time, until the whole seasonal movement was in progress. Those brigades
going the greatest distance were started first. Although cargoes left
the coast from early spring on to late summer, the great canoe movement
took place towards the fall. Canoe travel north and northwestward
from the Great Lakes had to be carefully timed, as goods had to be
accumulated at the base posts on the Lakes and the brigades placed in
movement at the last safe date which would permit them to reach their
destination before the first hard freeze-up. The base posts were those
where the run of the _maître canot_ ended and that of the _canot du
nord_ began, the places where reloading for the individual trading
posts in the Northland was necessary. The late start was usually
desirable in order to await the arrival at the base posts of all the
goods required, for movements of freight were uncertain before the days
of railroads and steamers.
[Illustration: Figure 143
DECORATIONS: FUR-TRADE CANOES. (_Watercolor sketch by Adney._)]
In the late 18th and early 19th centuries, before the whole canoe trade
fell under the control of the Hudson's Bay Company, it was the custom
to distribute 8 gallons of rum to each canoe for consumption during
the run, and it was also the custom for all hands to see how much of
this they could drink before starting out. This grandiose undertaking
usually began as soon as the local priest, who gave his blessing to the
canoemen, had left the scene. The magnificent drunk lasted one day and
the next morning the crew had to be underway. The first day's run, old
accounts repeatedly show, not only was short but was often beset by
difficulties.
The era of the bark trading canoe did not close with a dramatic
change. Its ending was a long, slow process. By the last decade of
the 19th century the bark trading canoe had disappeared from most of
the old routes, and even in the Northwest it had been almost wholly
displaced by York boats, scows, bateaux, and canvas or wooden canoes
of white-man construction. By the beginning of the first World War,
the _maître canots_ and _canots du nord_ were finished, except as
curiosities--hardly even as these, for not one was preserved in a
museum.
Indeed, so complete was the disappearance of the fur-trade canoe that
any attempt to record its design, construction, and fitting would have
been almost hopeless, had it not been for the notes, sketches, and
statements of such men as L. A. Christopherson, aided by a few models
and pictures, and for the memories of a few Indian builders who had
worked on the canoes.
_Chapter Six_
NORTHWESTERN CANADA
Indians of the Northwest Territories and the Province of British
Columbia in Canada, and the States of Alaska and Washington, built bark
canoes that may be divided into three basic models.
The first may be called the "kayak" model, a flat-bottom, narrow
canoe having nearly straight flaring sides and either a chine or a
very quick turn of the bilge. These bark canoes were low-sided and
were usually partly decked. A number of tribal groups built canoes of
this model, the variation being relatively minor. The rake and form
of the ends varied somewhat as did the amount of decking; there were
also some slight variations in structure and method of construction.
While these bark canoes had some superficial resemblance in general
proportions to the Eskimo kayaks, it is necessary to point out that
they did not, particularly in Alaska, have the same hull form as the
seagoing kayaks in that area. In fact, the single-chine form of the
Alaskan version of this canoe appears only in the kayaks of northern
Greenland and Baffin Island. The Alaskan seagoing skin kayaks are all
multi-chine forms that approximate a "round-bottom" hull. It has been
thought that the flat-bottom seagoing kayak form may have existed in
the Canadian Northwest, at the mouth of the Mackenzie; a kayak so
identified is in the collections of the U.S. National Museum (see p.
202), but there is now doubt among authorities as to the correctness of
this identification. As will be shown later, it seems probable that it
has been improperly assigned to the Mackenzie delta and is, in fact, an
eastern Eskimo model.
The second model used in the Northwest area was a narrow-bottom
flaring-sided bark canoe with elevated ends, having, perhaps, a faint
resemblance to the Algonkin-Cree canoes of the old type. Here too
there was some variation among the canoes of tribal groups, mostly
in the shape and construction of the ends and in the fitting of the
gunwales. Most of the canoes of this type had stem-pieces formed of a
plank-on-edge, but in a few examples the stem-pieces were bent. This
model was built by the same tribal groups in Canada that built the
kayak form, the explanation being that the kayak form was the hunting
while the second model was commonly the family or cargo canoe. In
Alaska, however, only the kayak-form was used and the family, or cargo,
canoe was merely an enlargement of it.
