Chapter 3 (1)

ST. FRANCIS-ABNAKI CANOE FOR OPEN WATER, a type that became extinct before 1890. From Adney's drawings of a canoe formerly in the Museum of Natural History, New York. Details of Abnaki canoes are also shown.]

In preparing the ribs, a common practice was the following: Assume, for example, that there are 10 ribs from the center to the first thwart forward; these are laid out on the ground edge-to-edge with the rib under the center thwart to the left and the rib under the first thwart to the right. On the rib to the left the middle thwart is laid so that its center coincides with that of the rib, and the ends of the thwart are marked on the rib. The same is done to the rib on the far right, over which the first thwart is laid as the measure. On each side of the centerline the points marking the ends of the thwarts are then joined by a line across the ribs, as they lie together, to mark the approximate taper of the canoe toward the ends, at the turn of the bilge. Each rib is taken in turn from the panel and with it is placed another from the stock on hand to be set in a matching position on the other side of the middle thwart, toward the stern; the pair, placed flat sides together, are then bent over the knee at, or outside of, the marks or lines. The ribs in the next portion of the canoe's length are shaped in the same manner, using the lengths of the first and second thwarts as guides. Thus, the ribs are given a rough, preliminary bend before being fitted inside the bark cover and stayed into place to season. This method allowed the bilge of the canoe to be rather precisely determined and formed during the first stages of construction. At the ends, of course, the ribs are sharply bent only in the middle. Since the full thwart length makes a wide bottom, by setting the length of the rib perhaps a hand's width less than that of the whole thwart, the narrow bottom is formed.

The rough length of the ribs was twice the length of the thwarts nearest them. Hackmatack was used for thwarts by the St. Francis Indians, rock maple being considered next best. Cedar was first choice for ribs, then spruce, and then balsam fir. Longitudinals were cedar or spruce. All canoe measurements were made by hand, finger, and arm measurements. Basket ash strips were often used in transferring measurements.

MODEL OF A ST. FRANCIS-ABNAKI CANOE UNDER CONSTRUCTION, showing method of moulding ribs inside the assembled bark cover.]

From what has been said, it will be seen that the construction practice of the St. Francis did not follow in all details that of their Malecite relatives. The intrusion of western practices into this group probably took place some time after the group's final settlement at St. Francis. As they gradually came into more intimate relations with their western neighbors and drifted into western Quebec, beyond the St. Lawrence, their canoe building technique became influenced by what they saw to the westward. As would be expected, the St. Francis Abnaki began early to use nails in canoe building, but, being expert workmen, they retained the good features of the old sewn construction to a marked degree up to the very end of birch-bark canoe construction in southern Quebec, probably about 1915. It should perhaps be noted that what has been discovered about the St. Francis Abnaki canoes refers necessarily to only the last half of the 19th century, since no earlier canoe of this group has been discovered. The changes that took place between the decline of the Penobscot style of canoe and that of the later Abnaki remain a matter of speculation.

ST. FRANCIS-ABNAKI CANOE.]

_Beothuk_

The fourth group of Indians, classed here as belonging to the eastern maritime area, are the Beothuk of Newfoundland. Historically, perhaps, these Indians should have been discussed first, as they were probably the first of all North American Indians to come into contact with the white man. However, so little is known about their canoes that it has seemed better to place them last, since practically all that can be said is the result of reconstruction, speculation, and logic founded upon rather unsatisfactory evidence. The tribal origin of the Beothuk has long been a matter of argument; they are known to have used red pigment on their weapons, equipment, clothes, and persons. A prehistoric group that once inhabited Maine and the Maritime Provinces appears to have had a similar custom; these are known as the "Red Paint People," and it may be that the Beothuk were a survival of this earlier culture. But all that can be said with certainty is that the Beothuk inhabited Newfoundland and perhaps some of the Labrador coast when the white man began to frequent those parts. The Beothuk made a nuisance of themselves by stealing gear from the European fishermen, and by occasionally murdering individuals or small groups of white men. Late in the 17th century, the French imported some Micmac warriors and began a war of extermination against the Beothuk. By the middle of the 18th century the Newfoundland tribe was reduced to a few very small groups, and the Beothuk became extinct early in the 19th century, before careful investigation of their culture could be made.

Their canoes were made to a distinctive model quite different from that of the canoes of other North American Indians. The descriptions available are far from complete and, as a result, many important details are left to speculation. Some parts of the more complete descriptions are obscure and do not appear to agree with one another. In spite of these difficulties, however, some information on the canoes is rather specific; by using this, together with a knowledge of the requirements of birch-bark canoe construction, and by reference to some toy canoes found in 1869 in the grave of a Beothuk boy, a reasonably accurate reconstruction of a canoe is possible.

Captain Richard Whitbourne had come with Sir Humphrey Gilbert to Newfoundland in 1580 and revisited the island a number of times afterward. In 1612 he wrote that the Beothuk canoes were shaped "like the wherries of the River Thames," apparently referring to the humped sheer of both; in the wherry the sheer swept up sharply to the height of the oar tholes, in profile, and flared outward, in cross section.

