to points of special interest. Order is enforced by a body
of men known as the Northwest Mounted Police, a detachment of which is stationed at Banff. This organization
has been wonderfully effective for many years past in
preserving the authority of the laws throughout the vast
extent of northwestern Canada by means of a number of
men that seems altogether insufficient for that purpose.
The small and scattered village of Banff occupies a
flat plain near the Bow River. This large stream, the
south branch of the Saskatchewan, one of the greatest
rivers of North America, is at this point not only deep
arid swift but fully one hundred yards in width. A fine
iron bridge spans the river and leads to the various hotels
all of which are south of the village. The permanent
population numbers some half thousand, while the various
stores, dwellings, and churches have a general air of neatness and by their new appearance suggest the fact that
the history of Banff extends back only one decade.
During the summer season, the permanent population
of Banff is sometimes nearly doubled by a great invasion
of tourists and travellers from far distant regions. Overland tourists from India, China, Ceylon, and England, the
various countries of Europe and the Dominion of Canada,
but chiefly from the United States, form the greater part
of this cosmopolitan assemblage, in which, however,
almost every part of the globe is occasionally represented.
Some  are bent   on   sport   with   rod   or gun ; others on Tourists. 3
mountaineering or camping expeditions, but the great
majority are en route to distant countries and make Banff
a stopping-place for a short period.
Arrived at Banff, the traveller is confronted by a
line of hack drivers and hotel employes shouting in loud
voices the names and praises of their various hotels.
Such sights and sounds are a blessed relief to the tourist,
who for several days has witnessed nothing but the
boundless plains and scanty population of northwestern
Canada. The chorus of rival voices seems almost a welcome back to civilization, and reminds one in a. mild
degree of some railroad station in a great metropolis.
On the contrary, the new arrival finds, as he is whirled
rapidly toward his hotel in the coach, that he is in a mere
country village surrounded on all sides by high mountains,
with here and there patches of perpetual snow near their
lofty summits.
Though the surrounding region, the adjacent mountains, and valleys represent nature in a wild and almost
primitive state, one may remain at Banff attended by all
the comforts of civilization. The several hotels occupy
more or less scattered points in the valley south from the
villaee. The one built and managed by the railroad
stands apart from the village on an eminence overlooking
the Bow River. It is a magnificent structure capable of
accommodating a* large number of guests. From the
verandas and porches one may obtain a fine panoramic
view of the surrounding mountains, and on the side
towards the river the view combines water, forest,  and The Canadian Rockies.
mountain scenery in a most pleasing manner. The Bow
River, some three hundred feet below, comes in from the left
and dashes in a snowy cascade through a rocky gorge, then,
sweeping away towards the east, is joined by the Spray
River, a mad mountain torrent deep and swift, but clear
as crystal, and with cold water of that deep blue color
indicating its mountain origin. The wonderful rapidity
with which these mountain streams flow is a source of
astonishment and wonder to those familiar only with the
sluggish rivers of lowland regions. Standing on the little iron bridge which carries the road across the stream
and looking down on the water, I have often imagined I
was at the stern of an ocean greyhound, so rapidly does
each ripple or inequality sweep under and away from the
eye. Though the water is less than a yard in depth, the
current moves under the bridge at the rate of from nine
to ten miles an hour.
The best point from which to get a good general idea
of the topography of Banff and its surroundings is from
the summit of a little hill known as Tunnel Mountain.
It is centrally located in the wide valley of the Bow, above
which it rises exactly 1000 feet, an altitude great enough
to make it appear a high mountain were it not dwarfed
by its mighty neighbors. The view from the summit is
not of exceeding grandeur, but is well worth the labor of
the climb, especially as a good path, with occasional seats
for the weary, makes the walk an easy one. The top of
the mountain is still far below the tree line, though the
earth is too thin to nourish a rich forest.    The soil was Banff Springs Hotel.  1  View from Tunnel Mountain.
all carried away in the Ice Age, for there are abundant
proofs that this mountain was once flooded by a glacier
coming down the Bow valley. The bare limestone of the
summit is grooved in great channels pointing straight up.
the Bow valley. In some places scratches made by the ice
are visible, and there are many quartz boulders strewed
about which have been carried here from some distant
region.
The meandering course of the Bow River, the village,
the hay meadows and grassy swamps, all form a pretty
picture in the flat valley below. The eastern face of
Tunnel Mountain is wellnigh perpendicular. The trail
leads along near the summit and allows thrilling views
down the sheer precipice to the flat valley of the Bow
River far below. The trees and prominent objects of the
landscape seem like toys, and the adjacent plains resemble
a colored map. There are no houses or dwellings in view
on this side, but a drove of horses grazing contentedly in
a pasture near the river, awaiting their turn to be sent out
into the mountains in the pack train of some sportsman
or mountaineer, gives life and animation to the scene.
On either side are two high mountains, conspicuous by
their unusual outlines and great altitude. The one to
the south is Rundle Mountain. It rises in a great curving
slope on its west side, and terminates in a rugged escarpment with precipitous cliffs to the-east, which tower in
wonderful grandeur more than 5000 feet above the flood
plains of the Bow River near its base.
On the opposite side is Cascade Mountain, which is
remarkable in being of almost identical height, and is in 6 The Canadian Rockies.
fact just two feet lower, as determined by the topographical survey. The name .of this mountain was given by
reason of a large stream which falls from ledge to ledge
down the cliffs of its eastern face in a beautiful cascade.
Both this and Rundle Mountain are composed of the old
Devonian and Carboniferous limestones, the strata of
which are plainly visible. The structure is that of a
great arch or anticline which has been completely overturned, so that the older beds are above the newer.
Several miles towards the east, the end of Devil's Lake
may be seen appearing through a notch in the mountains.
A fine road nine miles in length has been made to this
lake and is one of the most popular drives in the vicinity
of Banff. The lake is very long and narrow, about nine
miles in length by three fourths of a mile in extreme
breadth. The scenery is grand, but rather desolate, as the
bare mountain walls on either side of the lake are not
relieved by forests or abundant vegetation of any kind.
The lake is, however, a great resort for sportsmen as it
abounds in large trout, of which one taken last year
weighed thirty-four pounds. The name of the lake gives
illustration of the tendency among savages and civilized
people to dedicate prominent objects of nature to the
infernal regions or the master spirit thereof. There is no
apparent limit to the number of places named after the
Devil and his realm, while the names suggested by more
congenial places are conspicuous by their absence. The
original name, Lake Peechee, was given by Sir George
Simpson in honor of his guide.
-5iî*u_ Sir George Simpson. 7
The scattered threads of history which relate to this
part of the Rocky Mountains are suggested by these
names and indeed this lake has an unusual interest for this
reason. In a region where explorations have been very
few and far between, and where only the vague traditions
of warlike events among the Indians form a great part of
the history, each fragment and detail set forth by the old
explorers acquires an increased interest.
Previous to the arrival of the railroad surveyors, the
chief men on whom our attention centres are Sir George
Simpson, Mr. Rundle, and Dr. Hector.
The expedition of Sir George Simpson possesses much
of interest in every way. He claims to have been the first
man to accomplish an overland journey around the world
from east to west. After having traversed the greater
part of the continent of North America, he entered the
stupendous gates of the Rocky Mountains in the autumn
of 1841. He travelled with wonderful rapidity, and was
wont to cover from twenty to sixty miles a day, according
to the nature of the country. His outfit consisted of a
large band of horses, about forty-five in number, attended
by cooks and packers sufficient for the needs of this great
expedition. Nevertheless the long cavalcade of animals,
when spread out in Indian file along the narrow trails were
difficult to manage, and it not infrequently happened that
on reaching camp several horses proved to be missing, a
fact which would necessitate some of the men returning
fifteen or twenty miles in search of them.
Passing to the south of the Devil's  Head, a remark- The Canadian Rockies.
able and conspicuous mountain which may be recognized
far out on the plains, Sir George Simpson entered the
valley occupied by the lake. In this part of his journey
he was guided by a half-breed Indian named Peechee,-a
chief of the Mountain Crées. Peechee lived with his wife
and family on the borders of this lake, and Simpson
named it after him, a name, however, which never
gained currency. Dr. Dawson transferred the name to
a high mountain south of the lake, and substituted the
Indian name of Minnewanka, or in English, Devil's Lake.
The guide Peechee seems to have possessed much
influence among his fellows, and whenever, as was often
the case, the Indians gathered around their camp-fires
and gossiped about their adventures, Peechee was listened
to with the closest attention on the part of all. Nothing
more delights the Indians than to indulge their passion for
idle talk when assembled together, especially when under
the soothing and peaceful influence of tobacco,—a fact
that seems strange indeed to those who see them only
among strangers, where they are wont to be remarkably
silent.
A circumstance of Indian history connected with the
east end of the lake is mentioned by Sir George Simpson,
and admirably illustrates the nature of savage warfare.
A Crée and his wife, a short time previously, had been
tracked and pursued by five Indians of a hostile tribe into
the mountains to a point near the lake. At length they
were espied and attacked by their pursuers. Terrified by
the fear of almost certain death, the Crée advised his wife An Indian Legend. 9
to submit without defending herself. She, however, was
possessed of a more courageous spirit, and replied that as
they were young and had but one life to lose they had
better put forth every effort in self-defence. Accordingly
she raised her rifle and brought down the foremost warrior
with a well aimed shot. Her husband was now impelled
by desperation and shame to join the contest, and mortally wounded two of the advancing foe with arrows.
There were now but two on each side. The fourth warrior had, however, by this time reached the Cree's wife and
with upraised tomahawk was on the point of cleaving her
head, when his foot caught in some inequality of the
ground and he fell prostrate. With lightning stroke the
undaunted woman buried her dagger in his side. Dismayed by this unexpected slaughter of his companions,
the fifth Indian took to flight after wounding the Crée in
his arm.
Rundle Mountain, which has been already mentioned
and which forms one of the most striking mountains in
the vicinity of Banff, is named after a Wesleyan missionary who for many years carrried on his pious labors
among the Indians in the vicinity of Edmonton. Mr.
Rundle once visited this region and remained camped for
a considerable time near the base of Cascade Mountain,
probably shortly after Sir George Simpson explored this
region. The wt>rk of Mr. Rundle among the Indians
appears to have been highly successful, if one may judge
by the present condition of the Stoneys, who are honest,
truthful, and but little given to the vices of civilization. ^ fHMÊÊËmnrmKmnmm
nmswsmm
The Canadian Rockies.
Even to this day the visitor may see them at Banff
dressed in partly civilized, partly savage attire, or on rare
occasions decked out in all the feathers and beaded belts
and moccasins that go to make up the sum total of savage
splendor.
Our attention comes at last to Dr. Hector, who was
connected with the Palliser expedition. It is exceedingly unfortunate that the blue-book in which the vast
amount of useful information and interesting adventure
connected with this expedition is so clearly set forth
should be now almost out of print. There are no available copies in the United States or Canada and but very
few otherwise accessible. Dr. Hector followed up the
Bow River and passed the region now occupied by Banff
in the year 1858. He was accompanied by the persevering and ever popular botanist, Bourgeau. Under the
magic spell of close observation and clear description, the
most commonplace affairs assume an unusual interest in
all of Dr. Hector's reports. It is very evident that
game was much more abundant in those early days than
at present. For instance, Dr. Hector's men shot two
mountain sheep near the falls of the Bow River, which are
but a few minutes'walk from the hotel. Likewise when
making a partial ascent of the Cascade Mountain, Dr.
Hector came on a large herd of these noble animals, concerning which so many fabulous tales of their daring leaps
down awful precipices have been told. He also mentions
an interesting fact about the death of a mountain goat.
An  Indian had shot a goat when far up on the slope of Bow River and Cascade Mountain. iggg§g  ^""^fl^iJMJi^^^^g^^^^i^^g^B^yg^^^^^^^^^^^
mm 1
The Climate of Banff. 11
Cascade Mountain, but the animal, though badly wounded,
managed to work its way around to some inaccessible
cliffs near the cascade. Here the poor animal lingered
for seven days with no less than five bullets in its body,
till at length death came and it fell headlong down the
precipice.
The climate of Banff during the months of July and
August is almost perfection. The high altitude of 4500
feet above the sea-level renders the nights invariably cool
and pleasant, while the mid-day heat rarely reaches 8o° in
the shade. There is but little rain during this period
and in fact there are but two drawbacks,—mosquitoes and
forest-fire smoke. The mosquitoes, however, are only
troublesome in the deep woods or by the swampy tracts
near the river. The smoke from forest fires frequently
becomes so thick as to obscure the mountains and veil
them in a yellow pall through which the sun shines with
a weird light.
An effect of the high northern latitude of this part of
the Rocky Mountains is to make the summer days very
long. In June and early July the sun does not set till
nine o'clock, and the twilight is so bright that fine print
can be read out doors till eleven o'clock, and in fact there
is more or less light at midnight.
In June and September one never knows what to
expect in the way of weather. I shall give two examples
which will set forth the possibilities of these months,
though one must not imagine that they illustrate the
ordinary course of events.      In   the  summer   of   1895, The Canadian Rockies.
after having suffered from a long period of intensely hot
weather in the east, I arrived at Banff on the 14th of
June. It was snowing and the station platform was cov-,
ered to a depth of six inches. The next day, however,
I ascended Tunnel Mountain and found a most extraordinary combination of summer and winter effects. The
snow still remained ten or twelve inches deep on the
mountain sides, though it had already nearly disappeared
in the valley. Under this wintry mantle were many varieties of beautiful flowers in full bloom, and, most conspicuous of all, wild roses in profusion, apparently uninjured
by this unusually late snow-storm. I made a sad discovery near the top of the mountain. Seeing a little bird
fly up from the ground apparently out from the snow,
I examined more closely and observed a narrow snow-.
tunnel leading down to the ground. Removing the
snow I found a nest containing four or five young birds
all dead, their feeble spark of life chilled away by the
damp snow, while the mother bird had been, even when
I arrived, vainly trying to nurse them back to life.
This storm was said to be very unusual for the time
of year. The poplar trees in full summer foliage suffered
severely and were bent down to the ground in great
arches, from which position they did not fully recover
all summer, while the leaves were blighted by the frost.
As a general rule, however, mountain trees and herbs
possess an unusual vitality, and endure snow and frost or
prolonged dry weather in a remarkable manner. The
various flowers which were buried for a week by this late. Itl
The Mountains in Winter.
storm appeared bright and vigorous after a few warm
days had removed the snow.
Toward the end of September, 1895, there were two
or three days of exceptionally cold weather, the thermometer recording 6° Fahrenheit one morning. I made
an ascent of Sulphur Mountain, a ridge rising about
3,000 feet above the valley, on the coldest day of that
period. The sun shone out of a sky of the clearest
blue without a single cloud except a few scattered wisps
of cirrus here and there. The mountain summit is covered with a few straggling spruces which maintain a bare
existence at this altitude. The whole summit of the
mountain, the trees, and rocks were covered by a thick
mantle of snow, dry and powdery by reason of the severe
cold. The chill of the previous night had condensed a
beautiful frost over the surface of the snow everywhere.
Shining scales of transparent ice, thin as mica and some
half-inch across, stood on edge at all possible angles and
reflected the bright sunlight from thousands of brilliant
surfaces. This little glimpse of winter was even more
pleasing than the view from the summit, for the mountains near Banff do not afford the mountain climber grand
panoramas or striking scenery. They tend to run in long
regular ridges, uncrowned by glaciers or extensive snow-
fields.
A never failing source of amusement to the residents
of Banff, as well as to those more experienced in mountain
climbing, is afforded by those lately arrived but ambitious
tourists who   look up at the mountains as though they The Canadian Rockies.
were little hills, and proceed forthwith to scale the very
highest peak on the day of their arrival. A few years
ago some gentlemen became possessed of a desire to
ascend Cascade Mountain and set off with the intention
of returning the next day at noon. Instead of following
the advice of those who knew the best route, they would
have it that a course over Stoney Squaw Mountain, an
intervening high ridge, was far better. They returned
three days later, after having wandered about in burnt
timber so long that, begrimed with charcoal, they could
not be recognized as white men. It is not known whether
they ever so much as reached the base of Cascade Mountain, but it is certain that they retired to bed upon arriving at the hotel and remained there the greater part of
the ensuing week.
Cascade Mountain, however, is a difficult mountain
to ascend, not because there are steep cliffs or rough
places to overcome, but because almost every one takes
the wrong slope. This leads to a lofty escarpment, and
just when the mountaineer hopes to find himself on the
summit, the real mountain appears beyond, while a great
gulf separates the two peaks and removes the possibility
of making the ascent that day.
Banff, with its fine drives and beautiful scenery, its
luxurious hotels and delightful climate, will ever enjoy
popularity among tourists. The river above the falls is
wide and deep and flows with such gentle current as to
render boating safe and delightful. The Vermilion lakes,
with their low reedy shores and swarming wild fowl, offer Vermilion Lakes.
15
charming places for the canoe and oarsman, at least when
the mosquitoes, the great pest of our western plains and
mountains, temporarily disappear. Nevertheless, the climate of Banff partakes of the somewhat dryer nature of
the lesser and more eastern sub-ranges of the Rocky Mountains.    There  is  not sufficient moisture to nourish  the
the higher ranges to the west, which in imagination we
shall visit in the ensuing chapters.
RUNDLE MOUNTAIN AND BOW RIVER. nrnrîiiiiiimm
mmmtmmmsf8888
CHAPTER II.
Lake Louise—First Impressions—An Abode of Perpetual Winter—
The Chalet—Visitors—Stirring Tales of Adventure—Primeval Forests—
Forest Fires—Mosquitoes and Bull-Dog Flies—Mortal Combats between
Wasps and Bull-Dogs— The Old Chalet—Morning on the Lake—Approach
of a Storm—Sublimity of a Mountain Thunder-Storm—Cloud Effects—
The Lake in October—A Magnificent Avalanche from Mt. Lefroy—A
Warning of Approaching Winter.
LAKE LOUISE is one of the most beautiful
sheets of water in the Canadian Rockies. Many
who have travelled extensively say it is the
most charming spot they have ever beheld. The lake
is small, but there is a harmonious blending of grandeur
and quiet beauty in the surrounding mountains which
in some way makes a perfect picture out of lofty snow
peaks in the distance and dark forested slopes near at
hand.
The lake is a little more than a mile long and about
one fourth of a mile wide. The outline is remarkably
like that of the left human foot. Forests come down
nearly to the water's edge on all sides of the lake, but
there is a narrow margin of rough angular stones where
the ripples from the lake have washed out the soil and
16 First Impressions.
W
even undermined the trees in some places. The water is
a blue-green color, so clear that the stones on the bottom and the old water-logged trunks of trees, long since
wrested from the shores by storms and avalanches, may
be discerned even in several fathoms of water. The lake
is 230 feet deep in the centre, and the bottom slopes down
very suddenly from the shores.
The west shore makes a gently sinuous or wavy line,
forming little bays and capes. Ever new and artistic
foregrounds are thus presented, with the forest making a
retreating line of vegetation down the shore. Nothing
could be more beautiful than this border of the lake,
rough and tangled though it is, with a strange mingling
of sharp boulders and prostrate trees covered with moss
and half concealed by copses of alder bushes and flowering
-shrubs.
I shall never forget my first view of Lake Louise.
From the station, the old trail, constantly ascending as it
approaches the lake, leads its irregular course through
the forest. After a walk of nearly three miles, partial
glimpses of the lake and surrounding mountains were
obtained from among the tall spruce trees. A short
rapid descent of a small ridge placed us on the borders
of the lake.
It would be difficult indeed to give even a partial description of the scene. Imagine a cool morning with the
rising sun just beginning to touch the surface of a mountain lake. The air is tranquil and calm so that the glassy
.surface of the water mirrors the sky and mountains per- mmmnmHmmmmm
The Canadian Rockies.
fectly. In the realm of sound, too, all is repose but for the
call of birds near at hand among the balsam trees. From
the shores of the lake an either side rise great mountains,
showing cliffs and rocky ledges or long sweeping slopes
of forest to the tree line. Higher still are bare slopes,
crags, ledges, and scattered areas of snow. At the end
of the lake a great notch in the nearer mountains reveals
at a distance the wall-like, lofty mass of Mount Lefroy.
This most imposing snowy mountain stands square across
the gap, and with a sharp serrated cliff piercing the very
vault of heaven, shuts off the view and forms the most
conspicuous object of all. The lower part of the mountain is a vertical cliff or precipice where the longitudinal
strata are distinctly visible. Above, rise alternating slopes
covered with perpetual snow and hanging glaciers, the
white-blue ice of which is splintered by deep rents and
dark yawning crevasses. This mountain forms part of
the continental water-shed, for on the other side the melting snows finally reach the Pacific Ocean, while on the
near side the snows swept into the valleys by avalanches,
and melted by the warmer air of lower altitudes, find their
way at length into the Saskatchewan River and Hudson
Bay.
There is something wonderfully attractive about this
mountain. The pleasure grows as one continues to gaze
at the immense mass ; harsh and stern and cold though it
be, it excites awe and wonder as though here were the
rocky foundation and substratum of the globe. This is
the abode of perpetual winter, where ice and snow and Lake Louise.   wmmmmsmmmBasasBa 1
Mountain Flowers. 19
bleak rocks exist apart. Here all is grand but menacing,
dangerous, and forbidding. And these high mountains and
deep valleys, suggesting that some awful storm at sea had
become petrified into colossal waves to stand at rest forever, have been carved out by rain and running water, frost
and change of temperature, through the lapse of countless ages.
Our attention finally came to the quiet beauty of the
surrounding vegetation, where among the scattered skirmishers of the forest are flowering shrubs, and in the more
open grassy places forming the swampy borders of the
lake, are many bright flowers. The white mountain anemones in several varieties, the familiar violets, the yellow
columbine with beautiful pendent blossoms claiming relationship to its Eastern cousin with scarlet flowers, the
fragrant spiranthes, and orchids with pale-green flowers,
resembling insects on a leafy stem, may all be seen in
profusion near the north side of the lake. These humble
herbs, with their gaudy coloring, are the growth of a single season, but on all sides are copses of bushy plants
which endure the long winter, some of them clad in a garb
of evergreen and, like the annual plants, bearing elegant floral creations. The most conspicuous is the sheep
laurel, a small bush adorned with a profusion of crimson-
red flowers, each saucer-shaped, hanging in corymbs
among the small green leaves. Various shrubs with
white flowers, some small and numerous, others large and
scattered, make a contrast to the ever present laurel, while
the most beautiful of all is a species of mountain rhodo- %asg#ssaa
20
The Canadian Rockies.
dendron, a large bush, the most elegant among the mountain heaths, with large white flowers in clustered umbels.
In early July this bush may be found, here and there, scattered sparingly in the forest in full blossom at the level
of Lake Louise, but after this one must seek ever higher
on the mountain side as the advancing summer creeps to
altitudes where spring is later.
The early morning visitor turns with sharpened appetite to the hotel, if we may call it such,—a little Swiss chalet
of picturesque architecture built on an eminence in full view
of the lake. Here the tourist may live in rustic comfort for
a day, or for weeks, should he desire to prolong his visit.
Tourists come sparingly to Lake Louise. Unlike
Banff with its varied attractions, there is little here outside of nature, and few have the power to appreciate
nature alone. Of those who do come, only a small
number really see the lake with its forests and mountains
combined in exquisite attractiveness. They see the outlines of mountains, but know not whether they are near
or distant, nor whether their scale is measured in yards or
miles ; they see the water of the lake, but not the reflections in it, the ever changing effects of light and shade,
sun and shadow, ripple and calm. There are trees tall
and slender, but whether they be spruce or pine, larch or
hemlock, is all the same ; and as to the flowers—some are
differently colored from others.
A visitor to the lake once asked in good faith, apparently, if the mountains at the head of the lake were not
white from chalk ; another, why the water of the stream ï
Visitors. 21
—which leads out from the lake and rushes in roaring
cascades over its rocky channel toward the Bow River—
runs so fast down hill.
Fortunately, however, those who are not blessed with
that ever present source of pleasure, a love for nature, at
least to a slight degree, are exceptional. Nevertheless,
that most people lose much pleasure from a lack of close
observation is often painfully evident. I have seen, altogether, several hundred tourists arrive at the lake, coming
as they do in small parties, or singly, from day to day,
and have found it a very interesting study to observe
their first impressions as the lake bursts on their view.
Some remain motionless studying the details of the scene,
usually devoting their chief attention to the lake and forests, but less to the mountains, for mountains are the least
appreciated of all the wonders of nature, and are not fully
revealed except after years of experience. Others glance
briefly and superficially towards the lake, and rush hastily
into the chalet for breakfast, balancing their love for
nature against hunger for material things in uneven scale.
Some remain a week or ten days, but the great majority
spend a single day and leave, feeling that they have exhausted the charms of the place in so short a time. A single
day amid surroundings where there are such infinite possibilities of change in cloud and storm, heat and cold, the
dazzling glare of noon, or the calm romantic light of a
full moon, and the slow progress of tire seasons, gives but
one picture, a single mood from out a thousand, and it
may perchance be the very worst of all. The Canadian Rockies.
Upon climbing the steps to the open porch of the
chalet and entering the large spacious sitting-room, the.
eye falls at once on a fireplace of old-time proportions,
and within its walls of brick, huge logs are burning, with
more vigor indeed but hardly less constancy than the
ancient fires of the Vestal Virgins. Round this spacious
hearth visitors and guests gather, for the air at Lake
Louise is always sharp at morning and evening. Indeed,
frosts are not rare throughout the summer and may occur
any week even in July and August. The high altitude of
the lake, which is a little more than 5600 feet above
sea-level, is in great part the cause of this bracing
weather. On the hottest day that I have ever seen at the
lake in the course of three summers the thermometer
registered only 780.
The visitors who come to Lake Louise are of the same
cosmopolitan character and varied nationality as those at
Banff. Often of a cold night have I sat by the large fire,
our only source of light, and listened to tales of adventure
told by those who have visited the most distant and unfrequented parts of the earth. Englishmen, who have
spent the best years of their life in India, were among
our entertainers, and while beverages varying in nature
according to nationality or tastes of each were passed
around, I have heard thrilling accounts of leopard and
tiger hunts in the jungle, blood-curdling tales of treachery
and massacre or daring exploits in the Indian wars, and
rare experiences in unknown parts of Cashmere and
Thibet. Primeval Forests.
23
Though the great majority of visitors to the lake are
strangers, there are some half-dozen whose familiar faces
reappear each successive season ; like pilgrims they make
this region the termination of a long annual journey, and
here worship in " temples not built by human hands."
Among these lovers of nature, far distant England and
Ceylon are represented no less than the nearer cities of
the United States. The peculiar charms of this locality
present an inexhaustible treasurehouse of delightful experiences that grow by familiarity. One's impressions of
the beauty of the lake increase year by year as the full
meaning of each detail becomes more thoroughly appreciated.
A fact of great importance, which goes far to make up
the ensemble of the surroundings of Lake Louise, is the
perfect condition of the forests, which rise in uniform,
swelling slopes of dark-green verdure from the rocky
shores of the lake far up the mountain sides to those
high altitudes where the cold air suggests an eternal winter
and dwarfs the struggling trees into mere bushes. The
frequent forest fires, which have wrought so much destruction throughout the entire Canadian Rockies, have not as
yet swept through this valley. The great spruces and
balsams of this primeval forest indicate by their size that
for hundreds of years no fire has been through this region.
Some large tree stumps near the chalet show hundreds of
rings, and one that I counted started to grow in the year
1492, when Columbus set forth to discover the western
world. 24    s The Canadian Rockies.
• Nevertheless, on hot days after a long period of dry
weather, when the air is laden with the fragrant odor of
the dripping balsam and of the dry resin hardened in yellow
tears on the scarred trunks of the trees, and when the
dead lower branches hung with long gray moss seem to
offer all the most combustible materials, one feels certain
that the slightest spark would result in a terrible conflagration. Apparently, however, the past history of this
valley has never recorded a fire, whether started by careless Indian hunters or that frequent cause, lightning.
So far as I am aware, there are no layers of buried charcoal or reddened soil under the present forest which
would indicate an ancient fire.
Some years ago—apparently more than twenty,—a
fire destroyed the forest near the station of Laggan, which
is less than two miles from the lake in a straight line.
The fire approached within a mile of the lake and then
died out. There are two causes which will always tend
to preserve these beautiful forests if the visitors are not
careless and counteract them. The prevalent wind is out
of the valley toward the Bow valley, so that a fire would
naturally be swept away from the lake. Another cause
is the natural moisture of this upland region. The very
luxuriance of the vegetation indicates this, while in the early
morning the whole forest often seems reeking with moisture, even when there has been no rain for weeks. The
chill of night appears to condense a heavy dew under the
trees and moistens all the vegetation, so that the forest
rarely becomes so exceedingly dry as often happens in
wide valleys at lower altitudes. Mosquitoes and Bull-Dogs. 25
Though the scenery and climate at Lake Louise seem
almost ideally perfect during the summer time, nature
always renders compensation in some form or other, and
never allows her creatures to enjoy complete happiness.
The borders of the lake and the damp woods breed myriads
of mosquitoes, which conspire to annoy and torture both
man and beast. They appear early in spring and suddenly vanish about the 15th or 20th of August each year.
The chill of night causes them to disappear about ten
o'clock in the evening, not to be seen again until the
atmosphere begins to grow warm in the morning sun.
Another insect pest is a species of fly called the " bulldog," a name suggested by its ferocious bite. These
large insects are about an inch in length and are armed
with a formidable set of saws with which they can rapidly
cut a considerable hole through the skin of a man or the
hide of a horse. The bull-dogs frequent the valleys of the
Canadian Rockies, varying locally in their numbers, and
seem to prefer low altitudes and a considerable degree
of heat, for they are always most voracious and numerous
on hot dry days. These flies, when numerous, will almost
make a horse frantic. Their bite feels like a fiery cinder
slowly burning through the skin, but fortunately they do
not cause much trouble to man, for they are led by instinct
to seek the rough surfaces of animals and almost invariably
light on the clothes instead of the hands or face. They
have a most blood-thirsty and cruel enemy in the wasp,
and if it were not for the inexhaustible supply of the bulldogs, the wasps would annihilate the species. Nothing
in the habits of insects could be more interesting than the tmmmmmmm
mismfflftimsssm
26
The Canadian Rockies.
strange manner in which the wasps set out deliberately in
pursuit of a bull-dog fly, to overtake and seize the clumsy
victim in mid air. Both insects fall to the ground with a
terrible buzzing and much circling about while the mad
contest goes on. Meanwhile the wasp works with the
rapidity of lightning, and with its sharp powerful jaws dissevers legs and wings, which fall scattered in the melee,
till the bull-dog is rendered helpless and immovable.
Last of all, the wasp cuts off the head of its victim, then
leaves the lifeless and limbless body in order to continue
the chase.
I have seen a wasp thus dismember and kill one of
these large flies in less than thirty seconds. They seem
to perform their murderous acts out of pure pleasure, as
they do not linger over their prey after the victim is dead.
The water of Lake Louise is too cold to admit of
bathing except in a very brief manner. The temperature
of the water near the first of August is about 56°.
The old chalet, built in rustic fashion with unhewn
logs, was placed near the lake shore much closer than the
present building. One day in 1893, when every one was
absent, the building caught fire and burned to the ground.
Remarkably enough the forest did not take fire, though
some of the trees were close to the building.
Usually in the early morning, before the sun has
warmed the atmosphere and started the breezes of daytime into motion, the lake is tranquil and its surface
resembles a great mirror. About nine o'clock, the first
puffs of wind begin to make little cat's-paws at the far Approach of a Storm.
27
end of the lake, which widen and extend until finally the
whole water becomes rippled. A gentle breeze continues
to sweep down the lake from the snow mountains toward
the Bow valley all day long, and the water rarely becomes
smooth till after sunset. This is the usual order of events
in fair weather, a condition which may continue for several
weeks without a drop of rain.
The approach and progress of a storm, the wonderful
atmospheric changes attending it, and the ever moving
clouds obscuring the mountain tops reveal the lake in
the full grandeur of its surroundings. An approaching
storm is first announced by scattered wisps of cirrus
cloud, which move slowly and steadily from the west in
an otherwise blue sky. In the course of twenty-four
hours the cirrus clouds have become so thick that they
often resemble a thin haze far above the highest mountains. The sun with paled light can no longer pierce this
ever thickening hazy veil. The wind blows soft and warm
from out the south or southwest, and generally brings up
the smoke of forest fires from the Pacific coast, and renders the atmosphere still more obscure, till at length the
sun appears like a great ball of brass set in a coppery sky.
The trees and grass appear to change their color and
assume a strange vivid shade of green in the weird light
Sometimes light feathery ashes are wafted over the high
mountains south of the lake and settle down gently like
flakes of snow. The falling barometer announces the
coming storm, and presently another layer of clouds, the
low-lying cumulus, form just above the highest peaks and
111*3,,.      m mmmmmmEgmmesmmmsmmsmmimmm
The Canadian Rockies.
settle gradually lower till they touch the mountain tops.
Rain soon follows, the clouds settle till they almost rest
on the water of the lake, and the wind increases in violence.
Sometimes thunder-storms of considerable fury sweep
through the valley and among the mountains, one after
another for several days. A violent thunder-storm at
night among these lofty mountains is one of the grandest
phenomena of nature. The battling of the elements, the
unceasing roar of the wind in the forest, and the crash of
thunder redoubled by echoes from the rocky cliffs,—all conspire to fill the imagination with a terrible picture of the
majesty and sublimity of nature. From the lake there
comes up a low, hoarse murmur, not the roar of ocean
surf, but the lesser voice of a small mountain lake lashed
to fury and beating with its small waves on a rocky shore.
The noise of the forest, the sound of colliding branches
as the tall trees sway to and fro in the furious wind, and
the frequent crack and crash of dead forest giants overcome by the elements form the dull but fearful monotone,
above which the loud rumble of thunder rises in awful
grandeur.    These are the sounds of a mountain storm.
The bright flashes of lightning reveal a companion
picture, for in the momentary light succeeded by absolute
darkness the lake is revealed covered with foamy white
caps. The forests on the mountain side seem to yield to
the blast like a field of wheat in a summer breeze, and the
circling clouds sweep about the mountain slopes and conceal all but their bases.
Should the storm clear away during the daytime one Cloud Effects.
Î1
may witness grand cloud effects. The low-hanging masses
of clouds left behind by the battling elements slowly rise
and occasionally reveal small areas of blue sky among the
moving vapors. Gentle puffs of air sweep over the calm
surface of the water, making little areas of ripples here
and there, only to be succeeded by a tranquil calm, as if
the storm spirit were sending forth his dying gasps intermittently. While the air is thus calm below, the circling
- wisps of vapor high up on the mountain, rising and descending, show that the battle between the sun and the clouds is
still raging. From above the saturated forests, the rising
vapors condense and increase in size till at length, caught
in some counter-current, they are swept away or carried
downward, while the dissolving cloud spreads out in wisps
and streamers till suddenly it disappears into transparent
air,—a veritable cloud ghost. At length the mountain
tops appear once more, white in a light covering of new
snow, and, as the great masses of cumulus rise and disappear the sky appears of that deep blue-black color peculiar to mountain altitudes, while the sun shines out with
dazzling brilliancy through the clear atmosphere.
The last visit I made to Lake Louise was toward the
middle of October, 1895. A very snowy, disagreeable
September had been followed by a long period of milder
weather with much bright sunshine. The new snow,
which had been quite deep near the lake, had altogether
disappeared except high up on the mountain side. It
was the true Indian summer, a season with a certain mellow charm peculiar to  it alone,  characterized by  clear jo The Canadian Rockies.
sunny weather, a calm atmosphere, a low, riding sun, and
short days. Most of the flowers were withered. The
deciduous bushes, lately brilliant from frost, were rapidly losing their foliage, and the larches were decked in
pale yellow, far up near the tree line. However, the
greater part of the vegetation is evergreen, and the
spruces, balsams, and pines, the heaths, ericaceous plants,
and the mosses contrive to set winter at nought by wearing the garb of a perpetual summer in a region where
snow covers the ground three fourths of the year.
I could not resist the temptation as the morning train
rolled up to the station at Laggan to get off for the day
and make another visit to the lake. The sunrise had
been unusually brilliant and there was every promise of
a fine day. There is rarely much color at sunrise or sunset in the mountains. The dry clear atmosphere has
little power to break up the white light into rainbow
colors and give the brilliancy of coloring to be seen near
the sea-coast or in the lowlands. The tints are like the
air itself—pure, cold, and clear. With more truth they
might be called delicate shades or color suggestions.
They recall those exquisite but faint hues seen in topaz
or tourmaline crystals, or transparent quartz crystals,
wherein the minutest trace of some foreign mineral has
developed rare spectrum colors and imprisoned them
forever. Oftimes the snow of the mountain tops is thus
tinted a bright clear pink, beautifully contrasted against
the intensely blue sky. I have never seen a deep red on
the mountains or clouds at these altitudes.    The effect The Lake in October.
31
of forest-fire smoke is to give muddy colors : the sun
resembles a brazen globe, and the sky becomes coppery
in appearance.
After breakfast at the station house, I set off over the
hard frozen road toward the lake. I carried my camera
and luncheon on my back, my only companion being a
small dog which appeared ready for exercise. The air
was frosty and cold ; the low-riding sun had not as yet
struck into the forest trees and removed the rime from
the moss and leaves on the ground.
In somewhat less than an hour, I arrived at the lake.
All was deserted ; the chalet closed, the keeper gone,
and the tents
taken down.
E v en the
boats, which
usually rested near the
shore, had
been put
under cover.
The cold air
was perfectly
calm, and my
vapory breath
rose straight
upwards. The
mirror surface of the water was disturbed by some wild
fowl—black ducks and divers—which swarm on the lake
LAKE LOUISE LOOKING TOWARD CHALET. ^^^m^^^^&mmm^^ssmsmïs^^^
, 32 The Canadian Rockies.
at this season. Their splashings, and the harsh cries of
the divers came faintly over the water. It seemed
strange that these familiar haunts could appear so fearfully wild and lonely merely because man had resigned
his claim to the place and nature now ruled alone. All
at once a wild unearthly wail from across the water,
the cry of a loon, one of the most melancholy of all
sounds, startled me, and gave warning that activity alone
could counteract the effect of the imagination.
Accordingly I walked down the right shore of the lake
with the intention of going several miles up the valley and
taking some photographs of Mount Lefroy. The flat
bushy meadows near the upper end of the lake were
cold, and all the plants and reedy grass were white with
the morning frost. The towering cliffs and castle-like
battlements of the mountains on the south side of the
valley shut out the sun, and promised to prevent its
genial rays from warming this spot till late in the afternoon, if at all, for a period of several months. In the
frozen ground, as I followed the trail, I saw the tracks
of a bear, made probably the day before. Bruin had gone
up the valley somewhere and had not returned as yet, so
there was a possibility of making his acquaintance.
I was well repaid for my visit this day, as a magnificent
avalanche fell from Mount Lefroy. Mount Lefroy is a
rock mountain rising in vertical cliffs from between two
branches of a glacier which sweep round its base. A
hanging glacier rests on the highest slope of the mountain, and, ascending some distance, forms a vertical face of T
A Magnificent Avalanche. 33
ice nearly three hundred feet thick at the top of a great
precipice. The highest ridge of the mountain is covered
with an overhanging cornice of snow, which the storm
winds from the west have built out till it appears to reach
full one hundred feet over the glacier below. At times,
masses of ice break off from the hanging glacier and fall
with thundering crashes to the valley far below.
I was standing at a point some two miles distant
looking at this imposing mountain, when from the vertical
ice wall a great fragment of the glacier, some three hundred feet thick and several times as long, broke away, and,
slowly turning in mid-air, began to fall through the airy
abyss. In a few seconds, amid continued silence, for the
sound had not yet reached me, the great mass struck a
projecting ledge of rock after a fall of some half thousand
feet, and at the shock, as though by some inward explosion,
the block was shivered into thousands of smaller fragments
and clouds of white powdery ice. Simultaneously came
the first thunder of the avalanche. The larger pieces led
the way, some whirling around in mid-air, others gliding
downward like meteors with long trains of snowy ice dust
trailing behind. The finer powdered debris followed after,
in a long succession of white streamers and curtains resembling cascades and waterfalls. The loud crash at the
first great shock now developed into a prolonged thunder
wherein were countless lesser sounds of the smaller pieces
of ice. It was like the sound of a great battle in which the
sharp crack of rifles mingles with the roar of artillery.
Leaping from ledge to ledge with ever increasing velocity,
H
sssm 34 The Canadian Rockies.
the larger fragments at length reached the bottom of the
precipice, while now a long white train extended nearly
the whole height of the grand mountain wall 2500 feet
from base to top.
Imagine a precipice sixteen times higher than Niagara,
nearly perpendicular, and built out of hard flinty sandstone.
At the top of this giant wall, picture a great glacier with
blue ice three hundred feet thick, crevassed and rent into
a thousand yawning caverns, and crowding downwards,
ever threatening to launch masses of ice large as great
buildings into the valley below. Such avalanches are
among the most sublime and thrilling spectacles that
nature affords. The eye alone is incapable of appreciating
the vast scale of them. The long period of silence at first
and the thunder of the falling ice reverberated among the
mountain-walls produce a better impression of the distance
and magnitude.
I arrived at the lower end of the lake toward one
o'clock. The lake was only disturbed in one long narrow
strip toward the middle by a gentle breeze while all the
rest was perfectly calm. This was one of those rare days
of which each year only affords two or three, when the
lake is calm at midday under a clear sky. The mirror
surface of the water presented an inverted image of the
mountains, the trees on the shore, and the blue sky. The
true water surface and the sunken logs on the bottom of
the lake joined with the reflected objects in forming a
puzzling composite picture.
The brilliant sun had taken away the chill of morning Approaching Winter.
35
and coaxed forth a few forest birds, but there were no flowers or butterflies to recall real summer. It seemed as
though this were the last expiring effort of autumn before
the cold of winter should descend into the valley and with
its finger on the lips of nature cover the landscape with a
deep mantle of snow and bind the lake in a rigid layer of
ice. Even at this warmest period of the day the sun's rays
seemed inefficient to heat the atmosphere, while from the
cold shadows of the forest came a warning that winter was
lurking near at hand, soon to sweep down and rule uninterrupted for a period of nine long months.
i ^mmmmmmmmmmsm:
CHAPTER III.
Surroundings of the Lake—Position of Mountains and Valleys— The
Spruce and Balsam Firs—The Lyall's Larch—Alpine Flowers—The
Trail among the Cliffs—The Beehive, a Monument of the Past—Lake
Agnes, a Lake of Solitude—Summit of the Beehive—Lake Louise in the
Distant Future.
AMONG the mountains on all sides of Lake Louise
are many scenes of unusual beauty and grandeur.
While the lake itself must be considered the
focal point of this region, and is indeed wonderfully
attractive by reason of its rare setting, the encircling
mountains are so rough and high, the valleys separating
them so deep and gloomy, yet withal so beautiful, that
the scenery approaches perfection. The forces of nature
have here wrought to their utmost and thrown together
in apparently wild confusion some of the highest mountains in Canada and carved out gloomy gorge and rocky
precipice till the eye becomes lost in the complexity of it
all. Lakes and waterfalls reveal themselves among the
rich dark forests of the valleys, and afford beautiful foregrounds to the distant snow mountains which seem to
tower ever higher as one ascends.
