[[p. 246]] [This article has been prepared at the expense of the Smithsonian fund for the purpose of illustrating, in connexion with the meteorological observations now in progress under the direction of this Institution and the Patent Office, the climate of the continent of North America. A favor will be conferred by any person who may furnish materials for extending or correcting the list of trees herewith appended.]

     The list appended to this article has been prepared in order to show the present state of knowledge of the distribution of the most important and useful trees and shrubs of the country, and at the same time to elicit further information on this interesting subject.

     With these objects in view, it has been attempted to give, where known, the extreme points to which each species extends in every direction, and at the same time to show in what part of the country it attains its greatest development and abundance. The chief deficiency in the first part of the plan has been found to be the want of precise localities in the limits mentioned by botanists. For instance, a tree may be described as found in Virginia, and yet not occur for hundreds of miles in travelling through that large State, which includes parts of three very distinct botanical regions. Any positive locality is of more use in determining range than such statements as the foregoing, for it enables botanists to increase the known range by observations in their own districts. The various State collections and reports have given the most accurate information on this subject, but very much is still wanting, and, in fact, respecting the range of most trees east of the Mississippi, scarcely anything has been added to the published observations of the two Michaux.

     In regard to the region of greatest abundance, there is much of the same want of accurate knowledge. In some cases it has been necessary to judge of this from the nature of the country and the known preference of the tree for particular kinds of soil, or its ability to withstand cold. Although there is much uncertainty on this subject, it will probably be found that the chief facts now known are stated; since I have, at least, good authority for the occurrence, if not for the abundance, of each tree in the region where it is considered most characteristic. Thus, in the absence of statistical information, it was necessary to select for the Mississippi region: 1st. Trees of the swamps, which have their central or maximum of number and development in such a locality. 2d. Such as extend scarcely north of Georgia, and, presumptively, are more common in the warmer region near the Gulf of Mexico. The trees believed to be peculiar to any region have [[p. 247]] their name (abbreviated) in small capitals in the last column. The character (?) prefixed to a locality indicates doubt whether this or some allied species is really the one attributed to that place.

     Future information may lead to the addition to the general lists of some trees included in the lists of trees nearly peculiar to Florida and Mexico, or to the contrary change. The numbers and letters prefixed to the names are intended to refer to charts of the distribution of trees, now in course of preparation. As trees do not form a natural series distinct from other plants, and as size does not serve as a criterion for their separation from shrubs, it might be considered more scientific to have included all woody plants in this catalogue; but as that would have swelled the list entirely beyond bounds, I have made a selection, guided first by the size, and secondly by utility, independent of size. It may be advisable in future to add to or to omit some of the species. Thus, as to the grape vines, Prunus maritima and Cerasus pumila, I included these and a few more woody plants; not properly trees, for various reasons. Some of the vines grow a foot in diameter, and are of use for wood as well as fruit. The others are interesting as analogues of trees, growing under peculiar circumstances. They may, too, become worth cultivation for fruit; and one of my objects was to include all such as far as possible. The species of Crataegus and Prunus mostly come under this exception.

     The reason for giving the maximum heights is, that it is thought the cultivation of trees will become some day a matter of national interest, and I wish to show what they are under the best natural circumstances, supposing that, with cultivation, they will at least equal this standard. Some of the western plants are little more than shrubs; but as the western regions are comparatively poor in trees, I have stretched the limit a little there, since shrubs become more valuable where trees diminish in number.

     Nearly all the varieties mentioned by various authors are given; because, first, the difference between a species and a permanent variety is scarcely definable; second, because they are often as truly characteristic of botanical regions as the species themselves. In this article Populus canadensis is separated from P. monilifera for example, because Michaux could not identify the latter east of the Mississippi; and we look upon it as belonging rather to the Rocky Mountain than to the Apalachian province.

     As Michaux notices a difference in the beeches of Canada and the more southern United States, we preferred (as in the other supposed varieties) to consider them as species having distinct ranges, until enough good specimens can be procured to determine the fact. It is doubtful whether there are any trees extending entirely across the continent, within the limits of the United States, which are not more properly included among those of the Lacustrian or Mexican provinces.