John Gay, a member of the Company of Newfoundland Plantation, wrote in 1612 that Beothuk canoes were about 20 feet long and 4½ feet wide "in the middle and aloft," that the ribs were like laths, and that the birch-bark cover was sewn with roots. The canoes carried four persons and weighed less than a hundredweight. They had a short, light staff set in each end by which the canoes could be lifted ashore. "In the middle the canoa is higher a great deale, than at the bowe and quarter." He also says of their cross section: "They be all bearing from the keel to portlesse, not with any circular, but with a straight, line."

Joann de Laet, writing about 1633, speaks of the crescent shape of the canoes, of their "sharp keel" and need of ballast to keep them upright; he also states that the canoes were not over 20 feet long and could carry up to five persons.

The most complete description of the Beothuk canoe was in the manuscript of Lt. John Cartwright, R.N., who was on the coast of Newfoundland in 1767-1768 as Lieutenant of H.B.M. Ship _Guernsey_. However, some portions are either in error or the description was over-simplified. For example, Cartwright says that the gunwales were formed with a distinct angle made by joining two lengths of the main gunwale members at the elevated middle of the sheer. This hardly seems correct since such a connection would not produce the rigidity that such structural parts require, given the methods used by Indians to build bark canoes. The three grave models show that the sheer was actually curved along its elevated middle. It is possible that Cartwright saw a damaged canoe in which the lashings of the scarf of the gunwales had slackened so that the line of sheer "broke" there. Cartwright is perhaps misleading in his description of the rocker of the keel as being "nearly, if not exactly, the half of an ellipse, longitudinally divided." The models show the keel to have been straight along the length of the canoe and turned up sharply at the ends to form bow and stern. Cartwright also states the keel piece was "about the size of the handle of a common hatchet" amidships, or perhaps 1 inch thick and 1½ inches wide, and tapered toward the ends, which were about ¾ inch wide and about equally thick. The height of the sheer amidships was perhaps two-thirds the height of the ends.

A 15-FOOT BEOTHUK CANOE OF NEWFOUNDLAND with 42½-inch beam, inside measurement, turned on side for use as a camp. It gives headroom clearance of about 3 feet, double that of an 18-foot Malecite canoe with high ends. When the ends were not high enough to provide maximum clearance, small upright sticks were lashed to bow and stern. The shape of the gunwales would permit the canoe to be heeled to an angle (more than 35°) which would swamp a canoe of ordinary sheer and depth. (_Sketch by Adney._)]

Nearly all observers, Cartwright included, noted the almost perfect ~V~-form cross section of these canoes, with the apexes rounded off slightly and the wings slightly curved. From an interpretation of Cartwright's statements, it appears that after the bark cover had been laced to the gunwales, the latter were forced apart to insert the thwarts, as in some western Indian canoe-building techniques. The three thwarts are described as being about two fingers in width and depth. It is stated that the gunwales were made up of an inner and outer member and all were scarfed in the middle to taper each way toward the ends, the outer member serving as an outwale or guard. Cartwright also states that the inside of the bark cover was "lined" with "sticks" 2 or 3 inches broad, cut flat and thin. He refers also to others of the same sort which served as "timbers" so he is describing both the sheathing and the ribs as being 2 or 3 inches wide. He does not say how the thwarts were fitted to the gunwales, how high the ends were, how the ends of the gunwales were formed, nor does he give any details of the sewing used. However, the grave models suggest the form of sewing probably used and the approximate proportions of sheer.

An old settler told James Howley that the Beothuk canoes could be "folded together like a purse." Considering the construction required in birch-bark canoes, this is manifestly impossible; perhaps what the settler had seen was a canoe in construction with the bark secured to shaped gunwales, ready for the latter to be sprung apart by thwarts, as in opening a purse. Howley also obtained from a man who had seen Beothuk canoes a sketch which shows a straight keel and peaked ends, confirmed in all respects by the grave models or toys.

The toy canoes so often referred to here were found by Samuel Coffin in an Indian burial cave on a small island in Pilley's Tickle, Notre Dame Bay (on the east coast of Newfoundland), in 1869. Among the graves in the cave, one of a child, evidently a boy, was found to contain a wooden image of a boy, toy bows and arrows, two toy canoes and a fragment of a third, packages of food, and some red ochre. With one of the canoes was a fragment of a miniature paddle. One of the canoes was 32 inches long, height of ends 8 inches, height of side amidships 6 inches, straight portion of keel 26 inches and beam 7 inches, as shown by Howley.

In Newfoundland there was very fine birch but no cedar. There was, however, excellent spruce which would take the place of cedar. It seems certain, then, that all the framework of the Beothuk canoes was of spruce. It seems likely that they were never built of a single sheet of birch but were covered with a number of sheets sewn together, as in other early Indian birch-bark canoes. The canoe birch of Newfoundland grew to a diameter of 2 to 2½ feet at the butt, which would produce a sheet of birch of 6 to 7 feet width; the length would be decided by how far up the tree the Indian could climb to make the upper cut. As has been stated, the prehistoric Indians seemingly made little attempt to build birch-bark canoes of long lengths of bark, preferring to use only the bark obtainable near the ground and above the height of the winter snows.

The form of the Beothuk canoes, particularly the lack of bilge and the marked ~V~-form, has caused much speculation. One writer assumed that the form was particularly suited for running rapids. Actually, the Beothuk appeared to have used canoes for river travel very rarely, as few rivers in their country were suited for navigation. Instead, they seem to have been coast dwellers and to have used canoes for coastal travel and for voyages from island to island.