A brief description of the topography in the vicinity of
Lake Louise would be now in place.    Southwestward from w
The Summit Range. 37
the lake is a range of very high and rugged mountains
covered with snow and glaciers. This range is the crest
of the continent of North America, in fact the great
water-shed which divides the Atlantic and Pacific drainage. In this range are many peaks over 11,000 feet
above sea level, an altitude which is near the greatest
that the Rocky Mountains attain in this latitude. While
farther south in Colorado there are scores of mountains
13,000 or 14,000 feet high, it must be remembered that
no mountains in Canada between the International
boundary and the railroad have yet been discovered that
reach 12,000 feet. Nevertheless, these mountains of lesser
altitude are far more impressive and apparently much
higher because of their steep sides and extensive fields of
perpetual snow.
This great range, forming the continental water-shed
runs parallel to the general trend of the Rocky Mountains
of Canada, or about northwest and southeast. Several
spur ranges branch off at right angles from the central
mass and run northeast five or six miles. Between these
spur ranges are short valleys which all enter into the wide
valley of the Bow. Lake Louise occupies one of these
lesser valleys.
The several lateral valleys are all comparatively near
Lake Louise and differ remarkably in the character of the
scenery and vegetation. One is beautiful and richly covered with forests ; another desolate and fearfully wild. The
valley of Lake Louise contains in all three lakes, of which
the smallest is but a mere pool, some seventy-five yards
across. 38 The Canadian Rockies.
Far up on the mountain side to the north of Lake
Louise two little lakes were discovered many years ago.
They are now to the visitor who spends but one day,
almost the chief point of interest in this region. The trail
thither leads into the dense forest from near the chalet
and proceeds forthwith to indicate its nature by rising
steadily and constantly. The tall coniferous trees cast a
deep cool shade even on a warm day. So closely do the
trees grow one to another that the climber is entirely
shut out from the world of mountains and surrounded by
a primeval forest as he follows the winding path. Among
the forest giants there are two principal trees, the spruce
and the balsam fir. Each is very tall and slender and at
a distance the appearance of the two trees is closely similar. The spruce is the characteristic tree of the Rockies
and is found everywhere. It reaches a height of 75' or
100 feet in a single tapering bole, closely beset with small
short branches bent slightly downward, as though better
to withstand the burden of snow in winter. In open
places the lower branches spread out and touch the
ground, but in forests they die and leave a free passage
between the trees. The balsam tree is quite similar but
may be discerned by its smoother bark which is raised
from underneath by countless blisters each containing a
drop of transparent balsam. Here and there are a few
tall pines rivalling the spruces and firs in height but
affording a strong contrast to them in their scattered
branches and larger needles.
The ground is covered with underbrush tangled in a T
Mount Lefroy and Mirror Lake. &^^^S^3^^3§SË£^ __^éÊM
4$
"4'!^S5
jÉp^    Jj
EPS
M
1 HP?3i
§^^:s^||H|H
H^lÉiÉ  ^
The Lyall's Larch. 39
dense luxuriance of vegetable life and partly concealing
the ancient trunks of fallen trees long since covered with
moss and now slowly decaying into a red vegetable mold.
At length, after half an hour of constant climbing, a
certain indefinable change takes place in the forest. The
air is cooler, the trees grow wider apart, and the view is
extended through long vistas of forest trees. Presently
a new species of tree, like our Eastern tamarack, makes its "
appearance. It is the Lyall's larch, a tree-4hat endures
the rigors of a subalpine climate better than the spruces
and balsam firs, so that it soon becomes to the climber
among these mountains an almost certain indication of
proximity to the tree-line.
It is not far from the truth to say that the Lyall's
larch is the most characteristic tree of the Canadian
Rockies. It is not found in the Selkirk Range just west
of the main range, and while it has indeed been found as
far south as the International boundary, it has not been
discovered in the Peace River valley to the north. Restricted in latitude, it grows on the main range of the
Rockies only at a great altitude. Here on the borderland between the vegetable and mineral kingdoms it
forms a narrow fringe at the tree-line and in autumn its
needles turn bright yellow and mark a conspicuous band
around all the cliffs and mountain slopes at about 7000 feet
above sea level. Its soft needles, gathered in scattered
fascicles, are set along the rough and tortuous branches,
affording a scanty shade but permitting of charming
glimpses of distant mountains, clouds, and sky among its
1
I  I S™S»ra?^^^§^^^^^^^^^^^gg^^S|
40 The Canadian Rockies.
gray branches and light-green foliage. It seems incapable
of sending up a tall slender stem but branches out irregularly and presents an infinite variety of forms. Possibly
for this reason the larch cannot contest with the slender
spruces and firs of the valley, where it would be crowded
out of light and sun among its taller rivals.
Presently the trail leads from out the forest and
crosses an open slope where some years ago a great snow-
slide swept down and stripped the trees from the mountain side. Here, 1200 feet above Lake Louise, the air
feels sensibly cooler and indicates an Alpine climate.
«k/3 The   mountains now   reveal themselves   in
^^^        ^ar grander proportions than from below, as
-*■*-&»*■ they burst  suddenly on the view.     Nature
^/ftwnt^V nas already made compensation for the destroyed forest by clothing this slope with a
profusion of wild flowers, though much
different in character from those at Lake
Louise. Alpine plants and
several varieties of heather,
in varying shades of red or
pink and even white, cover the
ground with their elegant coloring.
One form of heath resembles almost perfectly the true heather of Scotland, and by its
abundance recalls the rolling hills and flowery highlands
of that historic land. The retreating snow-banks of June
and July are closely followed by the advancing column of
mountain flowers which must needs blossom, bear fruit,
1^ A Monument of the Past. 41
and die in the short summer of two months duration.
One may thus often find plants in full blossom within a
yard of some retreating snow-drift.
On reaching the farther side of the bare track of the
avalanche, the trail begins to lead along the face of craggy
cliffs like some llama path of the Andes. The mossy
ledges are in some places damp and glistening with
trickling springs, where the climber may quench his thirst
with the purest and coldest water. Wherever there is
the slightest possible foothold the trees have established
themselves, sometimes on the very verge of the precipice
so that their spreading branches lean out over the airy
abyss while their bare roots are flattened in the joints
and fractures of the cliff or knit around the rocky projections like writhing serpents.
More than four hundred feet below is a small circular
pond of clear water, blue and brilliant like a sapphire crystal. Its calm surface, rarely disturbed by mountain breezes,
reflects the surrounding trees and rocks sharp and distinct
as it nestles in peace at the very base of a great rock
tower—the Beehive. Carved out from flinty sandstone,
this tapering cone, if such a thing there be, with horizontal strata clearly marked resembles indeed a giant beehive. Round its base are green forests and its summit is
adorned by larches, while between are the smooth precipices of its sides too steep for any tree or clinging plant.
What suggestions may not this ancient pile afford Ï
Antiquity is of man ; but these cliffs partake more of the
eternal—existing forever.     Their nearly horizontal strata
I
Hug Égggaa
42 The Canadian Rockies.
were formed in the Cambrian Age, which geologists tell
us was fifty or sixty millions of years ago. Far back in
those dim ages when the sea swarmed with only the low7er
forms of life, the fine sand was slowly and constantly settling to the bottom of the ocean and building up vast
deposits which now are represented by the strata of this
mountain. Solidified and made into flinty rock, after the
lapse of ages these deposits were lifted above the ocean
level by the irresistible crushing force of the contracting
earth crust. Rain and frost and moving ice have sculptured out from this vast block monuments of varied form
and aspect which we call mountains.
Just to one side of the Beehiv*e a graceful waterfall
dashes over a series of ledges and in many a leap and
cascade finds its way into Mirror Lake. This stream
flows out from Lake Agnes, whither the trail leads by a
short steep descent through the forest. Lake Agnes is a
wild mountain tarn imprisoned between gloomy cliffs, bare
and cheerless. Destitute of trees and nearly unrelieved by
any vegetation whatsoever, these mountain walls present a
stern monotony of color. The lake, however, affords one
view that is more pleasant. One should walk down the
right shore a few hundred feet and look to the north.
Here the shores formed of large angular blocks of stone
are pleasantly contrasted with the fringe of trees in the
distance.
The solitary visitor to the lake is soon oppressed with
a terrible sensation of utter loneliness. Everything in
the surroundings is gloomy and silent save for the sound *r
Lake Agnes.
In early J^uly, i&pj.   L. A Lake of Solitude. 43
of a trickling rivulet which falls over some rocky ledges on
the right of the lake. The faint pattering sound is echoed
back by the opposite cliffs and seems to fill the air with a
murmur so faint, and yet so distinct, that it suggests something supernatural. The occasional shrill whistle of a
marmot breaks the silence in a startling and sudden manner. A visitor to this lake once cut short his stay most
unexpectedly and hastened back to the chalet upon hearing one of these loud whistles which he thought was the
signal of bandits or Indians who were about to attack
him.
Lake Agnes is a narrow sheet of water said to be unfathomable, as indeed is the case with all lakes before they
are sounded. It is about one third of a mile in length
and occupies a typical rock basin, a kind of formation that
has been the theme of heated discussion among geologists.
The water is cold, of a green color, and so pellucid that
the rough rocky bottom may be seen at great depths.
The lake is most beautiful in early July before the snowbanks around its edge have disappeared. Then the double
picture, made by the irregular patches of snow on the bare
rocks and their reflected image in the water, gives most
artistic effects.
From the lake shore one may ascend the Beehive in
about a quarter of an hour. The pitch is very steep but
the ascent is easy and exhilarating, for the outcropping
ledges of sandstone seem to afford a natural staircase,
though with irregular steps. Everywhere are bushes and
smaller woody plants of various heaths, the tough strong if"
44 The Canadian Rockies.
branches of which, grasped in the hand, serve to assist the
climber, while occasional trees with roots looped and knotted over the rocks still further facilitate the ascent.
Arrived on the flat summit, the climber is rewarded for
his toil. One finds himself in a light grove of the characteristic Lyall's larch, while underneath the trees, various
ericaceous plants suggest the Alpine climate of the place.
Though the climber may come here unattended by
friends, he never feels the loneliness as at Lake Agnes.
There the gloomy mountains and dark cliffs seem to surround one and threaten some unseen danger, but here the
broader prospect of mountains and the brilliancy of the
light afford most excellent company. I have visited this
little upland park very many times, sometimes with friends,
sometimes with the occasional visitors to Lake Louise,
and often alone. The temptation to select a soft heathery
seat under a fine larch tree and admire the scenery is irresistible. One may remain here for hours in silent contemplation, till at length the rumble of an avalanche from
the cliffs of Mount Lefroy awakens one from reverie.
The altitude is about 7350 feet above sea level and
in general this is far above the tree line, and it is only
that this place is unusually favorable to tree growth that
such a fine little grove of larches exists here. Nevertheless, the summer is very brief—only half as long as at Lake
Louise, 1700 feet below. The retreating snow-banks
of winter disappear toward the end of July and new
snow often covers the ground by the middle of September.     How could we  expect   it to be otherwise at this Summit of the Beehive. 45
great height and in the latitude of Southern Labrador ?
On the hottest days, when down in the valley of the Bow
the thermometer may reach eighty degrees or more, the
sun is here never oppressively hot, but rather genially
warm, while the air is crisp and cool. Should a storm
pass over and drench the lower valleys with rain, the air
would be full of hail or snow at this altitude. The view
is too grand to describe, for while there is a more extensive prospect than at Lake Louise the mountains appear
to rise far higher than they do at that level. The
valleys are deep as the mountains high, and in fact this
altitude is the level of maximum grandeur. The often
extolled glories of high mountain scenery is much overstated by climbers. What they gain in extent they lose
in intent. The widened horizon and countless array of
distant peaks are enjoyed at the expense of a much
decreased interest in the details of the scene. In my
opinion one obtains in general the best view in the
Canadian Rockies at the tree line or slightly below.
Nevertheless every one to his own taste.
The most thrilling experience to be had on the
summit of the Beehive is to stand at the verge of the
precipice on the east and north sides. One should
approach cautiously, preferably on hands and knees, even
if dizziness is unknown to the climber, for from the very
edge the cliff drops sheer more than 600 feet. A stone
may be tossed from this place into the placid waters of
Mirror Lake, where after a long flight of 720 feet, its
journey's end is announced by a ring of ripples far below. 46
The Canadian Rockies.
Lake Louise appears like a long milky-green sheet of
water, with none of that purity which appears nearer at
hand. The stream from the glacier has formed a fan-
shaped delta, and its muddy current may be seen extending far out into the lake, polluting its crystal water and
helping to fill its basin with sand and gravel till in the
course of ages a flat meadow only will mark the place of
an ancient lake.
There are even now many level meadows and swampy
tracts in these mountains which mark the filled-up bed of
some old lake. These places are called " muskegs," and
though they are usually safe to traverse, occasionally the
whole surface trembles like a bowl of jelly and quakes
under the tread of men and horses. In such places let
the traveller beware the treacherous nature of these
sloughs, for on many an occasion horses have been suddenly engulfed by breaking through the surface, below
which deep water or oozy mud offers no foothold to the
struggling animal.
At the present rate of filling, however, the deep basin
of Lake Louise will require a length of time to become
obliterated that is measured by thousands of years rather
than by centuries,—a conception that should relieve our
anxiety in some measure. n
CHAPTER   IV.
Organizing a Party for the Mountains—Our Plans for the Summer—
William Twin and lorn Chiniquy—Nature, Habits, and Dress of the
Stoney Indians—An Excursion on the Glacier— The Surface Debris and
its Origin—Snow Line—Ascent of the Couloir—A Terrible Accident—
Gating Down—An Exhausting Return for Aid—Hasty Organization of
a Rescue Party—Cold and Miserable Wait on the Glacier— Unpleasant
Surmises—" / Think You Die "—A Fortunate Termination.
PREVIOUS to the summer of 1894 my experiences
in the Canadian Rockies had made me acquainted
with but little more of their general features
and scenery than has been already described. This
was sufficient, however, to prove that a most delightful
summer could be spent among these mountains if a party
of young men were organized with some definite object in
view to hold the party together. Several of us accordingly assembled at one of our eastern colleges and discussed plans for the summer. Four men were persuaded
to go on this excursion after the glories of the region
had been duly set forth and the evidence corroborated so
far as possible by the use of photographs. We were to
meet at Lake Louise, where our headquarters were to be
at the chalet, as near the first of July as possible.
Though  the   individual   inclinations   of   the   various
members of our party might seem unlikely to harmonize
47 r
48 The Canadian Rockies.
together, we had nevertheless agreed on carrying out a
certain plan. One of the party was an enthusiastic
hunter, another eager for the glories of mountain ascents,
one a geologist, another carried away by the charms of
photography, while the fifth and last was ready to join in
almost any undertaking or enterprise whatsoever.
However, our common purpose joined us all together
to a certain degree. This was to explore and survey the
region immediately around Lake Louise, to ascend several
of the highest peaks, to get photographs of the best
scenery, and in general to learn all we could about the
environment of the lake.
Three of us arrived at the lake one fine morning early
in July. The beauty of the scenery seemed to make a
deep impression on my friends, and fortunately the clouds
which at first concealed the mountain tops lifted soon after
our arrival and produced very grand effects. At that time
there were two Stoney Indians at the lake, who were engaged in cutting a trail to a lately discovered point of
interest. One of these was named William Twin ; his
surname was probably derived from the fact that he had a
twin brother, whose name was Joshua. A Stoney Indian
who once acted as my guide was named Enoch ; and upon
being asked his surname he replied, M WTildman." These
curious cases afford good examples of the origin of names.
William was a fine-looking Indian. He came nearer to a
realization of the ideal Indian features such as one sees on
coins, or in allegorical figures, than almost any savage I ,
have ever seen. Stoney Indians. 49
Tom Chiniquy was the other of the two Indians, and
indeed the more important, as he is the eldest son of Chief
Chiniquy, who in turn is under Bears' Paw, the head chief
of all the Stoneys. An air of settled gravity, stern and
almost bordering on an appearance of gloom, betokened
his serious nature. I cannot but admire these Stoney Indians, free as they are from the vices of civilization, while
still retaining many of the simple virtues of savage life.
As we saw the Indians every day we soon became acquainted with them, especially as William could talk quite
intelligibly in English. The very first day of our arrival
at the chalet the sharp eyes of the Indians, which seemed
to be ever roving about in search of game, discovered a
herd of goats on the mountain side. In vain did we try
to see them, and at length, i>y means of a pair of powerful
field glasses, they appeared as small white spots without
definite forms, whereas to the Indians they were plainly
visible. William was disgusted with us, and said, " White
man no good eyes," in evident scorn.
With practice, our race can excel the Indians in every
undertaking requiring skill, patience, or physical endurance, with the exception of two things in which they are
infinitely our superiors. These are their ability to discover
minute objects at great distances, and to read those faint
and indefinite signs made by the passage of man or game
through the forests or on the hard plains, where a white
man would be completely baffled. A turned leaf, a bent
blade of grass, a broken twig, or even the sheen on the
grass, leads the swarthy savage unerringly and rapidly 5o
The Canadian Rockies.
along, where the more intelligent but less observant white
man can see absolutely nothing.
The Indian is said to be stolid and indifferent, while
the hard labor which the squaws are compelled to undergo
is always laid up against them as an evidence of their
brutal character. But on the contrary this is their method
of dividing labor, and a squaw whose husband is compelled
to work about their camps is the subject of ridicule among
the rest. The squaws do all the work which rationally
centres around the camp-fire, just as our wives preside
over our hearths and homes. The bucks provide the food,
and should privation occur they will cheerfully share their
last morsel with their wives and children, and, the more
honor to them, they will do the same by a white man.
The long and arduous labors of the chase, requiring the
severest physical exertion, exhaust the strength, often while
exposed to cold and rain for long periods of time. The
bucks rightly consider their labor ended when they reach
their camp, or | teepee " as they call them. Here the squaws
preside and perform all the labor of cutting and cooking
the meat, preserving and dressing the hides, and even
gathering the firewood. They cut the teepee poles and
set up their tents ; and when not occupied with these more
severe labors, they spend their time in making moccasins,
weaving baskets, or fancy sewing and bead-work.
After all, the poor Indian is our brother, and not very
unlike his civilized conqueror. One day William told me
that the year before he had lost his squaw and four children by the smallpox, and that it had affected him so that \\
Tom  Chiniquy.
By courtesy of Mr. S. B.  Thompson.
New Westminster, B. C.  1 m 1
Indian Pathos. 51
he could not sleep. In his own simple form of expression,
it was most pathetic to hear him speak of this sad event,
which evidently affected him deeply. - Me sleep no more
now," he would say, " all time think me, squaw die, four
papoose die, no sleep me. One little boy, me—love little
boy, me—little boy die, no longer want to live, me."
We had the satisfaction of rendering a great service
to William through his child, who was a bright and handsome little fellow. By some accident a splinter of wood
had become lodged in the boy's eye. We were at length
attracted by the peculiar actions of the little fellow, and
upon inquiry found that he must have been enduring
great pain, though without making a murmur of discontent. We took the matter in hand at once and sent him
down to Banff, where, under skilful medical attendance,
his eyesight, than which nothing is more dear to an Indian
and which was totally gone in the affected eye and partially so in the other, was restored in a great measure.
William was very grateful to us ever after, and on returning, some ten days later, delivered himself somewhat as
follows : " Me say very much obliged. Three white men
pretty good, I think."
The Stoneys are a remarkable tribe of Indians. Their
headquarters is at a little place called Morley, about twenty
miles east of the mountains on the plains. Here they are
under the religious instruction of the Rev. Mr. McDougal.
So far as the Indian is capable of receiving and following
the precepts of Christianity, the Stoneys seem to have
equalled or surpassed all other tribes.    They are said to 52
The Canadian Rockies.
be great Bible readers, and they certainly show some
familiarity with the Old Testament history, if we may
judge by tfoeir custom of adopting Bible names. They
have been taught a certain arbitrary code by which tney
can read and write in a simple manner, while many of
them talk English if not fluently at least intelligibly.
Their manner of dress is a concession to their own
native ideas and those of civilization, for while they invariably cling to moccasins and usually affect trousers cut
from blankets with broad wings or flaps at the sides, their
costume is not infrequently completed by some old discarded coat received by purchase or gift from the white
man. These Indians rarely wear hat or cap, but allow
their straight black hair to reach their shoulders and serve
in place of any artificial protection. On either side of
the face the hair is gathered into a braid so as to do away
with the inconvenience of constantly pushing back their
loose hair.
Dr. Dawson says that the Stoney Indians have very
few names for the mountains and rivers, and that they
have only inhabited this region for about forty years.
The greater part of the Indian names for various features
of the country are in reality Crée or their equivalents in
Stoney. The Stoneys have recently incorporated the
families of the Mountain Crées with their own. According to De Smet, both the Crées and the Stoneys migrated
southward from the Athabasca region a few years before
1849, and it is probable that they entered this region
about that time. «ç%~
Palliser's Opinion. 53
I cannot conclude this digression on the Stoney
Indians without quoting a few remarks from Captain
Palliser's reports. Though written nearly forty years
ago these facts are no less true than at that time.
J? The members of the Stone tribe are hard workers,
as their life is one requiring constant exertion and foresight. They travel in the mountains or in the forests
along their eastern base, in parties of six or seven families.
The young men are always off hunting in search of moose
or other kinds of deer, or of the Rocky Mountain sheep.
The old men busy themselves cutting out the travelling
tracks through the woods, while the women pack and
drive the few horses they use for earring their small
supplies. They generally use skin tents stretched on a
conical framework of poles, but their wigwams are much
smaller than those of the Plain Indians. The women
dress all the skins of the animals they kill into a soft
leather, which, when smoked, is the material used throughout the whole country for making moccasins, most of the
fine leather being obtained from the Stoneys. They are
excellent hunters, and though as a rule small and feeble
in body, are probably capable of more endurance than
any other class of Indians. They make trustworthy
guides, and, with a few exceptions, after some acquaintance with this tribe, you no more expect to be deceived,
or told lies, as a matter of course, than you would in a
community of white men."
So much for the Rocky Mountain Stoneys, or as they
are sometimes called, the Assiniboines.
Êttm 54 The Canadian Rockies.
The completion of our party did not take place at the
wished-for time, and for more than two weeks Mr. F. and
Mr. H., and I were alone at the chalet. We commenced
our surveying work by measuring a very accurate base line
on the lake shore, and began training by making various
moderate excursions on the mountain sides. On the third
day, however, after our arrival the whole plan of our party
came near having a most sudden and unwished-for termination, together with results which nearly proved fatal to
one of the party. The accident and its attendant circumstances proved the most exciting episode in all our experiences, and as it most clearly illustrates the chief danger
of climbing in the Canadian Rockies, I shall describe it
in detail.
It happened in this manner. On the 13th of July, Mr.
H., Mr. F., and I started to make an exploration of the
glacier that is plainly visible from the chalet and which,
some two miles distant, flows down from the snow fields
and hanging glaciers of Mount Lefroy. This glacier is
formed from two branches, which come in from the east,
and uniting into one great stream, terminate about one
mile above the head of the lake. The extreme length
from the snout measured to the highest part of the glacier is about three miles, while the average width is less
than one third of a mile.
The object of this excursion was in great part to gain
a little knowledge of the use of rope and ice-axe, which we
expected would be required in much of our subsequent
work.    There was no difficulty in the first part of this =v
Glacier Debris. 55
excursion, as a good trail leads round the lake and
some half-mile beyond. There we forded the icy stream
which comes from the glacier and pursued our way between the moraine and the mountain side for nearly a
mile on the east side of the glacier. Our next move was
to ascend the moraine, which was very steep and about a
hundred feet high at this point. On arriving at the sharp
crest of the moraine, we saw the great ice stream some
fifty feet below, and so thoroughly covered with debris
and boulders that the glacier was almost totally concealed.
The passage down the moraine was very disagreeable, as
the loose stones all scratched and polished by their former
passage under the glacier were now rolling from under
our feet and starting up great clouds of dust. Just below,
at the border of the glacier, the water from the melting
ice had converted the clay of the moraine into treacherous
pools of bluish-gray mud, veritable sloughs of despond.
At length, by the use of our ice-axes, we gained the firmer
ice and with it the advantage of far more pleasant walking. We found the whole surface of the glacier literally
covered with sharp stones and boulders of all sizes up
to those which must have measured ten feet square by
twenty feet long. They represented all sorts of formations, shales, limestones, and sandstones thrown down in
wild disorder over the entire surface of the ice. All this
material had been wrested from the mountain side far up
the valley by frost and avalanche, and was now slowly
moving toward the great terminal moraine. In one place
a large area of nearly half an acre was strewed with giant ^^^^^^^^^^^mmmmss^^wmmm^sjm
56 The Canadian Rockies.
blocks of a peculiar kind of rock different from all the
rest, which apparently had come thundering down the
mountain walls in one great rock-slide many years ago.
Large flat slabs of shale were seen here and there supported on pillars of ice, showing how much the general
surface of the glacier had wasted away under the influence of the sun's heat, while these pillars had been protected by the shade of the stone.
Advancing half a mile over the field of debris, we came
gradually to where there were fewer stones, and at length
reached almost pure ice. The question always arises
where do all the boulders and pebbles that cover the
lower parts of the glaciers come from ? In the upper
parts of the glaciers or névé regions, where the snow
remains perpetual and increases from year to year, the
stones from the mountain sides are covered as they fall,
and are at length buried deep and surrounded by ice as
the snow becomes compressed and solidified. As the
glacier advances down the valley and descends to lower
altitudes, a level is at length reached where the snowfall
of winter is exactly balanced by the melting of summer.
This is the snow line, or rather this is the best place in
which to locate such a variable level. Below this line the
surface of the glacier melts aWay more than enough to
make up for the winter fall of snow, and, as a result, the
stones and debris buried in the ice gradually appear on
the surface. In the Canadian Rockies near this latitude
the snow line on northerly exposures, as judged by this
method, is about 7000 feet above the sea, which is also
just about the level called tree line. On the Glacier. 57
In mountainous regions, where the climate is very dry,
as in Colorado or in certain parts of the Andes, there is
a great belt of several thousand feet between tree line
and snow line where there is not sufficient moisture to
allow of tree growth nor sufficient snowfall to form
glaciers at all. In the Canadian Rockies the climate is
moist enough to make these lines approach, and in the
Selkirk Range and regions of extreme humidity the snow
line is actually lower than the tree line.
We advanced slowly over the glacier and found much
of interest on every side. The surface of the ice was at
first comparatively smooth and channelled with small
streams of pure water which flowed along with utmost
rapidity but almost without ripples, as the smooth icy
grooves seem adapted to every whim of the flowing water.
At length the ice became more uneven and our passage
was interrupted by crevasses, around which we had to
thread our way by many a turn and detour. Most of
them were, however, partly filled or bridged by snow and
we found no particular difficulty in pursuing our way.
About one o'clock we found ourselves at the base of
Mount Lefroy, a little beyond the point where the two
branches unite, and we held a consultation as to the plan
of our farther advance. Mount Lefroy rises from the
glacier in precipitous cliffs on every side, and we were
even now under the shadow of its gloomy and threatening
rock wall. There is no apparent method of scaling this
mountain except by a long couloir or snow slope, which
rises from the glacier and ascends nearly 1000 feet to a
more gentle slope above the precipice.    It was our inten-
mmm 5& The Canadian Rockies.
tion to ascend this mountain, if possible, some time during
the summer but the results of our first exploration fora
favorable route rather inclined us to give up further
attempts.
The result of our consultation was the decision to
climb a short way up the couloir in order to see if it were
possible to reach the gentle slope above. If this proved
practicable, the ascent of the mountain was almost assured,
as no great difficulties presented themselves above. Accordingly we commenced the ascent, all roped together in
true Alpine fashion, and soon found the pitch so steep
that our ice-axes rendered us much assistance in cutting
steps. A number of great schrunds or horizontal crevasses often found on such slopes appeared to block our
way, but as we approached we found a passage round
every one. They were boat-shaped holes in the snow
some forty or fifty feet deep and about the same width.
The bottom of each appeared smooth and apparently of
firm snow, so that they were not in reality very dangerous
obstacles, as compared with the narrow and wellnigh
unfathomable crevasses of an ordinary glacier.
Nevertheless, when we had reached a point several
hundred feet above the schrunds and were on a steep slope
of snow, my companions advocated taking to the rock
ledges on the right of the snow, as they were altogether
inexperienced in mountain climbing and felt somewhat
nervous. We found the rock ledges practicable and quite
easy except for a great number of loose stones which
went  rattling  down   as  we  advanced.    We  were  in a A Terrible Accident. 59
gloomy narrow gorge filled with snow and hemmed in on
either side by cliffs which rose with almost vertical sides,
here and there dripping with water from the snows above.
Whenever we paused for a momentary rest and the
sliding, rattling stones ceased to fall, we were oppressed
by the awful silence of this cheerless place of rocks and
snow nearly 8000 feet above sea level.  .
It was while ascending these rock ledges that the accident occurred which came so near proving disastrous.
There were a series of ledges, from six to ten feet high
alternating with'narrow shelves where the slope was only
moderately steep. The whole place was strewed with loose
stones and boulders, some of which were so delicately
poised that the slightest touch seemed sufficient to send
them crashing down the cliff. At length a very dangerous
looking stone of large size could be seen on the next shelf
above us apparently just balanced in its precarious position,
for the light could be seen underneath its base. H. followed me in safety around this great boulder which must
have weighed more than half a ton. I was on the point
of ascending the next ledge with the assistance of H. when
we both heard a dull grating sound below, and turning,
beheld the great boulder starting to roll over, and F. just
below it and on the point of falling over the cliff. F.
fell about ten feet to the next shelf where he was partially
checked by the rope and prevented from falling farther.
But to our horror the boulder, which had now gained
considerable motion, followed after, and leaping over the
ledge, for a short but awful moment it seemed to hang in
mmm 6o The Canadian Rockies.
mid-air, and then came down on F. with terrible force. It
seemed impossible that there should be anything left of
our poor friend. With a horrible crash and roar the great
stone continued down the gorge, attended by a thousand
flying fragments till the rocky cliffs echoed again.
After a momentary pause, unable to move and riveted
to our places in horror, we hastily scrambled down to our
companion who lay on the cliff insensible and bleeding.
Our first efforts were to staunch his wounds with snow
and then a hasty examination proved that though his hip
appeared dislocated he had received probably no further
serious injury. This escape appeared almost miraculous
and it is probable that in the flying cloud of stones a
smaller piece just happened to come under the great
boulder and supported it partially at one end so that the
full force of the blow was not felt. It was now half-past
two in the afternoon and we were three hours' journey from
the chalet with a man on our hands absolutely incapable
of walking or even partially supporting his weight. It
was evident that one of us must needs hasten back to the
chalet for aid, but first it was necessary to get down the
long snow-slope to the glacier.
Fortunately our rope was fully sixty feet long and
after tying a loop under F.'s shoulders, I anchored myself
'securely with -my ice-axe in the snow, and then lowered
him rapidly but safely the length of the rope. H. then
went down to F. and held him while I descended, and
thus after twelve or fifteen repetitions of this proceeding
we all landed in safety on the glacier.    Having selected a To the Rescue.
61
place on the ice which was partially covered with a few
small stones, we took off our coats and placed our wounded
companion on this hard cold couch.
Carrying nothing but my ice-axe, I started for the
chalet at once. The first part of the journey, while threading the crevasses, was slow and somewhat dangerous without
the rope, but by running whenever practicable and pushing
my energies to the utmost, I reached the chalet in one
hour and ten minutes, or less than half the time required
by us to come up in the morning. Unfortunately no one
was at the chalet except Joe the cook. I however got
him started immediately to cut two long, stout poles and a
piece of canvas with which to make a litter. The two
Indians were on the mountain side near Mirror Lake
working on the trail and Mr. Astley, the manager of the
chalet, was guiding some visitors to Lake Agnes. There
was no other course open than to climb up after them,
though I was quite exhausted by this time. I found
William after twenty minutes of hard climbing and made
him understand the situation at once. One must use a
simple manner of speech as near like their own as possible,
so I said to him—" William, three white men go up big
snow mountain. Big stone came down, hurt one man.
Tom, Mr. Astley, you—all go up snow mountain, bring
white man back." William's face was a picture of horror,
and he asked jn anxiety—" Kill him ?" I said no, but that
he must hurry and get the other men. Dropping his axe,
he ran off for the others in all haste, while I returned to
the chalet and gathered sundry provisions and stimulants. 62 The Canadian Rockies,
The rescuing party of four men was started in about
thirty minutes, and taking the boat, rowed down the lake,
till at last the small black speck on the water disappeared
from our view as they neared the farther end.
A two-and-a-half mile ride on horseback brought me
to the railroad station, where I sent a telegram to Banff
for the Doctor. As there would be no train till the next
morning I made arrangements for a hand-car to bring
the Doctor up at once. A response soon came back that
he was just about to start on his long ride of thirty-eight
miles to Laggan.
Meanwhile poor F. and H. were having a miserable
time of it on the glacier. The long hours rolled by one
after another and no sign of aid or assistance was apparent. The days were still very long, but at length the
declining sun sank behind the great ridge or mountain
wall extending northward from Mount Lefroy. The glacier which imparts a chilly dampness even to the brilliancy
of a mid-day sun now rapidly became cold in the lengthening shadows, and the surface waters began to freeze,
while the deep blue pools of water shot out little needles
of ice with surprising rapidity.
As they had seen me no more after I had disappeared
behind a swelling mound of ice, they conjured up in their
imaginations the possibility that I had fallen into some
deep crevasse or had hurt myself on the treacherous
moraine. At length, urged to desperate resolves, they
formed a plan of leaving the ice by the nearest route,
at whatever hazard to life and limb, rather than die of Indian Consolation. 63
cold and exposure on the glacier. They had abundant
opportunity for studying the grand phenomena of this
Alpine region near at hand : the thundering avalanches
from the cliffs behind them, and the cracking, groaning
ice of the glacier as the great frozen stream moved slowly
over its rocky uneven bed.
At length, to their great joy, they discerned by means
of a field-glass which we had carried with us in the morning, the boat leaving the lake shore and slowly approaching. In half an hour the party reached the near end of
the lake and were then lost to view for nearly two hours,
till at length four little black dots appeared about a mile
distant moving over the ice toward them.
The rescuing party did not reach them till seven
o'clock, or more than four hours after the accident occurred. The return to the chalet was most exhausting
to the men, especially to the Indians, whose moccasins
afforded poor protection against the sharp stones and ice
of the glacier.
Two section men came up from Laggan and met the
party as they were returning, and afforded timely aid by
their fresh strength. Poor F. was carried in a canvas
litter hastily constructed and consequently not perfect
in its results, as it only served to lift him a very little
above the ground at the best and then where the ground
was very smooth. William observed his haggard face and
woe-begone appearance with concern and entertained the
invalid at frequent intervals by such remarks as, "You
think you die, me think so too."    The rescuing party ar- ztmmsmi
tSSBÇJÎ
The Canadian Rockies.
rived at the chalet shortly after midnight, while the Doctor
appeared an hour later. Each party had been travelling
for the last five hours toward the chalet, and while one
was accomplishing about three miles the other covered
more than forty.
Fortunately there were no injuries discovered that
would not heal in a few weeks, and through the influence
of mountain air and perfect rest, recovery took place much
more quickly than could be expected. 1
CHAPTER V.
Castle Crags—Early Morning on the Mountain Side—View from the
Summit—Ascent of the Aiguille—An Avalanche of Rocks—A Glorious
Glissade—St. Piran—Its Alpine Flowers and Butterflies—Expedition to
an Unexplored Valley—A Thirsty Walk through the Forest—Discovery
of a Mountain Torrent—A Lake in the Forest—A Mountain Amphitheatre— The Saddle—Impressive View of Mt. Temple—Summit of Great
Mountain—An Ascent in Vain—A Sudden Storm in the High Mountains
—Phenomenal Fall of Temperature—Grand Cloud Effects.
WHILE poor F. was recovering from his injuries, and before the two other men had
arrived, H. and I carried on the work of
surveying the lake, and made several interesting excursions on the adjacent mountain sides.
One fine cool morning, we went up the valley about
half a mile beyond the end of the lake, and commenced
an ascent of the sharp-crested ridge on the east side of
the valley. This ridge forms a connection between the
massive mountain on the left of the lake, known as Great
Mountain, and a very high summit, crowned with a fine
glacier, and named by some one Hazel Peak, which lies
about two miles due south of Lake Louise. This connecting ridge we called Castle Crags, a name readily suggested by the irregular forms and outlines of the sharp
5 65
j^àm 66 The Canadian Rockies.
needles and fingers, pointing heavenward, which adorned
its highest crest, and seemed to represent the battlements and embrasures of some great castle. Several
sharp columns of stone, with vertical sides, and narrow,
graceful forms, rose up from this great parapet built by
nature. Resembling feudal towers or donjons, they
seemed by their great altitude to pierce the blue vault
of heaven, and to dwarf by their proximity the snowy
crest of Hazel Peak, which, in reality, is several thousand
feet higher.
To ascend this ridge, and, if possible, gain the summit
of one of these needles, from which we hoped to obtain a
fine idea of the valley to the east, was the purpose of our
excursion. The ascent proved easy almost from the start.
On leaving the stream, which we crossed by means of
some great trees, long since overcome by age or storm,
and now serving as convenient bridges at frequent intervals, we commenced to ascend a long, even slope of limestone boulders, stable in position, and affording easy walking. The air was fresh and cool, for the morning sun
was just rising over the crest of Castle Crags, while the
rays of light seemed to skip from boulder to boulder, and,
gently touching the higher points, left the others in shade.
There were no bushes or tangled underbrush to impede
our way, and so we had abundant opportunity to enjoy
the beautiful flowers which cropped out in little patches
among the yellow, gray, and cream-colored limestones.
This was a mountain climb that proved thoroughly enjoyable, for all the conditions of atmosphere, of weather, and Morninsf on the Mountain Side.
6?
easy ascent were in our favor. There is a charm about
the early morning hours among the high mountains. The
bracing coolness of the air, as .yet still and calm after the
chill and quiet of night, the gradually rising sun and increasing light, the unusual freshness of the flowers and
green vegetation, in their sparkling bath of dew, and the
quiet calls of birds,—all seemed to herald the birth of a new
day, far richer in promise than any heretofore. The afternoon, with its mellow light and declining sun, is like the
calm, cool days of October, with its dusty foliage and sear
leaves, brilliant in autumnal colors, but ever suggesting
the approach of bleak winter, and pointing back to the
glories of the past. The morning points forward with a
different meaning, and hopefully announces the activity of
another day, even as spring is the threshold and the
promise of summer time.
As we advanced, and gradually increased our altitude,
the plants and flowers changed in variety, character, and
size, till at length we left all vegetation behind, and
reached the bottom of a long, gentle slope of snow. The
sun had not, as yet, touched the snow, and it was hard
and granular in the frosty air. The first snow on a
mountain climb is always pleasant to a mountaineer. To
him, as, indeed, to any one, the summer snow-bank has
no suggestion of winter, with its desolate landscapes and
cold blasts, but rather of some delightful experiences in
the mountains during vacation. These lingering relics of
winter have little power to chill the air, which is often
balmy and laden with the fragrance of flowers, in the im- 68
The Canadian Rockies.
mediate vicinity of large snow areas. The trickling
rivulet, formed from the wasting snows of the mountain
side, is often the only place where, for hours at a time, the
thirsty climber may find a cold and delicious draught.
Instead of destroying the flowers by their chilly influence,
these banks of snow often send down a gentle and constant supply of water, which spreads out over grassy slopes
below, and nourishes a little garden of Alpine flowers,
where all else is dry and barren.
Arrived at the top of the long snow-slope, we found ourselves already nearly 3000 feet above the valley and not
far below the crest of the ridge. A rough scramble now
ensued over loose limestone blocks, where we found the
sharp edges, and harsh surfaces of these stones, very hard
on our shoes and hands. Upon reaching the crest, we beheld one of those fearfully grand and thrilling views which
this portion of the Rocky Mountains often affords. The
most conspicuous object in the whole view was the glacier,
which descends from the very summit of Hazel Peak,
at an altitude of more than 10,000 feet, and sweeps down
in a nearly straight channel to the north, and in the course
of but little more than a mile descends 4000 feet. A
gloomy, narrow valley hems in its lower half, and on the
side where we were, the precipice rose, in nearly perpendicular sides from the ice, far heavenward to where we
stood. We launched a few large stones over the verge of
the beetling precipice, and watched them descend in a few
great leaps into the awful abyss, where they were broken
into a thousand fragments on projecting ledges, or else, Ascent of the Aiguille. 69
striking the glacier, continued their course till the eye
could no longer follow them.
We were standing just at the base of one of the
aiguilles which, from the valley, seem like sharp points of
rock, but, now that we were near, proved to be about sixty
feet high. This needle appeared to be precipitous and inaccessible on our first examination. But we discovered a
narrow crevice or gully on the west side which apparently
offered a safe method of ascent. I was soon near the top
of the needle, but at the most difficult part, where only
one small crack in the rock offered a good hand-hold, I
was warned not to touch one side where the cliff seemed
parted, and filled with loose material. Making a reconnaissance, I found the back of this same crag likewise separated
a little from the solid rock, and the crevice partially disguised by loose stones and dirt, which had settled in and
filled the hollow. This crag was about ten feet high and
six or seven feet square, and though it seemed impossible
to disturb so great a mass, I felt inclined to take the safer
course and leave it entirely alone, so I scrambled up by a
more difficult route.
Arrived on the top of the needle, I told H., who had remained below, to get under shelter while I should put this
' crag to the test. He accordingly found a projecting ledge
of rock a little to one side, while I sat down and got a
good brace and started to push with my feet against the
top of the crag. A slight effort proved sufficient, and
with a dull grating sound the great mass, which must have
weighed about twenty-five tons, toppled slowly over on its 7o The Canadian Rockies.
base, and then fell with a fearful crash against the sides of
the cliff, and commenced to roll down the mountain side
like a veritable avalanche. Through the cloud of dust
and flying stones I could faintly discern the features of my
friend below, apparently much interested in what was going
on. It was well that I had not trusted to this treacherous
stone.
After I had pushed down most of the loose stones, H.
came up and joined me on the summit of the aiguille.
This needle had a blunt point indeed, for it proved to be
a flat table about fifty feet long and ten feet wide.
We were 8,700 feet above sea-level, and the wind
was raw and chilly as it swept up from the valley and
over this ridge. The sun had but little power to temper
the air, and we soon started on our descent. In about five
minutes we reached the top of the long snow-slope, where
we enjoyed a glorious glissade and rapidly descended more
than a thousand feet. The best manner of glissading is
to stand straight up and slide on the feet, having one leg
straight and the other slightly bent at the knee. Trailing
the ice-axe behind as a precaution against too great speed,
or to check the motion in case of a fall, the mountaineer
can thus, in a few minutes, rapidly coast down long slopes
which may have required hours of toil to ascend. Nothing
in the experience of climbers is more exhilarating than a
good glissade down a long snow-slope. The rush of air,
the flying snow, and the necessity for constant attention
to balance—all give a sensation of pleasure, combined with
a spice  of  danger, without  which latter almost  all our St. Piran. 71
sports and pastimes are apt to be tame. Do not many of
our best sports, such as polo, horseback riding, foot-ball,
yachting, and canoe sailing, gain some of their zest from a
constant possibility of danger ?
A few minutes of rapid descent down the limestone
slope led us to a fine, small spring, which dashed in a
score of small streamlets over some rocky ledges covered
with moss and ferns. Here we sat down in the cool
shade of the cliffs and ate our lunch. The air was now
warm and still, because we were not far above the valley,
and here, instead of seeking the warmth of the sun as
we had done on the cold mountain summit, a brief three-
quarters of an hour before, we now enjoyed the shade
afforded by the rocks and forest near us. We reached
the chalet in time for a second lunch, and, as in our mountain exercise we never found any meal superfluous, we
were ready to present ourselves at the table at once.
