THE Geographical Distribution of Forest Trees in Canada (WITH A MAP) BY DR. ROBERT BELL, F.R.S. 1897 or* The EDITH and LORNE PIERCE COLLECTION of CANADI ANA Queen's University at Kingston Reprinted from The Scottish Geographical Magazine for June 189/ THE GEOGRAPHICAL DISTRIBUTION OF FOREST TREES IN CANADA. By Dr. Robert Bell. (Bead before the Society at Edinburgh and Glasgow, in March.) (With a Map.) Introduction. The writer has studied the geographical distribution of the forest trees of Canada for the last forty years, and during all that time he has had exceptional opportunities for observation, having been engaged each year in exploring, in connection with the Government Geological Survey, one part or another of the region represented on the accompanying map ; and, in addition to this extensive personal observation, he has collected much information on the subject from a great number of other travellers. The result has been to ascertain, with considerable accuracy, the limits of distribution of all our native trees from the Atlantic Ocean as far west as the prairie regions. Tree-Line Map. In 1873 he prepared a large map showing the northern limits of the principal trees in the four original provinces of the Dominion, and a reduction of this sheet was published in the report of the Montreal Horticultural Society and Fruit Growers' Association for 1879. The timber-tree map accompanying his official report for 1880, published by the Geological Survey in 1882, gives the general northern limits of thirty of our principal forest trees east of the Rocky mountains. The thirty species whose boundaries are shown with greater accuracy on the map accompanying the present paper embrace most of the same trees ; but a few substitutions have been made in order to introduce other species of more general interest without increasing the total number 282 SCOTTISH GEOGRAPHICAL MAGAZINE. represented, which is as large as can be exhibited with clearness on the scale of 80 miles to 1 inch. The same signs or forms of conventional lines have been retained for the several species in all the maps, and they have also been adopted where similar work has been done in adjacent parts of the United States. Our present map covers the most important and interesting portion of the great forest region of Canada. Beyond this area the northern species extend to Alaska, and there are valuable forests west of the prairies, in the Rocky mountains and British Columbia. But the northern group embraces only nine species, and the contours of their limits are not known to possess any peculiarities which cannot be described without a map. On the plains, trees of any kind occur only in favoured situations, and the different species are scattered in such a way that it would be difficult to represent the boundaries of their distribution on a map, while in British Columbia most of the trees are of different species from those east of the Rocky mountains, and they would require to be shown on a separate sheet. In Labrador. The map of the Labrador peninsula, published in the Scottish Geo- graphical Magazine for July 1895, gives the limits of each of the forest trees in that region so far as they are known up to the present time , but owing to the comparatively small amount of exploration which has yet been done in that peninsula, these tree-lines cannot be expected to be so accurate in detail as those of the rest of our map. The official reports of explorations made for the Geological Survey by the writer have generally contained notes on the forest trees of each region, and these have been utilised in preparing the present map. The writer is also indebted for various facts to nearly all his colleagues on the Geological Survey who are engaged in exploring work, and more especially to Messrs. Tyrrell, Low, and Macoun. Forest Wealth of North America. North America possesses a forest wealth which is perhaps unequalled in any other region on the globe. No fewer than 340 species of trees are known to be indigenous to the United States. Of these, 123 grow in Canada, 94 occurring east of the Rocky mountains, and the remaining 29 on the Pacific slope. Sixty-four of those east of the mountains are therefore unrepresented on the map, but the greater number of them are confined to small areas in southern Ontario, while those which ex- tend beyond this district are apparently of less interest with regard to their distribution than the thirty species which have been selected for presentation. Accuracy of the Tree-Lines. The contours of the limiting lines of each species have been carefully laid down from every source of information up to the present year, and, except in the Labrador peninsula, they cannot be far from the truth. If GEOGRAPHICAL DISTRIBUTION OF FOREST TREES IN CANADA. 283 there be any defect, it may be the omission of small outliers, patches, or colonies beyond the general boundary of a species, or of small inliers in the form of blank spaces within such boundary in which no trees of the species may occur ; but it is probable that if areas such as the former do exist they are of no great extent, while the occurrence of the latter will not alter the general boundaries. As a rule, we may assume that if we select any particular spot on the map we shall find near at hand specimens of all the trees whose limits lie to the northward of it. Former Erroneous Conceptions. In the southern portions of Quebec and Ontario, which were the first to be colonised by Europeans, it was observed that the tree-lines ran nearly east and west, and, before the subject had been investigated, it was supposed that the northern limits of the trees of Canada in general ran nearly in this direction ; but, as the present map shows, there are great divergencies in the contours of the limit-lines of many species, proving that these contours are governed by other causes than mere latitude. Only one out of