The third model may be called the "sturgeon-nose" type; in this the
ends were formed with a long, pointed "ram" carried well outboard
below the waterline as an extension of the bottom line of the canoe.
Primitive in both model and construction, it was built in a rather
limited area in British Columbia and in the State of Washington. The
last canoes built on this form were canvas-covered; in earlier times
spruce or pine bark was usually employed.
The birch in most of the Northwest is a small tree and the bark is of
poor quality for canoe building; hence, in many areas spruce bark was
commonly employed in its place; a single tribal group might build its
canoes of either, depending upon what was available near the building
site. However, near the Alaska coast, where kayak-form bark canoes were
used and good birch was usually not available, some tribes used seal or
other skins as a substitute. In the framework spruce and fir were most
commonly employed, but occasionally cedar was available and was used.
The canoe-building Indians in northwestern Canada were mostly of the
Athabascan family and included the Chipewyan or "Chipewans," the Slave
or "Slavey" (= Etchareottine), the Beaver (= Tsattine), the Dogrib (=
Thlingchadinne), the Tanana (= Tenankutchin), the Loucheux, the Hare (=
Kawchodinne), and others. Some of these tribal groups built not only
bark canoes but also dugouts. There were also some Eskimo people who
built bark canoes for river service, as well as skin canoes, on the
same model as the bark kayak-form.
In the vicinities of Lake Athabasca and Great Slave Lake, the Chipewyan
employed not only their own models of canoes but also that of the
western Cree. The latter had invaded Chipewyan territory before the
arrival of the first white men in the Northwest and undoubtedly had
influenced canoe-building technique during the long period of the fur
trade that followed. It is therefore not possible to say where the
influence of Chipewyan building techniques ends and that of the Cree
and the eastern Indians, as introduced through the fur-trade canoes,
begins. This raises the question whether the high-ended Athabascan
canoe is itself the result of influence. One may infer from Samuel
Hearne's description of his travels in this area, in his _Journey ...
to the Northern Ocean_,[1] that only the kayak-form then existed, for
this type is the only one he describes, and he describes it in great
detail. However, Alexander Mackenzie, in an entry in his journal for
June 23, 1789, refers to the "large canoe" in a manner indicating
that it was a local type. It may well be that then, as later, the
kayak-form and cargo canoe existed side by side, or it may be that
Mackenzie was referring to a large kayak-form canoe like the family
canoe of the Alaska Yukon Indians. Perhaps the reason that Hearne did
not mention the "large canoe" is that the people he met on his way to
the Coppermine River, and on his way back by way of Lake Athabasca to
Hudson Bay, did not then use canoes of the second model.
[1] See bibliography.
[Illustration: Figure 144
CHIPEWYAN 2-FATHOM hunter's canoe (top), with bent stem piece, and
Athabascan 2½-fathom canoe with plank stem piece. Plank and bent stem
pieces were both employed in Athabascan canoes. Spruce or birch bark
were used without alteration of the design or basic construction
methods.]
_Narrow-Bottom Canoe_
Because the variations in the second model, the Algonkin-Ojibway type,
are relatively slight, it will be easiest to describe this first. The
canoe is known to have been built extensively by the Chipewyan, Dogrib,
and Slave. The sizes most common were 16 to 22 feet over the gunwales,
with a beam of between 36 and 48 inches. The sheer was usually rather
straight, the sharp upward turn to the end taking place very close
to the gunwale ends. Most of the bottom was straight; the rocker, if
existing, occurred close to the ends of the canoe and was moderate.
The midsection was dish-shaped and nearly flat across the bottom, with
a rather slack, well-rounded bilge and almost straight flaring sides,
the amount of flare being usually great. The bottom apparently was
never dead flat athwartships, for in all known examples it was somewhat
rounded. Near the ends the sections were in the shape of a ~V~ with
apex rounded; the form of the ends was sharp and without hollow either
at the gunwale or at the level lines. The ends of the canoes were never
lofty and many had end profiles that were very long fore-and-aft and
showed a marked angularity. Inwales and outwales formed the gunwale
structure; some canoes also had gunwale caps which stopped well short
of the end profiles. The ends of the inwales were carried to the
stem-pieces; they were sharply tapered and curved to sheer, and were
elaborately cross-wrapped to secure them there. The end profiles were
formed of a thin plank-on-edge in most canoes, but some had stem-pieces
split into laminae in the usual fashion and bent. In all cases
headboards were employed; the heads were forced under the inwale ends
and against the inside face of the stem-piece. The gunwale lashings
were in groups, although some canoes exist in which the outwale was
omitted and the lashing was continuous; these canoes usually had
laminated bent stem-pieces and their stem lashing was identical with
that of the Algonkin-Ojibway fur-trade canoes. This departure, it
is reasonable to assume, was the result of outside influence on the
Athabascan technique. When the stem-piece was of thin plank, the bark
was usually fastened to it by multiple turns of two thongs passed, one
from each side, through the bark and through holes bored in the stem.