On the 28th of July, W. arrived at the chalet, and, as
A. had likewise appeared a few days previously, our party
of five was now complete.
One of the first points which we decided to occupy in
our surveying work was a high peak above Lake Agnes,
called Saint Piran. This mountain is very easy to ascend and on several occasions we found ourselves on the
summit for one purpose or another. The summit is far
above tree line and, indeed, almost reaches the upper limit
of any kind of plant growth. The rounded top is crowned
with a great cairn, about ten feet high, which has been
used as a surveying point some time in the past. *
wsm^mffimmmm
72 The Canadian Rockies.
During the midsummer months this mountain summit
is sparingly covered with bright flowers, all of an Alpine
nature, dwarfed in size and with blossoms enormously
out of proportion to the stems and leaves. There, are
several species of composites which rest their heads of
yellow flowers almost on the ground, and a species of dwarf
golden-rod about three inches high, with only two or
three small heads on the summit of the stem ; but the
most conspicuous is a kind of moss pink, which is in reality a mountain variety of phlox. This plant grows in
spreading mats upon the ground, with small, rigid, awl-
shaped leaves gathered in tufts along the stem, while here
and there are small bright blossoms of a pink color. Mr.
Fletcher, who has spent some time in this region investigating the flowers and insects, once found a plant of the
pink family on this mountain, which proved by its little
joints to be more than one hundred years old.
One day I came up here alone, and on reaching the
summit was surprised to find Mr. Bean, an entomologist,
busily at work collecting butterflies. Mr. Bean has lived at
Laggan for a number of years, and has made a most valuable collection of the insects, especially the butterflies and
beetles, of all this region. Remarkably enough, it is on just
such spots as this lofty mountain summit, 8600 feet above
tide, that the rarest and most beautiful butterflies assemble in great numbers, especially on bright, sunny days.
Here they are invited by the gaudy Alpine flowers, which
have devoted all their plant energy to large blossoms and
brilliant colors, so as to attract the various insects to them. Alpine Butterflies. 73
I was much interested in Mr. Bean's work, as he is the
first pioneer in this field and has made many valuable discoveries. He showed me one butterfly of small size and
quite dark coloring, almost black, which he said was a
rare species, first discovered in polar regions by the Ross
expedition, and never seen since till it was observed flitting about on this high peak, where arctic conditions prevail in midsummer. It is wonderful how the various
species vary in color, form, and habit ; some of the butterflies are very wild and shy, never allowing a near
approach by the would-be collector ; others are comparatively tame ; and while some fly slowly and in a straight
course, other species dart along most rapidly, constantly
changing direction in sharp turns, and completely baffle
all attempts at pursuit.
From the summit of this mountain we discovered a
small lake in the valley to the west, and, as no one at the
chalet had apparently ever visited the lake, or even known
of its existence, we decided to make an excursion to this
new region. Accordingly, a few days later, three of us
started by the trail toward Lake Agnes, and after reaching a point about 600 feet above Lake Louise, we turned
to the right and endeavored to make a traverse around
the mountain till we should gain the entrance to the
other valley. Our plan was not very good and the results
were worse. For about two miles, the walking was along
horizontal ledges of hard quartzite rock carpeted with
grass and heaths, and occasionally made very difficult by
the short dwarf spruces and larches which, with their tough ^^^^^^^^^gs^«»aj^4icfeia«â^aigai^wy^
74 The Canadian Rockies.
elastic branches, impeded our progress very much. The
day was unusually warm, and we were glad to reach at
length a small patch of snow, where we quenched our
thirst by sprinkling the snow on large flat stones, the heat
of which melted enough to give us a small amount of
muddy water. The roughness of the mountain and the
nature of the cliffs now compelled us to descend near
a thousand feet, and thus lose all the benefit of our first
ascent. We were constantly advancing westward, hoping
to come at length upon some stream that must descend
from the valley of the little lake. Every valley in these
mountains must have some stream or rivulet to drain
away the water resulting from the melting snows of winter
and the rains of summer, and we were certain that, if we
continued far enough, we would finally discover such a
stream. After our descent we proceeded through a fine
forest, densely luxuriant, and in some places much blocked
by prostrate trees and giant trunks, mossy and half decayed. The air seemed unusually dry, and our thirst,
which had been only in part appeased by our draught at
the snow-bank, now returned in greater severity than
ever.
Suddenly we heard a distant sound of water, which, as
we approached, grew still louder, till it burst into the full,
loud roar of a beautiful mountain stream. The water was
clear as crystal and icy cold, while nothing could exceed
the graceful beauty of the many leaps and falls of the
-stream as it dashed over its rocky bed. Here we took
lunch in a shady nook, seated on some rocky ledges at A Mountain Torrent.
75
the edge of the water, surrounded on all sides by deep
cool forests. How wild this little spot was ! Though
the railroad was less than two miles distant, probably no
white man had ever seen this pleasant retreat where we
were resting.
Had our excursion ended here, we
should have
been repaid
for all the toil,
heat, and
thirst we had
endured, by
this single experience.
Nor was
our pleasure
over,   for  the
stream, we knew, would prove a certain guide to the little
lake, and, with the anticipation of soon reaching some
enchanting bit of scenery when we should arrive at this
sheet of water, we pursued our way along the series of
falls and cascades by which our new-found stream leapt
merrily down the mountain slope. Such is the charm of
mountain excursions in these unexplored and little known
wilds, for here, nature is ever ready to please and surprise
the explorer by some little lake or waterfall or a rare bit
of mountain scenery.
A COOL RETREAT IN THE FOREST.
-1 0H
76 The Canadian Rockies.
Though we had stopped for luncheon at a place where
the dashing water made several cascades and falls of
exquisite beauty, we found a constant succession of similar spots, where I was often tempted to delay long enough
to take photographs. As the stream thus descended
rapidly, we found steep rock ledges, cut in giant steps and
overgrown with thick moss till they were almost concealed from view, on either side of the mad torrent. These
afforded us an easy method of ascent. The rocky formation of the stream bed revealed many different kinds of
stone, conglomerates, shales, and quartzites, in clearly
marked strata all gently dipping toward tte south.
At length the woods opened up on either side, while,
simultaneously, the slope decreased in pitch, and the stream
ran over a bed of loose, rounded stones and boulders in
the bottom of a shallow ravine. In a moment more we
reached the lake, much more beautiful than our first view
from St. Piran had led us to expect, but, also, much smaller
in area. It was a mere pool, clear and deep, but intensely,
blue in color and partially surrounded by a thin forest.
Passing round the shores and up the valley, we found ourselves in some beautiful meadows, or rather moors, wherein
streams of snow-water wandered in quiet, sinuous courses
and gathered at length into the stream that feeds the lake.
We came on a great number of ptarmigan—the high
mountain species of grouse characteristic of this region,—
which, with their young broods hardly able as yet to fly,
were the most abundant signs of life that we found in this
valley. The Saddle. 77
A vast amphitheatre or cirque, with lofty, bare walls
nearly free of snow, formed the termination of the valley.
We were not compelled, however, to return over the same
route as we had come, for we found an easy pass with a
long gentle slope of snow on our left. This led us over
the divide and, by a long steep descent, brought us to
Lake Agnes, where we took advantage of the trail down
the mountain side to the chalet.
Our attention was next turned toward the exploration
of the mountains and valleys to the east of Lake Louise,
which seemed to offer greater possibilities of grand scenery
than those on the opposite side. Accordingly, we made
several visits to a high upland park or alp, which was in
reality a sort of depression between Great Mountain and
a lesser peak to the east. This depression and the two
mountains, one vastly higher than the other, resemble in
outline, a saddle with pommel and crupper and suggested
a name for the place which seems eminently appropriate.
A trail now leads to the Saddle, and the place has proven
so popular among tourists that it is frequently in use.
The Saddle is a typical alp, or elevated mountain
meadow, where long, rich grass waves in the summer
breezes, beautified by mountain flowers, anemones, sky-
blue forget-me-nots, and scarlet castilleias. Scattered
larch trees make a very park of this place, while the
great swelling slopes rise in graceful curves toward the
mountain peaks on either side.
But this is only the foreground to one of the most
impressive views in the Rocky Mountains.    To the east- 7*
The Canadian Rockies.
ward about three miles, on the farther side of a deep valley, stands the great mass of Mount Temple, the highest
peak near the line of travel in the Canadian Rockies.
This mountain stands alone, separated from the surrounding peaks of the continental watershed to which it does
not belong. Its summit is 11,658 feet above the sea-level,
while the valleys on either side are but little more than
6000 feet in altitude. As a result, the mountain rises
over a mile above the surrounding valleys, a height which
approaches the maximum reached in the Canadian
Rockies. All sides of this mountain, except the south,
are so precipitous that they offer not the slightest possible
hope to the mountain climber, be he ever so skilful. The
summit is crowned by a snow field or glacier of small size
but of remarkable purity, since there are no higher cliffs to
send down stones and debris to the glacier and destroy its
beauty. On the west face, the glacier overhangs a precipice, and, by constantly crowding forward and breaking off,
has formed a nearly vertical face of ice, which is in one
place three hundred and twenty-five feet thick. I have
seen passengers on the trains who were surprised to learn
that the ice in this very place is anything more than a
yard in depth, and who regarded with misplaced pity and
contempt those who have any larger ideas on the subject.
Avalanches from this hanging wall of ice are rather
rare, as the length of the wall is not great and the glacier
probably moves very slowly. I have never had the good
fortune to witness one, though the thunders of these ice
falls are often heard by the railroad men who live at Lag- Mount Temple from the Saddle.   mum Mt. Temple.
79
gan, just six miles distant. They must indeed be magnificent spectacles, as the ice must needs fall more than 4000
feet to reach the base of the cliff. The compactness of
this single mountain may be well shown, by saying that a
line eight miles long would be amply sufficient to encircle
its base, notwithstanding the fact that its summit reaches
so great an altitude.
The strata are clearly marked and nearly horizontal,
though with a slight upward dip on all sides, and especially
toward the Bow valley, so that the general internal structure of the mountain is somewhat bowl-shaped, a formation very common in mountain architecture.
The surroundings of this great mountain are equally
grand. Far below in the deep valley, the forest-trees
appear like blades of grass, and in the midst of them a
bright, foamy band of water winds in crooked course like
a narrow thread of silver,—in reality, a broad, deep stream.
A small lake, nestling among the dark forests at the very
base of Mount Temple, is the most beautiful feature in
the whole view. The distance renders its water a dark
ultra-marine color, and sometimes, when the light is just at
the proper angle, the ripples sparkle on the dark surface
like thousands of little diamonds. On the right, an awful
precipice of a near mountain looms up in gloomy grandeur, like the cliffs and bottomless abysses of the infernal
regions pictured by Doré.    This we called Mount Sheol.
One may ascend from the Saddle to the summit of
Great Mountain in an hour. Mr. A. and I ascended this
mountain in 1893, before there was any trail to assist us, mmmmuimtma&mHtif^mmssBSï
80 The Canadian Rockies.
and we had a very hard time in forcing our way through
the tough underbrush, while below tree line.
In the course of a great many ascents of this peak I
have had several interesting adventures. The view from
the summit is so fine that I have made many attempts to
obtain good photographs from this point. One day, after
a period of nearly a week of smoky weather, the wind suddenly shifted, and, at about ten o'clock in the morning, the
atmosphere became so perfectly clear that the smallest
details of the distant mountains were distinct and sharp,
as though seen through a crystal medium. This was my
chance, and I proceeded at once to take advantage of it.
I had a large 8x10 camera and three plate-holders, which
all went into a leather case especially made for the purpose, and which was fitted out with straps, so that it rested
between my shoulders and left both hands free for climbing. It weighed altogether twenty-four pounds. With
lunch in my pocket, I set out from the chalet with all
speed, so as to arrive on the summit before the wind
should change and bring back the smoke.
I climbed as I had never climbed before, and though
the day was hot I reached the Saddle in an hour, and,
without a moment's pause, turned toward Great Mountain and commenced the long ascent of its rocky slope.
In fifty-five minutes more I reached the summit and had
ascended 3275 feet above Lake Louise. The air was
still clear and offered every promise of successful photographs, even as I was unstrapping my camera and preparing to set it up for work.    Suddenly, the wind shifted An Ascent in Vain.
81
once more to the south and brought back great banks of
smoke, which came rolling over the snowy crest of Mount
Lefroy like fog from the sea. In five minutes all was
lost. Mount Temple appeared like a great, shadowy
ghost, in the bluish haze, and the sun shone with a pale
coppery light. Such are the trials and tribulations of the
climber in the Canadian Rockies.
One day at the end of August, H. and I ascended this
mountain with our surveying instruments. The barometer had been steadily falling for several days, and already
there were cumulus clouds driving up from the southwest
in long furrows of lighter and darker vapors, which obscured the entire sky. " A few drops of rain on the summit compelled me to work rapidly, but, as yet, there was
no warning of what was in store.
After all the principal points were located we packed
up our instruments and commenced a rapid descent to
the Saddle. The slope is of scree and loose material,
which permits a rapid descent at a full run, so that one
may gain the Saddle in about fifteen minutes. Arriving
there I paused to get a drink at a small stream under
some great boulders, fed by a wasting snow-bank. H.
had gone off toward the other side of the pass to get his
rifle, which he had left on the way up.
Suddenly I heard a rushing sound, and, looking up,
saw a cloud of dust on the mountain side and the trees
swaying violently in a strong wind. A mass of curling
vapor formed rapidly against the cliffs of Great Mountain, and a dull moaning sound, as of violent wind, seemed The Canadian Rockies.
to fill the air. The sky rapidly darkened and black clouds
formed overhead, while below them the thin wisps of scud
rushed along and seemed white and pale by contrast.
I was no sooner up on my feet than the approaching
blast was upon me, and with such unexpected force did
it come that I was laid low at the first impulse. My hat
went sailing off into space and was never seen more.
The first shock over, I gained my feet again and started
to find H. The air changed in temperature with phenomenal rapidity, and from being warm and muggy, in
the space of about five minutes it grew exceedingly cold,
and threatened snow and hail.
Though everything betokened an immediate storm
and a probable drenching for us, I had time to notice a
magnificent sight on Mount Temple. As yet there were
no clouds on the summit, but, as I looked, my attention
was called to a little fleck of vapor resting against the
precipitous side of the mountain, half-way between summit and base. So suddenly had it appeared that I could
not tell whether it had grown before my eyes or was there
before. From this small spot the vapors grew and extended rapidly in both directions, till a long, flat cloud
stretched out more than a mile, when I last saw it. The
vapors seemed to form out of the very air where a
moment before all had been perfectly clear.
Realizing that the sooner we started the better chance
we should have of escape, we flew rather than ran down
the trail, and were only overtaken by the storm as we
approached the lake.    The temperature had dropped so A Sudden Storm.
83
rapidly that a cold rain and damp snow were falling when
we reached the lake. The boat had drifted from its
moorings, and was caught on a sunken log some distance
from the shore. I waded out on a sunken log, where I
expected at any moment to slip from the slimy surface
and take an involuntary bath in the lake. The boat was
regained by the time H. had arrived a few minutes later
and we reached the chalet thoroughly drenched.
Such sudden storms in the Canadian Rockies are rather
rare, and are almost always indicated in advance by a
falling barometer and lowering sky. I have never at any
other time observed such a sudden fall in temperature,
nor seen the clouds form instantaneously far down on the
mountain side as they had done in this storm. The sudden rush of wind, the curling vapors, and flying scud
afforded a magnificent spectacle on the Saddle, and one
that was well worth the drenching we suffered in penalty. m
CHAPTER  VI.
Paradise Valley—The Mitre Glacier—Air Castles—Climbing to the
Col—Dark Ice Caverns—Mountain Sickness—Grandeur of the Rock-
Precipices on Mt. Lefroy—Summit of the Col at Last—A Glorious Vision
of a New and Beautiful Valley—A Temple of Nature—Sudden Change
of Weather— Temptation to Explore the New Valley—A Precipitate
Descent—Sudden Transition from Arctic to Temperate Conditions—Delightful Surroundings—Weary Followers—Overtaken by Night—A
Bivouac in the Forest—Fire in the Forest—Indian Sarcasm.
THE valley to the east of Lake Louise and parallel
to it, we named Paradise Valley, on account of
the elegant park-like effect of the whole place
and the beauty of the vegetation. Our first entrance
into this region and the discovery of the valley were
partially accidental. In fact, we were making an expedition for the purpose of finding a practicable route
up Hazel Peak, on the day when we were diverted from
our original plan, and tempted to explore this hitherto
unseen part of the mountains.
It came about somewhat in this manner. On the 30th
of July, all but F., who was still lame from his accident,
left the chalet carrying rope and ice-axes, with the intention of making explorations on the southern slopes of
Hazel Peak. Our party, numbering four, left the chalet
84 The Mitre Glacier. 85
at a little after eight o'clock, with the intention of returning
no later than five in the afternoon. Our equipment, beside our Alpine implements, consisted of a camera, a prismatic compass, and that which proved no less necessary,
our lunches and a whiskey flask.
Taking the boat, we rowed to the other end of the lake,
and then followed the same route as our party of three
had taken on the disastrous expedition of July 13th, till
we came to the junction of the two glacier streams. Here
we turned toward the east, and followed the moraine of
the wide glacier between Mount Lefroy and Hazel Peak.
The whole valley between was floored by a smooth,
nearly level glacier, about a half mile wide and perhaps
two miles long. Presently we were compelled to get on
the ice as the moraine disappeared ; so we put on the
rope, and advanced with more caution. It was not long,
however, before W., who was next to last in our line,
broke through the bridge of a crevasse, despite our care,
and sank to his shoulders. This member of our party
was not versed in the art of snow-craft, and to him, every
occurrence common to mountain experiences, and Alpine
methods of procedure, were alike novel and terrible. In
consequence, this accident fell more severely on him, but
fortunately, he was extricated almost immediately by the
use of the rope.
At the head of our valley was a remarkable, symmetrical
mountain, resembling" in general outline a bishop's mitre.
From the glacier and snow-fields where we were walking,
there rose on either side of the Mitre, steep snow-slopes, 86
The Canadian Rockies.
which terminated in lofty cols about 8500 feet above
sea-level. That on the north side of the Mitre was exceedingly steep, and was rendered inaccessible by reason of
a great crevasse, extending from the precipices on either
side, clear across the snow-slope. This crevasse must
have been nearly twenty-five yards in width and of great
depth. At one side there still remained a thin bridge of
snow, suspended, as it were, in mid-air over the awful
chasm, as though to tempt climbers on to their instant
destruction, or perhaps to a lingering death from cold
and hunger.
The pass on our left appeared the more propitious and
seemed to offer a possible route to the summit of the
divide. We were anxious to get a view into the valley
beyond, even though it were but for a few moments. The
unknown regions on the other side of the pass had long
been for me a favorite pleasure-ground of the imagination.
Some fate had hitherto denied us any idea of the place
beyond the vaguest suggestions. Several ascents, or partial ascents, of mountains on all sides of this unknown
valley, had revealed the outlines of the surrounding mountains, but some intervening cliff or mountain range had
always, with persistent and exasperating constancy, shut
off all but the most unsatisfactory glimpses. Starting
from these substantial foundations of reality, my imagination had built up a wide circular valley, surrounded on all
sides by curious mountains of indefinite and ever changing outline and position. The picture always appeared
in a gloomy, weird light, as though under a cloudy sky, Air Castles.
87
or while the sun was near totally eclipsed. By some
curious analogy, this faint illumination was similar to that
which we always associate with the first creation of land
and water ; or far back in the geologic ages, when strange
and hideous reptiles,—some flying in the murky air, some
creeping amid the swampy growths of cycads, calamités,
and gigantic tree ferns,—excite a strange thrill of pleasure
and awe combined, as though the soul were dimly perceiving some new revelation of the universe, though but
vaguely. I n this weird, gloomy valley I wandered careless,
in my imagination, many days and at many times, among
forests infested by strange, wild animals, harmless like
those of Eden, and by the shores of ever new, ever changing lakes and rivers.
So strong had this picture become that I felt the most
intense anxiety to succeed in reaching the top of our pass,
and gain at length a view of the reality, even at the risk
of shattering these pleasant air castles, and annihilating, in
a single instant, one of my best mental pleasure-grounds.
There were many dangers to be risked, however, and
many obstacles to be overcome before this advantage
might be gained. The steep slope was rendered formidable by reason of many great schrunds* or horizontal
crevasses, caused by the ice of the glacier below, moving
downward. In the intense cold of winter the moving ice
becomes rigid and nearly stagnant, while the drifting
snows accumulate, so as partly to fill these rents in the ice
and bridge them over by cornices built out from one side
or the other.    When the increasing- warmth of summer 88 The Canadian Rockies.
causes the ice to become plastic and to move more rapidly,
these rents grow wider and the snow-bridges melt away
and eventually fall into the crevasses so as to leave impassable chasms, dangerous to approach. Fortunately, it was
not so late in the season that afHhe bridges were broken
down, else wé should have been completely defeated, for,
on either side, the glacier was hemmed in by dangerous
rock precipices. The south side of the glacier, moreover,
was subject to frequent rock falls from the disintegrating
cliffs of the Mitre. As we advanced over the extensive
névé, the slope increased gradually but constantly, and
soon became so steep that steps had to be cut, and great
care was necessary not to slip. We crossed some of the
schrunds by bridges of snow, where it was necessary to
proceed with great caution, and, by sliding the feet along,
apply the weight gently, lest the bridge should break
through. We passed round others by walking along the
lower edge or lip of the crevasse, which gave us a splendid
but almost terrifying view of the gloomy caverns, extending down through the snow and ice to unknown depths.
The dark-blue roofs of these crevasses were hung with
dripping icicles, while from far below could be heard thé
sound of rushing, sub-glacial streams. Three hours of
this slow, toilsome work were necessary to gain iooo feet
in altitude. We were now more than 8000 feet above the
sea, and the atmosphere was raw and cold. Large damp
flakes of snow and granular hail fell occasionally from a
cloudy sky, silently and swiftly, through a quiet atmosphere.    The whole horizon was bounded by high moun- Desolate Surrounding's.
89
tains, covered with glaciers and patches of snow, altogether
barren and destitute of vegetation. Not a single tree or
shrub, nor even a grassy slope at the far end of the great
amphitheatre of mountain walls by which we were hemmed
in, relieved the stern, cold monotony of the scene. So far
as we might judge by our surroundings, we might have
been exploring the lonely, desolate mountains of Spitz-
bergen, or some distant polar land, where frost and winter
rule perpetual. Our progress up the slope of the glacier was very slow, as each step had to be cut out with
the ice-axe. The pitch was so steep that a misstep might
have resulted in our all sliding down and making further
exploration of the schrunds below. The whole party
was, in consequence, more or less affected by these cheerless circumstances, and became much depressed in spirit.
As, however, the condition of the body is in great part
responsible for all mental and moral ailments, so it was
in our case. Had we been walking rapidly, so that the
circulation of the blood had been vigorous and strong,
both mind and body would have been in good condition,
and the cold air, the snow, and bleak mountains would
have been powerless to discourage. It is always at such
times that mountain climbers begin to ask themselves
whether the results are worth the efforts to attain them.
Any one who has climbed at all, as we learn by reading
the experiences of mountaineers, at many times has said
to himself: " If I get home safely this time I shall never
again venture from the comforts of civilization." The
ancients, when in the thick of battle, or at the point of mmmsmm
90 The Canadian Rockies.
shipwreck, were accustomed to vow temples to the gods
should they be kind enough to save them, but they usually forgot their oaths when safely home. Mountaineers
in like manner forget their resolves, under the genial
influence of rest and food, when they reach camp.
After many disappointments, we at last saw the true
summit of our pass or col not far distant, and only a few
hundred feet above us. xA more gentle slope of snow,
free of crevasses, led to it from our position.
! Now that we were confident of success, we took this
opportunity to rest by a ledge of rocks which appeared
above the surrounding snow field. Here we regained
confidence along with a momentary rest.
Nothing could surpass the awful grandeur of Mount
Lefroy opposite us. Its great cliffs were of solid rock,
perpendicular and sheer for about 2500 feet, and then
sloping back, at an angle of near fifty degrees, to heights
which were shut off from our view by the great hanging
glacier. We could just catch a glimpse of its dark precipices, where the mountain wall continued into the
unknown valley eastward, through a gorge or rent in
the cliffs south of the Mitre. A magnificent avalanche
fell from Mount Lefroy as we were resting from our
severe exertion, and held our admiring attention for several moments. Another descended from the Mitre and
consisted wholly of rocks, which made a sharp cannonade
as they struck the glacier below, and showed us the danger
to which we should have been exposed had we ascended
on the farther side of the slope. Discovery of a Beautiful Valley.
91
Having roped up once more, we proceeded rapidly
toward the summit of the col, being urged on by a strong
desire to see what wonders the view eastward might have
in store. This is the most pleasurably exciting experience in mountaineering—the approach to the summit of
a pass. The conquest of a new mountain is likewise
very interesting, but usually the scene unfolds gradually
during the last few minutes of an ascent. On reaching
the summit of a pass, however, a curtain is removed, as
it were, at once, and a new region is unfolded whereby
the extent of the view is doubled as by magic.
We were, moreover, anxious to learn whether a descent into this valley would be possible, after we should
arrive on the col. We were alternately tormented by the
fear of finding impassable precipices of rock, or glaciers
rent by deep crevasses, and cheered on by the hope of an
easy slope of snow or scree, whereby.a safe descent would
be offered.
Proceeding cautiously, as we approached the very summit, to avoid the danger of an overhanging cornice of snow,
we had no sooner arrived on the highest part than we beheld a valley of surpassing beauty, wide and beautiful, with
alternating open meadows and rich forests. Here and
there were to be seen streams and brooks spread out before our gaze, clearly as though on a map, and traceable
to their sources, some from glaciers, others from springs
or melting snow-drifts. In the open meadows, evidently
luxuriantly clothed with grass and other small plants,
though from our great height it was impossible to tell, the 92
The Canadian Rockies.
streams meandered about in sinuous channels, in some
places forming a perfect network of watercourses. In
other parts, the streams were temporarily concealed by
heavy forests of dark coniferous trees, or more extensively,
by hght groves of larch.
This beautiful valley, resembling a park by reason of its
varied and pleasing landscape, was closely invested on the
south by a half circle of rugged, high mountains rising precipitously from a large glacier at their united bases. This
wall of mountains, continuing almost uninterruptedly
around, hemmed in the farther side of the valley and terminated, so far as we could see, in a mountain with twin
summits of nearly equal height, about one mile apart.
The limestone strata of this mountain were nearly perfectly horizontal, and had been sculptured by rain and
frost into an endless variety of minarets, spires, and pinnacles. These, crowning the summits of ridges and slopes
with ever changing angles, as though they represented
alternating walls and roofs of some great cathedral, all
contributed to give this mountain, with its elegant contours
and outlines, the most artistically perfect assemblage of
forms that nature can offer throughout the range of mountain architecture.
On the north side of this mountain, as though, here,
nature had striven to outdo herself, there rose from the
middle slopes a number of graceful spires or pinnacles,
perhaps 200 or 300 feet in height, slender and tapering,
which, having escaped the irresistible force of moving
glaciers and destructive earthquakes, through the duration Discovery of Paradise Valley.
- ^  1 p^ A Temple of Nature.
93
of thousands of years, while the elements continued their
slow but constant work of disintegration and dissolution,
now presented these strange monuments ef an ageless
past. Compared with these needles, the obelisks and
pyramids of Egypt, the palaces of Yucatan, or the
temples of India are young, even in their antiquity.
When those ancient peoples were building, nature had
nearly completed her work  here.
Beyond the nearer range of mountains could be
seen, through two depressions, a more distant range,
remarkably steep and rugged, while one particularly
high peak was adorned with extensive snow-fields and
large glaciers.
Almost simultaneously with our arrival on the summit
of the pass, a great change took place in the weather.
The wind veered about, and the clouds, which hitherto
had formed a monotonous gray covering, now began to
separate rapidly and dissolve away, allowing the blue sky
to appear in many places. Long, light shafts of sunlight
forced a passage through these rents, and, as the clouds
moved along, trailed bright areas of illumination over the
valley below, developing rich coloring and pleasing contrasts of light and shade over a landscape ideally perfect.
This beautiful scene, which has taken some time to describe, even superficially, burst on our view so suddenly,
that for a moment the air was rent with our exclamations
and shouts, while those who had lately been most depressed
in spirit were now most vehement in their expressions of
pleasure.    We spent a half-hour on the pass and divided IP-
94 The Canadian Rockies.
up our work, so that while one took photographs of the
scenery, another noted down the angles of prominent
points for surveying purposes, while the rest constructed
a high cairn of stones, to commemorate our ascent of the
pass.
Whatever may have been the mental processes by which
the result was achieved, we found all unanimous in a decision to go down into the new valley and explore it, whatever might result. The cold, desolate valley on which wre
now turned our backs, but which was the route homewards, was less attractive than this unknown region of so
many pleasant features, where even the weather seemed
changed as we approached it.
It was now already two-thirty p. m. We were 8400
feet above sea-level and at an unknown distance from
Lake Louise, should we attempt the new route. Another great mountain range might have to be passed
before we could arrive at the chalet, for aught we knew.
There were, however, fully six hours left of daylight,
and we hoped to reach the chalet before nightfall.
A long snow-slope descended from where we were
standing, far into the valley. This we prepared to descend
by glissading, all roped together, on account of W., who
was this day enjoying his first experience in mountain
climbing. An unkind fate had selected him, earlier in the
day, to break through the bridge of the crevasse and now
doomed him to still further trouble, for we had no sooner
got well under way in our descent, before his feet flew out
from under him, and he started to slide at such a remark- A Precipitate Descent.
95
able rate that the man behind was jerked violently by the
rope, and, falling headlong, lost his ice-axe at the same
time. With consternation depicted in every feature, our
two friends came rolling and sliding down, with ever increasing speed, spinning round—now one leading, now the
other, sometimes head first, sometimes feet first. The
shock of the oncomers was too much for the rest of us to
withstand, and even with our ice-axes.well set in the soft
snow, we all slid some distance in a bunch. At length
our axes had the desired effect and the procession came
to a standstill. It required some time to unwind the tangled ropes wherein we were enmeshed like flies in a
spiders web, owing to the complicated figures we had
executed in our descent. Meanwhile, a committee of one
was appointed to go back and pick up the scattered hats,
ice-axes, and such other wreckage as could be found.
The end of the descent was accomplished in a better
manner, and in less than ten minutes we were 1500
feet below the pass. A short, steep scramble down
some rocky ledges, where strong alder bushes gave good
support for lowering ourselves, brought us in a few minutes to the valley bottom. At this level the air was warm
and pleasant as we entered an open grove of larch and
spruce trees. In the last quarter of an hour we had
passed through all the gradations from an arctic climate,
where the cold air, the great masses of perpetual snow,
and bleak rocks, made a wintry picture, to the genial climate of the temperate zone, where were fresh and beautiful meadows enlivened by bright flowers, gaudy insects. 96
The Canadian Rockies.
and the smaller mountain animals. Humboldt has truly
said : " In the physical as in the moral world, the contrast
of effects, the comparison of what is powerful and menacing with what is soft and peaceful, is a never failing
source of our pleasures and our emotions."
We followed a small, clear stream of an unusual nature. In some places it glided quietly and swiftly over a
sloping floor of solid stone, polished and grooved in some
past age by glaciers. A little farther on, the character of
the mountain stream suffered a change, and the water
now found its way in many sharp, angular turns and narrow courses by large square blocks of stone, for the most
part covered by a thick carpet of moss, while between
were deep pools and occasional miniature waterfalls.
Pursuing our way with rapid steps, for we were like
adventurers in some fairy-land of nature, where every
moment reveals new wonders, we came at length to an
opening in the forest, where the stream dashed over some
rocky ledges, that frost and age had rent asunder and
thrown down in wild disorder, till the stream bed was
fairly strewn wTith giant masses of sandstone. Some of
these colossal fragments were apparently just balanced on
sharp edges, and seemed ever ready to fall from their insecure positions. The variety and novelty of form presented by the falling water, as the streamlets divided here
and united there, some over, some under, the stone
bridges accidentally formed in this confusion of nature,
aroused our greatest admiration.
As we advanced down the valley towards the north, a
Weary Followers. 97
the outlines of the mountains changed, and we recognized
at length the bare slopes of the southern side of Mount
Temple, which at first seemed to us a strange mountain.
Meanwhile, we had approached very near to the base of
the beautiful mountain with the double peak and the
many pinnacles, and found that proximity did not render
it less attractive.
The stream which we followed had been joined by
many other rivulets and springs till it grew to be wide
and deep. At length a muddy torrent, direct from the
glacier at the head of the valley, added new volume and
polluted the crystal snow-waters of the stream which we
had followed from its very source.
For many hours we followed the banks of the small
river formed by these two branches, and found it an almost continuous succession of rapids, constantly descending, and with a channel swinging to right and left, every
few hundred yards, in a winding course.
H. and I led the way, and frequently lost sight of the
others who were beginning to tire and preferred a slower
pace. We waited on several occasions for them to come
up with us, though it seemed as if we should no more than
reach the chalet before nightfall, even by putting forth
our best efforts.
About-6.30 xx clock we came to a swampy tract, where
the trees grew sparingly, and gave the appearance of a
meadow to an expanse of nearly level ground, covered
with fine grass and sedges. Here, after a long wait for
our friends, who had not been seen for some time, we r
m
The Canadian Rockies.
decided to write a note on a piece of paper and attach it
to a pole in a conspicuous place where they could not fail
to see it. The mosquitoes were so numerous that it was
almost impossible to remain quiet long enough to write a
few words explaining our plans. On the top of the stick
we placed a small splinter of wood in a slit, and made it
point in the exact direction we intended to take.
Having accomplished these duties in the best manner
possible, we set out for the chalet with all speed, as we
did not relish the idea of making a bivouac in the woods
and spending a cheerless night after our long fast. It
was evident that we were now at the outlet of the valley,
and that, unless we should encounter very rough country
with much fallen timber, our chances were good for reaching the chalet before darkness rendered travelling impossible. It was likewise important to reach the lake on
account of those at the chalet, who might think that the
whole party had met with some accident on the mountain,
unless some of us turned up that night.
We accordingly walked as fast as our waning strength
permitted, and after surmounting a ridge about 800 feet
high, which formed part of the lower slopes of Saddle
Mountain, we found no great difficulty in forcing a passage through the forest for several miles, when we came
upon the trail to the Saddle. We reached the lake at 8.15
p.m., and after shouting in vain for some one to send over
a boat, we forded the stream and entered the chalet, where
a sumptuous repast was ordered forthwith, and to which
we did ample justice after our walk of twelve hours duration. A Bivouac in the Forest.
99
Our less fortunate friends did not appear till the next
morning. They discovered our note, but decided not to
take our route, as they thought it safer to follow the stream
till it joined the Bow River. They had not proceeded
far, however, beyond the place where we had left the note,
before they became entangled in a large area of fallen
timber and prostrate trees, where they were overtaken
by night and compelled to give up all hope of reaching
Lake Louise till the next day. In the dark forest they
made a small fire, and were at first tormented by mosquitoes and, later, by the chill of advancing night, so that
sleep was impossible. The extreme weariness of exhausted nature, crowned by hunger and sleeplessness amid
clouds of voracious mosquitoes, was only offset by the
contents of a flask, with which they endeavored to revive
their drooping spirits, and cherish the feeble spark of life
till dawn.
Fortunately, the nights in this latitude are short, and
at four o'clock they continued their way to the Bow
River, which they followed till they reached Laggan.
About six days later, a little column of smoke was observed rising from the forests towards the east, and from
Laggan we learned that the woods were on fire, and that
about forty acres of land were already in a blaze. A
large gang of section men were despatched at once with
water buckets and axes to fight the fire. The fire did not
prove so extensive, however, as at first reported, and in
about two days all the men were recalled.
William said to one of us :   " Me think two white man The Canadian Rockies.
light him fire " ; to which our friends replied that it was
impossible, as the fire had broken out nearly a week after
they had been there.
William replied, with the only trace of sarcasm I have
ever known him to use : " White man no light fire, oh
no, me think sun light him." CHAPTER  VII.
The Wild Character of Paradise Valley—Difficulties with Pack
Horses—A Remarkable Accident—Our Camp and Surroundings—Animal
Friends—Midsummer Flowers—Desolation Valley-^Ascent of Hazel Peak
—An Alpine Lake in a Basin of Ice—First Attempt to Scale Mt. Temple-^
Our Camp by a Small Lake—A Wild and Stormy Night—An Impassable
Barrier—A Scene of Utter Desolation—All Nature Sleeps—Difficulties
of Ascent— The Highest Point yet Reached in Canada—Paradise Valley in
Winter—Farewell to Lake Louise.
OUR delightful experience in Paradise Valley convinced us that a camp should be established in
it near the southern base of Mount Temple,
which we hoped to ascend. From this camp we intended to make branch excursions in all directions and
learn something of the mountains toward the east and
south. All this region, though so near the railroad,
had apparently never been explored by the surveyors,
and the early expeditions had of course never approached
this region nearer than the Vermilion Pass on the east
and the Kicking Horse Pass on the west. In all our
expeditions through these lonely but grand mountain
valleys, we never discovered any mark of axe or knife
on the trees, any charred pieces of wood to indicate a
camper's fire, nor any cairn or pile of stones to prove
some climber's conquest. The Canadian Rockies.
In fact, the impenetrable barrier of mountains at every
valley end dissolved the surveyor's hopes, even from a
distance, of finding any practicable pass through the maze
of lofty mountains and intervening valleys blocked with
glaciers and vast heaps of moraine. The lone prospector
would not be tempted by any sign of gold in the streams
to explore these valleys, though the Indian hunter may
have occasionally visited these regions in search of bears
or the mountain goat.
We first blazed a trail from the chalet to the entrance
of Paradise Valley. The route followed was merely the
best and most open pathway that we could find through
the forests, and though not more than three miles in
length, it required as many hours to reach the valley
entrance. Pack horses we obtained at the chalet, but no
man could be found who would consent to act as our cook
or assistant in managing the horses.
Our camp was at length established by the side of a
small rivulet on the lower slopes of Mount Temple, where
we found the altitude to be 6900 feet above sea-level. Our
experiences with pack animals were of a most exciting
nature and sometimes severely trying to our temper and
patience. The horses were not accustomed to this service and performed all sorts of antics, smashing the packs
among the trees, jumping high in air to clear a small
stream six inches wide, or plunging regardless into
rivers where, for a moment, the horse and packs would be
submerged in the water. There was one place about two
miles within the valley entrance that might well try the Difficulties with Pack Horses.
patience of Job himself. On one side of the stream,
was an impassable area covered with tree trunks crisscrossed and piled two or three deep by some snow-slide
of former years. On the other side of the stream,
which we were compelled to take, was a dense forest.
Below was a tangled growth of bush, and many fallen
trees, all resting on a foundation of large loose stones
covered six inches deep with green moss. Between these
stones were deep holes and occasional underground streams,
the water of which could be faintly heard below and which
had probably washed away the soil and left these angular
stones unprotected. To lead a horse through this place
required the greatest skill, patience, and even daring.
Without some one to lead the animal with a rope, the
poor beast would stand motionless, but to pick one's way
over the rough ground while leading the horse invariably
ended in disaster. The very first hole was enough to
frighten the horse, so that, instead of proceeding more
slowly, the animal usually made a mad rush forward regardless of the leader, who invariably fled and sought the protection of a tree, while the horse soon fell prostrate among
the maze of obstacles. In these frantic rushes many of
us were several times trampled on by the horse, and the
packs were smashed against the branches and trunks of
trees, or torn off altogether. This was an exceedingly
dangerous bit of ground, and it was remarkable that on so
many occasions we were able to lead our horses through
it without a broken leg.
One of our most remarkable adventures with a horse io4 The Canadian Rockies.
may indeed test the credence of the reader, but five men
can vouch for its actual occurrence. We were passing
along through the forest in our usual manner, which was
the outgrowth of much experience. First of all, one man
preceded and did nothing else but find the blaze marks
and keep on the ill-defined trail as well as possible.
About twenty-five yards behind came another man whose
duty it was to find the pathfinder, and if possible, improve
on his trail. Then came one of our party who led the
horse with a long head rope, while behind the horse were
two men whose duty it was to pick up whatever articles
fell out of the packs from time to time, and fasten them
on again.
- As we were proceeding in this manner, we came to
a slanting tree which leaned over the trail at an angle of
about thirty degrees. It was just small enough to be
limber, and just large enough to be strong. MoreovervJt
was too low for the horse to go under, and a little too
high for him to jump over. One might travel a lifetime
and never meet with just such another tree as this. In
less than ten seconds this tree had brought the horse and
two of our party to the ground and wrought consternation in our ranks.
As the horse approached the slanting tree, F., who
was leading, saw the animal rear high in the air to prepare
for a jump. He thought it best to get out of the way,
but in his haste stumbled and fell headlong into a bush.
Meanwhile the horse, a stupid old beast, prepared for the
effort of his life, and with a tremendous spring jumped A Remarkable Accident.
105
high in air, but unfortunately his fore-feet caught on the
small tree, which swung forward a little and then returning
like a powerful spring, turned the animal over in mid-air.
The horse landed on his back some five yards farther on,
and, with his four legs straight up in the air, remained
motionless as death. But this was not all, for the tree
swung back violentfy and struck H. on the nose, fortunately at the end of the swing, but with sufficient force to
knock him down.
When our two friends recovered, we turned our attention to the horse, which still remained motionless on his
back. " He is dead," said F., but, on rolling him over,
the poor animal got up and seemed none the worse for
his experience, except for a more than usual stupidity.
We camped about ten days in Paradise Valley in a
beautiful spot near the end. Here, on all sides except
towards the north, the place is hemmed in by lofty mountains. We saw the valley in all sorts of weather, in clear
sunshiny days, and when the clouds hung low and shut
out the mountains from view. On one or two occasions
the ground was white with snow for a short time, though
our visit was during the first part of August.
Many kinds of animals frequented the valley, and some
of the smaller creatures lived in the rocks on all sides of
our camp and became quite friendly. One of the most
interesting little animals of the Canadian Rockies is the
little pica, or tailless hare. This small animal abounded
in the vicinity of our camp and is in fact always found at
about 7000 feet altitude.    It is a hare about the size of io6 The Canadian Rockies.
a rat, which, with its round ears, it more resembles. These
little fellows have a dismal squeak, and they are very
impertinent in their manner of sitting up among the rocks
at the entrance to their holes, and gazing at their human
visitors, ever ready to pop out of sight at a sign of danger.
Chipmunks were likewise abundant and visited our camp
to pick up scattered crumbs from our table.
There is a species of rat with a bushy tail that lives
in the forests and rocky places of these mountains and
is the most arrant thief among all the rodents. Nothing
is too large for them to try and carry off, and they will
make away with the camper's compass, aneroid, or watch,
and hide them in some inaccessible hole, apparently with
the desire to set up a collection of curios.
The siffleur, or marmot, is the largest among these
rodents, and reaches the length of twenty-five or thirty
inches. These animals usually frequent high altitudes
at, or above the tree line, where they build large nests
among the rocks and lay up a store of provisions for winter time. They are very fat in the fall, but it is not known
whether they hibernate or not. Their note is a very loud
shrill whistle, which they make at a distance, but they never
allow one to approach very near, like the impudent picas.