     Collections of the leaves, fruits, bark, and wood of our native trees are particularly desirable, and from as many localities as possible, in order to determine both their range and abundance, and also to decide those knotty points as to true specific distinctions, which still perplex [[p. 248]] the most skilful botanists. The specimens from each tree should be kept carefully together, and the name of the locality and collector given in full. Without such collections no information as to the large genera of oaks, hickories, magnolias, and, in fact, most others, can be at all depended on or made use of. Collections from the extreme corners of the United States, and from any part of the western mountains, will be particularly useful in determining all these questions. A good way of preserving a complete set from each species of tree is to obtain two pieces of the thick bark of the trunk about a foot square, taking care not to rub off the mosses or lichens, which are often very characteristic of the tree. Other specimens of bark from the branches, sufficient to show all its changes in appearance, and twigs with leaves, flowers and fruits, may be pressed between the trunk bark, with sufficient paper of any kind intervening, to absorb all moisture. One change of this paper will usually be sufficient, (especially if the bark is dry;) and fruits, if large and hard, may be so fixed as to hang outside, wrapped in paper. Particular care is necessary to prevent mixture of specimens. Blocks of wood from the trunk and branches at various seasons are also desirable for experimenting upon.

     Observations as to the relative abundance of each tree at the various stations may be expressed numerically, thus: very rare, 1; occasionally met with, 2; not uncommon, 3; common, 4; very common, 5; abundant, 6, &c.; using numbers up to 10, and explaining them. Frequently several trees will be found so nearly alike in abundance as to require the same number. Notice should also be made of the nature of the country and soil--whether mountainous, rocky, gravelly, sandy, or swampy, which will help to determine the limits of the natural regions. The geological structure of the district is, however, of secondary importance.

     The columns of range may be used by observers filling the blanks or adding to the recorded range in either direction, but this must be done carefully and with a perfect knowledge of the species noted. The name of the county should be given as well as of the town, and is preferable if only one is stated. Such blanks, filled up, may be cut out and sent to the Smithsonian Institution, addressed to the Commissioner of Patents, with the writer's name. Meteorological observers will take a special interest in the subject, and in most cases can make the best notes from their habit of observing the connexions of peculiarity of climate and forest growth.

     In the annexed catalogue generic and general English names are in capitals. The most important synonyms are given in italics, as well as local or little-used English names.

     The following is a list of the principal authorities consulted in collation of the facts regarding the distribution of trees of the United States, and in the preparation of the map:

Michaux's Sylva.
Brown, Trees of America, vol. I.
Torrey and Gray's Flora of North America, vol. I, II.
Gray's Botany of the Northern States, ed. 1857.
Lewis and Clarke's Travels, (Pursh Botany.)
Long's Expedition. Botany by Dr. James, Say and Torrey, in Ann. Lyc. New York.
[[p. 249]] Pike's Expedition to Rocky Mountains in 1808, 1812.
Pacific Railroad Explorations, vols. I to VII, and Reports on Botany by Drs. Torrey and Gray, Newberry, Antisell, Bigelow, Cooper, and Suckley.
Collections in Smithsonian Institution from Nebraska, Kansas, California, Oregon, and Washington Territory, and other regions west of the Mississippi river.
Hooker's Flora Boreali-Americana, 2 vols. quarto.
Fauna Boreali-Americana, preface, by Dr. Richardson.
Darlington, Flora Cestrica.
Emerson, Trees and Shrubs of Massachusetts.
Lapham, Plants of Wisconsin.
Hoy, Trees of Wisconsin.
Wailes, Agriculture and Geology of Mississippi.
Kitchell and Cook, Geology of Cape May county, New Jersey.
Thompson, Natural History of Vermont.
Ray Society Reports on Botany, 1849, including Geyers' Notes of Journey across North America.
Agassiz, Lake Superior. (Narrative by J. E. Cabot.)
Torrey, Botany of New York.
Swallow, Geology of Missouri, 1st and 2d Reports, (list of trees.)
Hayden, Trees of Nebraska, in Warren's Explorations, 1859.
Elliott's Botany of Southern States, 1824.
Russell, Climate and Agriculture of United States, 1857.
Richardson's Journeys in Search of Franklin.
Reports on Botany.
Darby, Botany of Southern States.
Geyer, Botany of Oregon, in Hooker's Lond. Jour. Bot.
Gray, (Asa,) Statistics of Flora of Northern United States, in Sill. Jour., 2d series, vol. XXII, 1856, and XXIII, 1857.
Anderson, Synopsis of North American Willows, from Proc. Amer. Acad. Arts and Sciences IV, 1858.
Rafinesque, Florula Ludoviciana.
Trans. Amer. Phil. Soc., 2d series, vol. 3, Dr. Pickering on Distribution of Plants of United States.
Ibid, Nuttall on Plants of Arkansas.
Lieutenant Colonel Kearny and Major J. D. Graham, Top. Eng., East Boundary of Texas, with detailed maps, 1840.
Brevet Captain J. C. Fremont, Top. Eng., Explorations, 1842, 1843, 1844.
Major W. H. Emory, Top. Eng., Explorations of, 1846, 1847.
Lieutenant J. H. Simpson, Top. Eng., Canadian river and Navajo country, 1849.
Captain H. Stansbury, Top. Eng., Exp. to Salt Lake, 1849, 1850.
Lieutenant Colonel Johnston, Top. Eng., Explorations in Texas, 1849, 1850, 1851.
Captains Sitgreaves and Woodruff, Top. Engs., Boundaries of Creek country, (36° 30', de-viled maps,) 1850, 1851.
Captain L. Sitgreaves' Expl. Zuñi and Colorado rivers, 1851.
Captain R. B. Marcy, U.S.A., Expl. of Red river of Texas, 1852.
Lieutenant G. K. Warren, Top. Eng., Explorations in Minnesota and Nebraska, 1855, 1856, 1857.
Major W. H. Emory, U.S.A., Survey of Mexican Boundary, 1849, 1850, 1852, 1853, 1854, 1855.
J. R. Bartlett, U.S. Comm., Survey of Mexican Boundary, 1849, 1850, 1852, 1853, 1854, 1855.
Personal narrative.
A. D. Bache, Supt. Coast Survey, maps of Pacific coast and along Gulf of Mexico.
United States Land Office Surveys of Michigan, Indiana, Illinois, Wisconsin, Minnesota, Missouri and Arkansas, up to spring of 1858.
Journey to Pembina, on Red river of Minnesota, 1857, by R. Kennicott, of Illinois.
Nicollet's Report Expl. Upper Mississippi, 1843.
List of Plants by Dr. Torrey and C. Geyer.
Roemer's Texas, 1849.
A. Von Humboldt on Distribution of Plants, translated in Edin. Philos. Jour., vol. VI, VII, 1820.

     Many other works have been consulted which were not made note of at the time, and also results given of personal explorations and observations made from Maine to Virginia, and across the continent through Nebraska, Oregon, Washington, and California.

[[p. 268]] EXPLANATION OF THE MAP.

     This outline map of that part of America north of Mexico is intended to illustrate, so far as can be done on so small a scale, the distribution of forests, trees, &c., according to the results of the latest explorations and studies. In order to get it within the size of the page it was found necessary to make the principal meridian oblique to the sides of the map, but by noticing the corresponding numbers of the degrees of latitude the reader will easily perceive the true direction of the several points, east and west being, of course, nearly at right angles to the meridian. The equatorial projection is preferred, as showing best the true area of the various regions. The dark full lines crossing the continent are the outlines of the best determined and most important natural divisions, which may be called NATURAL PROVINCES. The broken lines represent approximately the subdivisions of these provinces, called REGIONS, and also parts of the province boundaries not accurately determined.

     The undulations of surface, mountains, and other circumstances cause great irregularities in the outlines of regions, especially in the western part of the continent; but these are neither well determined nor, if they were, are they capable of illustration on this scale. It is believed, however, that the most marked limits are represented, that is, those by which the greatest number of species, both of plants and animals, are bounded in their range.