the 340 species of trees above referred to is almost entirely confined to Canada, namely, the widely dispersed Banksian pine, whose southern boundary only touches the south shore of Lake Superior and takes in a small part of the State of Minnesota. Canada as a Whole. In the descriptions to follow, the geographical distribution of our forest trees is treated without reference to provincial boundaries, the in- troduction of which would only complicate matters and serve no good purpose. Eegion of Luxuriant Mixed Forest. In Canada and the United States the forests are more commonly called " woods " and " bush." The finest forests of North America, both as to variety of species and luxuriance of growth, were those of the middle and northern states, of which Ohio is about the centre. These splendid woods, which have been largely destroyed in the progress of the settlement of the country, extend into Pennsylvania, Kentucky, Ten- nessee, Indiana, Michigan, and southern Ontario. On any uncleared farm lot in the last-named region, one might find fifty or more kinds of native trees, all mingled together. This is in marked contrast to the poverty in species of the indigenous forests of the British Islands, where only about a dozen are represented, or indeed to the small number occur- ring in the forests of Europe in general. Botanic Limits of Trees. The tree-lines on the map represent the limits of each species from a botanic rather than an economic point of view ; for in some parts of its range the trees of any species may be so small or scattered as to have no commercial value. But as the public know so little as yet about such 284 SCOTTISH GEOGRAPHICAL MAGAZINE. matters, these lines may be of considerable importance in connection with questions affecting the future supplies of our various kinds of timber. Not so many years ago a vague impression prevailed in Canada that there was scarcely any limit to the northward and westward range of the white pine, which afforded our staple export timber. The author's tree-map, published in 1879, first dispelled this illusion, and caused some alarm among those interested in our forest wealth. One of the causes of this erroneous impression was the frequent allusions by travellers and the Hudson Bay Company's men, who came from Britain, to "pine" in the northern regions. To Canadians the word " pine," without any qualification, always means the white pine (Pinus strobus), and it appears to have been taken for granted that the persons just referred to used it in the same sense, whereas they referred to the spruce or balsam fir. Gradual and Abrupt Termination of Trees. In approaching their northern limits, some kinds of trees become gradually smaller and smaller, and are finally reduced to mere bushes before they disappear altogether, while others terminate abruptly or without any apparent diminution in the average size of the individual trunks. The latter habit is commoner in the southern than in the northern species, and it appears to prevail more in the eastern than in the western parts of Canada. As examples of this we may cite the black walnut, the chestnut, buttonwood, elm, yellow birch, basswood, hemlock, white and red pine, and white cedar. It is probable that those trees which bear large numbers of seeds, capable of being carried for some distance by the wind, such as the conifers and the poplars, have now reached the extreme northern limits of their growth; but some other species may be continuing to extend their borders. Indeed the general tendency appears to be to advance still farther north, as if many kinds of our trees had not yet had sufficient time to occupy all the territory congenial to their existence. In support of this view it may be mentioned that the few experiments that have been made in artificially transferring the more southern species northward have been successful through long distances. Among these experiments one of the most important is that of the Hon. Sir Henri Joly de Lotbiniere, who has found that the black walnut grows well and forms wood rapidly near the city of Quebec, although the nearest place in Canada where it is found growing naturally is in the Niagara district, about 500 miles to the south-west. At Carleton Place, thirty-five miles south-west of Ottawa City, and 200 miles from the Niagara district, the late Mr. Robert Bell raised black walnut-trees, which are now producing ripe nuts. These facts prove that the range of this, our most valuable timber, is capable of being extended over a much greater area than it yet occupies naturally. Pelee Island, in Lake Erie, is the most southern portion of Canada, and this is the only locality in the country where the honey-locust and the Kentucky coffee- tree are known to grow naturally ; but the former flourishes well GEOGRAPHICAL DISTRIBUTION OF FOREST TREES IN CANADA. 285 wherever it has been planted, throughout the lake peninsula of Ontario, and down the St. Lawrence nearly to Montreal, and small trees of it are also doing well at Ottawa. The Kentucky coffee-tree has also been found to grow to maturity at Ottawa. The negundo, or Manitoba maple, has not extended its range east of the western part of Lake Superior, and yet it is found to grow as well all the way to Montreal, 900 miles to the east of this, as it does in Manitoba. It is also success- fully cultivated at Lac la Biche, far to the north-west of its natural habitat. The black ash proves quite hardy when transported to Moose Factory on James Bay, more than 100 miles northward of its present range, and where it is exposed to the chilling influence of the sea air. There is little doubt that other kinds of trees, if tried, would be found to flourish beyond their present range. This ability of various native trees to grow far from their existing limits is probably not attributable to an improvement in the climate since they attained their present range, for