We saw very few of the mountain goats, though we
often came upon their fresh tracks in the mud near streams
or in the snow far up on the mountain sides. On several
occasions we could hear the patter and rattle of stones
sent down by the movements of some herd, though our
eyes failed to detect them. Midsummer Flowers. 107
Where the forests grew thick in the valley, the herbs
and flowering plants were always less numerous, but in the
meadows the ground was colored by mountain flowers of
beautiful shades and pretty forms. The tasselled heads of
the large anemones, long since gone to seed, were conspicuous everywhere, and they are always a beautiful object
among the meadow grass as the summer breezes make
gentle waves over these seas of verdure. Along the bare
rocky margins of the streams, where all else has been
forced to retire by occasional floods, two species of plants
make a most brilliant coloring and dazzle the eye with
discordant shades. They are the castilleias, or painter's
brush, with bright scarlet and green leaves clustered at
the top of a leafy stem, and the epilobiums, with reddish-
purple blossoms ; these two plants were often so close
together with their inharmonious color tones as to perplex
the observer in regard to nature's meaning. When nature
does such things we grow to like her apparent mistakes,
just as we love the bitter-sweet chords of Schumann, or
Grieg's harsh harmonies.
We made several excursions into the next valley to the
eastward, and beyond that, over the water-shed into British
Columbia. The valley to the east offered the greatest
contrast to Paradise Valley. It was somewhat wider, the
altitude was in general higher, so that a great part was
•above the tree line, while the awful wildness and confusion created by vast heaps of moraine and a large
glacier at the foot of a range of saw-edged mountains
made this place seem like a vale of desolation and death.
J* io8
The Canadian Rockies.
At the close of our camping experiences, we effected the
conquest of two mountains, Hazel Peak and Mount Temple, on two successive days. We first tried Hazel Peak,
and by fpllowing the route which had been previously selected, we found the ascent remarkably easy. On the
summit, the climber is 10,370 feet above sea-level,—higher
than the more celebrated Mount Stephen, often claimed
to be the highest along the railroad,—and surrounded by
more high peaks than can be found at any other known
part of the Canadian Rockies, south of Alaska. In fact
there are seven or eight peaks within a radius of six miles
that are over  11,000 feet high.
The view is, at the same time, grand and inspiring, and
has certain attractions that high mountain views rarely
present. The rock precipice and snow-crowned crest of
Mount Lefroy are separated from the summit of Hazel
Peak by one of the grandest and deepest canyons of the
Canadian Rockies, so that the distance from summit to
summit is only one mile and a half. The ascent of Hazel
Peak is certainly well worth the labor of the climb, as the
round trip may be easily accomplished from Paradise Valley in five hours, though the ascent is nearly 4000 feet.
On the north side, from the very summit, a fine glacier
sweeps down in steep pitch far into the valley below and
with its pure white snow and yawning blue crevasses
of unfathomable depth, forms one of the most attractive
features of this mountain. The most remarkable and
beautiful object that we discovered, however, was a small
lake or pool of water only a few yards below the summit Camp in Paradise Valley.  1  First Attempt on Mt. Temple. 109
of the mountain. Encircled on all sides by the pure snows
of these lofty altitudes, and embedded, as it were, in a blue
crystal basin of glacier ice, the water of this little lake was
colored deep as indigo, while over the surface a film of ice.
formed during the previous night, had not yet melted
away.
We returned to camp much elated with our success but
doubtful of the morrow, as no easy route had yet been discovered up the forbidding slopes of Mount Temple. The
year before, Mr. A. and I had been hopelessly defeated
even when we had counted most on success. Moreover,
the mere fact that, though this mountain was the highest
yet discovered anywhere near the railroad, it had never
been ascended by any surveyor or climber, made success
appear less probable, though it urged us on to a keener
ambition.
The attempt by A. and myself to ascend this mountain
in 1893 was probably the first ever made. During the
first week of August, we started from Laggan, having
with usa Stoney Indian, named Enoch Wildman, and one
horse to carry our tent and provisions. The day was unusually hot, and, as we forced our monotonous and tiresome
passage through the scanty forests of pine near the Bow
River, we suffered very much from heat and thirst. In
these mountain excursions, it is the best policy to wear
very heavy clothes, even at the disadvantage of being uncomfortable during the day, for the nights are invariably
cold, even at low altitudes. We did not camp until nightfall, when we found ourselves on the northern slope of The Canadian Rockies.
the mountain, 7000 feet above sea-level, by the side
of a small lake. The little lake occupied a depression
among giant boulders and the debris of the mountain.
At one end, a large bank of snow extended into and below
the water, which was apparently rising, as there were fragments of frozen snow floating about in the lake. The
banks sloped steeply into the water on all sides, and there
was not a single level spot for our camp, so that it was
necessary to build a wall of stones, near the water's edge,
for our feet, and to prevent ourselves from sliding into the
lake during the night.
The weather was wild and stormy, and the long night
seemed to drag out its weary length to an interminable extent of time, attended as it was by showers of rain and
hail and furious gusts of wind, which threatened to bring
our flapping tent to the ground at any moment.
Our camp-fire, which had been built on a scale appropriate to some larger race of men, was a huge pile of logs,
each fully ten feet long, and twelve or eighteen inches
through, but the wind blew so strong that the mass roared
like a vast forge during the early hours, and then died
away into an inert mass of cinders toward the chill of
morning.
The light of day revealed our wild surroundings. We
were under the northern precipice of Mount Temple, and
so close that we could see only the lower part of this inaccessible wall. A beautiful fall dashed down in a series
of cascades through a distance of about 1000 feet,
and fed our little lake.    Sometimes the strong wind, blow- An Impassable Barrier.
ing against the cliff, or sweeping upward, would make the
water pause and momentarily hang in mid-air, suspended,
as it were, on an invisible airy cushion, till gathering
greater volume, it would burst through the barrier and
fall in a curtain of sparkling drops.
Poor Enoch had suffered terribly from cold during the
night, and begged our permission to return to Laggan,
promising to come back the next day—" sun so high,"
pointing to its place in the early afternoon. He said in
his broken English : " No grass for pony here, too cold me ;
no like it me." So we took pity on him and sent him back
to more comfortable quarters while we rested in comparative quiet, it being Sunday.
Early Monday morning we had our breakfast and were
on foot at four o'clock. The gloom of early dawn, the
chill of morning, and the cloudy sky had no cheering effect
on our anticipations. Our plan was to traverse the mountain side till we should come to the southeast shoulder,
where we had once observed an outline of apparently
easy slope.
By eleven o'clock we had reached an altitude of nearly
10,000 feet without meeting with any very great
difficulty, but here we came suddenly to a vertical wall
of rock about 400 feet high and actually leaning
over in many places, a barrier that completely defeated
us, as the wall extended beyond our view and offered no
prospect of giving out. At the base of this cliff was a
steep, narrow slope of loose, broken limestone, and then
another precipice below.    Along this dangerous pathway The Canadian Rockies.
we continued for some distance, keeping close to the base
of the cliff. The loose stones, set in motion by our feet,
slid down and rolled over the precipice, where we could
hear them grinding to powder on the cliffs below.
Never in my life have I been so much impressed with
the stern and desolate side of nature. The air was bitter
cold and had the frosty ozone odor of winter. A strong
wind rushed constantly by us, and, as it swept up the
gorges of the precipice above, and over the countless projections of the cliffs, made a noise like the hoarse murmur of
wind in a ship's rigging, or the blast of some great furnace.
To the south and east, range beyond range of bare, saw-
edged mountains raised their cold, sharp summits up to a
cloudy sky, where the strong wind drove threatening
clouds in long trains of dark and lighter vapors. The intervening valleys, destitute of vegetation or any green
thing, were filled with glaciers and vast heaps of moraine,
and the slides of debris from the adjacent mountain side.
All was desolate, gloomy, cold, and monotonous in color.
Three thousand feet below, a small lake was still bound
fast in the iron jaws of winter, surrounded as it was by
the walls of mountains which shut out the light and
warmth of the summer sun. Inert, inanimate nature here
held perpetual rule in an everlasting winter, where summer; with its flowers and birds and pleasant fertility, is
unknown, and man rarely ventures.
Overcome with the terrors of this lonely place and the
hopelessness of further attempt to reach the summit,
where a snow-storm was now raging, we turned back.    As All Nature Sleeps.
i*3
we reached our camp we found Enoch just approaching,
according to his promise, and though the afternoon was
well advanced, we packed up and moved with all speed
toward Laggan. We reached Lake Louise at 10.30 p.m.,
after almost nineteen hours of constant walking.
Now, however, at our camp in Paradise Valley, the
conditions were somewhat different. We were at the
yery base of the mountain, and had learned much more
about it, in the year that had elapsed since our first
attempt.
The mountaineer has many discomforts mingled with
the keen enjoyment of his rare experiences. None is
more trying than the early hour at which he is compelled
to rise from his couch of balsam boughs and set forth
on his morning toil. At the chill hour before dawn, when
all nature stagnates and animate creation is plunged
in deepest sleep, the mountain climber must needs arouse
himself from heavy slumber and, unwilling, compel his
sluggish body into action.
This is the deadest hour of the twenty-four—the time
just before dawn. The breezes of early night have died
away into a cold and frosty Y:alm ; the thermometer sinks
to its lowest point, and even the barometer, as though in
sympathy, reaches one of its diurnal minima at this untimely hour. And if inanimate nature is thus greatly
affected, much more are the creations of the vegetable
and animal kingdoms. The plants are suffering from the
cold and frost ; the animals of daytime have not as yet
aroused themselves from sleep, while the nocturnal prowl- ii4 The Canadian Rockies.
ers have already ceased their quest of prey ançj. returned
to their dens. Even man is affected, for at this dead hour
the ebb and pulse of life beat slow and feeble, and the lingering spark of life in those wasted by disease comes at
this time most near going out.
At such an unseasonable hour, or more accurately at
four a.m., were we up, on the 17th of August preparing for our ascent of Mount Temple. There was no
trace of dawn, and the waning moon, now in her last quarter, was riding low in the southern sky, just above the
sharp triangular peak at the end of our valley.
At nine o'clock in the morning, we had gained the
summit of the pass between Mount Temple and Pinnacle
Mountain, where we were 9000 feet above sea-level.
The ascent so far had not been of an encouraging nature,
as we had encountered a long, loose slide where everything moved threateningly at each step. I have never
seen a more unstable slope. The stones and boulders
would slide, and begin to move at a distance of ten and
fifteen feet above the place where we stood, and on every
side also. F., who was one of the party, was terror-
stricken, for he now had a horror of moving stones of
any description.
The view from this pass was very extraordinary. To
the east stood the rugged, saw-edged mountains of the
Desolation Range, looming up in solemn grandeur through
an atmosphere bluish and hazy with the smoke of forest
fires. The air was perfectly calm and had the bracing
coolness of early  morning and high altitude, which the 1
Highest Point Reached in Canada.
115
rising sun tempered most gently. The weather conditions for accomplishing our ascent were perfect, but there
was little prospect of a fine view by reason of the smoke.
The outlook from the pass was indeed discouraging.
Cliffs and ledges with broken stones and loose debris
seemed to oppose all safe passage. Fortunately, as we progressed the difficulties vanished, and not till we reached an
altitude of about 10,000 feet did we encounter any real obstacles. We found a passage through the great rock wall
which had defeated us last year, by the aid of a little gully,
which, however, entailed some rather difficult climbing.
This arduous
work continued
throughout the
next 1000 feet,
when, at an altitude of 11,000
feet, we came to
the great slope
between the
southwest and
west arêtes and
found an easy
passage to the
summit.
Many a hearty cheer rent the thin air as our little
party of three reached the summit, for we were standing
where no man had ever stood before, and, if I mistake
not, at the highest altitude yet reached in North America
SUMMIT OF MOUNT TEMPLE. n6 The Canadian Rockies.
north of the United States boundary. The summit was
formed of hard bluish limestones, broken and piled up in
blocks, as on all high mountain tops. The cliffs toward
the east were stupendous and led the eye down to the
valley more than a mile below. The air was almost calm
and just above freezing, and the snow was melting quite
fast in the sun. The thermometer at the Lake Louise
chalet reached seventy-two degrees at the same time that
we were on the summit of Mount Temple, which proves
this to be almost the highest temperature that ever occurs
on this lofty point. It would be safe to say that the temperature on the top of Mount Temple never rises higher
than forty degrees.
If one is fortunate in a good selection of routes, the
ascent of Mount Temple will not be found difficult. But
the descent is very perplexing, for unless one remembers the intricate combination of gullies and ledges by
which the ascent is made, many precipitous cliffs will be
encountered down which it is impossible to descend.
This was our last exploit in Paradise Valley, and a
few days later the various members of our party, one by
one, bade farewell to the beautiful region of Lake Louise
with its many pleasant associations.
I remained there five or six weeks longer until winter
commenced in earnest and drove every one away. During
the first week of October I made a final visit to Paradise
Valley with Mr. Astley, the manager of the chalet, in order
to bring back our tent and the camping utensils. Snow
covered the ground in the shady parts of the woods, even Paradise Valley in Winte
117
at the entrance of the valley. The stream had fallen so
much that its rocky bed proved the best route up the
valley, especially for the horse. After an hour's journey
within the entrance we found ourselves at the base of
Mount Sheol, and not far above us could be seen a fine
herd of seven or eight mountain goats. They scampered
off on seeing us, but soon came to halt as they were
tempted by curiosity to have another look. These snow-
white goats are the most characteristic animals of the
Rockies and nearly correspond in habits with the more
cunning chamois of Switzerland. Like them it is a species
of antelope, though it resembles a goat to a remarkable
degree.
We found our camp buried in snow, the ridge-pole of
the tent broken down with the heavy burden, and everything so much disguised by the wintry mantle that we
had difficulty in finding the camping place. Even as we
were packing up the frozen canvas and blankets, the air
was full of falling snow and the mountains encircling the
valley were only revealed in vague and indefinite outlines,
while ever and anon could be heard the dull roar of snow-
slides sweeping down to the glacier.
About nightfall we were back at the entrance to the
valley, where the lower altitude gave us the advantage
of a ground nearly free of snow, though a fine rain sifted
down through the spruce needles almost constantly.
Herevwe camped in the dense forest, and our roaring
fire, built high with great logs, soon drove away the chill
and dampness of the rainy night.    The tent, our clothes, n8 The Canadian Rockies.
and the mossy ground were soon steaming, and the bright
glare of our camp-fire illumined the trees and gave us
good cheer, surrounded as we were by miles of trackless
forests in the blackness of night. A hearty supper and
a great pail of strong hot tea soon revived our spirits,
and on a soft couch of heaths and balsam boughs—more
luxurious than any bed of down—we bid defiance to the
darkness and storm in perfect comfort. The next day
the snow-flakes were falling gently and steadily, so that the
trees were covered even to their branchlets and needles
with the white mantle. The bushes, the mosses, and even
the blades of grass in the swampy marshes, as we pursued
our homeward way, were all concealed and transformed
into pure white images of themselves in snow.
A few days later I went up to Lake Agnes to hunt
for mountain goats, which frequent this place in great
numbers. The snow was two feet deep. The lake was
already nearly covered with ice, and I was compelled to
seek shelter behind a cliff against a bitterly cold wind,
driving icy particles of hail and snow against my face.
It was useless to prolong the contest longer. Winter
had resumed her iron sway in these boreal regions and
high altitudes, and in a few weeks Lake Louise too would
begin to freeze, and no longer present its endless change
of ripple and calm, light and shadow, or the reflected
images of rocks and trees and distant mountains. CHAPTER  V
The Selkirks—Geographical Position of the Range—Good Cheer of
the Glacier House—Charming Situation—Comparison between the Selkirks
and Rockies—Early Mountain Ascents—Density of the Forest—Ascent of
Eagle Peak—A Magnificent Panorama—A Descent in the Darkness—
Account of a Terrible Experience on Eagle Peak—Trails through the
Forest—Future Popularity of the Selkirks—The Forest Primeval—An
Epitome of Human Life—Age of Trees—Forests Dependent on Humidity.
WEST of that chain of the Rocky Mountains
which forms the crest or backbone of the
continent, lies another system of mountains
called the Selkirk Range. Having many features in common with the mountains to the east, this range has, nevertheless, certain constant characteristics of vegetation and
geological formation, so that the traveller who is but
slightly familiar with them should never be at a loss in
regard to his surroundings.
The position of this range in relation to the other
mountains of the great Cordilleran System is not difficult
to understand. The Selkirks may be said to begin in northwestern Montana between the Summit Range and the Bitter
Root Mountains, and, trending in a northwestward direction through British Columbia about three hundred mile's,
they approach the main range and apparently merge into
IIQ «pi
The Canadian Rockies.
it near the Athabasca Pass. The most remarkable feature of the range is the manner in which it compels the great
Columbia River to run northward for fifty leagues on its
eastern side, before it allows a passage to the west, so that
the northern portions of the range are entirely hemmed
in by this large river, flowing in opposite directions on
either side. Another feature of great interest in regard
to the drainage is the relation between the Columbia and
Kootanie rivers. The latter river is one of the chief tributaries to the upper Columbia, and flows southward to a
point one mile and a half from the head waters of the
Columbia, which it passes on its journey southward, while
the Columbia flows in the opposite direction.    The water
of the Kootanie is actually
higher than
that of the
Columbia a t
this point, and
as the two rivers are only
separated by a
low, level
plain, it was
once proposed
to cut a channel between,
and divert the Kootanie into the Columbia.
The traveller is always glad to find himself  at  the
GLACIER HOUSE. The Glacier House.
Glacier House in the heart of the Selkirks. This is
more especially true, if in previous years, he has visited
this charming spot and become in some degree familiar
with the place. The railroad makes a large loop round
a narrow valley and sweeps apparently close to the
great glacier of the Selkirks, a vast sea of ice that glistens in a silvery white sheen and appears to rise above
the forests as one looks southward. There is something
pre-eminently comfortable and homelike about the Glacier
House. The effect is indefinable, and one hardly, knows
whether the general style of an English inn, or the genuine hospitality that one receives, is the chief cause. One
always feels at home in this wild little spot, and scarcely
realizes that civilization is so far distant.
The rush of summer guests called for the erection of
an annex, so that there are now two hotels for the accommodation of tourists. The Glacier House is located near
the railroad, and occupies a small, nearly level, place at the
bottom of one of those deep and narrow valleys characteristic of the Selkirks. Those who have visited the Fran-
conia Notch in the White Mountains would be somewhat
reminded of that beautiful spot upon first seeing the
surroundings of Glacier. The ground in front of the
hotel has been levelled and is rendered beautiful by a
thick carpet of turf. In summer it is fragrant and almost
snowy in appearance from the multitude of white clover
blossoms. This garden spot in the wilderness is still further adorned by fountains, which break the continuity of
the greensward, and are fed by cascades that may be seen The Canadian Rockies.
descending the opposite mountain side in many a leap,
through a total distance of 1800 feet.
But this small area, that man has improved and rendered more suitable to his comfort, is surrounded on all
sides by a wilderness, perhaps better described as a little
explored range of mountains separated by deep gorges
and covered with dense forests. It is like the Alps of
Switzerland and the Black Forest combined. There are
snow-clad peaks, large glaciers, and névé regions of vast
extent in the higher altitudes, while the valleys below are
dark and sombre in their covering of deep, cool forests.
The main range of the Rockies presents no such rankness
of vegetable growth—mosses, ferns, and lichens covering
every available surface on tree trunks and boulders—nor
such huge trees as those found everywhere in the Selkirks.
Moreover, the mountains of the Selkirk Range probably average 1000 feet lower than in the corresponding
parts of the main range, but nevertheless they seem
white and brilliant in their mantles of everlasting snow
and sparkling glaciers. Finally, one observes that the railroad track is covered at frequent intervals by snow-sheds
of considerable length, constructed of heavy beams and
massive timbers, in order to withstand the terrible force
and weight of winter snow-slides and avalanches. In the
main range of the Rockies there are no snow sheds. The
question naturally arises—What is the reason of all these
differences from the more eastern ranges ?
The answer to the question is that the climate is more
humid.    The snowfall in winter is so great that it remains Humidity of Climate. 123
all summer at much lower altitudes than in the Rockies,
and supplies glaciers, which descend perhaps a thousand
feet nearer to sea-level. The moisture from this deep
covering of snow, saturates the ground as it melts in the
spring, and, in addition to frequent, heavy summer rains,
nourishes the rich forests of these mountains. Moreover,
the atmosphere is always slightly moister than it is to the
east, and does not tend to dry up the ground or evaporate
the mountain snows so rapidly as in the summit range.
The eastward movement of the atmosphere, carrying
up moisture from the Pacific, causes a great condensation
of clouds and a heavy rainfall as the air currents pass over
the Selkirks, and leaves the atmosphere robbed of a great
part of its moisture to pass over the next range to the east.
Almost all the differences between the Selkirks and
the Rockies proper, spring from the single cause of a
moister climate. The principal features of extensive snow
fields and luxuriant forests can be readily understood.
May not the deep, narrow valleys of the Selkirks be likewise explained from the more rapid action and greater
erosive power of the mountain streams in cutting down
their channels ?
Whatever may be the cause of all these phenomena,
the results are very apparent. Any one who has visited
the Selkirks for an extended period has, without doubt,
spent many a day within doors writing his diary or enjoying the pleasure of music or literature, while the rain is
falling constantly, and the clouds and vapors hang low on
the mountain sides.    The manner in which the clouds M
124 The Canadian Rockies.
come sweeping up the Illicellewaet valley at the base of
Mount Cheops and turn toward the flanks of Eagle Peak
or Mount Sir Donald is very impressive. Certainly the
cloud effects in the Selkirks are magnificent beyond all
description.
Nevertheless, it is not encouraging to have a friend
step off the train and announce the fact that he has been
enjoying fine weather for several days in the Rocky
Mountains, some fifty or sixty miles to the east, while you
have been confined to the house by a long period of rain.
Often, too, the climber or explorer becomes fretful
under long confinement, and, taking advantage of an
apparent clearing away of clouds and a promise of fair
weather, when far from the hotel, is caught in a sudden
downpour, and realizes the truth of that scriptural passage
which was apparently written concerning a similar region
—" They are wet with the showers of the mountains, and
embrace the rock for want of a shelter."
When the railroad first made this region accessible to
tourists, the Selkirks rapidly acquired a remarkable popularity, especially among mountain climbers. In this early
period several parties came over from England and other
countries of Europe with the express purpose of making mountain ascents. Such parties were those of Dr.
Green and the two Swiss climbers. Huber and Sulzer. A
good idea of the difficulties presented by the higher peaks
to skilled mountaineers may be had from the fact that
Dr. Green and his party only succeeded in reaching
the summit of one high peak, while Huber and Sulzer left Density of the Forests.
the Hermit Range in defeat, though they succeeded in
reaching the top of the sharp rock peak, Mount Sir
Donald, the Matterhorn of the Selkirks.
One of the chief difficulties to overcome is the penetration of the forest belt below the tree line. No one who
has not tried a Selkirk forest has any conception of its
nature in this respect. There are huge tree trunks lying
on or near the ground, which have been thrown down by
the precipitate fury of some winter snow slide, or have
fallen by the natural processes of death and decay. These
great obstacles are ofttimes covered with a slippery coating
of moss and lichens, while the ground is fairly concealed
by a rank growth of ferns, and plants in countless variety.
The density of the underbrush is rendered still more trying
to the mountaineer by reason of a plant of the Ginseng
family, which from its terrible nature is most fitly named
the Devil's Club, for it is armed with thousands of long
needle-like spines. This plant grows five or six feet high,
with a stout stem bearing a few leaves of large size. The
spines, which are an inch or more in length, project in
every direction like an array of quills on a porcupine, and
are strong enough to penetrate the skin and flesh with
surprising facility. The alder bushes attain a peculiar
growth in the Selkirks ; each bush consists of a bunch of
long slender stems, which spread out from the ground in
every direction, ofttimes with nearly prostrate branches,
which interlace and form a wellnigh impassable hedge.
The alder bushes are found most numerous on bare slopes
of the mountains, where snow slides have stripped down i26 The Canadian Rockies.
the forests ; or in ravines, where the crumbling earth gives
no certain foothold to larger and nobler trees.
In 1893, A. and I made an ascent of Eagle Peak. This
mountain lies just to the west from the great wedge-shaped
rock summit of Mount Sir Donald. The altitude of Eagle
Peak is, I believe, a little more than 9400 feet above
sea-level, and as the Glacier House is only 4400 feet,
the ascent involves a climb of 5000 feet. The name of the
mountain is derived from a great crag or cliff near the
summit, which appears to lean out from a ridge, and bears
a striking resemblance to the head of an eagle. When we
were making our ascent we came suddenly on the Eagle
itself, which now, on a nearer view, proved to be of colossal size, a great leaning tower, about sixty feet high.
Rising from one of the rocky ridges, it reached upwards
and outwards till the outermost point seemed to overhang
a bottomless abyss, perhaps twenty or twenty-five feet
beyond the verge of the precipice.
The ridge just below the summit is a scene of wild
confusion, for the rocky ledges have been split up and
wedged apart by frost and storms till they appear as giant
blocks of stone ten or fifteen feet high, between the crevices of which one may catch glimpses of the valley and
forests thousands of feet below.
The view from the summit of Eagle Peak is magnificent and well worth the labor of the climb. The proximity of Mount Sir Donald, which towers more than 1200
feet higher, causes its sullen precipices to appear strikingly
grand.    The great Illicellewaet névé, with its twenty square 1
Mount Sir Donald, from Eagle Peak.    v^
A Magnificent Panorama. 127
miles or more of unbroken snow fields, stretches out in the
distance and forms part of the eastern horizon. The
rugged appearance of the Hermit Range to the west, with
its sharp ridges and needles, is perhaps the most tumultuous part in all this wild sea of mountain peaks. It has
been stated on good authority that from Mount Abbott, a
far lower ridge on the farther side of the valley, more than
one hundred and twenty individual glaciers may be counted,
but there are even more within view from Eagle Peak.
We remained on the summit till nearly three o'clock,
and thereby took a great risk, as we learned afterwards to
our exceeding regret. Before leaving, however, we built
a high cairn and fixed several handkerchiefs among the
stones so as to render it, if possible, visible from the valley
below.
In our descent we found no trouble till we reached tree
line, when the gathering gloom of nightfall, made earlier
by a cloudy sky, aroused our apprehensions and led us to
a serious mistake. Thinking that it would be better to
follow the course of a stream, which had cut out a deep
ravine in the mountain side, as there would be more light,
for a time at least, we commenced our descent with all speed.
We soon found ourselves in a trap, as the sides of the ravine
grew constantly deeper and steeper as we descended, and
it was at length impossible to get out at all. Floundering
about among the long trailing branches of alders, our descent soon became a mixture of sliding, falling, and, indeed,
every method of progress save rational walking. The
darkness came on rapidly, as the days were short and the The Canadian Rockies.
twilight much curtailed, it being late in the summer. In
an hour it became so absolutely black that the foamy
course of the stream we followed was the only visible
object, as even the stars were concealed and their light
shut out by a heavy covering of dark cloud. Sometimes
the long, prostrate branches of the alders would catch our
feet in a most exasperating manner, and cause one or the
other to slide temporarily head-foremost, till some branch
or root could be seized in the hand and the progress
arrested. Once I saw a white object, just below me apparently, and thinking it might be a stone, was about to
lower myself in fancied security when suddenly I realized
that it was the foam of the stream some fifty feet below,
and that we were on the edge of a precipice ! At another
time I fell headlong through a bush and brought up against
some great obstacle around which I wound my leg, not
knowing whether it might be a huge grizzly or some other
denizen of the forest, when sure enough it moved away,
and rolled over my leg. It was a great boulder nearly a
yard in diameter.
This nocturnal descent was the most bitter experience
I have ever had in mountain climbing, as the anxiety and
worry consequent upon each movement were exquisitely
painful, and continued three hours. Arrived at the bottom of the slope at ten o'clock p.m., we found ourselves in
the mass of fallen logs and debris near the stream, and
likewise near the trail. Under the spell of a certain assurance that a few minutes more of toil would bring us
out to the trail, we thought nothing of falling into holes !
A Terrible Experience.
129
four or five feet deep, as we plunged about among the logs,
or, when walking on them, occasionally stepped off into
space.
We arrived at the Glacier House at 10:30 p.m., where
we were surrounded by anxious friends, and regaled by a
hot dinner of roasted chickens and all manner of good
things, such as one always finds at this most excellent
inn. At such times, more than at any other, one appreciates the thoughtfulness and care of a kind host.
Our experience on Eagle Peak, trying as it was, could
not equal that of two gentlemen who, in 1894, made an
attempt to scale the mountain. Unfortunately they failed
to reach the summit, and, worse still, were benighted
among the crags and cliffs at a high altitude, where
they spent the night in misery. Finding themselves in
their attempt unable to advance farther for some reason
or other, they were descending, when it suddenly occurred
to them that they were on a different ledge from any they
had seen hitherto. Nightfall was bringing rapidly increasing darkness, and it seemed impossible, at length, either
to proceed farther or even to retrace the steps by which
they had come. Here, then, on a narrow ledge overlooking a precipice, the awful depths of which were rendered
still more terrible in the obscurity of gathering gloom, and
with their feet dangling over the verge, they were forced
to remain motionless, and wear out the long night in cold
and sleepless suffering. The next morning a search party
was organized, and they were conducted back to the comforts of the Glacier House, much to the relief of their 130
The Canadian Rockies.
anxious friends, but nearly prostrated by their terrible
experience.
Later, we made an ascent of Mount Cheops, a striking
peak with a most perfect representation of a pyramid
forming its summit. The view is fine but not worth the
labor of the climb, as the ascent of the lower slopes seems
interminably long and tedious by reason of the underbrush
and steep slope. Like Eagle Peak, the summit revealed
no evidence of previous conquests, and it will probably be
a long time before any one will be so far led astray as to
make a similar attempt.
Trails and good foot-paths lead from the Glacier House
to points of interest in the vicinity. The chief resort is
the Great Glacier itself, where one may witness all the
phenomena of a large ice stream, or ascend to the vast
névé, and wander about on a nearly level, and apparently
limitless, snow field.
Mount Abbott is an easy and favorite climb, and is
often successfully attempted by women who are endowed
with considerable strength and endurance. On the way,
a small pool, called Marion Lake, is passed. It nestles
among the cliffs and forests on the mountain side far
above the valley. It is the only lake I know of in the
Selkirks. This is one of the remarkable differences between the Selkirks and the Summit Range of the Rockies :
the absence of lakes in one region, and their great number
in the other. The great majority of lakes in the Rockies
are very small and often do not deserve the name, as they
are mere pools a few yards across.    But their small size in
'•--""" Future of the Selkirks.
131
no way detracts from their beauty, and it is most unfortunate that the Selkirks possess so few of these, the most
charming of all features in mountain landscapes.
The Selkirks are but little known, because the dense
forests and the immense size of the fallen logs forbid the
use of horses almost altogether, and will ever prevent the
mountaineer from making extended journeys into the lesser
known parts of the mountains, unless trails are cutand kept
in good order. At present all provisions, blankets, and
tents must be packed on men's backs, a method that is
both laborious and expensive.
It must eventually result, however, that these mountains will prove a most popular resort for climbers and
sportsmen. The attractions for either class are very great.
For the mountaineer, they present all the grandeur and
beauty of the Swiss Alps, with difficulties of snow and
rock climbing sufficient to add zest to the sport. The
multitude of unclimbed peaks likewise offers great opportunities for those ambitious for new conquests. The
immense annual snowfall causes many of the higher peaks
to assume an appearance of dazzling beauty and brilliancy,
while the Alpine splendor of these higher altitudes is
strongly contrasted with the dark-green color of the
forested valleys.
For the sportsmen, too, there are abundant opportunities to hunt the larger game. On the mountains are
numerous herds of mountain goats and sheep, while the
forests abound in bears—the black bear and the grizzly or
silver tip.   During the berry season, these animals frequent The Canadian Rockies.
the valleys and are often seen by the railroad men even
near the Glacier House. One gentleman had the good
fortune to shoot a black bear from a window of the hotel
last year. Of course, there is practically no danger from
even the grizzly bear in this immediate vicinity, as they
have learned to fear man from being frequently shot at,
and have long since lost the ferocity which they sometimes
show in extremely wild and unfrequented regions.
No mention has yet been made of the kind of trees to
be found in a Selkirk forest. Almost all the varieties of
coniferous trees observed in the Rockies, except the
Lyall's larch, occur in the Selkirks, though each variety
attains much larger size. The cedar, the hemlock, the
Douglas fir, and the Engelmann's spruce are most conspicuous and form the chief part of the forest trees. Each of
these species here attains a diameter of from three feet
upward, even to six or seven, and a height of from 150
to 200 feet.
Nothing is more enjoyable than to take one of the
mountain trails and enter the depths of the forest, there to
rest in quiet contemplation where trees alone are visible
in the limited circle of view. On a quiet afternoon, when
all is calm and not a breath of air is stirring, the long,
gray moss hangs in pendent tufts from the lower branches
of the giant trees, and one feels that this is indeed another
Acadian forest of which Longfellow sings :
"This is the forest primeval.   The murmuring pines and the hemlocks,
Bearded with moss and in garments green, indistinct in the twilight,  !
 «   The Forest Primeval.
i33
Stand like Druids of eld, with voices sad and prophetic,—
Stand like harpers hoar, with beards that rest on their bosoms."
Such indeed is a Selkirk forest.
The idea that is at length developed in the mind, by a
long rest in one of these deep and sombre forests, is that
of the majesty, and silent, motionless power of vegetation.
The creations of the vegetable world stand on all sides.
They wellnigh cover the ground ; they limit the horizon,
and conceal the sky. The tall cedars have a shreddy bark
that hangs in long strips on their tapering boles and makes
the strongest contrast with the rough bark of the firs. What
could be more unlike, too, among evergreens, than the
spreading fanlike foliage of the cedars, the needle-like
leaves of the firs, and the delicate spray of the hemlocks ?
What a vast amount of energy has been preserved in
these forest giants ; with what a crash they would fall to
the ground ; and what a quantity of heat—which they have
stored up from the sun through hundreds of summers—
would they give out when burned slowly in a fireplace !
If we examine a single needle, or a thin shaving of wood,
under the microscope, and obtain a glimpse of the complexity of the cells and pores with which this vegetable
life is carried on ; or consider the wonderful processes by
which the flowers are fertilized, and the cones mature, so
that the species may never die out ; and then regard the
immensity of the whole forest stretching boundless in
every direction, all constructed from an infinity of atoms,
the mind and imagination are soon led beyond their
depth. 134 The Canadian Rockies.
Now let the pure, cold light of science, with its precise
and exact laws, fade away into the warm, mellow glow of
romance, till we picture the forest as an epitome of human
life, with its struggles, its suffering, and the slow but certain progress from infancy to old age and death. For
here, among the forest trees, are every age and condition
represented Beneath, are young trees, vigorous and full
of promise, hoping, as it were, some day to push their
highest branches above the general plane of tree tops and
share the life-giving sun, though, during the struggle,
many will surely weaken and die in the pale and inefficient
light beneath the older trees. Then there are the larger
trees in the full glory of their prime, with massive trunks,
straight and tall, giving promise of many years of life
yet to come ; and finally, the giants of the forest, their
branches torn off by storms or their trunks rent and
scarred by lightning. Everything about the oldest trees
betokens the slow decay and all-conquering death, which
is gradually sapping their life blood and pointing to their
certain, final destruction. The long, gray moss, gently
waving in the faintest breath of air, hangs from every
limb, and makes these venerable monarchs resemble
bearded patriarchs, which have stood here perhaps a
thousand years battling with the elements, the wind, and
the lightning, silent witnesses to the relentless progress of
the seasons.
Trees have, however, all the qualifications of living
forever. There is no reason why a tree should ever die,
were it not for some unnatural cause, such as the fury of Age of Trees. 135
a storm, the rending power of lightning, or the destructive
influence of insects and parasites. In California, in the
Mariposa Grove, some of the giant redwood trees are
twenty-five hundred years old. They began to grow
when Solon was making laws for the ancient Greeks.
These wonderful groves of California are, however, exceptional, and have survived by reason of the clemency of
the climate and the fact that the aromatic redwood is
avoided by insects. In m@st forests, the laws of chance
and probability rarely allow the sturdiest trees to run the
gamut of more than a few hundred years, and if they
attain a thousand years, it is their " fourscore—by reason
of strength."
In the Selkirks, one sees the ground covered with huge
tree trunks in all stages of decay, slowly moldering away
into a newer and richer soil ; some have yielded to the
natural processes of decay, others to accident or forest
fires, while in some places winter avalanches have cut off
the tops of the trees forty or fifty feet above the ground,
and left nothing but a maze of tall stumps where once
stood a noble forest.
The Selkirk forests are dense and sometimes almost
magnificent in their luxuriance, and vastly surpass the
forests of the eastern range in the variety of species, the
size of the trees, and the luxuriant rankness of vegetable
growth. At the same time they do not approach the
almost tropical vigor and grandeur of the Pacific Coast
forests, where a green carpet of moss covers the trunks
and   branches   of  the huge   trees, and  even   ferns  find 136
The Canadian Rockies.
nourishment in this rich covering, aided by the reeking,
humid atmosphere, on branches forty or fifty feet above
the ground. In such a forest, the ferns and brakes reach
a height of six or eight feet above the ground, the various
mosses attain a remarkable development, and hang in
long, green tresses, a yard in length, from every branch,
and exaggerate the size of the smaller branches, while the
beautiful tufts of the Hypnum mosses appear like the
fronds of small ferns, so large do they become.
The forests of the Summit Range, the Selkirks, and the
Pacific Coast are almost perfect indexes of the humidity
of the climate. The Selkirk forests are less vigorous
than those of the Pacific coast, but more so than the light
and comparatively open forests of the Summit Range,
where the climate is much drier. CHAPTER  IX.
Mount Assiniboine—Preparations for Visiting it—Camp at Heely's
Creek—Crossing the Simpson Pass—Shoot a Pack-Horse—A Delightful
Camp—A Difficult Snow Pass—Burnt Timber—Nature Sounds—Discovery of a Beautiful Lake—Inspiring View of Mount Assiniboine—
Our Camp at the Base of the Mountain—Summer Snow-Storms—Inaccessibility of Mount Assiniboine.
GREAT interest was aroused among tourists in
the summer of 1895, by the reports of a remarkable peak south of Banff named Mount
Assiniboine. According to current accounts, it was the
highest mountain so far discovered between the International boundary and the region of Mounts Brown and
Hooker. Besides its great altitude, it was said to be exceedingly steep on all sides, and surrounded by charming valleys dotted with beautiful lakes. The time required
to reach the mountain with a camping outfit and pack-
horses was said to be from five to seven days.
The romance of visiting this wild and interesting region,
hitherto but little explored, decided me to use one month
of the summer season in this manner. By great good fortune I met, at Banff, two gentlemen likewise bent on visiting the same region, and on comparing our prospective
plans, it appeared that mutual advantage would be gained 138 The Canadian Rockies.
by joining our forces. In this way we would have the
pleasure of a larger company, and at the same time
the opportunity of separating, should we come to a
disagreement.
The sixth of July was decided on as the date for our
departure. In the meantime, we made frequent visits to
the log-house of our outfitter, Tom Wilson, who was to
supply us with horses, our entire camping outfit, and
guides. Many years previously, Wilson had packed for
the early railroad surveyors, and had thus gained a valuable experience in all that concerns the management and
care of pack-animals among the difficulties of mountain
trails. In the past few years, he has been engaged in
supplying tourists with camping outfits and guides, for
excursions among the mountains.
The season of 1895 was very backward, and there was
an unusually late fall of snow at Banff, in the middle of
June. Moreover, the weather had remained so cold that
the snow on the higher passes still remained very deep,
and several bands of Indians, who attempted to cross the
mountains with their horses late in June, were repulsed by
snow six or eight feet deep.
The -weather continued cold and changeable during the
first week in July. In the meanwhile, however, our preparations for departure went on without interruption, and
Wilson's log-house, where the supplies and camp outfits
were safely stored, became a scene of busy preparation.
On every side were to be seen the various necessaries
of camp life :   saddles for the horses, piles of blankets, Preparation for Departure. 139
here and there ropes, tents, and hobbles. Great heaps of
provisions were likewise piled up in apparent confusion,
though, in reality, every item was portioned out and carefully calculated. Rashers of bacon and bags of flour
comprised the main bulk of the provisions, but there were,
besides, the luxuries of tea, coffee, and sugar, in addition
to large quantities of hard tack, dried fruits and raisins,
oatmeal, and cans of condensed milk. Pots and pails,
knives, forks, and spoons, and the necessary cooking
utensils were collected in other places. Our men were
already engaged for the trip, and were now busily moving
about, seeing that everything was in order, the saddle
girths, hobbles, and ropes in good condition, the axes
sharp, and the rifles bright and clean.
At length the sixth of July came, but proved showery
and wet like many preceding days. Nevertheless, our
men started in the morning for the first camp, which was
to be at Heely's Creek, about six miles from Banff. Our
prospective route to Mount Assiniboine was, first, over
the Simpson Pass to the Simpson River, and thence, by
some rather uncertain passes, eastward, toward the region
of the mountain.
Toward the middle of the afternoon we started on foot
for Heely's Creek, where our men were to meet us and
have the camp prepared. Passing northward up the valley, we followed the road by the famous Cave and Basin,
where the hot sulphur water bubbles up among the limestone formations which they have deposited round their
borders.    The Cave appears to be the cone or crater of 140
The Canadian Rockies.
some extinct geyser, and now a passage-way has been cut
under one wall, so that bathers may enjoy hot baths in
this cavern. A single opening in the roof admits the
light.
A short time after leaving these interesting places, we
had to branch off from the road, and plunge into a burnt
forest, where there was supposed to be a trail. The trail
soon faded away into obscurity among the maze of logs,
and, worse still, it now came on to rain gently but constantly. After an hour or more of hard work we came to
Heely's Creek.
The camp was on the farther side of the creek, and,
after  shouting  several   times,   Peyto,  our  chief  packer,
came dashing down on
horseback,
and conveyed us, one
at a time,
across the
deep, swift
stream.
Peyto made
an ideal picture of the
wild west,
mounted as
he was on
an     Indian Camp at Heely's Creek. 141
steed, with Mexican stirrups. A great sombrero hat pushed
to one side, a buckskin shirt ornate with Indian fringes on
sleeves and seams, and cartridge belt holding a hunting
knife and a six-shooter, recalled the romantic days of old
when this was the costume throughout the entire west.
Our encampment consisted of three tents, prettily
grouped among some large spruce trees. A log fire was
burning before each tent, and, on our arrival, the cooks
began to prepare our supper. This was my first night
in a tent for a year, and the conditions were unfavorable for comfort, as we were all soaked through by
our long tramp in the bush, and, moreover, it was still
raining. Nevertheless, we were all contented and happy,
our clothes soon dried before the camp fires, and after
supper we sang a few popular songs, then rolled up in
warm blankets on beds of balsam boughs, and slept peacefully till morning.
I was awakened at dawn by the cry of " Breakfast is
ready," and prepared forthwith to do it justice. The day
appeared cloudy but not very threatening. In an hour the
packers began their work, and it was wonderful to observe
the system and rapidity of their movements. The horses,
of which we had seven as pack-animals and two for
the saddle, were caught and led to the camp, where they
were tied to trees near by. All the provisions, tents, cook
boxes, bags, and camp paraphernalia were then made
ready for packing. There are three prime requisites in
skilful packing. They are : the proper adjustment of
the blanket and saddle so that it will neither chafe the 142
The Canadian Rockies.
back of the horse nor slip while on the march ; the exact
balancing of the two packs ; and the knowledge of the
" d i a m o n d
hitch." The
wonderful
combination
of turns and
loops which
go to make up
the diamond
hitch has always been surrounded with
a certain secrecy, and jealously guarded
by those initiated into the mysteries of its formation. It
was formerly so essential a part of the education of a
Westerner that as much as one hundred dollars have been
paid for the privilege of learning it. Without going into
details, it may be described as a certain manner of placing
the ropes round the packs, which, once learned, is exceedingly simple to tie on or take off, and it will hold the pack in
place under the most trying circumstances. The name is
derived from a diamond-shaped figure formed by the ropes
between the packs.