Their canoes were undoubtedly designed for open-water navigation, and the ~V~-form was particularly suitable for this. The draft aided in keeping the canoe on its course with either broadside or quartering winds, and if the Beothuks knew sail, the hull-form would have served them well. It is quite evident that the Beothuk canoes used ballast in the form of stones or heavy cargo. Stones would have been placed along the keel piece and covered with moss and skins. The strongly hogged sheer was useful in protecting cargo amidships from spray and, in picking up a seal or porpoise, the canoe could be sharply heeled without taking in water. The ~V~ sections fore and aft were suitable for rough-water navigation; because of its form and the weight of ballast, the canoe would pass partly over and through the wave-top without pounding. If a wave of such height as to overtop the gunwales just abaft the stem were met, the strongly flaring sides would give reserve buoyancy, causing the canoe to lift quickly as the wave reached up the sides.

The small sticks in the ends, mentioned by John Gay, served not only for lifting the canoe but also as braces to support the canoe at a given angle when turned over ashore to serve as a shelter. The Beothuk canoe, because of its form, was not well suited for portaging, and it must be concluded that little of this was done. In coastal voyages, the canoe would be unloaded and brought ashore each night to serve as a shelter.

It is believed that the gunwale lashing of these canoes was in groups, as in the Malecite. Howley questioned an old Micmac who had seen the Beothuk lashing; he likened it to the continuous lashing used by his own people, indicating some form of group wrapping, at least. It is probable that the group lashings were let into the gunwales by shallow notching at each group, a common Indian practice when no rail cap was used, to prevent abrasion from the paddle or from loading and unloading the canoe. The lacing of the ends appears to have been in the common spiral stitch, judging by the grave models. These, however, show a continuous wrapping at the gunwales, a common simplification found in Indian canoe models, representing either group or continuously wrapped gunwales indiscriminately.

The paddle of the Beothuks had a long, narrow blade, probably with a pointed tip and a ridged surface. The shape is nearly spatulate. The handle is missing from the grave model but was perhaps of the usual "hoe-handled" form without a top cross-grip.

From these descriptions and on the basis of common Indian techniques in birch-bark canoe construction, the form and methods of building the Beothuk canoe can be reconstructed. The drawing on page 97 shows the probable shape and appearance of the finished canoe. It seems likely that a level building bed was first prepared. The keel, probably rectangular in cross section, was then formed of two poles placed butt-to-butt, worked to shape, and scarfed. The fastening of the scarf was probably two or more lashings let into the surface of the wood. These lashings are assumed to have been of split-root material but may have been sinew. Possibly to strengthen the scarfs, pegs were also used, a technique consistent with the state of Beothuk culture. The keel probably had its ends split into laminae to allow the sharp bend required to form the bow and stern pieces; and it was probably treated with hot water and staked out to the desired profile. The main gunwales were similarly made and worked to the predetermined sheer which, in staking out, was hogged to a greater degree than was required in the finished canoe. The ends of the gunwales were apparently split into laminae to allow the shaping of the sharp upsweep of the sheer close to bow and stern. The outwales were probably formed in the same manner, after which the three thwarts were made and the material for ribs and sheathing prepared. The ribs were apparently bent to the desired shape, using hot water, and were either staked out or tied to hold them in form until needed.

BEOTHUK CANOE, APPROXIMATE FORM AND CONSTRUCTION]

The keel was then laid on the bed and a series of stakes, perhaps 4½ feet long, were driven into the bed on each side of the piece in opposing pairs at intervals of perhaps 2 or 3 feet. The stakes and keel piece were then removed and the bark cover laid over the bed. This may have been in two or three lengths, with the edges overlapped so that the outside edge of the lap faced away from what was to be the stern. The keel was then placed on the bark and weighted down with a few stones or lashed at the stem heads to the end stakes; then the bark was folded up on each side of the keel, and the stakes slipped back into their holes in the bed and driven solidly into place, perhaps with the tops angled slightly outward. The heads were then tied together across the work and battens placed along the stakes and the outside of the bark to form a "trough" against which the cover could be held with horizontal inside battens. These were secured by "inside stakes" lashed to each outside stake in the manner used in building eastern Indian canoes (see p. 45). The bark cover now stood on the bed in a sharp ~V~ form, with the keel supported on the bed, the ends of the bark supported by the end stakes, and both held down by stones along the length of the keel. An alternative would have been to fix heavy stakes at the extreme bow and stern of the keel and to lash the stem-heads firmly to these in order to hold the keel down on the bark.

Next the main gunwales, prebent to the required form, were brought to the building bed and their ends temporarily lashed to stem and stern. The bark was brought up to these, trimmed, folded over their tops, and secured by a few temporary lashings. Then the outwales were placed outside the bark with their ends temporarily secured, and a few pegs were driven through outwale, bark, and main gunwales, or a few permanent lashings were passed. The bark cover was next securely lashed to the gunwales and outwales combined, all along the sheer to a point near the ends. The excess bark was then trimmed away at bow and stern and the cover was laced to the end pieces to form bow and stern. This lacing must have passed through the laminations of the stem and stern pieces in the usual manner, avoiding the spiral lashing that held the laminae together. The ends of the gunwales and outwales were next permanently lashed together with root or other material and to the stem and stern pieces. This done, the gunwales were spread apart amidships, pressing the stakes outward still more at the tops. At this point the tenons may have then been cut in the main gunwales and the thwarts inserted. This method, incidentally, was used in building some western Indian bark canoes.