     In using this map the reader should compare it with a good and large map of North America. The regions in which trees are found (indicated by letters on the map) are:

A. The Algonquin, in which four or five species of trees seem to be more abundant than elsewhere, and therefore characteristic; none, however, are peculiar . . . 5

B. The Athabascan, in which about twelve are characteristic but none peculiar . . . 12

C.Canadian, having seventeen characteristic trees and one apparently peculiar . . . 15

D. Alleghany, in which there are twenty-four trees characteristic and six peculiar . . . 30

E. Ohio, having twenty-two characteristic and one peculiar . . . 23

F. Tennesseean, with thirty-four characteristic and three peculiar . . . 37

G. Carolinian, with eighteen characteristic and seven peculiar . . . 25

H. Mississippian, with thirty-two characteristic and seven peculiar . . . 39

I. Floridian, with about thirty-two characteristic and thirteen peculiar . . . 45

Total number of species in the eastern forest regions . . . 234

     D, E, F, G, H, may be considered as forming a natural province, and called the Apalachian. Florida appears rather to belong to the West Indian province.

     [[p. 269]] J, the Texan, K, the Illinois, and L, the Saskatchewan regions are characterized by prairies; and the forests, which occur only near their eastern borders and along the rivers, gradually decreasing towards the west, are composed entirely of species characteristic of the region eastward. No new ones occur, while a large number of trees disappear at the commencement of prairies. M, the Dacotah, and N, the Camanche regions, are composed of these great treeless and often arid plains which skirt the eastern slopes of the Rocky mountains. Trees occur only in narrow belts along the rivers, and for about three hundred miles east of the mountains most of the streams are entirely woodless.

     These five regions are, however, supplied with numerous peculiar plants and animals, mostly of quite distinct species from those of the forest-clad regions eastward. They are so well marked, in these respects, that, together, they may be considered as forming a natural PROVINCE, which, from its most remarkable character of plains, is called CAMPESTRIAN.

     O, P, Q, may be named, from the Mexican States which partly compose them, Tamaulipan, Choahuilan, and Chihuahuan regions; but too little is known of them to define their limits well. Their trees are given in the list of those along the Mexican boundary, and these are found almost exclusively along the Rio Grande, within the United States. Only seven or eight species are found, in addition to the few which extend from the Mississippi region and Florida. . . . 8

     R, the Arizonian region, though in great part treeless, furnishes a large number of additional species, some of which are probably more common in the adjoining Mexican States of Sonora and Lower California, while others, chiefly among the San Francisco mountains, seem to be nearly or quite peculiar.

     The former, thirteen in number, are included in the Mexican Boundary list, while the fourteen characteristic and ten (?) peculiar are in the catalogue of Western trees. . . . 37

     These numbers, except the total, must, however, be considered as far from settled, as the surrounding regions are but little explored; and in a region so poorly wooded, as most of it is, ten peculiar species could scarcely be expected. It will be remarked also, that most of these trees characteristic of the Mexican boundary are not over fifteen or twenty feet high, and in many cases these are extreme heights. Shrubs constituting the dense thickets called chapparal take the place of forests over great tracts.

     S, the Wasatch region, almost unexplored, is not known to have any trees peculiar to it; but as the mountain summits are usually well wooded, some may be hereafter found characteristic, if not peculiar. . . . 1

     T, the Padoucan, has apparently two peculiar pines, but its vegetation is scanty in forests, and but little explored. . . . 2

     U, the Utah region, being, as far as explored, almost woodless, it is rather surprising to find that the only tree which is abundant on some of its mountains is apparently almost peculiar to it--the Juniperus occidentalis. . . . 1

     V, the Shoshonee region, though, like the last, almost woodless on its plains, has some well-wooded mountain ranges, on which a larch [[p. 270]] and several pines are found. Two willows have as yet been found only in it. One pine is believed to be more common here than elsewhere. . . . 3

     In the four last regions the place of forests is also supplied by numerous peculiar shrubs, mostly very distinct from any of the trees, as well as from the chapparal of Mexico, though some of the same genera appear as in the latter. These are not all stunted trees, but of entirely different families, and formed especially for the arid regions they inhabit. R, S, T, U, V, together, form a peculiar province, which may be distinguished as the Rocky Mountain Province.