By eight o'clock our procession of ten horses was on
the march, and, after passing through a meadow where
every blade of  grass was hung with pendent drops of
PACKING   THE   BUCKSKIN. On the March.
r43
mingled rain and dew, now sparkling bright in the morning sun, we came to the trail. Our winding cavalcade
followed near the creek and gradually rose above its
roaring waters, which dashed madly over many a cascade
and waterfall in its rocky course. Our pathway rose constantly and led us through rich forests.
Peyto led the procession mounted on an Indian horse
called Chiniquy, not a very noble-looking beast, but a
veteran on the trail, and, by reason of his long legs, a
most trustworthy animal in crossing deep rivers. Then
followed the pack-horses with the men inter- *        /
spersed to take care of them, and the rear
was brought up by our second packer, likewise on horseback. The greater part of
the time, the gentlemen of the expedition
kept in the rear.
The flowers were in all the glory of their
spring-time luxuriance, and we discovered
new varieties in every meadow, swamp, and
grove. Beside the several varieties of anemones, the yellow columbines, violets, and
countless other herbaceous plants, we found,
during the march of this day, six kinds of
orchids. Among them was the small and
beautiful, purple Calypso, which we found
in bogs and damp woods, rearing its showy
blossom a few inches above the ground.
At the base is a single heart-shaped leaf. We were very
much pleased to find this elegant and rare orchid growing i44 The Canadian Rockies.
so abundantly here. There is a certain regal nobility and
elegance pertaining to the whole family of orchids, which
elevates them above all plants, and places them nearest
to animate creation. Whether we find them in high
northern latitudes, in cold bogs, or in dark forests, retreating far from the haunts of men, avoiding even their
own kind, solitary and unseen ; or perhaps crowded on the
branches of trees in a tropical forest, guarded from man by
venomous serpents, the stealthy jaguar, stinging insects
and a fever-laden air ; they command the greatest interest
of the botanist and the highest prices of the connoisseur.
We camped at about two o'clock, not far from the
summit of the Simpson Pass, in a valley guarded on both
sides by continuous mountains of great height.
We were surprised the next day, on reaching the summit, to find the pass covered with snow, heaped in great
drifts, ten or twenty feet deep, among the trees. The
Simpson Pass is only 6884 feet above tide, and, consequently, is below the tree line. Near the summit were two
small ponds still frozen over. A warm sun and a genial
south wind were, however, rapidly dissolving the snow
and reducing it to slush, while clear streams of water were
running in the meadows everywhere, regardless of regular
channels.
As we began our descent on the south side, a great
change came over the scene. Two hundred feet of descent
brought us from this snowy landscape to warm mountain
slopes, where the grass was almost concealed by reason of
myriads of  yellow lilies   in  full  blossom,   mingled  with The Simpson River.
H5
white anemones. These banks of flowers, resembling the
artificial creations of a hot-house, were sometimes surrounded on all sides by lingering patches of snow. Such
constant and sudden change is characteristic of mountain climates, where a few warm days suffice to melt
the snow and coax forth the flowers with surprising
rapidity.
The trail now descended rapidly, and led us through
forests much denser and more luxuriant than those on the
other side of the pass. Everything betokened a moister
climate, and the character of the vegetation had changed
so much that many new kinds of plants appeared, while
those with which we were familiar grew ranker and larger.
We had crossed the continental divide, from Alberta into
British Columbia.
Early in the afternoon we came to our camping place
on the banks of the Simpson River, where a great number
of teepee poles proved this to be a favorite resort among
the Indians. On all sides, the mountains were heavily
forested to a great height, and, far above, gray limestone
cliffs rose in bare precipices nearly free of snow.
On July the ninth, we made the longest and most arduous march so far taken. Our route, at first, lay down the
Simpson River for several miles. While the horses and
men followed the river bed almost constantly, making
frequent crossings to avail themselves of better walking
and short cuts, the rest of us necessarily remained on one
bank, and were compelled to make rapid progress to keep
up with our heavily laden horses. 146 The Canadian Rockies.
After we had proceeded down the winding banks of
the Simpson River for about two hours, our pass, a mere
notch in the mountains, was descried by Mr. B., who had
visited this region two years before in company with
Wilson. The pass lay to the east, and it was necessary
for every one to cross the river, which was here a very
swift stream nearly a yard in depth. We all got across in
safety, but had not advanced into the forest on the farther
side more than fifty yards, when one of my pack-horses
fell, by reason of the rough ground, and broke a leg. It
required but a few minutes to unpack the poor beast and
end his career with a rifle bullet. The packs were then
placed on old Chiniquy, the faithful beast hitherto used
by Peyto as a saddle-horse.
In less than fifteen minutes we were ready to proceed
again. The trail now led us up very steep ascents on a
forest-clad mountain slope for several hours. After this
we entered a gap in the mountains and followed a stream
for many miles, and at length pitched our camp late in the
afternoon, after having been on the march for nine hours.
Every one was rejoiced at the. prospect of a rest and
something to eat. Even the horses, so soon as their
packs and saddles were removed, showed their pleasure
by rolling on the ground before hastening off to a meadow
near by. Axes were busy cutting tent poles and firewood. Soon the three tents were placed in position, and
fires were burning brightly before each, while the cooks
prepared dinner.
This   place   was   most   delightful.    The   immediate A Delightful Camp. 147
ground was quite level and grassy. Near by was a clear
deep stream with a gentle, nearly imperceptible current,
which afforded a fine place fora cold plunge. The mountains hemmed in a valley of moderate width and presented
a continuous barrier on either side for many miles. The
general character of the scenery was like that of the
Sierra Nevadas, with high cliffs partly adorned with trees
and shrubs, down which countless waterfalls fell from
heights so great, that they resembled threads of silver,
waving from side to side in the changing currents of air.
On the mountain side south of our camp, there stood a
remarkable castle or fortress of rock, where nature had
apparently indulged her fancy in copying the works of
men. So perfect was the representation, that no aid from
the imagination was required to see ramparts, embrasures,
and turreted fortifications of a castle, in the remarkable
pinnacles and clefts cut out by nature from the horizontal
strata. The next morning, every one was more or less
inspired with a pleasing anticipation and excitement, as,
according to reports, we had not far to go before we
should get our first view of Mount Assiniboine. At the
end of our valley was a pass, from the summit of which
Mount Assiniboine could be seen. The trail led us
through a forest with but little underbrush, and presently
a beautiful lake burst on our view. Two of us, being
somewhat in advance of the pack train, caught a dozen
fine trout here in a very short time, and were only interrupted by the arrival of the horses and men. The fish
were so numerous that they could be seen everywhere on
1 w
The Canadian Rockies.
the bottom, and at the appearance of our artificial flies
on the water, several fish would rise at once.
In half an hour, the summit of our pass appeared over
the tree tops, and rose, apparently, 500 feet higher. The
state of the pass was, however such as to cool our enthusiasm decidedly. It was completely covered with snow to
a great depth, which made it seem probable that we would
not succeed in getting the horses over. As this could not
be proved from our position, we pushed on, determined
to overcome all difficulties. The snow began to appear,
at first, in small patches in shady places among the forest
trees, then in large drifts and finally, everywhere except
on the most exposed slopes. The trail had been lost for
some time, buried deep in the snow. Our progress was
not difficult, however, as the forest had assumed the
thin and open nature characteristic of high altitudes, and
it was possible to proceed in any direction. Our horses
struggled on bravely, and by dint of placing all the men
in front and breaking down a pathway, we managed to
effect passages over long stretches where the snow was
five or six feet deep. After the tree line had been
reached, we were more fortunate, as a long narrow
stretch, free of snow led quite to the top of the pass,
through the otherwise unbroken snow fields. A great
cornice of snow appeared on our right near the top of the
pass and showed a depth of more than forty feet.
Near the top of the pass the travelling was much
easier, and in a few minutes we were looking over the
summit across a wide valley to a range of rough moun- A Difficult Snow Pass.
149
■ ! Ill
tains hung with glaciers. Beyond them, and rising far
above, could be seen the sharp crest of Mount Assiniboine,
faintly outlined against
the sky in a
smoky atmosphere. The
intervening
wride valley
revealed a
great expanse
of burnt forest. The
dreary waste
of burnt timber was only
relieved by
two lakes,
several miles
distant, resting in a notch
among the
mou n tains.
The nearer
was about a
mile in length,
while slightly
beyond, and
at   a  higher   elevation,  was the   second,   a  mere   pool
9É»
APPROACHING THE PASS. 150 The Canadian Rockies.
of dark blue water, resting against the moraine of a
glacier.
In the valley, a meadow near a large stream seemed
to offer the best chances for a camp. In an hour we
reached this spot after a hard descent. Some of our
horses displayed great sagacity in selecting the safest and
easiest passages between and around the logs, and gave
evidence of their previous experience in this kind of work.
In order to rest the men and horses, after the arduous
marches of the past forty-eight hours, we decided to remain an entire day at this place. We were also anxious
to explore the two lakes, as they seemed tô offer fair
promise of beautiful scenery and interesting geological
formation. Our camp was surrounded on all sides by
burnt forests and charred logs, and so offered but little of
the picturesque. A partial compensation was enjoyed,
however, by reason of the great variety and number of
song birds which were now nesting in a small swamp near
by. This bog was clothed in a rich covering of grass, and
here our horses revelled in the abundance of feed, while
some small bushes scattered here and there afforded shelter
and homes for several species of birds. All day long and
even far into the night we were entertained by their melodies. The most persistent singer of all was the white-
crested sparrow, whose sweet little air of six notes was
repeated every half minute throughout the entire day, beginning with the first traces of dawn. Perhaps our attention was more attracted to the sounds about us because
there was so little to interest the eye in this place.    Smoke Nature Sonnets.
from distant forest fires obscured whatever there was in
the way of mountain scenery, while the waste of burnt
timber was most unattractive. A warm, soft wind blew
constantly up the valley and made dull moanings and
weird sounds among the dead trees, where strips of dried
bark or splinters of wood vibrated in the breeze. The
rushing stream, fifty yards from our camp, gave out a constant roar, now louder, now softer, according as the wind
changed direction and carried the sound towards or away
from us. The thunders of occasional avalanches, the loud
reports of stones falling on the mountain sides, were
mingled with the varied sounds of the wind, the rustling
of the grass, the moaning trees, and the songs of birds.
These were all pure nature sounds, most enjoyable and
elevating. Though but partially appreciated at the time,
such experiences linger in the memory and help make up
the complex associations of pleasures whereby one is led
to return again and again to the mountains, the forests,
and the wilderness.
Our time, which was set aside for this region, now
being consumed, we started on July the twelfth for the
valley at the base of Mount Assiniboine, where it was
probable that we should camp for a period of two weeks
or more. Our route lay toward the end of the valley and
thence around a projecting spur of the mountain which cut
off our view. In about two hours our horses were
struggling up the last steep slope near the summit of the
divide. I had delayed for a photograph of a small lake,
so the horses and men were ahead.    When at length I 152 The Canadian Rockies.
gained the top I found that a misplaced pack had caused
delay, and so I overtook the entire party on the borders of
a most beautiful sheet of water. The transformation was
nearly instantaneous. The burnt timber was completely
shut out from view by the low ridge we had just passed
over, and we entered once more a region of green forests.
The lake was long and narrow ; on the farther side,
hemmed in by rock slides and cliffs of the mountains, but
on the west side a trail led along the winding shore among
larch and spruce trees. In many shady nooks along the
banks of the lake were snow-drifts, under the trees or
behind protecting rocks. So long had winter lingered
this season that part of the lake was still covered with ice.
Large fragments of ice were drifting down the lake and
breaking among the ripples. Near the shore in some
places, the water was filled with thousands of narrow,
needle-like pieces of ice several inches long and perhaps
thick as a match, which, by their rubbing together in the
moving water, made a gentle subdued murmur like the
rustling of a silken gown. When ice is exposed to a
bright sun for several days, it shows its internal structure
by separating into vertical columns, with a grain like that
of wood. The ice needles which we saw had been formed
during the last stages of this wonderful process.
The Indians had made a most excellent trail round the
lake, as frequently happens in an open country. Wherever dense brush or much fallen timber occur, the trail
usually disappears altogether, only to be discovered again North Lake. ^ PV
1tBug|]
ÂfËt
EL
J*l$m
1] I
— First View of Mt. Assiniboine.
where there is less need for it. It is said that a trail,
once made, will be preserved by the various game animals of the country. In fact, there were quite recent
tracks of a mountain goat in the path we followed around
the lake.
The trail closely followed the water's edge and led us to
the extreme end of the lake and thence eastward, where,
having left this beautiful sheet of water, we passed through
a grove for a very short space and came at once to another
smaller, and possibly still more beautiful, sheet of water.
Simultaneously the magnificent and long-expected vision
of Mount Assiniboine appeared. It was a most majestic
spire or wedge of rock rising out of great snow fields,
and resembling in a striking manner the Matterhorn of
Switzerland.
It would be impossible to describe our feelings at this
sight, which at length, after several days of severe marching, now suddenly burst upon our view. The shouts of
our men, together with the excitement and pleasure depicted in every face, were sufficient evidence of our impressions. After a short pause, while we endeavored to
estimate the height and distance and gain some true
idea of the mountain, all moved on rapidly through
alternating groves and meadows to our camping place.
This was at length selected about a half mile from the
place where we first saw Mount Assiniboine. Here was a
lake nearly a mile long, which reached up nearly to the
base of the mountain, from which it was separated by a WA.«Ot**H
154
The Canadian Rockies.
glacier of considerable size. Our camp was on a terrace
above the lake, near the edge of a forest. A small stream
ran close to our tent, from which we could obtain water
for drinking and cooking purposes. The lake was in the
bottom of a wide valley, which extended northwards from
our camp for several miles, and then opened into another
valley running east and west. The whole place might be
described as an open plain among mountains of gentle
slope and moderate altitudes, grouped about Assiniboine
and its immediate spurs.
Our camp was 7000 feet above sea-level, and this was
the mean height of the valley in all this vicinity. On
mountain slopes this would be about the upper limit of
tree growth, but here, owing to the fact that the whole
region was elevated, the mean temperature was slightly
increased, and we found trees growing as high as 7400 or
7500 feet above sea-level. Nevertheless, the general character of the vegetation was sub-alpine. Many larches
were mingled with the balsam and spruce trees in the
groves, and extensive areas were destitute of trees altogether. These moors were clothed with a variety of bushy
plants, mostly dwarfed by the rigor of the climate, while
here and there a small balsam tree could be seen, stunted
and deformed by its long contest for life, and bearing many
dead branches among those still alive. These bleached
and lifeless limbs, with their thick, twisted branches resisting the axe, or even the approach of a wood-cutter, resembled those weird and awful illustrations of Doré, where Summit Lake, near Mount Assiniboine.
^ ^^^^m^^^^^8m8tmi}m(mnwmm^8^^8s^^^  mmm m— —^j IS  Camp Assiniboine. 155
evil spirits in the infernal regions are represented transformed to trees.
Curiously enough, the trees in the groves were more
or less huddled together, as though for mutual protection.
The outlying skirmishers of balsam or spruce were undersized, and often grew in natural hedges, so regular that
not one single branchlet projected beyond the smooth
surface, as if sensitive of the wind and cold. The vegetable world does not naturally excite our sympathy, but
this exhibition of, as it were, a united resistance against
the elements was almost pitiable.
Snow banks surrounded our camp and appeared everywhere in the valley. The lake was not entirely free of ice,
and large pieces of snow and ice, dislodged from the
shores, were drifting rapidly down the lake, driven on by
a strong wind and large waves. The whole picture resembled a miniature Arctic sea, where the curiously formed
pieces of ice, often T-shaped and arched over the water,
recalled the characteristic forms of icebergs.
It was at first impossible to explain where this never-
failing supply of ice came from. What was our surprise,
on making an exploration of the lake, to find that it had
no outlet and was rapidly rising ! The snow banks and
masses of ice near the glacier were being gradually lifted
up and broken off by the rising water, and so floated down
the lake.
We remained at Camp Assiniboine for two weeks.
During this time we ascended many of the lesser peaks 156 The Canadian Rockies.
in the vicinity, and made excursions into the neighboring valleys on all sides. The smoke only lasted one
day after our arrival, but, unfortunately, the weather
during the first week was very uncertain and fickle. A
succession of storms, very brief but often severe, swept
over the mountains and treated us to a grand exhibition
of cloud and storm effects on Mount Assiniboine. Sometimes the summit would be clear, and sharply outlined
against the blue sky, but suddenly a mass of black clouds
would advance from the west and envelope the peak in a
dark covering. Long streamers of falling snow or rain
would then approach, and in a few moments we would feel
the effects at our camp. During these mountain storms
the wind blows in furious gusts, the air is filled with snow
or sometimes hailstones, while thunder and lightning continue for the space of about ten minutes. The clouds and
storm rapidly pass over eastward, and the wind falls, while
the sun warms the air, and in a few minutes removes
every trace of hail or snow. Thus we were often treated
to winter and summer weather, with all the gradations
between, several times over in the space of an hour.
It seemed impossible to ascend Mount Assiniboine,
guarded as it was by vertical cliffs and hanging glaciers.
Only one route appeared on this side of the mountain, and
this lay up the steep snow-covered slope of a glacier,
guarded at the top by a long schrund and often swept by
rocks from a moraine above. It might be possible, having
gained the top of this, to traverse the great névé surround-
, Summit of Mt. Assiniboine.
157
ing the rock peak of Mount Assiniboine. From the snow
fields the bare rock cliffs rise about 3,000 feet. The angle
of slope on either side is a little more than fifty-one degrees,
a slope which is often called perpendicular, and, moreover,
as the strata are horizontal, there are several vertical walls
of rock, which sweep around the entire north and west
faces, and apparently make impassable barriers.
NORTH LAKE—LOOKING NORTHWEST. CHAPTER  X.
Evidence of Game—Discovery of a Mountain Goat—A Long Hunt—
A Critical Moment—A Terrible Fall—An Unpleasant Experience—
Habitat of the Mountain Goat—A Change of Weather—A Magnificent
Panorama—Set out to Explore the Mountain—Intense Heat of a Forest
Fire—Struggling with Burnt Timber—A Mountain Bivouac—Hope and
Despair—Success at Last—Short Rations—Topography of Mount Assiniboine— The Vermilion River—A Wonderful Canyon—Fording the Bow
River.
DURING our excursions we met with but little
game, though it was very evidently a region
where wild animals were abundant. The ground
in many places was torn up by bears, where they had dug
out the gophers and marmots. Large pieces of turf, often
a foot or eighteen inches square, together with great
stones piled up and thrown about in confusion around
these excavations, gave evidence of the strength of these
powerful beasts.
Higher up on the mountains we saw numerous tracks
of the mountain goat, and tufts of wool caught among the
bushes as they had brushed by them.
I was strolling through the upper part of the valley
late one afternoon, when my eye fell suddenly on a mountain goat walking along the cliffs about a quarter of a mile
158
	 A Goat Hunt.
159
distant. I had no rifle at the time and so returned to
camp for one, meanwhile keeping well covered by trees
and rocks. In a quarter of an hour I was back again and
saw the goat disappear behind a ledge of rock about a
half mile distant. The mountain goat always runs up in
case of danger, so that it is essential to get above them in
order to hunt successfully. I started forthwith to climb
to a ledge about 200 feet above the one on which
the goat appeared. This involved an ascent of some
600 feet, as the strata had a gentle dip southward
toward Mount Assiniboine, so that it was necessary to
take the ledge at a higher point and follow the downward
slope. I was well covered by intervening cliffs, and the
wind was favorable. It seemed almost a certainty that I
should get a shot by following this ledge for about a mile.
Accordingly I moved rapidly at first, and afterwards more
cautiously, expecting to see the goat at any moment. At
length I came to a narrow gorge, partially filled with snow,
where there were fresh tracks leading both up and down.
On a further study of the problem, I saw fresh tracks in
the snow of the valley bottom, and knowing that it would
be nearly useless to go up for the goat, I took the alternative chance of finding the animal below. After a hunt of
two hours I returned to camp completely baffled. Arrived
there, I caught sight of the goat standing unconcernedly
on a still higher ledge.
It was now late in the day, but after a good camp
dinner I set off again, determined to have that goat
if it was necessary to stalk him all night.    The animal i6o The Canadian Rockies.
was resting on a ledge near the top of a precipice fully
250 feet in height. I studied his position for at least
a quarter of an hour, carefully noting the snow patches
on the ledge above, so that it would be easy to recognize them on arriving there. Having made sure that
I could recognize the exact spot below which the goat
was located, I started to climb, and by a rough estimate calculated that I should have to ascend at least
1000 feet. After a few hundred yards, I was completely hidden from the goat in a shallow gully. Urged
on by the excitement of the hunt, I reached the ledge in
twenty minutes and turned southward. I now had to
scramble over and among some enormous blocks of stone
which had fallen from the mountain side and were strewn
about in wild disorder. Some were twenty feet high, and
between them were patches of snow which often gave way
very suddenly and plunged me into deep holes formed by
the snow melting back from the rock surfaces. Very soon
I came to a small pool of water and a trickling stream,
already freezing in the chill night air.
It was after nine o'clock, though there was still a bright
twilight in the northwest, somewhat shaded, however, by
the dark cliffs above. I proceeded very slowly, so as to
cool down somewhat and become a little steadier after the
rapid ascent. In about ten minutes I recognized the patch
of snow under which the goat was located, about one hundred yards ahead. I went to the edge of the precipice
cautiously, with rifle ready, and examined the ledges
below.    The up-draught, caused by the sun during the day- A Critical Moment. 161
time, just now changed to the downward flow of the night
air, chilled by radiation on the mountain side. This I
thought would arouse the goat, but just at that moment
my foot slipped and I dislodged a few pieces of loose
shingle, which went rattling down the cliffs. These
stones made the goat apprehensive of danger, in all
probability, for I had no sooner recovered my balance
than I caught sight of the white head and shoulders of
the animal about twenty-five yards below. The animal
stood motionless and stared at me in a surprised but impudent manner. I took aim, but could not keep the sight
on him long enough to make sure of a shot, as my rapid
climb had made my nerves a trifle unsteady. Fortunately,
the goat showed not the slightest disposition to move and
in a few seconds I got a good aim and fired. As soon
as the smoke cleared, I saw a dash of white disappearing, and then heard a dull thud far below. A
few seconds later I saw the animal rolling over and over
down the mountain side, where it finally stopped on a slide
of loose stones. I had to make a long detour in order to
get down to the animal, where I arrived in about half
an hour, and, remarkably enough, both horns were
uninjured, though the goat had fallen 125 feet
before striking. This good luck resulted from a
small snow patch at the base of the cliff, which had broken
the force of the fall, and here there was a perfect impression of the animal's body, eighteen inches deep, in the
hard snow, while the next place where he had struck was
about fifteen feet below. 162 The Canadian Rockies.
It was about 10:30 o'clock when I started for camp,
and so dark, at this late hour, that it was just possible
to distinguish the obscure forms of rocks and trees on
the mountain side. There was still another ledge to be
passed before I could get down to the valley, where
the only recognizable landmarks were occasional snow
patches, and a single bright gleam in the darkness—our
camp fire. I traversed northwards in descending, so
as to pass beyond the vertical ledge, and at length,
thinking that I had gone far enough, tried to descend.
The place was steep, but as there were a few bushes and
trees a safe descent seemed practicable. So I unslung
my rifle, and, after resting it securely in a depression, I
lowered myself till my feet rested on a projection of rock
below. At the next move there was great difficulty in
finding a rest for the rifle. At length I found a fair place,
and lowered myself again. One more step and I should
reach the bottom. Fortunately there was a stout balsam
tree at the top of the ledge, with great twisted roots above
the rocks, which would afford excellent hand-holds.
Grasping them, after placing the rifle in the lowest place,
I lowered myself again, but to my surprise I could not
touch the bottom, and, looking down, found that I was
hanging over a ledge twenty feet high with rough stones
below. Just then the rifle began to slip down, as in my
movements I had disturbed some bushes supporting it.
With one hand firmly grasping a stout root, and the toe
of my boot resting against the cliff, I took the rifle in my
other hand, and after a most tiresome struggle, succeeded A Difficult Descent. 163
at length in placing it secure for the moment. It was
now a hand-over-hand contest to get up. In going down
everything had seemed most firm and secure, but now it
was impossible to rely oh anything, as the bushes broke
away in my hand or were pulled out by the roots, and the
rocks all appeared loose or too smooth to grasp. Necessity, however, knows no law, and after a most desperate
effort I regained the top of the cliff. Not relishing any
more experiences of this nature, I groped my way along
for some distance and finally found an easy descent. On
reaching the valley, the snow patches here and there
afforded safe routes, illumined, as they were, by the
starlight. I reached camp after eleven o'clock tired but
successful.
My men started at five o'clock in the morning with ropes
and a pole to bring down the game. It was a fine young
male, and we found the meat a most pleasing addition to
our ordinary fare. Goat meat has always had a bad reputation among campers and explorers, by reason of its rank
flavor. This, however, probably depends on the age and
sex of the animal, or the season of year. In all those that
I have tried there was merely a faintly sweet flavor, which,
however, is not at all apparent if the meat is broiled or
roasted, and it is then equal to very fair beef or mutton.
The mountain goat inhabits the cliffs and snowy peaks
of the Rockies, from Alaska to Montana and Idaho, and
thence southward in certain isolated localities. Both sexes
are furnished with sharp black horns curving gracefully
backwards.    The muzzle and hoofs are jet black, but the 164 The Canadian Rockies.
wool is snow-white, long, and soft, making a beautiful rug
if the animal is killed in winter. Then the hair becomes
. very long, and the soft thick wool underneath is so dense
as to prevent the fingers passing through.
Though these strange animals resemble true goats to
a remarkable degree, and the old males sometimes have
beards in winter, they are really a species of antelope,
closely related to the chamois of Switzerland. They do
not resemble those animals in wariness and intelligence,
but are rather stupid and slow in getting out of danger.
They are, however, pugnacious, and, when brought to bay,
will often charge on the hunter and work fearful damage
with their sharp horns. The legs are exceedingly stout
and so thickly covered with long hair as to give the
animal a clumsy appearance. Their trails are almost
always to be found traversing the mountain sides, far
above the tree line, at the bases of cliffs, and often passing
over the lowest depression into the next valley. These
goat tracks are so well marked that they often help the
mountaineer, and sometimes lead him over places where
without, their guidance it would be impossible to go. The
gait of the animal when running is a sort of gallop, which
appears rather slow, but when one considers the nature of
the ground they traverse, it is very rapid. The most inaccessible cliffs, frozen snow fields, or crevassed glaciers
offer no barriers to these surefooted animals. I have
seen a herd of several goats bounding along on the face
of the cliffs, where it did not appear from below that there
could be any possible foothold.
When a herd of goats come to a gorge or passage of m
Head of Rocky Mountain Goal.
Shot July 18th, 18Q5.  il ■Ma Haunt of the Mountain Goat.
■65
any kind where loose stones
are liable to be dislodged on
those below, these skilful
mountaineers adopt the
same plan of progress practised by human climbers.
While the herd remains below, under the protection of
the cliffs, one goat climbs
the gully, and upon arriving
at the top another follows,
and thus, one by one, all
escape danger.
The   mountain   goat is
difficult to hunt by reason of
the    amount    of    climbing
necessary to get near them,
or above them.    They are
far   less   wary   than   the
chamois   of   Switzerland, or the Rocky
Mountain       sheep.
Nevertheless,    they
seem   to   be     endowed    with    a
wonderful   vitality, and are very
hard to kill.    A
goat  not fatally
wounded    will
HAUNT OF THE MOUNTAIN GOAT.' 166 The Canadian Rockies.
often jump from a cliff on which he is standing, and survive
a considerable fall. A friend of mine shot a goat near
Lake Louise, which, after the first bullet, rolled down a
cliff more than thirty feet high and landed on its feet at the
bottom, where it proceeded to walk off as though nothing
unusual had happened. The animal I shot near Mount
Assiniboine fell 125 feet, and then rolled 200 feet farther,
and was still alive when I reached him half an hour later.
These animals are by far the most numerous of the big
game in the Canadian Rockies, and are said to be increasing in numbers. Their habits of frequenting high altitudes and inaccessible parts of mountains will tend to
preserve them for many years from the relentless hunter.
After a week of fickle weather with five inches of new
snow on July 15th, there was a decided change for the
better, and the warm, bright days following one another
more regularly gave us the first taste of real summer that
we had. The massed drifts of snow diminished from day
to day and the ice disappeared from the lakes. Nature,
however, tempered her delights by ushering in vast numbers of mosquitoes and bull-dog flies to plague us. I was
engaged at this time in some surveying work, in order to
determine the height of Mount Assiniboine, and had to
exercise the utmost patience in sighting the instruments,
surrounded by hundreds of voracious foes, and often had
to allow my face and hands to remain exposed to their
stings for several minutes.
We obtained the most imposing view of Mount Assiniboine from the summit of a mountain about five miles Another View of Mount Assiniboine.
167
east of our camp. Standing at an altitude of 8800 feet,
there were eighteen lakes, large and small, to be seen in
the various valleys, which, together with the tumultuous
ranges of the Rocky Mountains on every side, some of
them fifty or sixty miles distant, formed a magnificent
pan orama.
From this
point, which
was nearly
due north of
Mount Assiniboine, the
m o u n t ai n
shows an outline altogether 31
different from
that seen at
our     camp.
Here it forms a magnificent termination of a stupendous
wall or ridge of rock, about 11,000 feet high, which runs
eastward for several miles, and then curving around to
the north, rises into another lofty peak nearly rivalling
Mount Assiniboine in height. A very large glacier sweeps
down from the névé on the north side of this lesser peak,
and descends in a crevassed slope to the valley bottom.
The valley just east of us was quite filled by three
lakes, the uppermost deep blue, the next greenish, and a
smaller one, farther north, of a yellowish color.
Our last exploit at Mount Assiniboine was  to walk
MOUNT ASSINIBOINE  FROM NORTHWEST. 168 The Canadian Rockies.
completely around the mountain. We had long desired
to learn something of the east and south sides of this
interesting peak, and to effect this Mr. B., Peyto, and I
started on July 26th, determined to see as much as possible in a three days' trip. Our provisions consisted of
bacon, hard tack, tea, sugar, and raisins. Besides this we
carried one blanket apiece, a small hand axe, and a camera.
As our success would depend in great measure on the
rapidity of our movements, we did not burden ourselves
with ice-axes.or firearms except a six-shooter. After bidding farewellto Mr. P. and the other men in camp, and
telling them to expect us back in three days, we left our
camp at eight o'clock in the morning. We walked for
three miles through the open valleys to the north and
east, and in about two hours stood at the top of the pass,
some 8000 feet above sea-level. From here we made a
rapid descent for about 2000 feet, to the largest lake of
this unexplored valley, which probably supplies one of the
tributaries to the Spray River. The change in the character of the vegetation was remarkable. The trees grew
to an immense size and reminded me strongly of a Selkirk
forest. We had a most difficult scramble here in the
pathless forest and up the opposite side of the valley.
The heat was oppressive, and we were glad to gain the
level of another more elevated valley where a cooler
atmosphere greeted us. We held our way eastward for
several miles through a fine upland meadow, where the
walking was easy and the surroundings delightful. By
noon we reached a small, shallow lake near the highest Intense Heat of a Forest Fire. 169
part of the divide, considerably below tree line. Here
we decided to rest and have lunch. Mr. B. had explored
this region with one of his men a few days previously,
and from him we learned that we should have to struggle^
with burnt timber in a few moments. The onward rush
.of the devastating fire had been stopped near the pass,
where the tree£ were small and scattered. After a short
descent we entered the burnt timber. I have never before
seen a region so absolutely devastated by fire as this.
The fire must have burnt with an unusually fierce heat,
for it had consumed the smaller trees entirely, or warped
them over till they had formed half circles, with their tops
touching the ground. The outcrops of sandstone and
quartz rocks had been splintered into sharp-edged, gritty
stones, covering the ground everywhere like so many
knives. The course of the valley now turned rapidly to
the south, so that we rounded a corner of the great mass
of mountains culminating in Mount Assiniboine. The
mountain itself had been for a long time shut out from
view by an intervening lofty ridge of glacier-clad peaks,
which were, in reality, merely outlying spurs.
The valley in which we were now walking had an unusual formation, for after a short distance we approached
a great step, or drop, whereby the valley bottom made a
descent of 400 or 500 feet at an exceedingly steep pitch.
Here it was difficult to descend even in the easiest
places. Arrived at the bottom of the descent it was not
very long before another appeared, far deeper than the
first.    The mountains on either side, especially a most 170 The Canadian Rockies.
striking and prominent peak on the east side of the
valley, which had hitherto appeared of majestic height,
seemed to rise to immeasurable altitudes as we plunged
deeper and deeper in rapid descent.
The burnt timber continued without interruption.
Our passage became mere log walking, as the extra exertion of jumping over the trees was worse than following
a crooked course on top of the prostrate trunks. This
laborious and exceedingly tiresome work continued for
three hours, and at length the charred trunks, uprooted
or burnt off near the ground, and crossed in every direction, were piled so high that we were often ten or twelve
feet above the ground, and had to work out our puzzling
passage with considerable forethought. At five o'clock
our labors ended. We made a camp near a large stream
which appeared to take its source near Mount Assiniboine. The only good thing about this camp was the
abundance of firewood, which was well seasoned, required
but little chopping, and was already half converted into
charcoal. Under the shelter of an overhanging limestone
ledge we made three lean-tos by supporting our blankets
on upright stakes. Black as coal-heavers from our long
walk in the burnt timber, seeking a refuge in the rocky
ledges of the mountains, and clad in uncouth garments
torn and discolored, we must have resembled the aboriginal savages of this wild region. Some thick masses of
sphagnum moss, long since dried up, gave us a soft covering, to place on the rough, rocky ground. Our supper
consisted of bacon, hard tack, and tea.    Large flat stones
h—^ A Mountain Bivouac.
laid on a gentle charcoal fire served to broil our bacon
most excellently, though the heat soon cracked the stones
in pieces.
At eight o'clock we retired to the protection of our
shelter. Overhead the starless sky was cloudy and threatened rain. The aneroid, which was falling, indicated that
our altitude was only 4,700 feet above the sea. We
arose early in the morning ; our breakfast was over and
everybody ready to proceed at seven o'clock. We were
now on the Pacific slope, and, according to our calculations, on one of the tributaries to the north fork of the
Cross River, which, in turn, is a tributary to the Kootanie.
We had a plan to explore up the valley from which our
stream issued, but beyond that, all was indefinite. It was
possible that this valley led around Mount Assiniboine so
that we could reach camp in two days. We were, however, certain of nothing as to the geography of the region
which we were now entering.
The clouds covered the entire sky and obscured the
highest mountain peaks. Worse still, they steadily descended lower and lower, a sign of bad weather. We had,
however, but this day in which to see the south side of
Mount Assiniboine, and consequently were resolved to do
our best, though the chances were much against us. For
three hours we followed the stream through the burnt timber, then the country became more open and our progress,
accordingly, more rapid. A little after ten o'clock we sat
down by the bank of the stream to rest for a few moments,
and eat a lunch of hard tack and cold bacon.    Such fare 172 The Canadian Rockies.
may seem far from appetizing to those of sedentary habits,
but our tramp of three hours over the fallen trees was
equivalent to fully five or six hours walking on a good
country road, and what with the fresh mountain air and a
light breakfast early in the morning, our simple lunch was
most acceptable.
A most pleasing and encouraging change of weather
now took place. A sudden gleam of sunlight, partially
paled by a thin cloud, called our attention upward, when
to our great relief several areas of blue sky appeared, the
clouds were rising and breaking up, and there was every
prospect of a change for the better.
Once more assuming our various packs, we pushed on
with renewed energy. On the left or south was a long
lofty ridge of nearly uniform height. On the right was a
stupendous mountain wall of great height, the top of which
was concealed by the clouds. This impassable barrier
seemed to curve around at the head of the valley, and, turning to the south, join the ridge on the opposite side.
.This then was a "blind" valley without an outlet. There
were two courses open to us. The first was to wait a few
hours, hoping to see Mount Assiniboine and return to
camp the way we came. The second was to force a passage,
if possible, over the mountain ridge to the south and so
descend into the North Fork valley, which we were certain
lay on the other side. The latter plan was much preferable,
as we would have a better chance to see Mount Assiniboine,
and the possibility of returning to camp by a new route.
After a short discussion, we selected a favorable slope
i Hope and Despair.
173
and began to ascend the mountain ridge. A vast assem-
, blage of obstacles behind us in the shape of two high
passes, dense forests, and a horrid infinity of fallen trees,
crossed bewilderingly, made a picture in our minds, constant and vivid as it was, that urged us forward. In striking contrast to this picture, hope had built a pleasing air
castle before us. We were now climbing to its outworks,
and should we succeed in capturing the place, a new and
pleasant route would lead us back to camp and place us
there—so bold is hope—perhaps by nightfall.
Thus with a repelling force pushing from behind and
an attractive force drawing us forward, we were resolved
to overcome all but the insuperable.
There was much of interest on the mountain slope,
which was gentle, and allowed us to pay some attention to
our surroundings. On this slope the scattered pine trees
had escaped the fire and offered a pleasant contrast to
the burnt timber. We passed several red-colored ledges
containing rich deposits of iron ore, while crystals of
calcite and siderite were strewed everywhere, and often
formed a brilliant surface of sparkling, sharp-edged rhombs
over the dull gray limestone. Among the limestones and
shales we found fossil shells and several species of trilo-
bites.
In an hour we had come apparently to the top of our
ridge, though of course we hardly dared hope it was the
true summit. As, one by one, we reached a commanding
spot, a blank, silent gaze stole over the face of each. To
our dismay, a vertical wall of rock, without any opening 174
The Canadian Rockies.
whatever, stood before us and rose a half thousand feet
higher. Thus were all our hopes dashed to the ground
suddenly, and we turned perforce, in imagination, to our
weary return over the many miles of dead and prostrate
tree trunks that intervened between us and our camp.
The main object of our long journey was, however,
at this time attained, for the clouds lifted and revealed the
south side of Mount Assiniboine, a sight that probably no
other white men have ever seen. I took my camera and
descended on a rocky ridge for some distance in order to
get a photograph. Returning to where my friends were
resting, I felt the first sensation of dizziness and weakness,
resulting from unusual physical exertion and a meagre
diet. I joined the others in another repast of raisins and
hard tack, taken from our rapidly diminishing store of
provisions.
Some more propitious divinity must have been guiding.,
our affairs at this time, for while we were despondent at
our defeat, and engaged in discussing the most extravagant routes up an inaccessible cliff, our eyes fell on a well
defined goat trail leading along the mountain side on our
left. It offered a chance and we accepted it. Peyto set
off ahead of us while we were packing up our burdens,
and soon appeared like a small black spot on the steep
mountain side. Having already passed several places that
appeared very dangerous, what was our surprise to see him
now begin to move slowly up a slope of snow that appeared nearly vertical. We stood still from amazement,
and argued that if he could go up such a place as that,
he could go anywhere, and that where he went we could
h^H Success at Last.
follow. We rushed after him, and found the goat trail
nearly a foot wide, and the dangerous places not so bad
as they seemed. The snow ascent was remarkably steep,
but safe enough, and, after reaching the top, the goat trail
led us on, like a faithful guide pointing out a safe route.
We could only see a short distance ahead by reason of
the great ridges and gullies that we crossed. Below us
was a steep slope, rough with projecting crags, while, as we
passed along, showers of loose stones rolled down the mountain side and made an infernal clatter, ever reminding us
not to slip. At one o'clock we stood on the top of the
ridge 9000 feet above sea-level, having ascended 4300 feet
from our last camp.
The valley of the north fork of the Cross River lay
far below, with green timber once more in sight, inviting
us to descend. After five minutes delay, for another
photograph, we started our descent, very rapidly, at first,
in order to get warm. We descended a steep slope of
loose debris, then through a long gully, rather rough, and
rendered dangerous by loose stones, till at length we
reached the grassy slopes, then bushes, finally trees and
forests, with a warm summery atmosphere. Here,
beautiful asters and castilleias, and beds of the fragrant
Linneas, delicate, low herbs with pale, twin flowers, each
pair pendent on a single stem, gave a new appearance to
the vegetation. In still greater contrast to the dark coniferous forests of the mountain, there were many white
birch trees, and a few small maples, the first I have ever
seen in the Rockies. In a meadow by the river we feasted
on wild strawberries, which were now in their prime. i76 The Canadian Rockies.
Near the river we discovered a trail, the first we had
seen so far on our journey around Assiniboine. After an
hour of walking we came to a number of horses, and soon
saw on the other side of the river a camp of another party
of gentlemen, who were exploring this region, and had
been out from Banff twenty-four days. We forded the
river, and found it a little over our knees, but very swift.
A very pleasant half hour was spent at this place, enjoying their hospitality, and then we pushed on. We
were now going westward up the valley, which held a
straight course of about six miles, and then turned
around to the north. The trail being good, we
walked very rapidly till nightfall in a supreme
effort to reach our camp that night. Having now
been on our feet almost continuously for the past
fifteen hours, we had become so fatigued that a very
slight obstruction was sufficient to cause a fall, and every
few minutes some one of the party would go headlong
among the burnt timber. We had barely enough provisions for another meal, however, and so we desired to
get as near headquarters as possible. At length, nightfall having rendered farther progress impossible, we found
a fairly level place among the prostrate trees, and, after a
meal of bacon and hard tack, lay down on the ground
around a large fire. The night was mild, and extreme weariness gave us sound sleep. After four hours of sleep, we were
again on foot at four o'clock in the morning. We marched
into camp at 6:30, where the cooks were just building the
morning fires, and commencing to prepare breakfast. Topography.
177
We were without doubt the first to accomplish the
circuit of Mount Assiniboine. By pedometer, the distance was fifty-one miles, which.we accomplished in forty-
six hours, or less than two days.
Mount Assiniboine is the culminating point of a nearly
square system of mountains covering about thirty-five
square miles. According to my estimates from angles
taken by surveying instruments made on the spot, the
mountain is 11,680 feet in height. Later on, however, I learned from Mr. McArthur, who is connected
with the Topographical Survey, and who has probably
climbed more peaks of the Canadian Rockies than any
other two men, that, according to some angles taken on
this mountain from a great distance, the height is 11,830
feet.
Three rivers, the Spray, the Simpson, and the North
Fork of the Cross, drain this region, and as the two latter
flow into the Columbia, and the former into the Saskatchewan, this great mountain is on the watershed, and
consequently on the boundary line between Alberta and
British Columbia. About two-thirds of the forest area
round its base has been burned over, and this renders
the scenery most unattractive. The north and northwest
sides, however, are covered with green timber, and studded
with lakes, of which one is two miles or more in length.
There are in all thirteen lakes around the immediate base
of the mountain, and some are exquisitely beautiful.
The great height and striking appearance of Mount
^Assiniboine will undoubtedly, in the future, attract moun- The Canadian Rockies.
taineers to this region, especially as a much shorter route
exists than the one we followed. If the trail is opened
along the Spray River, the explorer should be able to
reach the mountain, with horses, in two days from Banff.