The usual steps of completing a birch-bark canoe would then follow--the insertion of sheathing, held in place by temporary ribs, and then the driving home of the prebent ribs under the main gunwales, with their heads in the spaces between the group lashings along the gunwales and against the lower outboard corner of the main gunwale member, which was probably beveled as in the Malecite canoe. The sheathing may have been in two or three lengths, except close to the gunwale amidships where one length would serve. On each side of the keel piece a sheathing strake was placed which was thick on the edge against the keel but thin along the outboard edge, in order to fair the sheathing into the keel piece.

At some point in this process, the bark cover was pieced out to make the required width, and gores were cut in the usual manner. In spreading the gunwales, the bow and stern would have to be freed from any stakes, as these would tend to pull inboard slightly as the gunwales were spread in the process of shaping the hull. The ribs could have been put in while green and shaped in the bark cover by use of battens and cross braces inside, as were those of the St. Francis canoes.

The sewing of the bark cover at panels and gores would take place before the sheathing and ribs were placed, of course. A 15-foot canoe when completed would have a girth amidships of about 65 to 68 inches if the beam at the gunwales were 48 inches, and a bark cover of this width could be taken from a tree of roughly 20 inches in diameter. Hence, there may have been little piecing out of the bark for width. In the form of the Beothuk canoe as reconstructed there is nothing that departs from what is possible by the common Indian canoe-building techniques. The finished canoe would, in all respects, agree with most of the descriptions that have been found and would be a practical craft in all the conditions under which it would be employed.

These were the only birch-bark canoes supposed to have made long runs in the open sea clear of the land. In them the Beothuk are supposed to have made voyages to the outlying islands, in which runs in open water of upward of 60 miles would be necessary, and they probably crossed from Newfoundland to Labrador.

The ~V~-form used by the Beothuk canoe was the most extreme of all birch-bark canoe models in North America, although, as has been mentioned, less extreme ~V~-bottoms were used elsewhere. The Beothuk canoe may have been a development of some more ancient form of bark sea canoe also related to the ~V~-bottom canoes of the Passamaquoddy. The most marked structural characteristic of the Beothuk canoe was the keel; the only other canoe in which a true keel was employed was the temporary moosehide canoes of the Malecite.

The Beothuk keel piece may have sometimes been nearly round in section like the keel of the Malecite moosehide canoe (p. 214). The two garboard strakes of the sheathing may have been shaped in cross section to fair the bark cover from the thin sheathing above to the thick keel and at the same time allow the ribs to hold the garboards in place. They could, in fact, be easily made, since a radial split of a small tree would produce clapboard-like cross sections. This construction would perhaps comply better with Cartwright's description of the keel than that shown in the plan on page 97.

The sheer of the Beothuk canoe is an exaggerated form of the gunwale shape of the Micmac rough-water canoe but this, of course, is no real indication of any relationship between the two. Indeed, the probable scarfing of the gunwales of the Beothuk canoe might be taken as evidence against such a theory. On the other hand, the elm-bark and other temporary canoes of the Malecite and Iroquois had crudely scarfed gunwale members, as did some northwestern bark canoes.

Most of the building techniques employed by Indians throughout North America are illustrated by these eastern bark canoes, yet marked variation in construction details existed to the westward, as will be seen.

_Chapter Five_

CENTRAL CANADA

The Indians inhabiting central Canada were expert builders of birch-bark canoes and produced many distinctive types. The area includes not only what are now the Provinces of Quebec (including Labrador), Ontario, Manitoba, and the eastern part of Saskatchewan, but also the neighboring northern portions of Michigan, Wisconsin and Minnesota in the United States. The migrations of tribal groups within this large area in historical times, as well as the influence of a long-established fur trade, have produced many hybrid forms of bark canoes and, in at least a few instances, the transfer of a canoe model from one tribal group to another. It is this that makes it necessary to examine this area as a single geographic unit, although a wide variation of tribal forms of bark canoes existed within its confines.

The larger portion of the Indians inhabiting this area were of the great Algonkian family. In the east during the 18th and 19th centuries, however, some members of the Iroquois Confederacy were also found, and in the west, from at least as early as the beginning of the French fur trade, groups of Sioux, Dakota, Teton, and Assiniboin. From the fur trade as well as from normal migratory movements there was much intermingling of the various tribes, and it was long the practice in the fur trade, particularly in the days of the Hudson's Bay Company, to employ eastern Indians as canoemen and as canoe builders in the western areas. These apparently introduced canoe models into sections where they were formerly unknown; as a result, the tribal classification of bark canoes within the area under examination cannot be very precise and the range of each form cannot be stated accurately. It was in this area, too, that the historical _canot du maître_ (also written _maître canot_), or great canoe, of the fur trade was developed.

Most of central Canada, except toward the extreme north in Quebec and toward the south below the Great Lakes, is in the area where the canoe birch was plentiful and of large size. There the numerous inland waterways, the Great Lakes, and the coastal waters of James and Hudson Bays make water travel convenient, and natural conditions require a variety of canoe models. Hence, when Europeans first appeared in this area they found already in existence a highly developed method of canoe transportation. This they immediately adopted as their own, and in the long period lasting until very recent times, during which the development of the northern portion of this area was slow, the canoe remained the most important means of forest travel.