[Illustration: Figure 145
ATHABASCAN CARGO OR FAMILY CANOES WITH BENT STEM PIECES, Chipewyan
2½-fathom (top) and Dogrib 3-fathom. These canoes were covered with
spruce or birch bark.]
The end profile varied with the tribe of the builder. Chipewyan canoes
had a very long end profile fore-and-aft; the heel was angular, and the
outline of the stem then swept forward in an easy curve to a height
about two-thirds the depth of the canoe amidships, then began to tumble
in a little, the curve becoming gradually sharper until the head was
reached. The stem-head in its fore-and-aft length was almost one third
the height of the ends and was roughly parallel to the bottom of the
canoe directly beneath it. Because of the rocker of the bottom, the
after end of the head was thus lower than the fore end. The sheer
was faired up to the after end of the head in a short, quick curve.
Usually the outwales were cut off short of this point, but in some
canoes they were brought up along with the inwales to the stem-head.
Wedges were used in making up the gunwale-end lashings in both the
Chipewyan and Dogrib canoes; these served to tighten the lashings and
formed a sort of breasthook. In a few examples of the Athabascan type,
the stem-pieces were of cedar root without lamination; this use of
the roots enabled the angular form of the plank-on-edge stems to be
retained. It cannot be determined whether the root stem-pieces were
part of the old Athabascan technique or were an importation from the
western Cree. The lashing in these canoes followed the forms used
in the fur-trade canoes--long-and-short turns in groups generally
triangular in shape, with a spiral turn between groups.
[Illustration: Figure 146
PLANK-STEM CANOES OF HYBRID FORMS, 3-FATHOM Slavey (top) and 2½-fathom
Algonkin-type Athabascan, probably the results of the influence of
fur-trade canoe-building.]
The canoes of the Dogrib were practically identical with those of the
Chipewyan except that the end profiles were usually slightly deeper
fore-and-aft; also the Dogrib canoes were perhaps more often of birch
bark, judging from the remaining canoes and models. The form of the
ends in the Dogrib canoes was such that they often appeared higher than
they really were, as the stem-heads stood some distance above the ends
of the sheer, an effect which was heightened by the small fore-and-aft
depth of the stem-heads.
The large canoes of the Slave had the same hull characteristics as
the others but differed in end profiles and did not have rail caps.
In the Slave canoe, the ends were formed of thin plank and in profile
were almost upright and slightly curved. The stem line came out from
the bottom in a sharp, almost angular curve and ascended with a slight
sweep to a point about level with the gunwale amidships (in some, to
within a few inches of the stem-head); from there a tumble-home carried
it to the stem-head, which was short fore-and-aft and slightly crowned,
the inboard end dropping vertically downward inside the gunwales. The
headboards were under the gunwale ends. Inwales and outwales were both
carried to the stems but the end lashings were quite short. There were
no rail caps. The bark cover was lashed to the stem with an in-and-out
stitch from side to side through holes in the plank. The sheer was
brought up nearly to the top of the stem in a rather long, easy sweep
beginning inboard at the endmost thwart.
The gunwale members in all these Athabascan canoes were quite light
compared with their Eastern counterparts. A reinforcing strip of bark
was placed under the outwales so as to hang down below them some four
inches or so amidships and less toward the ends; this was sometimes
decorated with a painted zigzag stripe or with widely spaced circles.
The end lashings of the gunwales were protected by short bark deck
pieces inserted under the caps. The edges of these deck pieces
were trimmed flush with the outboard edges of the caps, so that no
_wulegessis_ resulted.