Mount Assiniboine, especially when seen from the north,
resembles the Matterhorn in a striking manner. Its top
is often shrouded in clouds, and when the wind is westerly,
frequently displays a long cloud banner trailing out from
its eastern side. The mountain is one that will prove
exceedingly difficult to the climber. On every side the
slope is no less than fifty degrees, and on the east, approaches sixty-five or seventy. Moreover, the horizontal
strata have weathered away in such a manner as to form
vertical ledges, which completely girdle the mountain, and,
from below, appear to offer a hopeless problem. In every
storm the mountain is covered with new snow, even in
summer, and this comes rushing down in frequent avalanches, thus adding a new source of danger and perplexity
to the mountaineer.
The day of our arrival in camp was spent in much-
needed rest. Our time was now up, and it was necessary,
on the next day, to commence our homeward journey,
and, as our winding cavalcade left the beautiful site of
our camp under the towering walls of Mount Assiniboine,
many were the unexpressed feelings of regret, for in the
two weeks spent here we had had many delightful experiences, and had become familiar with every charming view
of lakes and forests and mountains.
In two days we reached the fork where the Simpson . Crossing the Vermilion River.
179
and Vermilion rivers unite. It was our intention to
follow up the Vermilion River and reach the Bow valley
by the Vermilion Pass. The Vermilion River is at this
point a large, deep stream flowing swiftly and smoothly
The valley is very wide and densely forested, with occasional open places near the river. For three days we
progressed up the river, often being compelled to cross it
on account of the dense timber. At one place, after
several of the horses had gained a bar in the middle of
the river, one of those following, got beyond his depth
and was swept rapidly down, and appeared in great danger
of being drowned. Fortunately, the animal was caught
by an eddy current, and by desperate swimming at length
gained the bar. The poor beast was, however, so much
benumbed by the cold water that he could not climb upon
the bar, but the men dashed in bravely, and by pulling on
head and packs, and even his tail, the animal finally
struggled into shallow water. Standing up to our knees
in the water, with a deep channel on either side of us and
an angry rapid below, our prospects were far from encouraging.
I mounted old Chiniquy behind Peyto and we plunged
in first. " It 's swim sure this time," said Peyto to me, as
the water rose at once nearly to the horse's back, and the
ice-cold water, creeping momentarily higher, gave us a
most uncomfortable sensation. The current was so swift
that the water was banked up much higher on the upstream side. Such crossings are very exciting, for at any
moment the horse may stumble on the rough bottom or i8o The Canadian Rockies.
plunge into a deep hole. Chiniquy had a hard time of
it and groaned at every step, but got us across all right.
The rest all followed, not, however, over-confident at our
success, to judge by their anxious looks. All got across
except one pack-horse, which, after a voyage down stream,
we finally caught and pulled ashore.
There was evidence of much game in this valley, as
we saw many^ tracks of deer, caribou, and bears. One
day, just as we stopped to camp, a doe started up and ran
by us. We camped on August 2nd at a beautiful spot
near the summit of the Vermilion Pass. A large stream
came in from the northwest, and we set out to explore it
for a short distance, as, before leaving Banff, we had heard
of a remarkable canyon near this place.
Not more than an eighth of a mile from the junction
of the two streams the canyon commences. At first, the
stream is hemmed in by two rocky walls a few feet in
height, but as one ascends, the walls become higher and
higher, and the sound of the roaring stream is lost in the
black depth of a gloomy chasm. To one leaning over the
edge of the beetling precipice, this wonderful gorge appears like a bottomless rift or rent in the mountain side,
and so deep is it and so closely do the opposite, irregular
walls press one towards another, that it is impossible to
see the waters below from which a faint, sullen murmur
comes up.
Most wonderful of all, the canyon at length comes to
a sudden termination, and here the whole mighty stream
plunges headlong, as it were, into the very bowels of the A Wonderful Canyon.
earth. The boiling stream, turned snow-white by a short
preliminary leap, makes a final plunge downwards and
is lost to sight in a dark cavernous hole, perhaps 300
feet deep, whence proceeds a most awful roar, like that
of ponderous machinery in motion. The ground, which
is here a solid quartzite formation, fairly trembles at.
the terrible concussion and force of the falling waters,
while cold, mist-laden airs ascend in whirling gusts from
the awful depths. Niagara is majestically and supremely
grand, but this lesser fall, where the water plunges into a
black bottomless hole, is by far the more terrifying.
On the fourth of August we reached the summit of
the Vermilion Pass. On the summit we passed several
small lakes
in the forest.
The water was
of a most beautiful color, far
more vivid
than any I
have^ hitherto
seen. In the
shallow places
where the bottom could be
easily seen,
the water assumed a bright, clear, green color, and in the deeper
places, according to the light and angle of view, the color
LAKE ON VERMILION PASS. The Canadian Rockies.
varied to darker hues of all possible shades and tints.
The rich colors of sky and water in the Rocky Mountains
is one of the most beautiful features of the scenery, but
likewise one that can only be appreciated by actual
experience.
Our horses were plagued by great numbers of bull-dog
flies as we entered the Bow valley. It seems as though
these insects were more numerous in the valley of the
Bow, and its various tributaries, than in those parts of the
mountains drained by other rivers.
At four o'clock we reached the Bow River, and forded
it where the width was about one hundred yards, and the
depth four feet. My camera and several plates were
flooded in this passage, which was, however, effected in
safety.
A march of one hour more, along the tote-road, brought
us to the station of Castle Mountain, once a thriving village in the railroad-construction days, but now presenting
a forlorn and deserted appearance. The section men
flagged the east-bound train for us, and we arrived in
Banff that evening, after having been in camp for twenty-
nine days. CHAPTER XI.
The Waputehk Range—Height of the Mountains— Vast Snow Fields
and Glaciers—Journey up the Bow—Home of a Prospector—Causes and
Frequency of Forest Fires—A Visit to the Lower Bow Lake—Muskegs—
A Mountain Flooded with Ice—Delightful Scenes at the Upper Bow
Lake—Beauty of the Shores—Lake Trout— The Great Bow Glacier.
THE Summit Range of the Rocky Mountains as they
extend northward from the deep and narrow
valley of the Kicking Horse River has a special
name—the Waputehk Range,—derived, it is said, from a
word which in the language of the Stoney Indians means
the White Goat.
From the summit of one of the peaks in this range,
the climber beholds a sea of mountains running in long,
nearly parallel, ridges, sometimes uniting and rising to a
higher altitude, and again dividing, so as to form countless
spurs and a complicated topography. In this range each
ridge usually presents a lofty escarpment and bare precipitous walls of rock on its eastern face, while the opposite slope is more gentle. Here the Cambrian sandstones
and shales and the limestones of later ages may be seen
in clearly marked strata tilted up, generally, toward the
east, though many of the mountains reveal contortions
and faults throughout their structure, which indicate the The Canadian Rockies.
wellnigh  inconceivable  forces that  have  here  been   at
work.
The Waputehk Mountains have remained to this day
but very little known, and almost totally unexplored, in
their interior portions. No passes are known through
this range between the Kicking Horse Pass on the south
and the Howse Pass on the north. Then another long
interval northwards to the Athabasca Pass is said by the
Indians to offer an impassable barrier to men and horses.
The continuity of the range is well indicated by the fact
that, for a distance of one hundred miles, these mountains
present only one pass across the range available for horses.
The several ridges which form this range rise to a
very uniform altitude of 10,000 or 11,000 feet. On Pal-
liser's map of this region, one peak north of the Howse
Pass, Mount Forbes, is accredited with an altitude of
13,400 feet, and the standard atlases have for many years
placed the altitude of Mount Brown at 16,000 feet, and
that of Mount Hooker at 15,700 feet, but there is much
reason to doubt that any mountains attain such heights
in this part of the Rocky Mountains.
A heavy snowfall, due to the precipitation brought
about by this lofty and continuous range, as the westerly
winds ascend and pass over it, and the existence of many
elevated plateaus, or large areas having gentle slopes,
have conspired to make vast névé regions and boundless
snowfields among these mountains. From the snowfields, '
long tongues of ice and large glaciers descend into the
valleys, and thus drain away the surplus material from the The Waputehk Range.
Looking across the range from near Hector.    The Waputehk Range. 185
higher altitudes. No other parts of the Rocky Mountains,
south of Alaska, have glaciers and snowfields which may
compare in size or extent with those of the Waputehk
Range.
The desolate though grand extent of gray cliffs and
boundless snowfields, extending farther than the eye can
reach, when seen from a high altitude, gives no suggestion
of the delightful valleys below, where many beautiful
lakes nestle among the green forests, and form picturesque
mirrors for the surrounding rugged mountains. On the
shores of one of these mountain lakes, in the genial warmth
of lower altitudes, where the water is hemmed in, and encroached upon, by the trees and luxuriant vegetation
fostered by a moist climate, the explorer beholds each
mountain peak as the central point of interest in every
view. Each cliff or massive snow-covered mountain then
appears an unscalable height reaching upward toward the
heavens,—a most inspiring work of nature, raising the eyes
and the thoughts above the common level of our earth.
When seen from high altitudes, a mountain appears merely
as a part of a vast panorama or a single element in a wild,
limitless scene of desolate peaks, which raise their bare,
bleak summits among the sea of mountains far up into the
cold regions of the atmosphere, where they become white
with eternal snow, and bound by rigid glaciers.
Having become much interested in reports of the vast
dimensions of the glaciers in the Waputehk Mountains,
and the beauty of the lakes, especially near the sources of
the Bow River and the Little Fork of the Saskatchewan, I 186
The Canadian Rockies.
started on August 14th, 1895, with the intention of visiting
those regions and spending some time there.    My outfit
consisted     of
ÊË&iËÊBSLl?*'   'f^m^mi ifi^ ""' ffi*^
five horses, a
^0t^^^^^ !R£jÉ^'
cook,   and    a
V                                                        *È* S^>Si
packer.   I had
jËÈÊ^
engaged Pey
TfSgpft ï-^v^ -** *■   < *. ;■>% ?|    'ta\5*^P^V^|^|^JfeS.\Mr"5
to for the lat
^^î^^w^^s^^sftj"
ter service, as
he   had been
^^^Mli»1^^^®^
most efficient
^^^^^^^^^^^^^^^^^^^^^^
on our trip to
MountAssini-
SI^^^^Bi^Wftfl^B^^^K
boine.     We
left Laggan a
READY TO  MARCH.
little     before
noon.    Not far from the station, there commenced an old
tote-road, which runs northward for many miles toward the
source of the Bow River. This tote-road had been hastily
built for wagons, previous to the construction of the railroad through the Kicking Horse Pass, for at one time it
was thought the line would cross the range by the Howse
Pass.
Thus for several miles we enjoyed easy and rapid travelling. The weather was mild and pleasant, and my men
seemed pleased at the prospect of another month or so in
camp.
In the course of a few miles we came to the house of
an old prospector.    As this was the farthest outpost of *
Home of a Prospector. 187
civilization, and the old man was reported to be an interesting character, I entered the log-house for a brief visit.
The prospector's name was Hunter. I found him at
home and was cordially welcomed. Here, in a state of
solitude and absolute loneliness, with no lake or stream to
entertain, and surrounded by a bristling maze of bleached
bare sticks looking like the masts of countless ships in a
great harbor, this man had spent several years of his life,
and, moreover, was apparently happy. On his table I saw
spread about illustrated magazines from the United States
and Canada, newspapers, and books. The house was
roughly but comfortably finished inside, and furnished
with good chairs and tables evidently imported from
civilization.
This isolated dwelling and its solitary inhabitant
reminded me somewhat of Thoreau at Walden Pond.
Like this lover of nature, Hunter enjoys his hermit life,
which he varies occasionally by a visit to the village of
Laggan. Hunter had the better house of the two men,
but Thoreau must have had much more to entertain him,
in his garden, and the beautiful lake with its constant
change of light and shadow, and the surrounding forests
full of well-known plants and trees, where his bird and
animal friends lived in undisturbed possession.
No sooner had we taken leave of this interesting home
of the old prospector, than the trail plunged into the
intricacies of the burnt timber, and our horses were
severely tried. Peyto and another man had been at work
on this part of the trail for two days, very fortunately for The Canadian Rockies.
us, as without some clearing we should not have been
able to force our way through.
The fire had run through after the tote-road was built,
so that the fallen timber now rendered it nearly impassable in many places. The forest fires have been much
more frequent since the country has been opened by the
whites, but it would be a great mistake to conclude that
before the arrival of civilized men the country was clothed
by an uninterrupted primeval forest. When we read the
accounts of Alexander Mackenzie, and the earliest
explorers in the Rocky Mountains, we find burnt timber
frequently mentioned.
However, these accounts only cover the last one hundred years, and records of geology must be sought previous
to 1793. Dr. Dawso-n mentions a place near the Bow
River where forest trees at least one hundred years old
are growing over a bed of charcoal made by an ancient
forest fire. Another bank near the Bow River, not far
from Banff, reveals seven layers of charcoal, and under
each layer the clay is reddened or otherwise changed by
the heat. Thus the oldest records carry us back thousands of years. The cause of these ancient fires was
probably, in great part, lightning, and possibly the
escaping camp fires of an aboriginal race of men.
Forest fires in the Canadian Rockies only prevail at
one season of the year—in July, August, and September,
—when the severe heat dries up the underbrush and fallen
timber. Earlier than this, everything is saturated by the
melting snows of winter, while in autumn the sharp frosts Forest Fires. 189
and heavy night dews keep the forests damp. According
to the condition of the trees, a forest fire will burn sometimes slowly and sometimes with fearful rapidity. . When
a long period of dry, hot weather has prevailed, the fire,
once started, leaps from tree to tree, while the sparks soar
high into the air and, dropping farther, kindle a thousand
places at once. The furious uprush of heated air causes
a strong draught, which fans the fire into a still more
intense heat. Sometimes whirlwinds of smoke and heated
air are seen above the forest fires, and at other times the
great mass of vapor and smoke rises to such a height that
condensation ensues, and clouds are formed. In the summer of 1893, a forest fire was raging about five miles east
of Laggan. Standing at an altitude of 9000 feet, I had
a grand view of the ascending smoke and vapors, which
rose in the form of a great mushroom, or at other times
more like a pine tree,—in fact, resembling a volcanic eruption. Judging by the height of Mount Temple, the clouds
rose about 13,000 feet above the valley, or to an altitude
of 18,000 feet above sea-level. It was a cumulus cloud,
shining brilliant in the sunlight, but often revealing a coppery cast from the presence of smoke. The ascending
vapors gave a striking example of one of the laws of
rising air currents. The tendency of an ascending column of air is to break up into a succession of uprushes,
separated by brief intervals of repose, and not to rise
steadily and constantly. The law was clearly illustrated
hy this cloud, which, at intervals of about five or six
minutes,   would    nearly    disappear    and    then    rapidly
	 190
The Canadian Rockies.
form again and rise to an immense height and magnitude.
In the course of a few years after a forest fire has
swept along its destructive course, the work of regeneration begins, and a new crop of trees appears. Sometimes the growth is alike all over the burnt region, young
trees springing up spontaneously everywhere, and sometimes the surrounding green forests send out skirmishers,
and gradually encroach on the burnt areas. Curiously
enough, however, a new kind of tree replaces the old
almost invariably. Out on the prairie the poplar usually
follows the coniferous trees, but in the Rockies, where the
poplar can not grow at high altitudes, the pines follow
after spruce and balsam, or vice versa. The contest of
species in nature is so keen that the slightest advantage
gained by any, is sufficient to cause its universal establishment. This is probably due to the fact that the soil becomes
somewhat exhausted in the particular elements needed by
one species of tree, so that when they are removed by an
unnatural cause, new kinds have the advantage in the
renewed struggle for existence. Thus we have a natural rotation of crops illustrated in the replacement of forest trees.
While we have been considering the causes and effects
of forest fires, our horses and men have been struggling
with the more material side of the question, and as the
imagination leaps lightly over all sorts of obstacles, let us
now overtake them as they arrive at a good camping place
about eight miles from Laggan. Here the Bow is no
longer worthy the name of a river, but is rather a broad, The Lower Bow Lake. 191
shallow stream, flowing with moderate rapidity. Towards
evening Peyto shot a black duck on the river, and I caught
a fine string of trout, so that our camp fare was much
improved.
The next day we marched for about three hours
through light forests and extensive swamps, finally pitching our camp near the first Bow Lake. The fishing was
remarkably fine in this part of the river. From a single
pool I caught, in less than three minutes, five trout which
averaged more than one pound each. We camped in this
place for two days in order to have time to explore about
the lake. This first Bow Lake is about four miles long, by
perhaps one mile wide, and occupies the gently curving
basin of a valley which here sweeps into that of the Bow.
There is something remarkable in the unusual manner in
which the Bow River divides itself into two streams some
time before it reaches this lake. The lesser of these two
streams continues in a straight course down the valley,
while the larger deviates to the west and flows into the
lower end of the lake, only to flow out again about a
fourth of a mile farther down, at the extreme end of the
lake. The island thus formed is intersected everywhere
by the ancient courses of the river, which are now marked
by crooked and devious channels, in great part filled with
clear water, forming pools everywhere. This whole region
must have once formed part of a much larger lake, as ior
several miles down the valley there are extensive swamps,
almost perfectly level and underlaid by large deposits of
fine clay. The Canadian Rockies.
The drier places in these muskegs are covered with a
growth of bushes or clumps of trees, gathered together on
hummocks slightly elevated above the general level. A
rich growth of grass and sedge covers the lower and
wetter places, which often assume all the features of a
peat bog, with a thick growth of sphagnum mosses, while
the ground trembles, for many yards about, under the
tread of men and horses.
The next day Peyto and I crossed the river on one of
our best horses known as the " Bay," and after turning
him back towards the meadow, we started on a tramp
around the lake. We followed the west shore for the
entire distance. The last half mile was over a talus slope
of loose stones, broken down from the overhanging mountain, and now disposed at a very steep angle. There was
a barely perceptible shelf or beach about six inches wide,
just at the edge of the water, which we gladly took
advantage of while it lasted.
The glacial stream entering the lake has built out a
curious delta, not fan-shaped as we should expect, but
almost perfectly straight from shore to shore. This delta
is a great gravel wash, nearly level, and quite bare of
trees or plants, except a few herbs, the seeds of which
have lately been washed down from higher up the valley.
All this material has been carried into the lake since the
time when, in the great Ice Age, these valleys were
flooded with glaciers several thousand feet in depth.
As we turned the corner near the end of the gravel
wash, the glaciers  at the  head  of  the valley began  to Mount Daly.  ~-ss  A Mountain Flooded with Ice. 193
appear, and in a few more steps we commanded a magnificent view of a great mountain, literally covered by a vast
sheet of ice and snow, from the very summit down to our
level. As we looked up the long gentle slope of this
mountain, we could hardly realize that it rose more than
5000 feet above us. The glacier which descended into
the valley was not very wide, but showed the lines of flow
very clearly. Six converging streams of ice united to
form the glacier on our right, while the glacier on the left
poured down a steep descent from the east, and formed a
beautiful ice cascade, where the sharp-pointed séracs, leaning forward, resembled a cataract suddenly frozen and
rendered motionless. As if by way of contrast, a beautiful little waterfall poured gracefully over a dark precipice
of rock on the opposite side of the valley, and added
motion to this grand expanse of dazzling white snow.
The loud-roaring, muddy stream near where we stood, is
one of the principal sources of the Bow, and, after depositing its milky sediment in the lake, the waters flow out
purified and crystal clear, of that deep blue color characteristic of glacial water. On a smaller scale this lake is
like Lake Geneva, with the Rhone entering at one end,
muddy and polluted with glacial clays, and flowing out
at the other, transparently clear, and blue as the skies
above it.
After a partial ascent of Mount Hector on the next
day, we moved our camp and continued our progress up
the Bow River for about two hours. Here we camped on
a terrace near the water, surrounded on all sides by a very 194
The Canadian Rockies.
light forest im a charming spot. On the following day the
trail led us for two miles through some very bad country,
where the horses broke through the loose ground between
the roots of trees, and in their efforts to extricate themselves were often in great danger of breaking a leg.
Fortunately, however, this was not of long duration. The
trail soon improved and became very clearly marked like
a well made bridle-path. It led us along the banks of
the Bow, through groves of black pine, with a few spruces
intermingled. The ascent was constant, though gradual,
and our altitude was made apparent by the manner in
which the trees grew in clumps, and by the fact that the
forests were no longer densely luxuriant, but quite open,
so that the horses could go easily among the trees in any
direction.
In about three hours after leaving camp, our horses
entered an open meadow where the trail deserted us, but
there was not the slightest difficulty in making good
progress. To the south, a great wall of rock rose to an
immense height, one of the lower escarpments of the
Waputehk Range, and as we progressed through the
pleasant moors a remarkable glacier was gradually revealed, clinging to the cliffs in a three-pronged mass.
As, one by one, these branches of the glaciers were disclosed, they appeared first in profile, and owing to the very
steep pitch down which the ice was forced to descend,
the glacier was rent and splintered into deep crevasses,
with sharp pinnacles of ice between, which appeared to
lean out over the steep descent and threaten to fall at
any moment. The Upper Bow Lake.
*95
The absence of trees to the north of us, and the
general depression of the country in that direction, gave
us every indication that we were approaching the Upper
Bow Lake, nor were our surmises incorrect, for in a few
minutes more of progress, after seeing the glacier, glimpses
of water surface were to be had in the near distance among
the trees. I went ahead of our column of horses and
selected a beautiful site for our camp, on the shore of the
lake, only a few yards from the water. The surrounding
region was certainly the most charming I have seen in
the Rocky Mountains. The lake on which we camped
was nearly cut off from the main body of water to the
north, by a contraction of the shores to a narrow channel.
In fact, it might be regarded as a land-locked harbor of
the Upper Bow Lake. Just below our camping place the
waters were contracted again, and descended in a shallow
rapid to another lake, resting against the mountain side
on the south. This latter lake is about three or four
feet lower than the others, and appeared to be about
two-thirds of a mile in length.
This region, for the artist with pencil and brush,
would be a fairy-land of inexhaustible treasures. The
shores along these various lakes were of a most irregular
nature, and in sweeping curves or sudden turns, formed
innumerable coves and bays, no less pleasing by reason
of their small extent. Long, low stretches of land, adorned
with forest trees, stretched straight and narrow far out
into the two larger lakes, their ends dissolving into chains
of wooded islands, separated from the mainland by shal- The Canadian Rockies.
low channels of the clearest water. In every direction
were charming vistas of wooded isles and bushy shores,
while in the distance were the irregular outlines of the
mountains, their images often reflected in the surface of
the water. The very nature of the shores themselves,
besides their irregular contours, varied from place to place
in a remarkable manner. In one locality the waters
became suddenly deep, the abrupt shores were rocky, and
formed low cliffs ; in other places the bottom shelved off
more gradually, and there would be a narrow beach of
sand and small pebbles, ofttimes strewed with the wreckage of some storm,—a massive tree trunk washed upon
the beach, or stranded in shallow water near the shore.
There were, moreover, many shallow areas and
swampy tracts where a rich, rank growth of water grasses
and sedges extended into the lake. Such border regions
between the land and water were perhaps the most
beautiful and attractive of all the many variations of
these delightful shores. The coarse, saw-edged leaves
of the sedges, harsh to the touch, are pliant in the
gentlest breath of wind. The waving meadows of green
banners, or ribbons, rising above the water, uniform in
height, and sensitive to the slightest air motion, rustle
continuously as the breezes sweep over them, and rub
their rough surfaces together.
From this region, wherein were combined so many
charming views of nature, with mountain scenery, lakes,
islands, and forests, all of the most attractive kind, it
proved impossible to move our camp for several days. if I l\\mHmntt™imam!imHm*Mm\tMWntimmBE&fflmL BHHHHHHH
Upper Bow Lake.
Looking east. pp
n fsm
Ifflïfit tHHtllPHWlGi
mm
y Ml BHMHHHS
mmmmem
^PSBBSfcC*
Excursions.
197
During the time that we remained here, our explorations and wanderings took us along all the shores and
islands, and up the neighboring mountain slopes. On
one of the islands opposite our camp we discovered a
small pool of singular formation. The pool was nearly
circular, and about ten yards in diameter. The bottom
was funnel-shaped, and in the very centre was^a black
circle—in fact a bottomless hole—apparently connected
by dark subterranean channels with the depths of the
adjacent lake. Its borders were low and swampy, where
the spongy ground quaked as we moved about, and
trembled so much that we feared at any moment to be
swallowed up. In fact the whole pool became rippled by
the movements of its banks.
The glacier opposite was the object of another trip,
and this, too, proved interesting. The névé on the flat
plateau above discharges its surplus ice for the most part
by hanging glaciers, which from time to time break off
and fall down the precipice. We were often startled
both day and night by the thunder of these avalanches.
Two tongues of ice, however, effect a descent of the
precipice where the slope is less steep, and though much
crevassed and splintered by the rapid motion, they reach
the bottom intact. Here the two streams, together with
the accumulations of ice constantly falling down from
above, become welded into a single glacier, which terminates only a short distance from the lake. The most
unusual circumstance about this glacier is the fact that
the   ice  is  much   higher  at  the very end   than a little The Canadian Rockies.
farther back, so that a great, swelling mound of ice,
about 200 feet thick, forms the termination.
About one fourth of a mile below the end of the
glacier, on an old moraine ridge now covered over with
luxuriant forest, we saw a towering cliff of rock rising
above the trees. This proved, on a closer examination,
to be a separate boulder, which must have been carried
there by the ice a long time ago. It was of colossal proportions, at least sixty feet high, and nearly as large in
its other dimensions. From the top we had an extensive
view of the lakes and valleys ; while at its base we found
on one side an overhanging roof, making so complete a
shelter, that it was not difficult to imagine this place to
have been used by savages, in some past age, as a cave
dwelling.
Many years ago, not less than one hundred, the forests
on the slopes to the east of the valley had been devastated by a fire. The long lapse of time intervening had,
however, nearly obliterated the dreary effects of this
destruction. The trees had replaced themselves scatter-
ingly among the dead timber, and attained a large size.
The fallen trunks showed the great length of time they
had lain on the ground by the spongy, decomposed condition of the wood. Many of the trunks had dissolved
into red humus, the last stage of slowly decomposing
wood, and the fragments were disposed in lines, bare of
vegetation, indicating where each tree had found its final
resting-place.
The swampy shores and large extent of water surface BMiHBBIËIMHB
Insect Life.
199
in this region fostered many varieties of gnats, mosquitoes, and other insects, though, fortunately, not in such
great numbers as to be very troublesome. In fact, the
season of the year was approaching that period when the
mosquitoes suddenly and regularly disappear, for some
unexplained reason. I have always noticed that in the
Canadian Rockies the mosquitoes become much reduced
in numbers between the 15th and 20th of August, and
after that time cause little or no trouble. In order, however, that there may be no lack of insect pests, nature
has substituted several species of small flies and midgets,
which appear about this time and follow in a rotation of
species, till the sharp frosts of October put an end to all
active insect life. Some of these small pests are no less
troublesome than the mosquitoes which have preceded
them, though they afford a variation in their manner of
annoyance, and are accordingly the more endurable.
Along the reedy shores of the lake and sometimes
over its placid surface, when the air was quiet toward
evening, we often saw clouds of gnats hovering motionless in one spot, or at times moving restlessly from place
to place, like some lightless will-o'-the-wisp, composed of
a myriad of black points, darting and circling one about
another. Nature seems to love circular motion : for just
as the stars composing the cloudy nebulae revolve about
their centres of gravity in infinite numbers, moving forever, through an infinity of space ; so do these ephemeral
creations of our world pass their brief lives in a ceaseless
vortex of complicated circles. The Canadian Rockies.
On one occasion we built a raft to ferry us across the
narrow part of the lake so that we might try the fishing
on the farther side. The raft was hastily constructed,
and, after we had reached deep water, it proved-to be in .a
state of stable equilibrium only when the upper surface
was a yard under water. After a thorough wetting we
finally reached the shore, and proceeded to build a more
trustworthy craft.
On the 21st of August we moved our camp down to
the north end of the lake. Here the nature of the scenery
is entirely changed. Whereas the lower end of the lake
abounds in land-locked channels and wooded islands, so
combined as to make the most pleasing and artistic pictures from every shore, the other part of this lake presents regular shore lines, and everything is formed on a
more extensive scale. The north side of the lake is
curved in a great arc, so symmetrical in appearance that
it seems mathematically perfect, and the eye sweeps along
several miles of shore at a single glance as though this
were some bay on the sea-coast.
As we neared the north end of the lake, a valley was
disclosed toward the west, and an immense glacier appeared descending from the crest of the Waputehk Range.
Even at a distance of three or four miles, this glacier
revealed its great size. The lower part descended in
several regular falls to nearly the level of the lake. In
the lower part, the glacier is less than a mile in width, but
above, the ice stream expands to three or four miles,, and
extends back   indefinitely,  probably ten miles or more. lilllljliilljlllljllillllllljllljllllllllllll
Upper Bow Lake.
Looking west.    Companions. 201
This Great Bow Glacier had the same position relatively
to the lake, as the glacier we visited at the Lower Bow
Lake held to that body of water.
A better knowledge of these lakes revealed a striking
similarity between them. Each lake occupies a curving
valley, which in each case enters the Bow valley from the
south. The two lakes are about the same size and
nearly the same shape, a long gentle curve about five
times longer than broad. At the head of each, though
at slightly different distances, are large glaciers. The
glacial streams have likewise formed flat gravel washes,
or deltas, which have encroached regularly on the lake
and formed a straight line from shore to shore, perfectly
similar one to another. A further resemblance might be
observed in the presence of two talus slopes from the
mountain sides, in each case on the south side of the
lake, near the delta. The Lower Bow Lake is about
5500 feet above sea-level, while the upper lake is a little
more than 6000 feet. The increased altitude has the
effect of making the forest more open, and the country
more generally accessible, in the region of the upper lake.
From one point on the shores of the upper lake, five
large glaciers may be counted, the least of which is two
miles long, and the greatest has an unknown extent, but is
certainly ten miles in length.
Our camp was pleasantly located in the woods not
far from the water. After Peyto had put up the tent
and got the camp in order, with the horses enjoying a fine
pasture, he set off to explore the lake shore toward the The Canadian Rockies.
Great Glacier.     He returned to camp about five o'clock
carrying a fine lake trout which he had caught.    This fish
was taken
near the shore,
and was probably a small
one compared
with those
which live in
deeper water ;
nevertheless,
it measured
twenty - three
inches in
length, and
weighed
about seven
pounds. The Bow lakes have a reputation for abounding
in fish of a very large size. So far as I am aware, no
boat has ever sailed these waters, and there is no certainty what size the fish may reach in the deeper parts of
the lake. Judging by trout which have been caught in
Lake Minnewanka, near Banff, it is very probable that
they run as high as thirty or forty pounds.
The next day, Peyto and I took a lunch with us and
spent the entire day exploring and photographing the
glacier and its immediate neighborhood. The ice is not
hemmed in by any terminal moraine, but shelves down
gradually   to a thin edge.     In   fact the termination of
CAMP AT UPPER BOW LAKE. mmrnmism
mmmmmmmmmm
mmmm
The Great Bow Glacier.
the glacier resembles somewhat the hoof of a horse, or
rather that of a rhinoceros, the divided portions being
formed by crevasses, while long thin projections of ice
spread out between. It is a very easy matter to get
on the glacier, and quite safe to proceed a long way
on its smooth surface. We had some fine glimpses
of crevasses so deep that it was impossible to see the
bottom, while the rich blue color of the ice everywhere
revealed to us marvels of colored grottoes and hollow-
sounding caverns, their sides dripping with the surface
waters. There is something peculiarly attractive, perhaps
from the danger, pertaining to a deep crevasse in a glacier.
One stands near the edge and throws, or pushes, large
stones into these caverns, and listens in awe to the hollow
echoes from the depths, or the muffled splash as the missile finally reaches a pool of water at the bottom. There
is a suggestion of a lingering death, should one make a
false step and fall down these horrible crevasses, where,
wedged between icy walls far below the surface, one could
see the glimmering light of day above, while starvation
and cold prolong their agonies. A party of three mountaineers thus lost their lives on Mount Blanc in 1820, and
more than forty years later their bodies were found at the
foot of the Glacier des Bossons, whither they had been
slowly transported, a distance of several miles, by the
movement of the ice. The most dangerous crevasses are
not those of the so-called "dry glacier," where the bare ice is
everywhere visible, but those of the névé regions where the
crevasses are concealed, or obscured by the overlying snow. 204
The Canadian Rockies.
Not far from the foot of the glacier the muddy stream
flows through a miniature canyon, with walls near together,
cut out of a limestone formation. The water here rushes
some quarter of a mile, foaming and angry, as it dashes
over many a fall and cascade. Where the canyon is deepest an immense block of limestone about twenty-five feet
long has fallen down, and with either end resting on the
canyon walls, it affords a natural bridge over the gloomy
chasm. As probably no human being had ever crossed
this bridge, we felt a slight hesitation in making the
attempt, fearing that even a slight jar might be sufficient
to dislodge the great mass. It proved, however, quite
safe and will undoubtedly remain where it is for many
years and afford a safe crossing-place for those who visit
this interesting region. mÊmmmmz
CHAPTER XII.
Sources of the Bow—The Little Fork Bass—Magnificence of the
Scenery—Mount Murchison—Camp on the Divide—A High Mountain
Ascent—Future of the Bow Lakes—Return down the Bow—Search for a
Bass—Remarkable Agility of Back-Horses—The "Bay" and the. "Pinto "
—Mountain Solitudes—Mount Hector—Difficult Nature of Johnston
Creek—A Blinding Snow-Storm—Forty-Mile Creek—Mount Edith Pass.
A FINE trout stream entered the lake near our camp. "
This was, in fact, the Bow River. It held a meandering course a short distance before entering
the lake, through a level meadow, or rather an open
region, thickly grown over with alder bushes and other
shrubby plants.
We were delayed at this camp by a period of unsettled
weather with occasional storms and strong winds, so that
three days were required to finish our explorations. At
length, on the 24th of August, we broke camp, and followed the Bow valley northwards towards the source of the
river. The valley preserves its wide character to the head
of the pass, and is unusual among all the mountain passes
for several reasons. The ascent to the summit is very gradual and constant, the valley is wide, and the country is
quite open. In about two hours we came to the summit,
and, after along level reach, the slope insensibly changed
and the direction of drainage was reversed.
205 20Ô
The Canadian Rockies.
This was a most delightful region. The smooth valley
bottom sloped gradually upward toward the mountains on
the east and west, and insensibly downward toward the
valleys north and south, thus making an extensive region
with gentle slopes curving in two directions, which in some
way impresses the mind with a sense of quiet grandeur
and indefinite liberty. But chiefly this region of the divide
is made charming by a most beautiful arrangement of the
trees. There are no forests here, nor do the trees grow
much in groves or clumps, but each tree stands apart, at
a long interval from every other, so that the branches
spread out symmetrically in every direction and give perfect forms and beautiful outlines. Between are smooth
meadows, quite free of brush, but crowded with flowering
plants, herbs, and grasses, so that the general impression
is that of a gentleman's park, under the control and care
of a landscape gardener, rather than of the undirected
efforts of nature.
I shall never forget the first view we had into the
valley of the Saskatchewan. Approaching a low ridge
at the south side of the valley, suddenly there is revealed a magnificent panorama of glaciers, lakes, and
mountains, unparalleled among the Canadian Rockies for
its combination of grandeur and extent. To the south,
one beholds the end of an immense glacier, at the termination of which there are two great arched caverns in the
ice. From out these issue two roaring glacial streams,
the source of the Saskatchewan River, or at least of its
longest tributary called the Little Fork.    Lofty mountains ^^gy^yj^L^^f^fffff^ff^r.* jj=^xfify^x^a3g^^^^^^g^^s^j^^g
Source of the Little Fork of the
Saskatchewan River.
_  7b
> 1
11
1
i
*
M
f J 1
^^^m^m
1
fr'^tl
ill
> '
M
H^Vtt
§§:
x»    Jtfl
wÈÊÈ ^M
WBmEMbuM I
1  mm^mmmmm^mmmmmmsmmmmÊmBB&smm^s^sm^
The Little Fork Pass. 207
hem in this glacier on either side, only revealing a portion
of the vast névé which may be seen extending southward
for six or seven miles.
To the north and, as it were, at our feet, though in
reality a thousand feet below, lay a large and beautiful
lake with irregular outlines. This lake reaches several
miles down the valley of the Little Fork, which here
extends northward so straight and regular, that the view
is only limited at the distance of thirty miles by the long
range of mountains on its east side. Dr. Hector, who
came through this region in the fall of 1858, comments
on the magnificent extent and grandeur of this view.
Through a notch in a mass of mountains to the north,
there appeared the extreme summit of Mount Murchison,
a very sharp and angular rock peak, which the Indians
regard as the highest mountain of the Canadian Rockies.
According to some rough angles taken by Dr. Hector,
this mountain has an altitude of 13,500 feet. In Palliser's
Papers a sketch of this mountain, as seen from the summit of the Pipestone Pass, makes the rock peak much
more sharp and striking in appearance even than that
of Mount Assiniboine, or of Mount Sir Donald in the
Selkirks.
We continued our journey over the pass and descended
into the valley of the Little Fork for several miles. The
trail was very good, though the descent was remarkably
steep. We camped by a small narrow lake, in reality
merely an expansion of the Little Fork. Behind us was
an area of burnt timber, but southward the forests were 208
The Canadian Rockies.
in their primeval vigor and the mountains rose to impressive heights above. The weather became rather dubious,
and during the night there was a fall of rain, followed by
colder weather, so that our tent became frozen stiff by
morning.
It seemed best to return the next day to the summit of
the pass, where everything conspired to make an ideal
camping place. Accordingly, the men packed the horses
and we located our camp on the crest of the divide, 6350
feet above sea-level. The tent was pitched in a clump of
large trees surrounded on all sides by open meadows,
where one could wander for long distances without encountering rough ground or underbrush. Near the camp
a small stream, and several pools of clear water, were all
easily accessible.
The next day I induced Peyto to ascend a mountain
with me. He was not used to mountain climbing, and
had never been any higher than the ridge that we were
compelled to cross when we were walking around Mount
Assiniboine, which was less than 9000 feet in altitude.
The peak which I had now in view lay just to the northeast from our camp on the pass. It appeared to be
between 9000 and 10,000 feet high, and offered no
apparent difficulties, on the lower part at least. We left
camp at 8:30 a.m. and passed through some groves of
spruce and balsam, where we had the good fortune to see
several grouse roosting among the branches of the trees.
Peyto soon brought them down with his six-shooter, in
handling which   he always displays remarkable accuracy Storm in Little Fork Valley. m ______  A High Mountain Ascent.
209
and skill. Many a time, when on the trail, I have seen
him suddenly take his six-shooter and fire into a tall tree,
whereupon a grouse would come tumbling down, with his
neck severed, or his head knocked off by the bullet.
A hawk scented our game and came soaring above us
so that we had to hide our birds under a covering of stones,
as of course we did not care to take them with us up the
mountain. We found not the slightest difficulty in the
ascent till we came near the summit. The atmosphere
was remarkably clear, and some clouds high above the
mountains rendered the conditions very good for photography. At an altitude of 9800 feet we came to the summit of the arete which we were climbing, and saw the
highest point of the mountain about one-third of a mile
distant, and considerably higher. Fortunately, a crest of
snow connected the two peaks, and with my ice-axe I
knocked away the sharp edge, and made a path. In a
few minutes we were across the difficult part and found
an easy slope rising gradually to the summit. We reached
it at 11:30, and found the altitude 10,125 feet. The view
from the great snow dome of this unnamed mountain was
truly magnificent. The Waputehk Range could be seen
through an extent of more than seventy-five miles, while
some of the most distant peaks of the Selkirks must have
been more than one hundred miles from where we stood.
To the east about ten miles was the high peak of Mount
Hector, almost touching the clouds.
In the northern part of the Waputehk Range we saw
some very high peaks, though the clouds covered every-
Wïm The Canadian Rockies.
thing above 11,000 feet. There seemed to be a storm in
that direction, as snow could be discerned falling on the
mountains about thirty miles distant. The general uniformity of height, and the absence of unusually high peaks,
a characteristic feature of the Canadian Rockies, were very
clearly revealed from this mountain.
Peyto was overwhelmed with the magnificent panorama, and said that he now appreciated, as never before,
the mania which impels men to climb mountains. The
storm which we saw in the west and north passed over
us toward evening, in the form of gentle showers. On the
next day, however, the weather was perfectly clear and calm.
On the 26th of August our horses were packed and
our little procession was in motion early in the morning,
and we were wending our way down the Bow River. I
cannot take leave of this region, however, even in imagination, without a word in regard to the unusual attractiveness of this part of the mountains.
In the first place there are magnificent mountains and
glaciers to interest the mountaineer, and beautiful water
scenes, with endless combinations of natural scenery for
the artist ; moreover, the streams abound in brook trout
and the lakes are full of large lake trout, so numerous as
to afford endless sport for fishermen. The botanist, the
geologist, and the general lover of science will likewise
find extensive fields of inquiry open to him on every side.
The time of travelling required by us to reach the
Upper Bow Lake was about nine hours, and this was with
heavily laden pack-horses.      Hitherto,   only  those  con- mmmmmmmmmmmm
Mount Hector and Slate Mountains.
From" summit of a mountain near Little  Fork Pass, 10,123 feel *n
altitude.  mmmmm
 ^SBB-.sj mm $^m
mimimmÊmgmmmgmmm
Visions of the Future.
nected with the early explorations, or the railroad surveys,
have visited this lake, but I cannot look forward to the
future without conjuring up a vision of a far different
condition of things. In a few years, if I mistake not, a
comfortable building, erected in a tasteful and artistic
manner, will stand near the shores of this lake on some
beautiful site. A steam launch and row-boats or canoes
will convey tourists and fishermen over the broad waters
of the lake, and a fine coach road will connect this place
with Laggan, so that passengers may leave Banff in the
morning and, after a ride of two hours by railroad, they
will be transferred to a coach and reach the Upper Bow
Lake in time for lunch ! If a good road were constructed
this would not be impossible, as the distance from Laggan
is only about twenty miles, and the total ascent iooo feet.
With such visions of the future and the more vivid
memory of recent experiences in mind, we took leave of
the beautiful sheet of water, and continued on our way
down the Bow valley. It was not our purpose, however,
to return to Laggan directly, for Wilson had planned an
elaborate route, by which some of the wilder parts of the
• mountains might be visited. This route would lead us
over a course of about eighty or one hundred miles
through the Slate Mountains and Sawback Range, and
eventually bring us to Banff.
We were to follow a certain stream that enters the
Bow from the north, but as we were now, and had been
for many days, outside the region covered by Dawson's
map, it was impossible to feel certain which stream we The Canadian Rockies.
should take. On our way up the Bow River, Peyto had
made exploring excursions into several tributary valleys,
but in every case these had proved to be hemmed in by
precipitous mountain walls, and guarded at the ends by
impassable cliffs or large glaciers.
The second day after leaving the lake we came to a
large stream which had not been examined hitherto.
Though we were far from certain that this was the stream
that had been indicated by Wilson, it seemed best to
follow up the valley and see where we should come out.
After ascending an exceedingly steep bank, we found easy
travelling in a fairly open valley. One fact made us apprehensive that there was no pass out of the valley.
There was no sign of a trail on either side of the stream,
and none of the trees were blazed. Indian trails exist in
almost every valley where an available pass leads over the
summit, and where there are no trails the probability is
that the valley is blind, or, in other words, leads into an
impassable mountain wall. The valley curved around in
such a manner that we could not tell what our prospects
were, but at about two o'clock we reached a place far
above timber line,—a region of open moors, absolutely
treeless,—surrounded by bare mountains on every side.