In the northeastern portion of the area, including the Province of Quebec (with Labrador) from a line drawn from the head of James Bay eastwardly through Lake St. John and the Saguenay River Valley to the St. Lawrence and thence northward to the treeline in the sub-Arctic, dwelt the eastern branch of the far-ranging Cree tribe. Those living on the shores of Hudson and James Bays, along the west side of the Labrador Peninsula, were known as the Eastern, Swamp, or Muskeg Cree. To the north, at the Head of Ungava Bay, around Fort Chimo, and to the immediate southward, were the Nascapee, or Nascopie, supposedly related to the Eastern Cree. In southern Labrador and in Quebec along the north shore of the Gulf of St. Lawrence and for some distance inland, dwelt another related tribal group now known as the Montagnais.

Although the most recent canoe forms employed by these three Indian groups were very much the same, this may not have been the case earlier. A common canoe model in this area was the so-called "crooked canoe," in which there was a very marked fore-and-aft rocker to the bottom without a corresponding amount of sheer; as a result the canoe was much deeper amidships than near the ends. Another common model had a rather straight bottom fore and aft, with some lift near the ends and a corresponding amount of sheer. Between these was a hybrid which had some fore-and-aft rocker in the bottom and a very moderate sheer. Not until the 1870's was any detailed examination made of the canoes in this area; then it appeared that the crooked canoe might be the tribal model of the eastern Cree only, while the Nascapee employed a straight-bottom model, but it is possible that the examination was limited and that Nascapee use of the crooked canoe was simply not observed. By 1900, however, the crooked model was in use not only by the eastern Cree and the Nascapee but also by the Montagnais.

MONTAGNAIS CROOKED CANOE. (_Canadian Geological Survey photo._)]

In the area around Fort Chimo and at the northern ranges of the eastern Cree and of the Montagnais the lack of good birch bark made it necessary to make up the bark cover out of many small pieces. This not only was laborious but made a rough and rather unsightly cover. Hence, some of the northern builders, particularly the Nascapee, substituted spruce bark, which was available in quite large sheets. The use of the spruce bark, however, did not cause any of these people to depart markedly from the model or the method of constructing birch-bark canoes, as it did for the Indians in the maritime area.

At the time (1908) when Adney was carefully observing the canoes in this area he found that both crooked and straight-bottom canoes were being used by all three tribal groups, but with a variation in midsection form among individual builders. Both types were built with a midsection that had a wide bottom and vertical sides, or, as an alternative, a narrow bottom and flaring sides. The end profile of all these canoes showed chin. In some crooked canoes the profile was apparently an arc of a circle, but in most canoes the form was an irregular curve. The stem met the gunwale in a marked peak rounded very slightly at the head, as the result of the method by which the stem was constructed, but in the hybrid model used by the Nascapee the ends were low and not much peaked and the quick upward rise of the sheer near the ends was lacking. In cross section all these canoes became ~V~-shaped close to the ends, regardless of the midsection form. For the straight-bottom canoe and in the hybrid form this resulted in very sharp level lines, but the very great rocker of the crooked canoe brought the ends well above the normal line of flotation, so that this type was quite full-ended at the level line in spite of the ~V~-section.

It is apparent upon examining the crooked canoe that there was actually less variation in its form, in spite of differences in midsection shape, than in that of the straight-bottom canoe, owing to its very great depth amidships in proportion to its width. This proportion made necessary a very moderate flare in the narrow-bottom midsection and resulted in a rather wall-sided appearance, even in this model. The hybrid form, which fell between the extremes of the crooked canoe and the straight-bottom canoe, had a narrow-bottomed flaring-sided midsection, and its relatively moderate depth made obvious the flare in the topsides and thus created a distinctive model.

BIRCH-BARK CROOKED CANOE, UNGAVA CREE. (_Smithsonian Institution photo._)]

_Eastern Cree_

The construction of canoes of the eastern Cree and related tribes seems generally like that of the Micmac craft. Instead of the gunwale method employed in the Maritime area, a building frame was used, and as a result the gunwales were longer than the bottom. In constructing the crooked canoe, the building frame must be heavily sheered, and there is evidence that the building bed was depressed amidships, rather than raised as was usual in the east. The great amount of rocker in the bottom in this form of Cree canoe made it necessary to block up the ends of the building frame to a very great height, and there was no need to raise the building bed at midlength, since the rocker extended the full length of the bottom. The bark cover had to be gored at closely spaced intervals to allow the rocker to be formed, and even in the straight-bottom model, the quick rise of the bottom near the ends required closely spaced gores there. In the straight-bottom model, however, the building bed was raised at midlength, as in eastern canoe-building, and the building frame was ballasted to a cupid's-bow profile, when on the bed, so as to achieve the combination of straight bottom amidships with sharply rising ends.