In spruce-bark canoes, because the bark was stiff the ribs were spaced
6 to 8 inches, whereas in birch-bark canoes the ribs were spaced about
as usual, 1 to 2 inches edge to edge. In the Dogrib and Slave canoes
the ribs were without taper; in the Chipewyan there was usually a
slight taper from the bottom to the gunwale end. The ends of the ribs
were forced under the gunwales in the usual manner employed in the
east, the gunwales being rectangular in cross-section, with the lower
outboard corner beveled.
The thwarts were all parallel-sided, but tapered toward the ends, in
elevation. The thwart ends were tenoned into the inner gunwale and
usually had two holes in each end for the lashings.
In the bark cover the horizontal sewing was often over root battens. In
many canoes rawhide was used in much of the lashing and sewing, and in
the last-built bark canoes the end lashings of the gunwales were often
protected by a decking formed of a small triangular sheet of metal,
obtained from a large can and crimped along its edges so as to clamp
the bark and main gunwales. When this metal deck-piece was used, the
cap and outwale ended against the inboard edge of it.
For use in open water these canoes were often fitted with a blanket
square-sail. The sapling serving as a temporary mast stood in a hole
in the second thwart, and was stepped on a block, or board, pegged or
lashed to the ribs.
The sheathing of all canoes of this class was of the same form--wide,
short strakes amidships, narrower short strakes afore and abaft. The
midship strakes were often quite short and their ends were over the
longer end strakes. The end strakes were, of course, tapered toward the
stems. The placing of the strakes was often irregular, with the result
that the butts were somewhat staggered. Some canoes had four strakes to
the length, but three appears to have been most common.
The large canoe was employed on the large lakes of the Mackenzie
region; smaller canoes of the same general form, 14 to 16 feet in
length and 30 to 40 inches in beam, were used on the large rivers
and streams. In the smaller canoes of this class, the flare of the
topsides was often less than in the larger craft. The Cree in this
area, particularly to the south of Great Slave Lake, also employed the
Athabascan form. This class of canoe, in general, appears to have been
strongly affected by outside influence; consequently this description
must be understood to cover existing canoes and models, not pure
Athabascan canoe building.
The usual construction methods were employed in building this class
of canoe; the stakes around the building frame were set vertically,
and when the bark cover was lashed to the gunwale members (inwale and
outwale together) the gunwales were spread and the thwarts inserted in
their tenons. Skill was required in preshaping the gunwale members,
which, as in the fur-trade canoes, had to be arched in sheer amidships
to allow for the change in sheer caused by spreading the gunwales in
construction. The building bed was also arched at midlength to allow
for the lifting of the ends that occurred in spreading the gunwales
with the bark cover attached.
A typical large Chipewyan canoe of this class was 21 feet 4 inches in
overall length, 43 inches beam and 14 inches in depth amidships. A
smaller Dogrib canoe of the same class was 14 feet 7 inches in overall
length, 31¼ inches beam, and 11½ inches in depth. However, these
smaller canoes appear to have been relatively uncommon, and the average
large canoe was about 20 feet long.
_Kayak-Form Canoe_
The kayak-form canoe was widely employed in the Northwest and was
highly developed in both model and construction. It was essentially a
portage and hunting craft, ranging in length from 12 to 18 feet and in
beam from about 24 to 27 inches, with a depth between 9 and 12 inches.
In areas where the kayak form was used as a family and cargo canoe,
the length would be as great as 20 or 25 feet and the beam might reach
30 inches. Except in the family or cargo canoe, which had none, there
was usually some decking at the ends, most of it forward. Some tribal
groups built the kayak form with its greatest beam at midlength, but
the most common form had its greatest beam abaft midlength and its
greatest depth there likewise. Many of the kayak forms had unlike end
profiles, so that there was a distinct bow in appearance as well as in
fact.