Our tent was pitched in a ravine near a small stream.
Immediately after lunch, Peyto and I ascended iooo
feet on a mountain north of the valley with the purpose of discovering a pass. From this point we saw
Mount Hector due southland the remarkable mountain
named Mount  Molar, nearly due east.    Three possible Camp at Little Fork Pass.  wm&
mmmMmsMmffîfânmmmmm
wmm  A Difficult Place.
outlets from the valley appeared from our high elevation.
Peyto set off alone to explore a pass toward the north,
in the direction of the Pipestone Pass, while I made an
examination of a notch toward the east. Each proved
impossible for horses, if not for human beings. The third
notch lay in the direction of Mount Hector, and together
we set out to examine it. A walk of about two miles
across the rolling uplands of this high region brought us
to the pass. It was very steep, but an old Indian trail
proved that the pass was available for horses. The trail
appeared more like those made by the mountain goats
than by human beings, for it led up to a very rough and
forbidding cliff, where loose stones and long disuse had
nearly obliterated the path. We spent some time putting
the trail in repair, by rolling down tons of loose stones,
and making everything as secure as possible.
The next morning was threatening, and gray, watery
clouds hung only a little above the summit of the lofty
pass, which was nearly 8000 feet above sea-level. I
started about an hour before the outfit, as I desired to observe the horses climbing the trail. I felt considerable anxiety as they approached. All my photographic
plates, the result of many excursions and mountain ascents in a region where the camera had never before
been used, were placed on one of the horses, for which
purpose one of the most sure-footed animals had been
selected. In case of a false step and a roll down the
mountain side, the results of all this labor would be lost.
The horses, however, all reached the summit in safety. 214
The Canadian Rockies.
These mountain pack-horses reveal a wonderful agility
and sagacity in such difficulties as this place presented.
In fact, the several animals in my pack-train had become
old friends, for they had been with me all summer. Peyto,
as packer, always rode in the saddle, for the dignity of
this office never allows a packer to walk, and besides, from
their physical elevation on a horse's back they can better
discern the trail. A venerable Indian steed, long-legged
and lean, but most useful in fording deep streams, was
Peyto's saddle-horse. The bell-mare followed next, led
by a head-rope. The other horses followed in single file,
and never allowed the sound of the bell to get out of hearing. There were two horses in the train that were endowed with an unusual amount of equine intelligence and
sagacity.      The   larger  of  the two   was  known as   the
| Bay," and
the other was
called " Pinto," the latter
being a name
given to all
horses having
Irregular
white markings. These
animals   were
THE      BAY."
well proportioned, with thick necks and broad chests, and, though
of  Indian   stock,   they  probably  had   some   infusion   of m
4g
ï
Intelligence of Pack Horses.
Spanish blood in their veins, derived from the conquest
of Mexico.
The Pinto was remarkably quick in selecting the best
routes among fallen timber, or in avoiding hidden dangers,
but the Bay was far more affectionate and fond of human
company. In camp, all the horses would frequently leave
the pasture and visit the tent, where they would stand
near the fire to get the benefit of the smoke when the flies
were thick, or nose about in the hope of getting some salt.
On the trail, it was always very interesting to watch the
Bay and Pinto. They would unravel a pathway through
burnt timber in a better manner than their human leaders,
and would calculate in every case whether it were better
to jump over a log or to walk around it. But one day I
was surprised to see the Bay jump over a log which
measured 3 feet 10 inches above the ground. With a
heavy, rigid pack this is more of a feat than to clear a
much greater height with a rider in the saddle. Sometimes
when the trail was lost we would put the Pinto ahead to
lead us, and on several occasions he found the trail for us.
The summit of the pass revealed to us one of those
lonely places among the high mountains where silence
appears to reign supreme. We were in an upland vale,
where the ground was smooth and rolling, and carpeted
with a short growth of grass and herbs. On either side
were bare cliffs of limestone, unrelieved by vegetation or
perpetual snow. Here no birds or insects broke the silence
of the mountain solitude, no avalanche thundered among
the mountains, and even the air was calm and made no 2i6 The Canadian Rockies.
sound in the scanty herbage. All was silent as the desert,
or as the ocean in a perfect calm. The dull tramp of our
horses, and the tinkling of the bell, were the only sounds
that interrupted the death-like quiet of the place. It is
said that such places soon drive the lost traveller to insanity, but in company with others these lonely passes
afford a delightful contrast to the life and motion and
sound of lower altitudes.
As we advanced and commenced to descend, the north
side of Mount Hector began to appear. It was completely
covered with a great ice sheet and snow fields. Mount
Hector is a little more than 11,000 feet in altitude, and
gives a good example of how the exposure to the sun
affects the size of glaciers in these mountains. On the
south and west sides of Mount Hector there is almost no
snow, while the opposite slopes are flooded by a broad
glacier many miles in area, and brilliant in a covering of
perpetual snow.
At the tree line a trail appeared, and led us in rapid
descent to the valley. The scenery on all sides was
magnificent. Many waterfalls came dashing down from
the melting glaciers of Mount Hector and joined a torrent
in the valley bottom. The great cliffs about us, and the
lofty mountains, visible here and there through avenues in
the giant forest trees, were illumined by a brilliant sun,
ever now and again breaking through the clouds. About
eleven o'clock we stopped to have a light lunch, as was our
custom on all long marches. Peyto loosed the girdle of
the horses, slipped off the packs, and turned the animals Little Pipestone Creek.
into a meadow near by. Meanwhile our cook cut firewood and made a large pot of tea, which always proved
the most acceptable drink when a long march had made
us somewhat weary. These brief rests of about forty
minutes in the midst of a day's march always proved very
beneficial to men and horses.
A long straight valley led us southwards for many
miles. In every clear pool or stream, trout could be seen
darting about and seeking hiding-places, though we had
no time to stop and catch them. At about one o'clock we
reached the Pipestone Creek and obtained a view of
Mount Temple and other familiar peaks about fifteen
miles to the south.
We camped near the stream in a meadow, not far from
the Little Pipestone Creek. As the march of this day
had brought us back to the region covered by the map, we
had little apprehension of losing our way in the future.
The next day we followed up the Little Pipestone
Creek and enjoyed a fine trail through a dense forest.
We camped near the summit of a pass south of Mount
Macoun, which I partially ascended after lunch. The
rugged peak named Mount Douglas lay due east, and
presented some very large and fine glaciers.
Our camp was on a little peninsula jutting out into a
lake, with water of a most brilliant blue color. The sunset colors this evening were heightened by the presence of
a little smoke in the atmosphere, which gave a deep
copper color to the western sky, while the placid lake
appeared vividly blue in the evening light. 2i8 The Canadian Rockies.
L
The following day, which was the first of September,
we continued south over a divide and into the valley of
Baker Creek, which we followed for several hours, and
then took a branch stream which comes in from the
east, and finally camped in a high valley. We were now in
the Sawback Range, where the mountains are peculiarly
rugged, and the strata thrown up at high angles. The
weather was giving evidence of an approaching storm,
and before we had made camp the next day in Johnston's
Creek, rain, began to fall.
Hitherto the nature of the country since leaving the
Upper Bow Lake had been such as to render the travelling very easy and delightful, but from this point on,
we met with all sorts of difficulties. In the lower part of
Johnston's Creek, and in the valley of a tributary which
comes in from the northeast, the trail was covered by
fallen timber, and our progress was very slow and tedious.
Moreover, the weather now became very bad, and we
were caught near the summit of a pass between Baker
Creek and Forty-Mile Creek in a heavy snow-storm, so
that the trail was soon obliterated and the surrounding
mountains could not be seen. Fearing that we might
lose our bearings altogether, Peyto urged forward the
horses at a gallop, so that we might get over the pass
before the snow gained much depth.
The descent into the valley of Forty-Mile Creek was
very steep, and we camped among some large trees with
several inches of snow on the ground. The next day we
urged our horses on again and followed down the valley ^^^^^^^m^m^^^^^^^^^^mm^^^^^^^^m
Mount Edith Pass.
219
of Forty-Mile Creek. In some parts of the valley we
found absolutely the worst travelling I have anywhere
met with in the Rockies. The horses were compelled to
make long detours among the dead timber, and the axe
was frequently required to cut out a passage-way. Frequent snow showers swept through the valley, and,
though very beautiful to look at, they kept the underbrush covered with damp snow and saturated our clothes
with water.
In the afternoon we reached the summit of the Mount
Edith Pass, and once more caught sight of the Bow
valley and the flat meadows near Banff. A fine wide
trail or bridle-path, smooth and hard, led us down toward
the valley. The contrast to our recent trails was very
striking. We walked between a broad avenue of trees,
each one blazed to such an extent that all the bark
had been removed on one side of the tree, and some
were practically girdled. This was very different from
our recent experience where we had only found a small
insignificant axe-mark on some dead tree, about once in
every quarter mile, or often none at all during hours of
progress.
On the fifth of September we reached Banff late in
the evening, and found that the valley was free of new
snow by reason ! of its lower altitude. We had been out
for twenty-three days and had covered, in all, about one
hundred and seventy-five miles. CHAPTER   XIII.
HISTORICAL.
Origin and Rise of the Fur Trade— The Coureurs des Bois and the
Voyageurs—Perils of the Canoe Voyages—The Hudson Bay Company and
the Northwest Company—Intense Rivalry—Downfall of the Northwest
Company—Sir Alexander Mackenzie—His Character and Physical Endowments — Cook's Explorations — Mackenzie Starts to Penetrate the
Rockies—The Peace River—A Marvellous Escape —The Pacific Reached
by Land—Perils of the Sea and of the Wilderness.
THE history of the early explorations in the Canadian Rockies centres about the fur trade. From
theMate of the very earliest settlements in Canada,
the quest of furs had occupied a position of chief importance, to which the pursuits of agriculture, grazing, or
manufacture had been subordinate. The search for gold,
which throughout the history of the world has ever been
one of the most powerful incentives to hardy adventure
and daring exploit, did not at first occupy the attention of
those who were ready to hazard their lives for the sake of
possible wealth quickly acquired.
The unremitting and often ruthless destruction of the
fur-bearing animals, in the immediate vicinity of the settlements, caused them to become exceedingly scarce, and at
length to disappear altogether.    But fortunately it was not Voyageurs.
difficult to induce the Indians to bring their furs from
more distant regions, until at length even those who lived
in the most remote parts of Canada became accustomed to
barter their winter catch at the settlements.
As the trade gradually became more extensive, there
sprang up two slightly different classes of men, the
coureurs des bois, or wood rangers, and the voyageurs,
each of Canadian birth, but who, by reason of constant
contact with the Indians and long-continued separation
from the amenities and refinements of civilized life, came
at length to have more in common with the rude savages,
than with the French settlers from whom they were sprung.
Many of these wilderness wanderers married Indian wives,
and, moreover, their plastic nature, a result of their
French extraction, helped them quickly to assume the
manners and customs of the swarthy children of the
forest. The voyageurs, like the coureurs des bois, were
accustomed to take long canoe voyages, under the employ
of some fur company, or even of private individuals ; sometimes alone, but more often several banded together, carrying loads of ammunition, provisions, and tobacco from the
settlements and returning with their canoes laden down
with beaver, marten, and other furs collected among the
Indians. The vast domain of Canada is so completely
watered by a network of large streams, rivers, and lakes,
more or less connected, that it is not difficult to make
canoe voyages in almost any direction throughout the
length and breadth of this great territory. It is indeed
possible to start from Montreal and journey by water to The Canadian Rockies.
Hudson   Bay, the  Arctic   Ocean,   or   the base  of the
Rocky Mountains.
The voyageurs were a hardy race, possessed of incredible physical strength and untiring patience, remarkable for
an implicit obedience to their superiors, and endowed with
a happy, careless nature, regardless of the morrow, so long
as they were well-off to-day. While making their long
and arduous journeys, the voyageurs would arouse their
flagging spirits with merriment and laughter, or awaken
echoes from the wooded shores and rocky cliffs along the
rivers and lakes, by their characteristic songs, to the accompaniment of the ceaseless and rhythmic movement of their .
paddles.
How much of romance and poetry filled up the measure of their simple lives ! Nature in all its beauty and
grandeur was ever around them, and nature's people—the
Indians—were those with whom they most associated.
They loved all men, and all men loved them, whether
civilized or barbarian. The stranger among them was
called Cousin, or Brother, and the great fur barons, the
partners in the fur companies, on whom they gazed with
awe and admiration, as they travelled in regal state from
post to post, and to whom they bore almost the relation of
serf to feudal lord, they called by their Christian names.
The melodies which they chanted in unison as they glided
along quiet rivers, with banks of changing outlines and
constant variety of forest beauty, would hardly cease as
they dashed madly down some roaring, snow-white rapid
beset with dangerous rocks, where a single false stroke
would be fatal.    For many days continuously they were mmmmmmm
glgggggggggg£g£ll
Perils of the Canoe Voyages.
223
wont to travel, with short time for sleep, working hour
after hour at the paddle, or making the toilsome portages,
when they were accustomed to carry on their backs loads
of almost incredible weight. Nevertheless, on any opportunity for relaxation, they were ever ready for revelry,
music, and the dance, which they would prolong throughout the night.
The usual dress of the voyageur consisted of a coat or
capote cut from a blanket, a cotton shirt, moccasins, and
leather or cloth trousers, held in place by a belt of colored worsted. A hunting knife and tobacco-pouch, the
latter a most indispensable adjunct to the happiness of
the voyageur, were suspended from his belt. Sometimes
they would be absent from the settlements twelve or
fifteen months, and many never returned from their perilous trips. Some were drowned while attempting to run
dangerous rapids. Others were overtaken by the approach of winter, or were stopped by ice-bound rivers
impossible to navigate, and perished miserably from exposure and starvation.
Those who returned, however, would be amply rewarded by the wealth suddenly acquired from the result
of their long toil. The dissipation of their gains in the
course of a few weeks, accompanied by all manner of
revelry, licentiousness, and mad extravagance, was their
compensation for long periods of privation. At length,
their means being exhausted, a longing for the old manner of life returned, and with renewed hopes they would
recommence their long journeys into the wilderness.
The value of the fur trade soon aroused the attention 224 The Canadian Rockies.
of a number of wealthy and influential traders, and in
1670 a charter was granted to Prince Rupert and a company of fourteen others^ to I the sole trade and commerce" throughout all the regions watered by streams
flowing into Hudson or James Bay. This region was
henceforth known as Rupert's Land. In addition to the
right of trade, the Hudson Bay Company had the authority of government and the dispensation of justice throughout this vast territory.
During the winter of 1783-4, however, a number of
Canadian merchants, previously engaged in the fur trade,
joined their several interests, and formed a coalition which
assumed the name of the Northwest Company.
This organization, governed, as it was, by different
principles from that of the Hudson Bay Company, soon
became a powerful rival. The younger men in the Northwest Company were fired with ambition and assured of an
adequate reward for their services. While for many years
their older rivals had slumbered, content with the limits
of their territory, the more enterprising Northwest Company, with infinite toil and danger, extended their posts
throughout the interior and western parts of Canada, and
opened up a new and hitherto undeveloped country.
Another great advantage that the Northwest Company
had over the Hudson Bay Company resulted from their
employment of the suave and plastic voyageurs, in whose
blood the French quality of ready adaptability to surroundings was especially well shown in their dealings with
the Indians, with whom they had the greatest influence. Rivalry of Hudson Bay Company.
On the other hand, the greater part of the Hudson
Bay canoe men were imported from the Orkney Islands.
What with their obstinate, unbending nature, and mental
sluggishness, these men presented a most unfavorable
contrast to the genial voyageurs.
The establishment of the Northwest Company aroused
the utmost jealousy and animosity of the Hudson Bay
Company. While the various parties were engaged in
dealings with the Indians, there not infrequently occurred open conflicts, bloodshed, and murder among the
agents, in their attempts to outwit and circumvent one
another.
At length the partners of the Northwest Company in
the interior of Canada, realizing that all the profits were
more than balanced by their endless and painful contest,
determined to open a negotiation with their rivals, and
for this purpose sent two delegates to London with full
authority to close whatever agreement would be for the
best intefests of the company. Just at this time the
directors of the two companies were about to sign a contract most favorable to the Northwest Company. Unfortunately, on the eve of this event, the two delegates
from Canada made their appearance, and instead of communicating at once with their own directors, they showed
their papers to the officers of the Hudson Bay Company.
The Hudson Bay Company took advantage of the opportunity, and, instead of receiving terms from the other, now
proceeded to dictate them. The outcome of this unfortunate   manœuvre was,   that the  Northwest  Company 226 The Canadian Rockies.
became merged in that, of the Hudson Bay Company,
together with the privileges and trade of all of the vast
territory which the Northwest Company had developed
by superior enterprise. Thus, in 1821, the Northwest
Company  ended its career.
The Hudson Bay Company's territory was at length,
from time to time, encroached upon as the colonies
of British Columbia, Vancouver's Island, and Manitoba were established. Finally, in 1869, the Company
ceded all their governmental and territorial rights to the
Dominion, receiving ,£300,000 in compensation. Their
forts or posts, together with a small amount of land in
the immediate vicinity, were reserved by them. The
Hudson Bay Company still exists as a commercial organization, carrying on a thriving business in many of the
principal cities and towns of Canada.
So much by way of introduction to the exploration of
the Canadian Rockies.
Let us now turn to Sir Alexander Mackenzie, the
hardy explorer who first crossed the continent of North
America, after penetrating the grim and inhospitable array
of mountains which had hitherto presented an impassable
barrier to all further westward progress..
Mackenzie was born in the northern part of Scotland,
in the picturesque and historic town of Inverness. The
year of his birth is usually set down as 1755. In his .
youth he emigrated to Canada, and found employment
as a clerk to one of the partners in the great Northwest
Fur Company.    Later on he went to Fort Chipewyan, mm£mmm§^^^&mm$mmmm^mg
Mackenzie's Character.
227
on Lake Athabasca, and became one of the principal partners in the Northwest Company.
Mackenzie was endowed by nature with a powerful
physique and a strong constitution, which enabled him
to undergo the unusual hardships of his explorations in
the wilderness. Beside these physical qualifications, he
was inspired with the ambition necessary to the formation of great plans, and with an enterprising spirit which
impelled him to carry them through to a successful termination. Great versatility of idea enabled him to
oppose every novel and sudden danger with new plans,
while a rugged perseverance, indomitable patience, and
a boldness often bordering on recklessness, carried him
through all manner of physical and material obstacles.
In his dealings with the Indians and his own followers,
he showed an unusual tact, a quality which more than
any other contributed to his success. Nothing so quickly
saps the strength and tries the courage of the explorer,
be he ever so bold and persevering, as cowardice and
unwillingness among his followers.
Nevertheless, Mackenzie was not a scientific explorer.
Outside of the manners and customs of the various tribes
with which he came in contact, only the most patent and
striking phenomena of the great nature-world impressed
him. No better idea of his views on this subject could
be obtained than from a passage in the preface to his
Voyages ;
" I could not stop," says Mackenzie, "to dig into the
earth, over whose surface I was compelled to pass with 228 The Canadian Rockies.
rapid steps ; nor could I turn aside to collect the plants
which nature might have scattered on the way, when my
thoughts were anxiously employed in making provision
for the day that was passing over me. I had to encounter
perils by land and perils by water ; to watch the savage
who was our guide, or to guard against those of his tribe
who might meditate our destruction. I had, also, the passions and fears of others to control and subdue. To-day,
I had to assuage the rising discontents, and on the morrow, to cheer the fainting spirits of the people who accompanied me. The toil of our navigation was incessant,
and oftentimes extreme ; and, in our progress overland,
we had no protection from the severity of the elements,
and possessed no accommodations or conveniences but
such as could be contained in the burden on our shoulders,
which aggravated the toils of our march, and added to the
wearisomeness of our way.
" Though the events which compose my journals may
have little in themselves to strike the imagination of those
who love to be astonished, or to gratify the curiosity of
such as are enamoured of romantic adventures ; nevertheless, when it is considered that I explored those waters
which had never before borne any other vessel than the
canoe of the savage ; and traversed those deserts where
an European had never before presented himself to the
eye of its swarthy natives ; when to these considerations
are added the important objects which were pursued, with
the dangers that were encountered, and the difficulties
that were surmounted to attain them, this work will,  I ggj/^n^g^^^^f^g^^ggg^^^
Previous Explorations.
229
flatter myself, be found to excite an interest and conciliate
regard in the minds of those who peruse it"
Thus Mackenzie writes in the preface to his journal.
Nevertheless, there is no evidence throughout his works
that he was learned or even interested in the sciences
of botany or geology. The scientific mind becomes
so much absorbed in the search for information, when
surrounded by the infinite variety of nature's productions, especially in regions hitherto unknown, that
mere inconvenience, physical suffering, or imminent peril
is incapable of withdrawing the attention from the
chosen objects of pursuit. Whoever reads Humboldt's
narrative of travels in the equinoctial regions of South
America, especially—that part which pertains to his voyage
on the Orinoco, will appreciate the truth of this. The
stifling, humid heat of a fever-laden atmosphere, the ever
present danger of sudden death from venomous serpents,
ferocious alligators, or the stealthy jaguar, the very air
itself darkened by innumerable swarms of mosquitoes and
stinging insects, with changing varieties appearing at
every hour of the day and night, were unable to force this
great naturalist to resign his work.
Unfortunately, the explorer and the naturalist are not
often combined in one person, notwithstanding that the fact
of being one, implies a tendency toward becoming the other.
Mackenzie mentions one or two attempts previous to
1792 to cross the Rocky Mountains. No record of these
expeditions is available, a circumstance that implies their
termination in failure or disaster. 230 The Canadian Rockies.
Up to this time the Rocky Mountains, with their awful
array of saw-edged peaks covered with a dazzling white
mantle of perpetual snow, had stood as the western limit
of overland exploration, beyond which no European had
ever passed. The Pacific Coast had already been explored
by Captain Cook in 1778, and a few years later so accurately charted by Vancouver, that his work is still
standard among navigators. The eastern border of the
Rockies was vaguely located, but between these narrow
strips there remained a vast region, four hundred miles
wide, extending to the Arctic Ocean, about which little.or
nothing was known.
As in the case of other unexplored regions, there were
vague and conflicting rumors among the Indians concerning the dangers of these upland fastnesses, accounts of
hostile tribes, men partly human, partly animal in form
and nature, and colossal beasts, endowed with fabulous
strength and agility, from which escape was next to impossible. These Indian tales, though in great part the
product of imagination or superstition, unfortunately did
but partial justice to the reality, for although the reported
dangers and terrors were mythical, there were real and
material obstacles in the form of mountain ranges bewildering in their endless extent and complexity, between
which were valleys blocked by fallen timber, and torrential
streams rendered unnavigable by roaring rapids or gloomy
canyons of awful depth. In fact, this region was one of
the most difficult to penetrate and explore that the world
could offer at that time. The Peace River.
231
Nevertheless, Mackenzie now turned his attention
toward this region, resolved to traverse and explore it till
he should reach the Pacific. Moreover, he was confident
of success, perhaps realizing his many qualifications for
such an enterprise, and certainly encouraged by the
remembrance of the difficulties he had overcome during
his former voyage, in 1789, to the mouth of that great
river which bears his name.
Leaving Fort Chipewyan on Lake Athabasca, he soon
reached that great waterway, the Peace River, and with
several canoes began to stem the moderate current of this
stream, which is at this point about one fourth of a mile
in width and quite deep.
The origin of names is always interesting, and that of
the Peace River is said to be derived from a circumstance
of Indian history. The tribe of Indians called the Knis-
teneux, who originally inhabited the Atlantic seaboard
and the St. Lawrence valley, migrated in a northwesterly
direction. In the course of this tribal movement, after
reaching the centre of the continent, they at length came
in contact with the Beaver Indians, and a neighboring
tribe called the Slaves, at a point some fifty leagues due
south from Lake Athabasca. The Knisteneux drove
these tribes from their lands, the Slave Indians moving
northward down'the Slave River to Great Slave Lake,
fromvwhich circumstance the lake derives its name. The
term Slave was not applied to indicate servitude, but by
way of reproach on their unusual barbarity and destitution.    The Beaver Indians moved in another direction, The Canadian Rockies.
more to the westward, and on the ratification of peace
between them and the Knisteneux, the Peace River was
assigned as the boundary between them.
After proceeding for three weeks up the Peace River,
Mackenzie camped for the winter at a point previously
decided on, and early in the following spring recommenced his "voyage," as these inland water journeys
are called. Mackenzie was accompanied by Alexander
Mackay, one of the officers of the Northwest Company.
The crew consisted of six Canadian voyageurs, and the
party was completed by two Indians, who, it was intended, should act as interpreters and hunters. A single
canoe, twenty-five feet long and not quite five feet in extreme breadth, served to carry the entire party, in addition to three thousand pounds of baggage and provisions.
It would be entirely aside from our purpose to narrate
in detail the many interesting adventures and narrow
escapes of the party. A single incident will serve to
throw some light on the perils and toils that were encountered. At the time of the incident in question, they
had crossed the watershed by following the south branch
of the Peace River to its source, and were now descending a mad torrent which runs westward, and is tributary
to the Fraser River, which latter Mackenzie mistook for
the Columbia.
It was on the morning of the 13th of June, and the
canoe had proceeded but a short distance, when it
struck, and, turning sidewise, broke on a stone. Mackenzie and all the men jumped into the water at once, A Marvellous Escape. 233
and endeavored to stop the canoe and turn it round. But
almost immediately she was swept into deeper water,
where it became necessary for everybody to scramble
aboard with the greatest celerity. In this uncertain contest, one of the men was left in mid-stream to effect a
passage to shore in the best way he could.
" We had hardly regained our situations," writes Mackenzie, " when we drove against a rock, which shattered
the stern of the canoe in such a manner that it held only
by the gunwales, so that the steersman could no longer
keep his place. The violence of this stroke drove us to
the opposite side of the river, which is but narrow, when
the bow met with the same fate as the stern. At this
moment the foreman seized on some branches of a small
tree, in the hope of bringing up the canoe, but such was
their elasticity that, in a manner not easily described, he
was jerked on shore in an instant, and with a degree of
violence that threatened his destruction. But we had no
time to turn from our own situation to inquire what had
befallen him ; for, in a few moments, we came across a
cascade, which broke several large holes in the bottom of
the canoe, and started all the bars, except one behind the
scooping seat If this accident, however, had not happened, the vessel must have been irretrievably overset
The wreck becoming flat on the water, we all jumped out,
while the steersman, who had been compelled to abandon
his place, and had not recovered from his fright, called
Out to his companions to save themselves. My peremptory commands superseded the effects of his fear, and 234 The Canadian Rockies.
they all held fast to the wreck ; to which fortunate resolution we owed our safety, as we should otherwise have
been dashed against the rocks by the force of the water,
or driven over the cascades. In this condition we were
forced several hundred yards, and every yard on the verge
of destruction ; but, at length, we most fortunately arrived
in shallow water and a small eddy, where we were enabled
to make a stand, from the weight of the canoe resting on
the stones, rather than from any exertions of our exhausted strength. For, though our efforts were short,
they were pushed to the utmost, as life or death depended
on them."
At this juncture, the Indians, instead of making any
effort to assist the others, sat down and shed tears, though
it is considered a mortal disgrace among Indians to weep
except when intoxicated.
On the 22d of July, after encountering countless trials
and the dangers of savage foes, no less than the obstacles
of nature, Mackenzie reached an arm of the sea in latitude
520 20' 48", where on a rocky cliff he inscribed this brief
legend in vermilion : " Alexander Mackenzie from Canada
by land, the 22d of July, one thousand seven hundred
and ninety-three."
The next day, when alone, he was nearly murdered by
a band of Indians, but escaped by his agility and by a
fortunate momentary hesitation on the part of the savages.
Mackenzie's return journey was over the same route
that he had first taken, and required but four weeks to
traverse the mountains. Perils of the Wilderness. 235
In reading a detailed account of this voyage, one is
impressed with the many perils encountered, no less than
the ofttimes remarkable and fortunate escapes from them.
It is so with the journals of nearly all great travellers.
They recount an endless succession of dangers and adventures by sea and land, from which, though often in the
very jaws of death by reason of the operations of nature
and the elements, the- traveller ever eventually escapes,
apparently in defiance of the laws of chance and probability. But we must bear in mind the great host of travellers who have never returned, and whose unfinished^
journals are lost forever to mankind.
The remotest corners of the earth have been mute
witnesses to these tragedies. The inhospitable, rock-
bound shores of lonely islands, or low-lying sands of coral
reefs, where the ceaseless ocean billows thunder in everlasting surf, have beheld the expiring struggles of many a
bold navigator. The colossal bergs and crushing ice of
polar seas ; hurricanes and typhoons in tropic latitudes ;
the horrors of fire at sea ; the broad wastes of continents ;
trackless desert sands, where, under a scorching sun,
objects on the distant horizon dance in the waving air,
and portray mirage pictures of lakes and streams to the
thirsty traveller ; deep, cool forests bewildering in the
endless maze of trees ; piercing winter storms, with cutting winds and driving snows ; the blood-thirsty pack of
famishing wolves ; rivers, dangerous to navigate, with impetuous current swirling and roaring in fearful rapids,—
all these have their records of death and disaster. 236
The Canadian Rockies.
But of them all, man has ever been the worst
destroyer. The hostile savage, the mutinous crew, or
treacherous guide have proved far more cruel, revengeful, and cunningly destructive than the catastrophes of
nature, whose mute, dead forces act out their laws in
accordance with the great plan of the universe, unguided
by motives of hate, and envy, and the wicked devices of
human passions. HffHimtt~r        " ''\J£fffifiîtt^SJ§}*&
CHAPTER XIV.
HISTORICAL.
Captain Cook's Explorations—The American Fur Company—First
Exploration of the Fraser River—Expedition of Ross Cox—Cannibalism
—Simplicity of a Voyageur—Sir George Simpson's Journey—Discovery
of Gold in 1858— The Palliser Expedition—Dr. Hector's Adventures—
Milton and Clieadle—Growth of the Dominion—Railroad Surveys—
Construction of the Railroad—Historical Periods—Future Popularity
cf the Canadian Rockies.
THE early explorations of Captain Cook had an
almost immediate effect on the development of
the fur trade. Upon the publication of that
wonderful book, Cooks Voyages round the World, wherein were shown the great value and quantity of furs
obtainable along the northwest coast of America, a considerable number of ships were fitted out for the purpose
of carrying on this trade. Three years after, or in 1792,
there were twenty American vessels along the Pacific
Coast, from California northward to Alaska, collecting
furs, especially that of the sea otter, from the natives.
Of these " canoes, large as islands, and filled with white
men," Mackenzie had heard many times from the natives
met with on his overland journey across the Rocky Mountains.     Mackenzie's journal was not published  till   1801.
237 j&êt
238 The Canadian Rockies.
In this book, however, he outlines a plan to perfect a well
regulated trade by means of an overland route, with posts
at intervals along the line, and a well established terminus
on the Pacific Coast. Should this plan be carried out, he
predicted that the Canadians would obtain control of the
fur trade of the entire northern part of North America,
and that the Americans would be compelled to relinquish
their irregular trade.
While the agents of the American Fur Company, a
rival organization controlled and managed by Mr. John
Jacob Astor, were preparing to extend their limits northwards from their headquarters at the mouth of the Columbia, the Northwest Company was pushing southward
through British Columbia, and had already established a
colony called New Caledonia near the headquarters of
the Fraser River. Thus Mr. Astor's scheme of gaining
control of the head waters of the Columbia River was
anticipated. The war of 1812 completely frustrated his
plans, when the post of Astoria fell temporarily into the
hands of the English.
A very good idea of the hardships of life at one of
these western posts, together with a brief account of the
first exploration of the Fraser River, may be obtained
from a letter written in 1809 by Jules Quesnel to a friend
in Montreal. The letter is dated New Caledonia, May
1st, 1809, and after a few remarks on other matters, Mr.
Quesnel goes on to say : " There are places in the north
where, notwithstanding the disadvantages of the country
in general, it is possible sometimes to enjoy one's self ;
but here nothing is to be found but hardship and loneli-
<JA^ mmmmimmMMÊmmmMMmmmm
Exploration of the Fraser River. 239
ness. Far away from every one, we do not have the
pleasure of getting news from the other places. We live
entirely upon salmon dried in the sun by the Indians, who
also use the same food, for there are no animals, and we
would often be without shoes did we not procure leather
from the Peace River.
" I must now tell you that I went exploring this summer writh Messrs. Simon Fraser and John Stuart, whom
you have met, I believe. We were accompanied by twelve
men, and with three canoes went down the river, that
until now was thought to be the Columbia. Soon finding
the river unnavigable, we left our canoes and continued
on foot through awful mountains, which we never could
have passed had we not been helped by the Indians, who
received us well. After having passed all those bad
places, not without much hardship, as you may imagine,
we found the river once more navigable, and got into
wooden canoes and continued our journey more comfortably as far as the mouth of this river in the Pacific Ocean.
Once there, as we prepared to go farther, the Indians of
that place, who were numerous, opposed our passage, and
we were very fortunate in being able to withdraw without
being in the necessity of killing or being killed. We were
well received by all the other Indians on our way back,
and we all reached our New Caledonia in good health.
The mouth of this river is in latitude 490, nearly 30 north
of the real Columbia. This trip procured no advantage to
the company, and will never be of any, as the river is not
navigable. But our aim in making the trip was attained,
so that we cannot blame ourselves in any manner."
I f
240 The Canadian Rockies.
This letter throws some light on the history of this
period, and shows whence the names of certain rivers and
lakes of British Columbia were derived. It would be in
place here to say that when Mackenzie first came to the
Fraser River, after crossing the watershed from the Peace
River, he entertained the idea that he was on the Columbia.
A few years later, the agents of the fur companies
had established certain routes and passages across the
mountains, which they were accustomed to follow more
or less regularly in their annual or semi-annual journeys.
One of the largest of these early parties to traverse the
Rockies was under the management of Mr. Ross Cox,
who was returning from Astoria in the year 1817. There
were, in all, eighty-six persons in his party, representing
many nationalities outside of the various Indians and
some Sandwich Islanders.
A striking incident in connection with this expedition
illustrates the hazard and danger which at all times attended these journeys through the wilderness. The party
had pursued their way up the Columbia River, and were
now on the point of leaving their canoes and proceeding on
foot up the course of the Canoe River, a stream that flows
southward and enters the Columbia not far from the
Athabasca Pass. The indescribable toil of their passage
up the Columbia, and the many laborious portages, had
sapped the strength of the men and rendered some of
them wellnigh helpless. Under these circumstances, it
seemed best that some of the weakest should not attempt
to pursue their journey farther, but should return down
<V «§s»iiÉœ§i
^^H
ÎÊÊÊ
'mmmmëm-ù
A Terrible Adventure.
241
the Columbia. There were seven in this party, of whom
only two were able to work, but it was hoped that the
favorable current would carry them rapidly towards Spokane, where there was a post established. An air of foreboding and melancholy settled upon some of those who
were about to depart, and some prophesied that they
would never again see Canada, a prediction that proved
only too true. In Ross Cox's Adventures on the Columbia River the record of their disastrous return is thus
vividly related :
" On leaving the Rocky Mountains, they drove rapidly down the current until they arrived at the Upper
Dalles, or narrows, where they were obliged to disembark. A cod-line was made fast to the stern of the
canoe, while two men with poles preceded it along the
banks to keep it from striking against the rocks. It had
not descended more than half the distance, when it was
caught in a strong whirlpool, and the line snapped. The
canoe for a moment disappeared in the vortex, on emerging from which it was carried by the irresistible force of
the current to the opposite side, and dashed to pieces
against the rocks. They had not had the prudence to
take out either their blankets or a small quantity of provisions, which were, of course, all lost. Here, then, the
poor fellows found themselves, deprived of all the necessaries of life, and at a period of the year in which it was
impossible to procure any wild fruit or roots. To return
to the mountains was impossible, and their only chance
of preservation was to proceed downwards, and to keep 242 The Canadian Rockies.
as near the banks of the river as circumstances would
permit. The continual rising of the water had completely inundated the beach, in consequence of which
they were compelled to force their way through an almost
impervious forest, the ground of which was covered with
a strong growth of prickly underwood. Their only nourishment was water, owing to which, and their weakness
from fatigue and ill-health, their progress was necessarily
slow. On the third day poor Maçon died, and his surviving comrades, though unconscious how soon they
might be called to follow him, determined to keep off the
fatal moment as long as possible. They therefore divided
his remains in equal parts between them, on which they
subsisted for some days. From the swollen state of their
feet their daily progress did not exceed two or three
miles. Holmes, the tailor, shortly followed Maçon, and
they continued for some time longer to sustain life on his
emaciated body. It would be a painful repetition to
detail the individual death of each man. Suffice it to
say that, in a little time, of the seven men, two only,
named La Pierre and Dubois, remained alive. La Pierre
was subsequently found on the borders of the upper lake
of the Columbia by two Indians who were coasting it in
a canoe. They took him on board, and brought him to
the Kettle Falls, whence he was conducted to Spokane House."
" He stated that after the death of the fifth man of the
party, Dubois and he continued for some days at the spot
where he had ended his sufferings, and, on quitting it, Cannibalism.
243
they loaded themselves with as much of his flesh as they
could carry ; that with this they succeeded in reaching
the upper lake, round the shores of which they wandered
for some time in vain, in search of Indians ; that their
horrid food at length became exhausted, and they were
again reduced to the prospect of starvation ; that on the
second night after their last meal, he (La Pierre) observed something suspicious in the conduct of Dubois,
which induced him to be on his guard ; and that shortly
after they had lain down for the night, and while he
feigned sleep, he observed Dubois cautiously opening his
clasp knife, with which he sprang on him, and inflicted
on his hand the blow that was evidently intended for his
neck. A silent and desperate conflict followed, in which,
after severe struggling, La Pierre succeeded in wresting
the knife from his antagonist, and, having no other resource left, he was obliged in self-defence to cut Dubois's
throat ; and that a few days afterwards he was discovered
by the Indians as before mentioned. Thus far nothing
at first appeared to impugn the veracity of his statement ;
but some other natives subsequently found the remains of
two of the party near those of Dubois, mangled in such
a manner as to induce them to think that they had been
murdered ; and ,as La Pierre's story was by no means
consistent in many of its details, the proprietors judged it
advisable to transmit him to Canada for trial. Only one
Indian attended; but as the testimony against him was
merely circumstantial, and was unsupported by corroborating evidence, he was acquitted." g&w
244
The Canadian Rockies.
Meanwhile the greater part of this expedition continued their way through the mountains by the Athabasca
Pass. Here, when surrounded by all the glory and -
grandeur of lofty mountains clad in eternal snow and
icy glaciers, and amid the frequent crash and roar of
descending avalanches, one of the voyageurs exclaimed,
after a long period of silent wonder and admiration—g I '11
take my oath, my dear friends, that God Almighty never
made such a place."
On the summit of the Athabasca Pass they were on the
Atlantic side of the watershed, and here let us take leave
of them while they pursue their toilsome journey across
the great plains of Canada to the eastern side of the
continent.
All of these early expeditions were undertaken in the
interests of the fur trade, and carried out by the agents
of the various fur companies, except for occasional bands
of emigrants on their way to the Pacific Coast, the accounts
of whose journeys are only referred to by later writers in
a vague and uncertain manner.
The expedition in 1841 of Sir George Simpson, however, to which reference has been made in a previous
chapter, is in many respects different from all the others.
The rapidity of his movements, the great number of his
horses, and the ease and even luxury of his camp life indicate the tourist and traveller, rather than the scientist, the
hardy explorer, or the daring seeker after wealth in the
wilderness. His narrative is the first published account
of the travels of any white man in that part of the moun-
K mugTO|B^^^^^^ga^^^^^^fey^^
Discovery of Gold.
245
tains now traversed by the Canadian Pacific Road, though
he mentions a party of emigrants which immediately
preceded him in this part of his journey. The rapidity
with which Sir George Simpson was wont to travel may
be appreciated from the fact that he crossed the entire
continent of North America in its widest part, over a
route five thousand miles in length, in twelve weeks of
actual travelling. The great central plains were crossed
with carts, and the mountainous parts of the country
with horses and pack-trains.
In 1858, gold was discovered on the upper waters of
the Fraser River, and a great horde of prospectors and
miners, together with the accompanying hangers-on,
including all manner of desperate characters, came rushing
toward the gold-fields, from various parts of Canada and
the United States. This year may be considered as
marking the birth of a new enterprise and the comparative decline of the fur trade ever after.
About this time, or, more precisely, in 1857, Her
Majesty's Government set an expedition on foot, the
object of which was to examine the route of travel between eastern and western Canada, and to find out if this
route could be shortened, or in any other manner improved upon. Moreover, the expedition was to investigate the large belt of country, hitherto practically unknown,
which lies east of the Rocky Mountains and between the
United States boundary and the North Saskatchewan
River. The third object of this expedition was to find a
pass,  or passes,  available for horses across the Rocky 246 The Canadian Rockies.
Mountains south of the Athabasca Pass, but still in British
territory.
As this was an excellent opportunity for the advancement of science without involving great additional
expense, four scientists, Lieut. Blackiston, Dr. Hector,
Mr. Sullivan, and M. Bourgeau, were attached to the
expedition. The party were under the control and
management of Captain John Palliser.
The third object of this expedition is the only one
that concerns the history of explorations in the Canadian
Rockies. In their search for passes, Captain Palliser and
Dr. Hector met with many interesting adventures, of
which it is, of course, impossible to give more than the
merest outline, as the detailed account of their journeys
fills several large volumes. In August, 1858, Captain
Palliser entered the mountains by following the Bow
River, or south branch of the Saskatchewan. He then
followed a river which comes in from the south, and
which he named the Kananaskis, after an Indian, concerning whom there is a legend of his wonderful recovery from
the blow of an axe, which merely stunned instead of
killing him outright.
When they approached the summit of the pass, a lake
about four miles long was discovered, round the borders
of which they had the utmost difficulty in pursuing their
way on account of the burnt timber, in which the horses
floundered about desperately. One of the animals, wiser
than his generation, plunged into the lake before he
could be caught and proceeded to swim across.     Unfor- The Palliser Expedition.
247
tunately this animal was packed with their only luxuries,
their tea, sugar, and blankets.
On the very summit of the pass is a small lake some
half an acre in extent, which overflows toward the Pacific,
and such was the disposition of the drainage at this point
that while their tea-kettle was supplied from the lake,
their elk meat was boiling in water from the sources of
the Saskatchewan.
A few days later, Captain Palliser made a lone mountain ascent near one of the Columbia lakes, but was
caught by night in a fearful thunder-storm so that he
could not reach camp till next day. His descent through
the forests was aided by the frequent and brilliant flashes
of lightning.
A little later they met with a large band of Kootanie
Indians, who, though very destitute and miserable in
every other way, were very rich in horses. Captain Palliser exchanged his jaded nags for others in better condition, and despairing of pursuing his way farther, as the
Indians were at war and would not act as guides, he
started, on the first of September, to return across the
mountains, and reached Edmonton in three weeks.
In the meantime Dr. Hector made a branch expedition which has some incidents of interest in connection
with it. He was accompanied at first by the indefatigable botanist, M. Bourgeau, and by three Red River men,
besides a Stoney Indian, who acted as guide and hunter
for the party. Eight horses sufficed to carry their instruments and necessary baggage, as it was not considered 248 The Canadian Rockies.
necessary to take much provision in those parts of the
mountains which he intended to visit.