The gunwales were formed of the main gunwale member and a light gunwale cap, no outwale being employed. They were joined at the ends and, after hot water had been applied, were staked out with posts under the ends to obtain the required sheer. The thwarts were then tenoned into the main gunwales, though occasionally a canoe was built with "broken" gunwales, that is, the thwart-ends were let flush into the top and covered by the caps. Some builders did not spread the gunwales and place the thwarts until after the bark cover was lashed at the sheer; others used the eastern methods of assembling the gunwale structure prior to securing the bark cover at sheer. The bark cover was attached to the main gunwales with a continuous lashing, as in the Micmac canoes, but the bark was not always brought over the top of the gunwales. As a result, some canoes had a batten placed under the lashing, near the edge of the cover, to prevent the lashing from tearing away. Due to the lack of good root material, the lashing was often of rawhide. For all horizontal seams in the side panels of the bark cover, rawhide sewing over a root batten was used. The ends of the gunwales were supported by sprung headboards; in some canoes these were bellied toward the ends to such a degree that they almost paralleled the end profiles.

NASCAPEE 3-FATHOM CANOE, EASTERN LABRADOR. Similar canoes, with slight variations in model and dimensions, were used by all Ungava Indians: the Montagnais and the Eastern, or Swamp, Crees.]

MONTAGNAIS 2-FATHOM CANOE OF SOUTHERN LABRADOR AND QUEBEC, showing old decoration forms. Drawing based on small model of a narrow-bottom canoe built for fast paddling.]

CROOKED CANOE, 2½-FATHOM, OF THE UNGAVA PENINSULA, used by the Ungava-Cree, Montagnais, and Nascapee. Also built with a wide bottom and a slight tumble-home in the topsides.]

HYBRID MODEL OF THE NASCAPEE-CREE CANOE, 2-FATHOM, built of spruce or birch bark, with details of canoes and paddles.]

The ends were formed by means of the same technique used for Micmac canoes; no inside stem-piece was employed and the bark cover was stiffened by outside battens covered by the lashing. In the Cree canoes, however, the stem battens were "broken" sharply at the sheer to form a slightly rounded peak where the end met the gunwale caps. The "break" in the battens was made by bending them very sharply, so that they were almost fractured. The Cree practice also differed from that of the Micmac, although not universally, by passing the lower end of the stem batten through the bark cover at the point where the stem met the bottom. The slit thus made was sealed with gum or, more recently, covered with cloth impregnated with gum. The stems were lashed in various ways; the most common was a spiral form up to the sheer. Near the gunwale caps crossed stitches or small, closely spaced wrappings were also employed. The tops of the battens, forming the peak of the stem, were brought along under the rail caps, in line with the gunwale lashings inboard, and secured with a continuous lashing for about 6 inches. In the northern parts of the area under discussion the stem lashing was often of rawhide.

EASTERN CREE CROOKED CANOE of rather moderate sheer and rocker. (_Canadian Pacific Railway Company photo._)]

Gunwale caps were wider than the gunwales and thus gave some protection to the lashing there. The ends of the gunwale caps were heavily tapered to allow the sharp bends necessary to carry them out on the stems. They were pegged or nailed to the gunwales, but at the ends were lashed; usually with two or three small group lashings over and under the stem battens, below the caps.

The most recent canoes had canvas covers instead of bark. Nails, tacks, and twine for sewing were used; otherwise they were built as the Indians built birch- and spruce-bark craft, and not as white men built canvas canoes and boats.

The framework of the canoes was usually spruce or larch. Toward the south and along the St. Lawrence some white cedar was used, and in the south maple was sometimes used for thwarts. The ribs of the canoes inspected by Adney were usually about 3 inches wide, and a short taper brought them to about 2 inches at the ends, where they were cut square across. They were spaced about 1 inch apart edge-to-edge amidships and somewhat further apart toward the ends of the canoe. The canoes usually had an odd number of ribs, as the first was placed under the thwart amidships. The last three ribs at the ends were "broken" at the centerline to allow them to take the necessary ~V~-section there; but the fourth rib from each end was only sharply bent. In some canoes the heel of the very narrow headboard was stepped on the sheathing against the endmost rib, in others it was stepped, as in the Micmac canoes, on a frog which rested against the endmost rib.

STRAIGHT AND CROOKED CANOES, EASTERN CREE.]

In more recent times the sheathing was laid in one of two ways, according to the preference of the builder, but the existence of the two styles suggests that each was once a tribal-group method. One method of shaping the bottom sheathing was to employ a center, or keelson, piece in two lengths, the butts being overlapped amidships, parallel-sided except toward the stems, where it was tapered to fit the ~V~-sections rather closely. The next strake outboard was short and was in the form of a shallow triangle with its base along the middle portion of the first strakes and about one-third the length of the bottom. Its apex was under the middle thwart. The next strake outboard was in two lengths lapped amidships, parallel sided along the arms of the triangular strake, and snied off at the ends to fit along the sides of the first strake. Another strake outboard of this was similar in form and position, but longer. Thus seven strake widths would complete the bottom sheathing. The side sheathing was narrow and slightly tapered; each strake in two lengths overlapped slightly amidships. The ends of the topside sheathing ran well into the ends, in most canoes, where they apparently served as stiffening. The second method of sheathing employed parallel-sided strakes throughout, laid side by side on the bottom, with the ends snied off to fit the form of the bark bottom. The existence of a model canoe made about 1850 (see p. 91) supports the theory that the first method was originally the Montagnais tribal construction and that the more primitive second method was probably Cree or Nascapee.