There was much variety in end profile, and the canoes of each tribal
group were usually identifiable by this means. The kayak-form bark
canoes of the lower Yukon and neighboring streams had a short
overhang, formed in a curved rake and alike or very nearly so, at bow
and stern. On the upper Yukon and adjoining streams the canoes had much
rake at both ends, the rake being straight from the bottom outward for
some distance, then curving rather markedly. The bow rake was usually
greatest, but the stern might be higher by one or two inches. The
bottom was without rocker, being straight or even slightly hogged in
most of these canoes. The sheer was straight to the point where the
rake began, then rose in a easy sweep to the ends. The end decks on
the upper Yukon canoes were short, those on lower Yukon canoes were
much longer; on the latter the bow deck was nearly a third the length
of the canoe, on the former about a fifth. In the Mackenzie Basin,
the kayak-form canoes had a moderate rake, curved in profile, at bow
and stern and a rather low stem-head; the depth at the stern was
noticeably greater than at the bow, and the deck forward was commonly
a little less than a fourth the length of the canoe. In these canoes
the greatest beam in most cases was abaft midlength, and this was also
true of the lower Yukon canoes. On the upper Yukon and in some of these
canoes on the lower Mackenzie, the greatest beam was amidships and the
depth at bow and stern were equal.
[Illustration: Figure 147
ESKIMO KAYAK-FORM BIRCH-BARK CANOE FROM ALASKAN COAST, with long
foredeck batten-sewn to the gunwales, no afterdeck, and rigid bottom
frame.]
The variation in depth at bow and stern in some of the kayak-form
canoes seems to have been related to the position of the greatest beam;
when the beam was abaft the midlength, the greatest depth was aft,
whereas when the greatest beam was amidships, the depth at the ends was
equal. With the beam abaft midlength, the weight of the paddler trimmed
the canoe by the stern somewhat, hence greater depth aft than forward
was necessary to make the canoe run easily and turn readily in smooth
water. In the sea kayaks of the eastern Eskimo, on the other hand, the
depth and the draft were greatest forward, to bring them head to the
sea when paddling ceased. The Alaskan sea kayaks were commonly of equal
draft at bow and stern or might have a slightly greater draft aft than
forward.
A third variation of the kayak form existed in British Columbia in
early times, and apparently was employed by the Beaver, Nahane, and
Sekani. It was an undecked bateau-shaped canoe having a fair sheer in
a long sweep from end to end, the stem profiles were nearly straight,
the ends were raked rather strongly, and the bow was somewhat higher
than the stern. The beam was greatest slightly abaft midlength. It is
estimated that canoes of this type, which has long been extinct and now
can only be reconstructed from a model, were about 14 feet 8 inches
long and 30 to 36 inches in beam, and probably were built of both
spruce and birch.
The gunwales of the kayak-form canoes were formed by inwales and
outwales; no caps were employed. In the Alaskan types and in the
extinct British Columbia bateau variation, the gunwale lashings were
continuous, but in the Mackenzie models the lashings were in groups.
Inwales and outwales in all the kayak forms ran to the stem-pieces,
which were plank-on-edge of a thickness that varied according to
tribal practice. No headboards were employed. The gunwale members were
rectangular in cross-section and were bent square with the flare of the
sides. The ends sometimes were swelled and rounded, and in the bateau
variation the gunwales, in cross section, appear to have been rounded.
Six thwarts appear in most of the kayak forms but the Loucheux model
had five and the bateau variation seems to have had but three.
[Illustration: Figure 148
ATHABASCAN HUNTING CANOES OF THE KAYAK FORM, showing characteristic
hull shape. These canoes were light, handy, and fast.]
Reinforcing bark was placed under the outwales in all Mackenzie Basin
canoes, but not in the Alaskan or in the bateau variation. The ribs in
all these canoes were small, usually about ½ inch square, and widely
spaced, about 9 to 14 inches on centers. No ribs were placed in the
rake of the ends. The ends of the ribs were chisel-pointed and were
forced between the inwale and outwale, against the inside of the bark
cover. In some canoes, however, the ribs near the ends of the canoe
were forced into short splits on the underside of the inwale. The
thwart ends might also be forced into short splits on the inside face
of the inwales or might be tenoned there; in any case a single lashing
was used at the thwart ends. Thwarts were parallel-sided in plan and
slightly tapered toward the ends in elevation; no shoulders were used.
In the bateau variation, a heavy thwart was placed directly under the
middle thwart with its ends against the side battens, apparently to
act as a spreader. Each end was notched over the side battens and was
held by two lashings to the bottom crosspiece below it. This structure
was probably made necessary by the fragile construction of this form
of canoe. In all kayak forms there was no complete sheathing--the one,
two, or three narrow battens to a side above the chine were held in
place only by the sprung ribs (without lashings); in the bateau form,
however, the side batten was lashed to each frame after the manner of
of an Eskimo sea kayak.