there is no meteorological proof of such improvement, and, on the contrary, there is evidence that the verge of the forest is retreating southward both in North America and Asia. It is more likely due to the cause above suggested, namely, that sufficient time has not yet elapsed to have permitted the fullest possible territorial expansion of all the species. When frost alone operates to check the northward extension of a tree, there would appear to be no reason why such a species should not grow as far in that direction as it can ripen its seeds, even once in a number of years. When early frost happened to coincide with good seed years, many seasons might intervene between crops of ripe seeds, and the natural progress of northward extension would be very slow. The red oak, which requires two seasons to ripen its fruit, would suffer a double disadvantage. The writer has noticed that beech, elm, and black ash-trees, when growing in the most northern situations, seldom bear any fruit. As a rule, each species is at its best in the central part of the area it occupies, both as to breadth and longitudinal extension. Following this rule in cases where the northern limits of certain species only come a short distance into Canada, the finest specimens of these will be found to the southward, in the United States. Some Features of the Tree-Lines. One of the most striking characteristics of the tree-limits, as shown on the accompanying map, is the approximate parallelism of many of them to the general trend of the various sea-coasts, including those of Hudson Bay. There is, therefore, obviously some connection of the one circumstance with the other, and this connection becomes still more apparent when we also take into consideration the lines of former sea- coasts which must have existed when the land was at lower levels. A study of certain tree-lines shows that their present contours are in some way connected with conditions which existed in late geological times, and that any explanation of them must be involved with questions of 286 SCOTTISH GEOGRAPHICAL MAGAZINE. Pleistocene geology, such as the elevation and partial subsidence of the continent, with consequent changes in the ocean currents and in the temperatures of both sea and land ; also with the shifting of glacial centres, followed by the presence or absence of ice in certain regions at different times. The distribution of the Banksian pine and the balsam poplar will serve to illustrate the above inference. The former seems to shun proximity to the sea, while the latter seeks it. If we trace the boundary of the Banksian pine throughout its whole course, we shall observe that it nowhere touches the salt water, while the balsam poplar is found everywhere near to the sea as far north as it can grow, but it is apparently wanting over a great area in the central part of Labrador, which is probably too far inland for it to nourish. The southern lobe of this vacant space extends into the Ottawa valley in precisely the same south-westerly direction that was followed by the land ice from the old centre of glaciation in central Labrador to this valley. The erratic course of the balsam poplar line north-west of Hudson Bay may be con- nected with changes which took place in the later movements of the glacier ice in that region. Continuation of Pleistocene Conditions. Around the shores of James Bay the land is only moderately elevated for a considerable distance inland, especially on the west side, where a level tract extends for more than 100 miles from the sea. This border of low land affords abundant evidence of having been submerged in post- Pliocene times, and the line of the Banksian pine still skirts its margin all the way round, not having had time as yet to invade the old sea- bottom to any appreciable extent. It will be observed that the southern boundary of the Banksian pine reaches its extreme limit between the lower part of the Ottawa river and Georgian Bay of Lake Huron, where it intersects the 45 th parallel. In explanation of this decided southward curve, we find unusually high ground to the north-west of Ottawa and throughout much of the tract between this city and Georgian Bay, which would stand out as a promontory when the sea covered the country to the south and east, as it is known to have done at about the same time as it submerged the district around the head of James Bay. An outlier of the Banksian pine occurs in New Brunswick, and another in Nova Scotia, both being .at some distance from the open sea. Peculiarities of the White Cedar Line. The boundary of the white cedar is perhaps the most remarkable tree- line upon the map. In the central part its course is tolerably direct, but in both the east and west it turns southward and crosses the other tree- lines at right angles. If it had followed the average course of the boundaries of the other trees with which it is associated, it would have included the island of Anticosti, the whole of the maritime provinces, and the greater part of Newfoundland, in all of which the conditions of climate, soil, etc., are evidently as favourable for the growth of this tree as they are in New Brunswick and the Gaspe peninsula, where it is quite GEOGRAPHICAL DISTRIBUTION OF FOREST TREES IN CANADA. 