The ribs were preformed and fitted to the canoe after drying out. They were bent to the desired shape in pairs and tied with a thong across the ends to hold their shape while drying. Some builders inserted a strut inside the bent ribs, parallel to the thong, protecting the surface of the inner rib by a pad of bark placed under each end of the strut. The pair of ribs might also be wrapped with a bark cord to help hold them together. To aid in handling, one pair of ribs might be nested inside another. As in eastern canoes the ribs under the gunwales were driven into place. At the ends they were canted toward the center, so that in the straight-bottom models they stood nearly perpendicular to the rocker of the bottom there; in the crooked canoe the ribs were all somewhat canted in this manner.

MONTAGNAIS CANVAS-COVERED CROOKED CANOE under construction. (_Canadian Geological Survey photo._)]

The paddles used in this area were made with parallel-sided blades, the end of the blade being almost circular. The handle might be fitted with a wide grip at the head or it might be pole-ended. It is impossible to say how early sails were used to propel canoes, but it is probable they were introduced by the fur traders. Square sails were being used on the coastal canoes at the time the earliest reference was made to these canoes, in the 1870's.

Little is known about the decorations employed by the eastern Cree. The Montagnais birch-bark model canoe of about 1850 (see p. 91) has three small circles placed in a triangular position on the bow and a band along the bottom of the side panels. The circles and the bands are in red paint, but may have been intended to represent the dark inner rind left after scraping the winter bark cover. The use of decoration in this area after 1850 has not been noted in any available reference.

As a rule, the straight-bottom canoes were small, commonly between 12 and 18 feet overall, and the most popular size was 14 to 16 feet overall. A canoe of this size was usually employed as a hunters' canoe for forest travel, though it might be used occasionally along the coasts. These canoes were light and, in this respect, resembled the Micmac models shown in Chapter 4.

The original purpose of the crooked canoe is in question. Those travelers who saw this canoe in use on the Hudson Bay side of the Labrador Peninsula believed that it was designed for use in rough, exposed water. While it would be a desirable form for beach work in surf, the high ends would make paddling against strong winds very difficult. On the other hand the Montagnais used the crooked canoe for river navigation, particularly where rapids were to be run, and for this work it appears to have been well adapted. The crooked canoe was commonly built larger than the straight-bottom model, between 16 and 20 feet in length overall, and was a vessel of burden rather than a hunting canoe. Canoes up to 28 feet in length have been mentioned by travelers in this area but investigation indicates strongly that these were not the tribal form but the _canot du nord_, or north canoe of the Hudson's Bay Company traders.

Along the southern borders of their territory and to the westward the eastern Cree often built and used canoes modeled on those of their neighbors, the Têtes de Boule and the Ojibway. Hence the tribal classification does not hold good in these localities. Also, the eastern Cree were employed by the Hudson's Bay Company as builders of forms of the _maître canot_ and _canot du nord_ that are unlike their typical tribal model.

_Têtes de Boule_

The Têtes de Boule, particularly the western bands, were skilled canoe builders and had long been employed by the Hudson's Bay Company in the construction of large fur-trade canoes. Apparently made up of bands of Indians inhabiting lower Quebec, in the basin of the St. Maurice River and on the Height of Land, these bands had come down to the lower Ottawa River to trade with the local Algonkin tribe there in early times. They were known to the Algonkins, who had had some contact with civilization, as "wild Indians." They also came into close trading relations with the French colonists, as the Ottawa River was the early French canoe route between Montreal and Lake Superior. Because they cut their hair short, unlike the other Indians, these northern bands were nicknamed "Bull Heads," or "Round Heads," by the French traders, and the tribesmen soon came to accept this rather than their own designation of "White Fish People" as the tribal name. In more recent times, the name has been applied to groups of Indians living in western Quebec Province, near Lake Barrière and Grand Lake Victoria, but these do not consider themselves related to the St. Maurice bands.

It seems apparent that the canoe models of all these groups had been altered as a result of long contact with other tribal groups. Although the St. Maurice and the western bands were apparently not of the same tribal stock, their relations with the Algonkin may have brought about the use of a standard model by all.

FIDDLEHEAD OF SCRAPED BARK on bow and stern of a Montagnais birch-bark canoe at Seven Islands, Que., 1915.]

DISK OF COLORED PORCUPINE QUILLS decorating canoe found at Namaquagon, Que., 1898. Within the 4-inch disk may have been an 8-pointed star.]

The Têtes de Boule lived in an area where very superior materials for birch-bark canoe construction were plentiful. This, with the need for canoes imposed by the numerous waterways and the demand for canoes from white traders, made many of the tribesmen expert builders. Their small canoes, ranging from the 8-to 12-foot hunter's canoes to the 14-to 16-foot family canoes, were very similar in profile to the canoes of the St. Francis Abnaki. The Têtes de Boule canoes, however, were commonly narrower on the bottom, and in their construction a building frame was always used. The Têtes de Boule model was straight along the bottom for better than half the length and then rose rather quickly toward the ends. Similarly, the sheer was moderate amidships and increased toward the ends. The stems showed a chin and were much peaked at the gunwale ends. Most commonly the midsection had a flat bottom athwartships and a well-rounded bilge, giving the topsides, near the gunwale, a very slight outward flare. Some Têtes de Boule canoes had rather ~V~-section ends in which the endmost rib was "broken" at the centerline. As a result the lines were sharp and the canoes paddled very easily.