Some reference has already been made to Dr. Hector's experiences in .the vicinity of Banff, and we shall
only give one or two of the more interesting details of
his later travels. He left the Bow River at the Little
Vermilion Creek, and followed this stream over the Vermilion Pass. The name of this pass is derived from the
Vermilion Plain, a place where the ferruginous shales
have washed down and formed a yellow ochre. This
material the Indians subject to fire, and thus convert it
into a red pigment, or vermilion.
Perhaps the most interesting detail of Dr. Hector's
trip is that which occurred on the Beaverfoot River, at
its junction with the Kicking Horse River. The party
had reached the place by following down the Vermilion
River till it joins the Kootanie, thence up the Kootanie
to its source, and down the Beaverfoot. Here, at a place
about three miles from where the little railroad station
known as Leanchoil now stands, Dr. Hector met with
an accident which gave the name to the Kicking Horse
River and Pass. A few yards below the place, where the
Beaverfoot River joins the Kicking Horse, there is a fine
waterfall about forty feet high, and just above this, one of
Hector's horses plunged, into the stream to escape the
fallen timber. They had great difficulty in getting the
animal out of the water, as the banks were very steep.
Meanwhile, Hector's own horse strayed off, and in attempting to catch it the horse kicked him in the chest, The Kicking; Horse River.
249
fortunately when so near that he did not receive the full
force of the blow. Nevertheless, the kick knocked Hector down and
rendered him
senseless for
some time.
This was the
more unfortunate, as they
were out of
food, and had
seen no sign
of game in the    ffSr
vicinity. His
men ever after
called the river
the Kicking
Horse, a name
that has remained to this
day despite its
lack of euphony.
To the
transcontinental traveller,
one of the most beautiful and inspiring points along the entire railroad is the descent of the Kicking Horse Pass from
the station of Hector to Field.     Here, in a distance of
FALLS OF  LEANCHOIL. 250
The Canadian Rockies.
eight miles, the track descends iooo feet, in many a curve
and changing grade, surrounded by the towering cliffs of
Mount Stephen and Cathedral Peak, while the rich forests of the valley far below are most beautiful in swelling slopes of dark green. Certainly, whoever has ridden
down this long descent at breakneck speed, on a small
hand-car, or railway velocipede, while the alternating
rock cuts, high embankments, and trestles or bridges of
dizzy height fly by in rapid succession, must feel at the
same, time a grand conception of the glories of nature
and the triumphs of man. In striking contrast to this
luxury of transportation was the old-time method of travelling through these mountains. The roaring stream
which the railroad follows and tries in vain to descend in
equally rapid slope is now one of the most attractive
features of the scenery of the pass.
When Dr. Hector first came through this pass he had
an adventure with one of his horses on this stream. They
were climbing up the rocky banks of the torrent when the
incident occurred. The horses had much difficulty in getting up, and in Hector's own words, | One, an old gray, that
was always more clumsy than the others, lost his balance
in passing along a ledge, which overhung a precipitous
slope about 150 feet in height, and down he went, luckily
catching sometimes on the trees ; at last he came to a
temporary pause by falling right on his back, the pack
acting as a fender. However, in his endeavors to get up,
he started down hill again, and at last slid on a dead tree
that stuck out at right angles to the slope, balancing him- ^^^^^^^^^^m^^^^^^Ê^mm^^^^^^^^s^
Short Rations. 251
self with his legs dangling on either side of the trunk of
the tree in a most comical manner. It was only by making a round of a mile that we succeeded in getting him
back, all battered and bruised, to the rest of the horses."
That night they camped at one of the lakes on the
summit of the pass, but were wellnigh famished. A
single grouse boiled with some ends of candles, and odd
bits of grease, served as a supper to the five hungry men.
The next day they proceeded down the east slope and
came to a river that the Indian recognized as the Bow.
About mid-day the Stoney Indian had the good fortune to
shoot a moose, the only thing that saved the life of the
old gray that had fallen down the rocky banks of the
Kicking Horse River, for he was appointed to die, and
serve as food if no game were killed that day.
Here we shall take leave of Dr. Hector and the Palliser expedition, and only briefly say that Hector followed
the Bow to its source and thence down the Little Fork to
the Saskatchewan and so out of the mountains. The next
year Dr. Hector again followed up the Bow River and
Pipestone River to the Saskatchewan, and thence over the
Howse Pass to the Columbia, where he found it impossible
to travel either west or northwest, and was forced to proceed southward to the boundary.
The main objects of the Palliser expedition were in a
great measure accomplished, though the Selkirk Range of
mountains was not penetrated by them, and no passes
discovered through this formidable barrier. The vast
amount of useful scientific material collected by the mem- 252 The Canadian Rockies.
bers of this expedition was published in London by the
British Government, but it is now, unfortunately, so
rare as to be practically inaccessible to the general
reader.
The account of an expedition across the Rockies in
1862, by Viscount Milton and Dr. Cheadle, is perhaps the
most interesting yet published. It abounds in thrilling
details of unusual adventures, and no one who has read
The Northwest Passage by Land will ever forget the
discovery of the headless Indian when they were on the
point of starvation in the valley of the North Thompson,
or the various interesting details of their perseverance and
final escape where others had perished most miserably.
The object of this expedition was to discover the most
direct route through British territory to the gold mines of
the Caribou region, and to explore the unknown regions
in the vicinity of the north branch of the Thompson
River.
A period of very rapid growth in the Dominion of
Canada now follows close upon the date of this expedition. In 1867, the colony of Canada, together with New
Brunswick and Nova Scotia, united to form the new
Dominion of Canada, and, in 1869, the Hudson Bay Company sold out its rights to the central and northwestern
parts of British North America.
In the meantime the people of the United States had,
been  vigorously  carrying  on  surveys, and  preparing  to
build railroads across her vast domains, where lofty mountain passes and barren wastes of desert land intervened
■V Railroad Surveys.
253
between her rich and populous East and the thriving and
energetic West, but in Canada no line as yet connected
the provinces of the central plains with her eastern possessions, while British Columbia occupied a position of
isolation beyond the great barriers of the Rocky Mountains.
On the 20th of July, 1871, British Columbia entered
the Dominion of Canada, and on the same day the survey
parties for a transcontinental railroad started their work.
One of the conditions on which British Columbia entered
the Dominion was, that a railroad to connect her with the
east should be constructed within ten years.
More than three and one half millions of dollars were
expended in these preliminary surveys, and eleven different
lines were surveyed across the mountains before the one
finally used was selected. Nor was this vast amount of
work accomplished without toil and danger. Many lives
were lost in the course of these surveys, by forest fires,
drowning, and the various accidents in connection with
their hazardous work. Ofttimes in the gloomy gorges and
canyons, especially in the Coast Range, where the rivers
flow in deep channels hemmed in and imprisoned by precipitous walls of rock, the surveyors were compelled to
cross awful chasms by means of fallen trees,or, by drilling
holes and inserting bolts in the cliffs, to cling to the rocks
far above boiling cauldrons and seething rapids, where a
fall meant certain death. The ceaseless exertion and frequent exposure on the part of the surveyors were often
unrewarded by the discovery of favorable routes, or passes Épi?
ill !
III i
254 The Canadian Rockies.
through the mountains. The Selkirk Range proved
especially formidable, and only after two years of privation and suffering did the engineer Rogers discover, in
1883, the deep and narrow pass which now bears his name,
and by which the railway seeks a route across the crest of
this range, at the bottom of a valley more than a mile in
depth.
The romance of an eagle leading to the discovery of a
pass is connected with a much earlier date. Mr. Moberly
was in search of a pass through the Gold Range west of
the Selkirks, and one day he observed an eagle flying up
a narrow valley into the heart of these unknown mountains. He followed the direction of the eagle, and, as
though led on by some divine omen, he discovered the
only route through this range, and, in perpetuation of this
incident, the name Eagle Pass has been retained ever
since.
But all these surveys were merely preliminary to the
vast undertaking of constructing a railroad. At first, the
efforts of the government were rewarded with only partial
success, and at length, in 1880, the control and management of railroad construction was given over to an organization of private individuals. In the mountain region
there were many apparently insuperable obstacles, to
overcome which there were repeated calls for further
financial aid. However, under the able and efficient control of Sir William Van Home, the various physical difficulties were, one by one, overcome, while his indomitable
courage and  remarkable energy inspired   confidence in ^^a^Mig»rrfp^^^^m^q
Historical Periods. 255
those who were backing the undertaking financially.
Moreover, he had a thorough knowledge of railroad construction, together with unusual perseverance and resolution, combined with physical powers which enabled him
to withstand the nervous strain and worry of this gigantic
enterprise.
In short, after a total expenditure of one hundred and
forty million dollars, the Canadian Pacific Railroad, which
is acknowledged to be one of the greatest engineering
feats the world has ever seen, was completed, five years
before the stipulated time.
With the opening of the railroad came the tourists
and mountaineers, and the commencement of a new period
in the history of the Canadian Rockies.
The short period of one hundred years which nearly
covers the entire history of the Canadian Rockies may be
divided into four divisions. The first is the period of the
fur trade, which may be regarded as beginning with the
explorations of Sir Alexander Mackenzie in 1793, and
lasting till 1857.
From 1858 to 1871 might be called the gold period,
for at this time gold-washing and the activity consequent
upon this new industry were paramount.
The next interval of fifteen years might be called the
period of railroad surveys and construction,—a time of
remarkable activity and progress,—and which rationally
closes in 1886, when the first trains began to move across
the continent on the new line.
The last period is that of the tourists, and though as 256 The Canadian Rockies.
yet it is the shortest of all, it is destined without doubt
to be longer than any.
Every one of these periods may be said to have had a
certain effect on the growth and advance of this region.
The first period resulted in a greater knowledge of the
country, and the opening up of lines of travel, together
with the establishment of trading posts at certain points.
The second period brought about the construction of
wagon roads in the Fraser Canyon leading to the Caribou
mining region and to other parts of British Columbia.
These roads were the only routes by which supplies and
provisions could be carried to the mining camps. The
method of gold mining practised in British Columbia has
hitherto been mostly placer mining, or mere washing of
the gravels found in gold-bearing stream beds.
With the commencement of the railroad surveys, a
great deal of geographical information was obtained in
regard to the several ranges of the Rocky Mountain
system, and the culmination of this period was the final
establishment of a new route across the continent, and the
opening up of a vast region to the access of travellers.
Year by year there are increasing numbers of sportsmen
and lovers of wild mountain life who make camping expeditions from various points on the railroad, back into the
mountains, where they may wander in unexplored regions,
and search for game or rare bits of scenery.
The future popularity of these mountains is in some
degree indicated by the fact that those who have once
tried even a brief period of camp life among them almost Future F
opularity.
257
invariably return, year after year, to renew their experiences. The time will eventually come when the number
of tourists will warrant the support of a class of guides,
who will conduct mountaineers and sportsmen to points of
interest in the wilder parts of the mountains, while well
made roads will increase the comfort and rapidity of travel
through the forests. CHAPTER  XV.
The Pleasures of the Natural Sciences—Interior of the Earth—
Thickness of the Crust—Origin arid Cause of Mountains— Their Age and
Slow Growth—System in Mountain Arrangement—The Cordilleran System— The Canadian Rockies—Comparison with Other Mountain Regions—
Climate—Cause of Chinook Winds—Effect of High Latitude on Sun and
Moon—Principal Game Animals—Nature of the Forests—Mountain
Lakes—Camp Experiences—Effect on the Character.
THOSE who have spent a few weeks or months in a
mountain region, such as that of the Canadian
Rockies, must soon come to feel an interest in
those more striking features of the wilderness which
have been constantly revealed. The special character of
the mountains, which* have given so much pleasure ; the climate, on which, in a great measure, every action depends ;
the fauna, which adds so much of interest to the environment ; and the flora, which increases the beauty of every
scene—must all excite some degree of interest in those
who have passed a short period of time surrounded by
nature in her primeval state.
They spend their time to little advantage who do not ,
thus become interested in the wonders of nature.    A very
slight knowledge of the habits and kinds of birds and
animals, the principal characteristics of trees and plants,
258 msmmmimmmmmi^   .■■$&&&
Interior of the Earth.
259
the nature of minerals, the structure and formation of the
earth's crust, and the laws which govern the circulation of
currents in the atmosphere will, in every case, offer wide
and boundless fields of research and pleasure. The
camper, the huntsman, the explorer, and the mountaineer,
armed with such information, will be prepared to spend
the many hours of enforced idleness, which frequently
occur by reason of fickle weather or a smoky atmosphere,
in an interesting and profitable manner.
In the preceding chapters, the details of the flora and
fauna, together with digressions on other topics, have
been, from time to time, set forth in connection with
various exploring excursions.
It is the purpose of this-chapter, however, to discuss, in
a general and very brief manner, such questions as have
a special interest, and to present them in a somewhat
more systematic manner than was possible, or natural, in
connection with accounts of adventures.
To begin then with the foundation of things, the question first arises as to the origin and cause of mountains.
Astronomy teaches us that the earth is a mass of molten
or semi-viscid matter, covered with a crust which has
formed from the cooling of the exterior. As to the relative or absolute thickness of this crust, there is much diversity of opinion, but the great majority of estimates
ranges between the limits of one hundred and one thousand miles.
The general features of the earth and the formation of
mountains—subjects which lie in the province of geology 26o The Canadian Rockies.
—likewise point to a comparatively thin crust covering a
molten interior. Some geologists contend that the centre
is likewise solid, and that there is a partially molten layer
between the centre and crust. Now as the earth gradually cools by radiation, its volume diminishes, and the solid
crust not having the strength to hold up its own weight, is
forced to adapt itself to the contracting interior. The
pressure thus brought to bear on the thin 'shell causes
wrinkles or folds, so that the earth's surface is raised in
some places and depressed in others. Moreover, the strata
are folded, fractured, and thrown one over another as
they are compressed, till at length lofty mountain ranges
are formed, with all the phenomena of faults, flexures, and
the wonderful contortions of the originally horizontal
beds, that are to be observed in all mountain regions.
In some respects the mountains on the earth are comparable to the wrinkles on a drying apple, but in size, the
highest peaks of the Himalayas and Andes have been
compared more justly to the minute roughness on an egg
shell.
Thus the mountain ranges of the world which appear
so vast and lofty are exceedingly small and insignificant
as compared with the great mass of the earth. The
strength of the earth's crust seems incapable of supporting
the weight of even these relatively small masses, for the
highest peaks in the world never exceed an altitude of
five and one half miles, a height which, if represented on
a globe of ordinary size, would hardly be observable.
All the great mountain ranges of the world have been wmmm
wmë^Mm
!gM^^M^g^^i&Mf^t^»if^
Growth of Mountains.
261
raised to their present altitude since the Tertiary Age,
but, nevertheless, we must conceive of mountain growth
as a very slow and gradual process, a few feet or yards of
elevation each century. That mountain chains have been
upheaved at one or two violent convulsions of nature, is
not in accordance with reason or geological facts. Faults
are often found with a displacement of the strata through
several thousand feet, a fact that has been used to prove
a sudden catastrophe. But it should be held in mind that,
after the strata were once fractured and made to slide one
on another, the sliding would tend to be repeated at long
intervals in this same place. Even then a yielding of but
a few inches would be attended by a violent earthquake.
Beside the comparatively low altitude and very slow
growth of mountain chains, there is a system in their arrangement which adds simplicity to the study of this subject. Dana calls attention to the fact that the great
mountain chains of the earth are arranged along the borders of continents, and are proportional in height to the
size of the oceans near them. The continents of North
and South America reveal this law in a striking manner.
The stupendous chain of the Andes in South America, and
the more extensive Rocky Mountains in North America,
stand opposite to the vast Pacific Ocean, and run nearly parallel to its shores, while the lesser systems on the eastern
borders of each continent face the lesser area of the Atlantic
Ocean. Moreover, almost all mountain chains show evidence of a pushing force from the direction of the sea,
and a resisting- force from the direction of the land. Ill I
tin I  \ii
262 The Canadian Rockies.
The erosion of valleys commenced as soon as the
strata were elevated above the sea-level, and thus the
valleys of the world, being mostly those of erosio-n, are
older than the mountains themselves.
Turning now to the Rocky Mountains or the Cor-
dilleran System of North America, we observe that the
chain extends from the region of the City of Mexico to
the Arctic Ocean, and westward into the Alaskan Peninsula and the Aleutian Islands, a total distance of about
five thousand miles. The Rocky Mountain system
attains its greatest width in the latitude of Colorado,
where it extends one thousand miles from east to west.
Thence northward, the range becomes narrower toward
the International boundary. From this point the system
is only about four hundred miles in width, and the eastern
range follows a line parallel to the Pacific Coast, nearly
to the Arctic Circle.
Having thus very briefly glanced at the cause of mountain chains, the system in their arrangement, and the
area covered by the Rocky Mountains of North America,
let us turn our attention more particularly to the main
features of the chain in its extension through Canada. In
all, there are four ranges of mountains composing the
Canadian Rockies. The most easterly is the highest and
most important, and is, besides, the watershed between
the Atlantic and Pacific drainage. Next to the west lie
the Selkirk and Gold ranges, which must be grouped
together. Near the Pacific Coast is a third range called
the Coast Range, while Vancouver Island and the chain ^^^^^mmmm^^mmsmsms^^^^^mii^^^
o^immmuiQftft^ g
Altitude of the Canadian Rockies.
of islands extending north represent a fourth range of
mountains. Between the two inner of these four ranges,
there is a plateau region with an average altitude of 3500
feet.
Our attention centres with peculiar interest on the
watershed or Summit Range, as in these mountains are
found the grandest scenery and the most lofty peaks, and
they are withal the most accessible to the traveller. On
the eastern side, the Rocky Mountains rise abruptly from
the plains and reach altitudes of 9000 to 11,000 feet. The
plain is here, according to Dr. Dawson, about 4350 feet in
altitude, while on the western side of the range the altitude
of the Columbia valley is only 2450 feet, or nearly 2000
feet lower. The Summit Range is from forty to fifty miles
wide in this portion of its course, and is made up of about
five sub-ranges. The rivers and streams follow the valleys
between these ranges, and find their way out of the mountains by occasional, transverse valleys, cutting through
the ranges at right angles, so that every stream has a zigzag course.
It would lead us too far to discuss the formations represented in the strata, and it is more important to learn
the altitudes of the mountains above the valleys, and their
other physical features, since these characteristics have
a more direct bearing on the scenery and on the general
nature of the mountains. The highest peaks of the Canadian Rockies rise from 5000 to 7000 feet above the
valleys, and rarely surpass 11,000 or 12,000 feet altitude
above sea-level.    Thus they cannot compare in magnitude 264
The Canadian Rockies.
with the Himalayas, the Andes, or even the Swiss Alps.
They, however, are more accessible than the Himalayas,
are far more attractive than the Andes, and afford much
greater variety of scenery, together with more beauty of
vegetation, than the Alps. No picturesque hamlets adorn
these valleys, no herds of cattle with tinkling bells pasture on these hillsides, and no well-made roads or maps
guide the tourist to every point of interest ; but, on the
other hand, the climber may ascend mountains never
tried before, the explorer may roam in wild valleys hitherto practically unseen by white men ; and the camper
may fish or hunt where no one besides the savage Indian
has ever lowered a baited hook or joined in the stealthy
chase.
Before leaving the discussion of geology, it would be
well to call attention to the wonderful effects of ancient
glacial action, everywhere in evidence among these mountains. The countless lakes were, almost without exception,
formed in the Quaternary ice invasion. A few of the
lakes occupy rock basins, and more are dammed by old
terminal moraines, while the vast majority are held in by
ridges of drift formed underneath the glaciers where they
joined together at the confluence of valleys. Mention
has already been made of the evidence of ice action
on the summit of Tunnel Mountain, near Banff, showing that the ice was at least 1000 feet in thickness,
but on the neighboring mountains there are further evidences that the ancient glaciers flooded this valley to
a depth  of   2700 or  2800  feet.     Such  evidences   may
I Mre»raKB»ffiJB^^
Climate. 265
be traced up the valley of the Bow to its source,
where the upper surfaces of the glaciers were no less
than 8500 or 9000 feet above sea-level, though these ice
streams were about the same thickness as at Banff,
because the valleys are much higher at this point.
Throughout the eastern range, all the valleys were
flooded, while only the mountain tops rose above the
fields of ice, and the creeping glaciers moved slowly down
the valleys and discharged in a great sheet of ice upon
the plains to the east.
The climate of the Canadian Rockies is exceedingly
cold in winter and temperate in summer, but the air is at
all times so dry that changes of temperature are not felt
as in lowland regions. The rainfall in summer is light,
and rarely attended by heavy showers. The amount of
snow and rainfall varies locally in a remarkable manner, by
reason of the mountains themselves. Thus the maximum
winter depth of the snow in the Bow valley may be two or
three feet, when up in the higher regions, only five or six
miles distant, the depth will approach fifteen or twenty
feet. That mountains have a great influence on the
climate and the amount of rainfall, is universally admitted. In fact, climate and mountains are mutually
dependent one on the other. A range of mountains near
the sea coast, if the circulation of the atmosphere carries
the moist air over them, will cause a great precipitation
of rain and snow, and, vice versa, the amount of precipitation decides the erosive power of streams, and consequently, the altitude and form of the mountains. 266 The Canadian Rockies.
One of the most interesting features of the Canadian
Rockies is the Chinook wind. These peculiar winds
occur at all seasons of the year but are most noticeable
in winter. At such times, after a period of intense frost,
a wind springs up from the west, directly from the mountains, the temperature rises, and the snow disappears as if
by magic. The air is so dry that the snow and moisture
evaporate at once, leaving the ground perfectly free of
moisture, where a few hours before was a deep covering
of snow. Identical winds called Foehn winds occur in
Switzerland, and in other mountain regions of the world.
The explanation of these winds has been stated by Ferrel
and others, but it is difficult of demonstration to those
who do not understand the laws governing condensation
and evaporation of moisture in our atmosphere. Most of
these laws may be clearly illustrated by an experiment
not very difficult to perform. A stout glass cylinder,
closed at one end, is fitted with a closely fitting plunger.
Now if a tuft of cotton, moistened with ether, be placed
in the cylinder, and the plunger be suddenly and forcibly
pushed in, the cotton will take fire. The compression of
the air raises the temperature so that the cotton ignites.
The experiment might have been reversed, and the
plunger pulled suddenly outwards so as to rarefy the
enclosed air. In this case the temperature of the air
would have been much reduced, and, if there were sufficient moisture, it would condense on the sides of the
cylinder or form a cloud of vapor. These experiments
are exceedingly valuable, as they demonstrate the laws ^^^j^mmMffi^^ësMmd^^^^^^^
fëreO
Chinook Winds.
267
of temperature under changing pressure. Moreover, it
shows how cold air discharges its moisture in the form
of a mist, and thus illustrates the formation of the
clouds in the upper cold regions of our atmosphere.
Now the circulation of the air in the Canadian Rockies
is, in general, from the Pacific Ocean across the mountains in an easterly direction. It is, of course, interfered
with by the circular cyclonic storms which, from time to
time, pass over the mountains. But when one or both
causes of air motion compel the wind to blow from the
west towards the east, the moist currents are forced to
ascend and flow over the mountains. In this case the air
becomes colder as it rises, mist and clouds are formed,
and rain or snow-falls, especially on the mountains themselves. As the air descends on the eastern side it becomes
warmer in the increasing pressure, and the clouds evaporate
and disappear. Now this air is much drier than when it
left the other side of the mountains, because a great deal
of rain and snow have been precipitated from it. Moreover, the latent heat given out as the clouds form, raises
the temperature of the air above the normal temperature
of those altitudes. This air gains heat as it descends, and
is subjected to the increasing pressure of lower altitudes,
and it finally appears as a warm and very dry wind on the
east side of the mountains. Such a wind evaporates the
snow, and causes it to disappear in a remarkably rapid
manner.
The cause of Chinook winds is thus  not difficult of
explanation, if one understands the effects of atmospheric ggjyr
268 The Canadian Rockies.
pressure and condensation. The latent heat given out by
the condensing vapors and falling rain is of course equal
to the heat furnished by the sun, when it was evaporating
the surface waters of the ocean, and rendering the air full
of invisible water vapor.
The aspect of the sky and clouds is one of the most
beautiful features of the mountains. Except when obscured by the smoke of forest fires, the sky is at all
times of that deep hue rarely seen near the sea-coast
or in lowland regions. The dark blue extends without
apparent paleness to the very horizon, while the zenith
is of such a deep color, especially when seen from the summit of a lofty mountain, as to suggest the blackness
of interstellar space. Against such a background, the
brilliant cumulus clouds stand out in striking contrast,
and every internal movement of the forming or dissolving
vapors, as they rise, and descend, or curl about, is distinctly seen, because the clouds are so near.
The high latitude of this region has, of course, a considerable effect on the length of the days. Near the summer solstice the twilight is faintly visible all night, and the
sun is below the horizon only a little more than six hours.
The moon, however, is rarely visible in the summer months,
because when near the full it occupies that part of the
ecliptic opposite the sun, which, in this latitude, is much
depressed. In consequence, the full moon runs her short
arc so near the horizon that the high mountains shut out
all view of her. In winter, these conditions are reversed,
and the moon shines from the clear and frosty sky with mmmmsmë
m&
Game Animais
269
unusual brilliancy, for many hours continuously, while the
low-lying sun leaves many of the deeper mountain valleys
without the benefit of his slanting rays for several months
together.
It would be impossible to enumerate even the principal varieties of game animals, birds, and fish that inhabit
this region. The mountain goat and sheep have been
mentioned in previous chapters, and many of the interesting animals frequently met with have been described
in more or less detail. The ordinary explorer or camper
will see very little of the larger game, as he moves along
with a noisy train of pack-horses and shouting men to
drive them. He may occasionally see a bear, or catch
sight of an elk or caribou, but the wary moose and the
other members of the deer tribe will rarely or never be
seen without an organized hunt. The camper will come
to rely on the smaller game to give variety to his camp
fare. Chief among these will be the grouse, of which
there are six species in the Canadian Rockies. One
variety is tame, or rather very stupid, and may be
knocked down with stones, or snared with a strong elastic noose at the end of a pole. These birds are so numerous in the forests that one may always rely on getting a
brace for dinner, after a little search, and I have even seen
them walking about on the main street of Banff, where,
of course, they are protected by law. Most of the mountain streams abound in trout, except where a high waterfall below has intercepted their coming up the stream.
The larger lakes likewise afford fine fishing, and in many 270
The Canadian Rockies.
cases swarm with lake trout of a remarkable size. The
camper will often obtain wild fowl, the black duck, mallards, and teal, in his excursions. Outside of these
game birds and fish, there is little left for him to rely
on, unless he chooses to dine on marmots and porcupines. These are often extolled by travellers as most
excellent eating, but I have tried them both, and would
prefer to leave my share to others, while there is anything else on hand.
The vegetation of the Canadian Rockies deserves a
few remarks. The principal trees are all conifers. There
are about six or seven species of these in the eastern
range, and several more in the Selkirks. The paucity in
the variety of deciduous trees in the Rocky Mountains,
and the great number of conifers on the Pacific slope of
North America, are in striking contrast to the wonderful
number of deciduous species in the forests east of the
Mississippi River. In the latter region, the number of
species of forest trees is nowhere exceeded in the world,
outside of tropical regions. Another remarkable fact in
this connection was stated by Gray. He calls attention
to the fact that there is a greater similarity, and affinity
of species, between the Atlantic Coast trees and those of
far distant Japan, than with those of the Pacific slope.
In the Canadian Rockies, trees cease to grow at altitudes above 7500 feet, under the most favorable circumstances, and the average tree line is in reality about
7000 feet. Bushes of the heath family and Alpine plants,
however,   reach  much   higher, while  dwarfed  flowering 'wMMMnmMmmmm
WllM\WtlW8Êi
*m
Upper Bow Lake.
Looking south.  rnrp^HHi.L.L JHr'ni1' 'inTTTTiii i ihii iwl
J&mmMHu! ilisg^gffiâtgtrfmrt i mm sssaa  &smmm&
Pleasures of Camp Life. 271
herbs may be found in blossom as high as 8700 or 8800
feet. I once found a small mat of bright yellow sedums
on the summit of a mountain, 9100 feet above sea-level,
but this was an exceptional case. Above this altitude,
various stone-gray, bright yellow, or red lichens, are the
only sign of vegetable life. Nevertheless, in such cheerless regions of high altitudes, one sees a considerable
variety of insect life—butterflies, wasps, mosquitoes, and
spiders. The latter insects may sometimes be seen crawling about on the snow after winter has commenced, and
naturalists have often described them as one of the
most abundant insects on barren, volcanic islands of
the Atlantic Ocean, where there is scarcely a trace of
vegetation.
The pleasures of camping in the Canadian Rockies
are almost infinite in their variety. They vary with the
locality and the scenic interest of the surroundings, and
suffer a constant change of mood and aspect with the
changing weather. There is an exhilarating buoyancy in
the mountain air that conspires to make all things appear
as though seen through some cheerful medium, and where
nature is so lavish with countless things of rare interest
on every side, one comes at length to regard all other
places unworthy of comparison. The formation of these
mountains is such as to present an infinite variation of
outline and altitude, such as one observes in almost no
other mountain region of the world. The mountaineer may
stand on the summit of a lofty peak and behold a sea of
mountains extending fifty or one hundred miles in every 272 The Canadian Rockies.
direction, with no plains or distant ocean to suggest a limit
to their extent. Such a vast area, nearly half a thousand miles in width, and thousands of miles in length, presents an extent of mountain ranges such as are found in no
other part of the world.
The exquisite charm and beauty of the lakes, so
numerous in every part of the mountains, is one of the
chief delights of the camper. Some are small and solitary, perched in some amphitheatre far up among the
mountains, surrounded by rocky walls, and hemmed in by
great blocks of stone. Here, no trees withstand the
Alpine climate, and the water surface is free of ice only
during a short season. A few Alpine flowers and grasses
wave in the summer breezes, while the loud whistling
marmots, and the picas ever sounding their dismal notes,
live among the rocks, and find shelter in their crevices.
Other lakes, at lower altitudes, are concealed among
the dark forests, and, with deep waters, richly colored,
appear like gems in their seclusion. Here the wild
duck, the diver, and the loon resort in search of food,
for the sedgy shores abound with water rice, and the
waters with fish.
Most of the mountain lakes are small, and hide in
secluded valleys, but many are large enough to become
rough and angry in a storm, and have beaten out for themselves narrow beaches of gravel and shores lined with sand.
Even the sounds of the mountains and the forests
give constant pleasure. There is every quality and volume of sound, from the loud rumble of thunder, or the
^ fflgHSBISiggfti iH&jjtu gttjffinffl?^
Emerald Lake and Mount Field.  mstsmimtiè  mî^^&^Mffi^WffMiîM
Camp Experiences. 273
terrible crash of avalanches, re-echoed among the mountains, to the sharp, interrupted report of falling rocks, the
roar of torrents, or the gentle murmur of some purling
stream. The sighing of the wind in the forests, the
susurrant pines and spruces, the drowsy hum of insects,
the ripple of water on the shores of a lake, and the myriad
sounds of nature—half heard, half felt—conspire to make
up the sum of the camper's pleasure ; though in a manner
so vague and indescribable that they must needs be experienced to be understood.
Nor are all the experiences of camp life attended by
pure enjoyment alone. Mountain adventures comprise
a multitude of pleasures, mingled many times with disappointment and physical suffering. They comprise all the
scale of sensations, from those marked by the pains of
extreme exhaustion, physical weakness, hunger, and cold,
to those of the greatest exhilaration and pleasure. Fortunately, the sensations of pleasure are by far the more
abundant, while those of pain almost invariably follow
some rash act or error in judgment.
The effect on the health and strength is, of course,
one of the chief advantages of camp life. But there is
another beneficial result brought about by this manner
of life that is more important, though less often taken
into consideration. This is the effect that camp life has
on the character. In the first place, one learns the value
of perseverance, for without this quality nothing can be
accomplished in such a region as the Canadian Rockies.
The explorer will realize this when he comes to a long
J I
274
The Canadian Rockies.
stretch of burnt timber, where his horses flounder in a
maze of prostrate trees ; and the climber will feel the
need of continued resolution when, after a long and arduous climb to an apparent summit, he reaches it only to
find the slope extending indefinitely upwards.
The quality of patience under toil and aggravation
while on the march—patience with tired horses and weary
men—patience under the distress of wet underbrush, or
uncomfortable quarters, or, indeed, when tormented by
mosquitoes, is one of the prime requisites of life in the
wilderness.
While these qualities are more or less common to
every one, they are much developed in mountain camp
life. But, perhaps, the ability to judge quickly and well
is that characteristic which is most needed among the
mountains, and the one which is attended by the most
suffering if it is not brought into play. If the explorer
or mountaineer decides on the time of day when he must
turn back, and then, under the temptation of seeing a
little more, or of reaching another summit, delays his
return, let him not bewail his fate if he is caught by darkness in the forest and is compelled to pass a sleepless,
hungry night. The laws of nature are inexorable,, and
while we obey them there is abundant opportunity of
pleasure, but if we expose ourselves to the grinding of
her vast machinery, one must suffer the consequence.
The storm will not abate merely because we are exposed
to it, nor will our strength be renewed merely because
we are far from camp. Camp Necessaries.
Let the camper surround himself with all the luxuries
that are possible without trespassing on the bounds of
reason. Let him have a good cook and a good packer ;
horses that are used to the trail ; a fine camp outfit ; comfortable blankets and good tents ; a full supply of cooking utensils, knives, forks, and spoons ; above all, let him
take an abundant supply of provisions, comprising a large
variety of dried fruits and the various cereals, and let
each article be of the best quality.
Under such circumstances there is no risk of danger,
no opportunity for discomfort, especially if every action
is controlled by a moderate amount of judgment ; but, on
the other hand, the rich experiences among the mountains
will prove a store of physical and mental resources, the
memory of which will tempt him to revisit these regions
year after year. •&SÏ pKuaiii
&s»S3ië£&irë
225iœi23»5fc mjwwu" -■   -         - g|gaafiggt»sc«ff^^3Sffiggtgg
INDEX.
Abbott, Mount  130
Agnes, Lake  42
"   depth of  43
"        "   in winter  ri8
"        "   solitude of  42
Air circulation in Canadian Rockies  267
Alders in Selkirks  125
Alpine insects, varieties of  271
"     plants  271
American-Fur Company  238
Anemones  107
Assiniboine, another name for Stoneys  53
Assiniboine, Mount, altitude of  177
f       features of  178
"                "        first circuit of  168
"          "    view of  153
"        outline  156
"                "        south side of  174
Astley, Mr  61
Athabasca Pass  244
Atmosphere, eastward movement of  123
Avalanche from Mount Lefroy  33
Balsam fir  38
Banff, altitude  11
"    climate. 11, 15
' '    location  1
4 '    population  2
' '     Springs Hotel  3
"     surroundings  3
"     topography of  4
Barometer, diurnal minima of »  113
Bean, Mr  72
Bear's Paw, chief of Stoneys  49
Beehive, the 41, 44
"       altitude of *  44
277  W^i- 278
Index.
PAGE
Blackiston, Lieut  246
Blind valleys  172
Bourgeau, M  246
Bow Lakes  201
"       "    future popularity of  211
' '   Lake, Lower  191
"     Upper  195
' '   River  2
British Columbia  253
Brown, Mount, altitude of  184
Bull-dog flies  25
Butterflies, habits of  72
Caledonia, New  238
Calypso borealis  143
Cambrian Age, reference to  42
Canada, highest point reached in  115
Canadian National Park  1
"        Pacific Road, cost of  255
' '        Rockies, comparisons of  264
Cannibalism, anecdote of  242
Canoe River  240
Caribou mining region  256
Cascade Mountain, ascent of  14
"          description of  5
"              "          origin of name  6
Castilleias 1  107
Castle Crags      65
Cave and basin at Banff  159
Chalet at Lake Louise  22
"    old  26
Character, effect of camp life on ,  273
Cheops, Mount  130
Chiniquy, Tom  49
Chinook winds, cause of  266
Chipmunks  106
Cirque '.  77
Climate of Canadian Rockies  265
Cloud effects 29, S2
Coast Range  262
Condensation of clouds  267
Condition, physical  S9
Continental watershed 18,37
Contrast of surroundings  95-
Cold weather in September  13
Colorado, altitude of mountains in  37
Color, sunset and sunrise  30 K^^ixîoSS^So^i
«6
Index.
279
PAGE
Columbia River.  120
Columbine, yellow  ig
Cook, Captain  230
"           "     explorations of  237
Cordilleran System  262
Coureurs des bois  221
Cox, Ross  240
Crées, Mountain  52
Crevasses, dangers of  203
Cross River  171
Daly  Mountain  193
Dawson, Dr., on Stoney Indians  52
Desolation Valley  107
Devil's Club  125
' '      Head  7
Lake  6
"         "    Indian legend of  8
Diamond hitch  142
Dominion of Canada  252
Eagle Pass  254
' '      Peak  126
"        "    later attempts on  129
k Earth, interior of  259
Edith, Mount, Pass  219
Epilobium  107
Experiences in camp  273
Exploration, pleasure of 75, 96
Forbes, Mount, altitude of  184
Forest fires, ancient  188
"        "    causes of  188
"     fire smoke      11
Forests, near Lake Louise  38
"      of Pacific Coast  135
' •      regeneration of  190
"      Selkirk  125
Forest trees, replacement of  190
Fraser, Simon  239
' '      River, first exploration of  238
Fur trade, origin of  220
Glacier, House  121
"      debris  55
"      thickness of ice in  78
Glissading, method of  70
Goat, Rocky Mountain 117, 163, 164
Gold, discovery of  245 280
Index.
Golden-rod, Alpine species       72
Gold Range     262
Great Mountain 77, 80
' '    Slave Lake, origin of name     231
Green, Dr      124
Grouse     269
Hazel Peak, altitude      108
Hector, Dr 10, 246
Hector,  Mount     216
Heely's Creek     139
Hermit Range      127
Hooker, Mount, altitude of      184
Huber and Sulzer      124
Hudson Bay Company 224, 226
Ice Age.
"   pillars
, ability to follow trails.
frankness	
gratitude	
habits of	
idea concerning weeping
Kootanie	
loquacity	
pathos	
, 264
56
49
63
51
50
234
247
29
, 212
Kananaskis Pass, legend of,  246
Kicking Horse Pass, discovery of  250
"           "     River, origin of name  249
Kootanie River, direction of flow  120
Laggan, distance from Banff. '.  62
"       distance to Lake Louise  24
" "       " Mount Temple  79
Lake Louise, altitude of  22
"        depth of  17
' '        early morning at  26
''        forests about ,  23
"        highest recorded temperature at  22
"        in October  31
"        past history of forests at  24
"        prevalent wind at  24
' '        size and shape of  16 Index.
m£6
Lake Louise, summer temperature of water  26
"        "        topography of region near  36
"        "        visitors at  22
Lakes in Canadian Rockies      272
Lake trout, size of 6, 202
Laurel, sheep  19
Lefroy, Mount, avalanche from  90
"            "      description of  18
"      precipices of 34, 90
Linnea borealis  175'
Little Fork Pass, altitude of  208
Lyall's larch  39
Mackay, Alexander  232
Mackenzie, Sir Alexander  226
"         River, discovery of  231
Mackenzie's plan for an overland route  238
Maple trees  175
Marion Lake  130
Mariposa Grove  135
Marmots  43, 106
Milton and Cheadle  252
Minerals on mountain sides  173
Minnewanka Lake      8
Mirror Lake 42, 45
Moon, effect of latitude on  268
Morley  51
Mosquitoes, annual disappearance of     25, 199
Mountains, age of  261
' '        altitude of  263
' '        comparative size of  260
' '        origin and cause of  259
"        system in arrangement  261
Mountaineers, tribulations of  113
Mounted Police, Northwest  2
Murchison, Mount  207
Muskegs  46
Névé regions  56
Northwest Company... >  224
' '        downfall of  225
' '          Mounted Police  2
Outfit for camp  275
Pacific Coast reached by Mackenzie  234
Pack-horses, difficulties with  102
"        "      nature of  214
"        "      remarkable experience with  105 f4
282 Index.
PAGE
Palliser expedition  245
Paradise Valley 91, 105
"          "     discovery of  91
"          "     in winter  117
"          "     location  84
Patience, need of, in camp life  274
Peace River, origin of name  23 r
Peechee  8
Perseverance, need of, in camp life  274
Peyto, William  140
Phlox, alpine  72
Pica, tailless hare  105
Pinnacle Mountain 92, 97
Plateau region  263
Pleasure of camp life   271
Ptarmigan  76
Quesnel, Jules  238
Rat, wood  106
Rhododendron  20
Roger's Pass  254
Rundle Mountain  5
Rundle's early visit to Banff region  9
Rundle the missionary      9
Rupert, Prince  224
Saddle, the  77
Saskatchewan  206
Scenery at high altitudes 45, 89, 108
Schrunds, cause of  87
Selkirk Range, humidity of  123
Selkirks, early popularity of  124
"       forest trees  132
' '       geographical position  119
Sheep, mountain  10
Simpson Pass  144
"        River    . 145
' '        Sir George  244
Simpson's, Sir George, expedition  7
Sky, color of, in mountains  268
Smoke of forest fires  81
Snow line, determination of  56
"      "     in Colorado and the Andes  57
"    patches, effect on vegetation   67
"    storm in June  12
Solitude of high altitudes  215
Sounds, forest and mountain  273 Index.
281
Spray River  4
Spruce trees  38
Stones, loose, danger of, in Canadian Rockies 59, 69
Stoney Indians, characteristics of  51
"           "        dress of ,  52
"            "        nature of  9
"            "         Palliser's account of  53
"           "         religion of  52
Storms, approach of 27, 81
1 '      mountain  156
St. Piran, flowers and butterflies 72, 73
' '       summit of  71
Stuart, John  239
Sullivan,  Mr ,  246
Sulzer and Huber  124
Surveys for railroad  253
Temple, Mount, altitude  78
"            "         first ascent of  115
"            "         avalanches from  78
i '            "         first attempt to ascend  109
"            "         maximum temperature on summit  116
"            "         north side of  no
"            "        strata of  79
"            "         summit of  116
Thirst, method of quenching  74
Thunderstorms in mountains  28
Tourists at Banff  2
Tree line  270
Trees, age of  135
"          "at Lake Louise.     23
Tunnel Mountain  4
Twilight, length of  n
Twin, William  48
Valleys, age of  262
"      blind  212
Van Home, Sir William  254
Vegetation of Canadian Rockies  270
Vermilion Pass ,.  181
Plain  248
"          River  179
Vitality of mountain trees and herbs ,  12
Voyageurs  221
Waputehk Range  183
Wasps and bull-dog flies  25
Wildman, Enoch 48, 119
Wilson, Tom  138  »œssf*s&s*j
sslayg    ^rn?£|gHiaœiicfbr&^ fsitv of British Columbia Library
DATE DUE rtftfr&mgfett&cg^ggn
ml psoas
Rt. Wt>5
l£«î7  of the LAKE LOUISE region showing part of
the Summit Range of the Canadian  Rocky Mountain
SURVEYED   AND   DRAWN   BY
WALTER   D .WILCOX.
SCALE     1    AND    1-2    INCH   -   i    MILE
-f-i!i»
GLACIERS
LAKES
TRAILS 	
CONTINENTAL WATERSHED	
INDREW B GRAHAM  PHOTO'UTHO. WASHIN GTON. 0 C
COPYRIGHT BY G- P. PUTNAM'S SONS.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/cdm.bcbooks.1-0340553/manifest

Comment

Related Items