28/ abundant, and even larger and finer than usual. The most probable explanation of this singular feature is to suppose that when the tree had established itself in the latter regions, Nova Scotia was an island, and that the seeds of the cedar had no good means of crossing a considerable breadth of water. There is evidence that a strait did separate Nova Scotia from New Brunswick where the cedar line passes, but since it became dry land the tree has made little progress towards occupying the peninsula. The white cedar occurs in the western part of Prince Edward Island, and in the form of a few clumps on the eastern side of the Bay of Fundy. If these occurrences are due to seeds having floated thither, this fact would show that they are not killed by immersion in salt water, and that no seeds have yet floated to the more distant shores of the cedarless regions above mentioned, or that, if some have chanced to do so, they did not reach suitable ground in which to grow. The cones of the cedar open and shed their seeds before they themselves fall off the trees, and it is possible that the naked seeds, on being blown into the sea, become water- logged and sink. It could then only be by a rare accident that seeds capable of germinating would be carried across any considerable breadth of sea water. When stranded upon a distant beach, they would have little chance of being carried from it to a suitable soil for their pro- pagation, as, owing to their unpleasant taste and odour, they have no attraction for any bird or beast. These suggestions are made in order to show that there may be some simple reason for the absence of the cedar from the greater part of the lands surrounding the Gulf of St. Lawrence. The currents in the Gulf, and the strong tide which must have passed through the strait formerly separating Nova Scotia from New Brunswick, may have prevented the seeds of the cedar crossing to those parts from which the tree is absent ; but if this were so, it becomes difficult to understand why the white pine, yellow birch, and other trees which are more southern in their general habit than the species in ques- tion, should be found in the cedarless areas. If some one were to plant the white cedar in Anticosti, Newfoundland, Cape Breton, and the eastern parts of Nova Scotia and Prince Edward Island, the experiment would be of much interest ; and the writer ventures to predict that the trees would do well. To the west of James Bay the white cedar attains a higher latitude than anywhere else, owing probably to the occurrence there of limestone rocks which afford suitable conditions for its growth. From this latitude it drops suddenly to the south of English river, and then turns westward to the head of Lake Winnipeg — in both cases running directly across the other tree-lines — and from the latter point it continues south into the United States, keeping on the east side of the Eed river valley. The pronounced bend to the south side of English river may have some significance in connection with the last phase of glacial action in this region. A remarkable outlier of small white cedars occurs around Cedar lake on the Saskatchewan river near its mouth, and at a distance of nearly 290 miles northward from the nearest point on the main boundary of the 288 SCOTTISH GEOGRAPHICAL MAGAZINE. species at the south end of Lake Winnipeg. The existence of the white cedar at this particular locality, so far removed from its general limits, may have been rendered possible by a combination of conditions such as the required soil, moisture, warmth, etc., which may not be all united at any intermediate locality ; or perhaps some better explanation may be found to account for this singular outlier. Perhaps it is due to human agency, as the Indians often carry cedar boughs in the bottoms of their canoes as a lining, for which purpose they are very suitable on account of their flatness. It is probable that if care were taken to plant the cedar in suitable situations, it would be found to grow at almost any other locality on the western side of Lake Winnipeg, for it appears probable that this and some other species of our trees have not yet reached in all quarters the limits of their possible extension from the southward ; in other words, it is not by any means certain that the present limits of all our trees have become permanently established. Some of them may be still slowly progressing from the southward in one part, and possibly retreating in another, or some may be advancing or retiring all along the line. The slow rate at which most kinds of trees have spread themselves over the country is evidence, either that more than 10,000 or 20,000 years have elapsed since the glaciers retired from their most southern extension to the Arctic regions, or that they were never confluent across the greater part of the intervening country. Means for the Diffusion of Trees. The seeds of some kinds of trees are endowed with much better means of transportation than others, as, for example, the down of the poplar seeds and the wings of those of the conifers, and hence such species may have reached the utmost limits of their possible growth. Some species are prolific in individual trees, while others are nowhere common, being only sparingly scattered in all parts of their area of distribution, and as they produce proportionately less seed, they will make slower progress in occupying new ground. With some species "seed years" occur seldomer than with others, and this is another cause of dis- parity in the rate of dispersion of different species. The number of seeds produced by some of the nut-bearing trees is extremely small com- pared with that of most others, and no provision is made for scattering the nuts by the wind. But other means are not wanting, and one of the most interesting is that of the squirrels and the birds of prey. It is a common thing for a squirrel, when running off with a nut to a distance from the parent tree, to be attacked by a hawk, when he will immediately drop his burden to facilitate his escape, and this nut may have an opportunity to germinate the following spring. Again, if he should fall a prey to the bird, all the " caches " of nuts which he has made in the ground may grow the next year. Outliers or Colonies. The following may be mentioned as examples of the known outliers or BANKSIAN PINE (Pinus Banksiana). (N.B.—The central tree only.) WHITE PINE (Pinus strobus). OLD TREES AND SECOND GROWTH. RED PINE. (Pinus resinosa.) Photographed by Dr. /.'. BeW.] O e O S 02 ^, H ^ w g J PLi =