A FLEET OF 51 BIRCH-BARK CANOES of the Têtes de Boule Indians, assembled at the Hudson's Bay Company post, Grand Lake Victoria, Procession Sunday, August 1895. (_Photo, Post-Factor L. A. Christopherson._)]

For construction of the Têtes de Boule canoe, which was marked by good structural design and neat workmanship, the building bed was slightly raised at midlength, as was the general practice of the St. Francis builders. The building frame was usually about 6 inches less in width amidships, inside to inside, than were the gunwales, and from 15 to 18 inches shorter. The building frame was made quite sharp toward the ends so that, viewed from above, it rather approached a diamond form; this produced the very sharp lines that are to be seen in many examples of the Têtes de Boule canoes. The building frame was of course removed from the canoe as soon as the gunwales were in place and the bark cover lashed to them.

The gunwale structure, comprised of main gunwale members, caps, and outwales, was the same as in the Malecite canoes. The main gunwales were rectangular in cross-section, some being almost square, with the lower outboard corner bevelled off. Compared to those of eastern canoes of equal length, the main gunwales were unusually light; their depth and width rarely exceeded 1 inch, and in very small hunter's canoes these were often only about ¾ inch. Toward the ends, they tapered to ½ inch, or even slightly less. The ends of the main gunwales, usually of the common half-arrowhead form, were held together by rawhide or root thongs passed back and forth through horizontal holes in the members. After being thus lashed together, they were securely wrapped with thongs which usually went over gunwales and outwales and through the bark cover.

The gunwale caps, also light, were usually between ¼ and ½ inch thick and from 1 to 1½ inches wide. At the ends they were tapered in width and thickness, often to ³⁄₁₆ by ½ inch, so as to follow the quickly rising sheer there. The ends of the gunwales, caps, and outwales required hot-water treatment to obtain the required curve of the sheer. The caps were pegged to the gunwales and were secured at each end with two or three groups of lashings which passed around the outwales as well, and through the bark cover.

The outwales were likewise light battens between ¼ and ½ inch thick and from ¾ to 1¼ inches deep, the depth near the ends being tapered to ⅜ to ¾ inch so as to sheer correctly.

The bark cover had four or five vertical gores on each side of the middle thwart, the gore nearest each stem being commonly well inboard of the end thwarts. The side panels were usually deep amidships and narrowed toward the ends. A root batten was used under the stitching of the longitudinal seams of the side panels, which were sewn with a harness-maker's stitch. The top edge of the bark cover was brought over the top of the main gunwales, as in the Malecite canoes, and was secured by group wrappings passing over the gunwales and outwales, under the caps. These groups were not independent, the root thong being carried from group to group outside the bark in a long pass under the outwales. The groups of seven to nine turns were roughly an inch apart in many small canoes, and perhaps 1½ inches in the large craft. In the last birch-bark canoes in which no nails or tacks were used, wrappings of root thongs began with a stop knot, but this does not appear to have been the earlier practice.

TÊTES DE BOULE CANOE.]

The Têtes de Boule canoes had inside stem-pieces split, according to the size of the canoe, in four to six laminations and lashed with a bark or root thong in an open spiral in some canoes but close-wrapped in others. The stem-piece was as in the Malecite canoes, except that it ended under the rail cap, and did not pass through it as in the Eastern canoes; the heel was notched to receive the heel of the headboard. The bark was usually lashed through the stem, as in the Malecite construction. However, in some Têtes de Boule canoes, the stem close to the heel was not laminated and the bark was lashed to the solid part by an in-and-out stitch passing through closely spaced holes drilled in the stem piece. Above this, the lashing was the usual spiral which, in at least a few instances, was passed through the bark just inboard of the stem piece. Near the top of the stem the lashings sometimes were rather widely spaced and passed inboard of the stem-pieces; at other times, however, these lashings were more closely spaced and passed through the stem.

Ordinarily, at the ends of the canoe no _wulegessis_, or covers of bark, were used under the gunwale caps, although in one example examined a small cover had been inserted over the gunwale ends and under the caps, it did not extend below the outwales to form a _wulegessis_. In some canoes the bark cover was pieced up at the peak of the stems by a panel whose bottom faired into the bottom of the side panels.

A variety of methods was used to fit the gunwale caps at the ends of the canoe. Some builders carried the cap out beyond the gunwale ends, flat, over the edges of the bark cover and the top face of the outwale, but others tilted the cap outboard and downward. The ends of the caps came flush with the face of the stems. In an apparently late variation, the gunwales, instead of ending in the half-arrowhead, were snied off the inside and a triangular block was inserted between the ends. The gunwales were then pegged or nailed to the block and the whole secured with a root wrapping around them, before the outwales were in place. The first turn began by passing the root through a hole in the block near its inboard end, with a stop knot in the root.

The ends of the gunwales were supported by a narrow headboard sharply bellied toward the end of the canoe. The top of the headboard was notched to stand under the main gunwales; the center portion often was carried high and ended with a cylindrical top that was slightly swelled like the handle of a gouge or chisel. The heel was sometimes held in the stem-piece notch with a root lashing.

TÊTES DE BOULE CANOES.]