     W, the Californian region, is the most highly favored of the western groups in the variety, though not of the extent of its forests, which are confined chiefly to the mountains, while most the species occur in groups, scattered singly through the prairie. Eighteen trees are considered characteristic, and eighteen peculiar; making, in all, . . . 36

     Seven others are found near San Diego, and are probably characteristic of Lower California.

     X, the Oregonian, is densely wooded over nine-tenths of its surface, prairies occurring chiefly in the valleys of the southern part, and gradually disappearing towards the north. Thirteen trees are considered characteristic, and nine peculiar, though the unexplored character of the regions northward may make the number larger than it really is, . . . 22

     Y, the Kootanic region, is described as very densely wooded almost everywhere, and certainly that part which extends south of latitude 49° is covered with forests almost equal in size and density to those of the Oregonian. Eight trees are believed to be among those characteristic of it, and some will probably be found peculiar. . . . 8

     Z, the Yukon region, is believed to be generally wooded, but almost nothing is known of it or its peculiar products. Some trees have been found as far north as Kotzebue's Sound, (2) the northern limit of trees which are not known farther south, but may be Asiatic. Six of those in the catalogue are attributed to this region, as they are found only as stragglers within the boundary of the United States. . . . 6

     The trees of the western regions (including the Mexican boundary group) are, therefore, 131 species. The great wooded regions of the northwest--X, Y, and Z--apparently form a province distinct from those east or south of them.

     W partakes of the characters of those of Lower California and Arizona, and with the former may form a province to be called the NEVADIAN. Further exploration is, however, necessary to decide their mutual relations.

     Although neither time nor space will permit here a full statement of the grounds on which the dividing lines are laid down, yet a short statement of the principal ones will be necessary, in order to enable observers to collect and increase information on this interesting branch of Physical Geography. Trees represent only certain families and genera of plants, and their distribution alone would not, therefore, indicate that of other plants and of animals; but forests have a wonderful influence on both, and it is well known to naturalists that neither rivers, lakes, nor mountains separate the range of so many species as the line where a continuous forest skirts a woodless country.

     The full line 1--2, broken towards the west, shows the northern [[p. 271]] limit of forest growth. It is laid down from the information given by Richardson, Belcher, and other arctic explorers, and is one of the best defined boundaries of organized beings on the continent. North of it there are no trees and but few woody plants; the vegetation is mostly peculiar; and the animals, numerous in species, in but few instances extend south of it. Both are generally identical entirely around the north pole, and this therefore forms part of the Arctic Province; while many peculiar species as well as marked geographical boundaries separate it from the Asiatic portion and, it may be called the Esquimaux Region. This line coincides well with the yearly isothermal of 17° 5', summer 50°, winter -15°, in all that portion between Hudson bay and the Mackenzie river, but it seems probable that other circumstances besides temperature regulate the forest limit near the east and west ends. Neither the climate nor vegetation of these parts is yet well enough known to admit even of speculations on this subject.

     A, B, and C, may, together, be considered as forming a Province, which, from the number of its great lakes, may be called the Lacustrian. It is well characterized by its forests, almost unbroken, consisting chiefly of evergreen Corifera, which afford food and shelter to a large number of characteristic animals. It comprises most of the fur countries of North America. It is only near the Alleganies that its limits are obscure, but the broken lines there indicate the average range of its characteristic species. The next line southward is 3--4, running nearly parallel to 1--2, and separating the water-sheds of Hudson Bay and the Gulf of St. Lawrence from the basins of the Great Lakes and Lake Winnipeg. It forms the northern limit to a large number of trees, &c., but can only be considered approximate towards its eastern end. It corresponds very well with the isotherm of 37° 5', summer 60°, winter 15°. The careful observations of Richardson at its western end show that there are about 24 species of trees common north of it, while many cease to grow very abruptly in the vicinity of Lake Winnipeg.

     The country between 1--2 and 3--4 is divided for convenience into two regions, A the Algonquin, and B the Athabascan. Though the peculiar products of each are little known, there seem to be indications even among the trees of considerable difference, especially in the proportions. The Algonquin is almost unexplored, but will probably be found to have the greatest numbers of several of the trees mentioned in the catalogue, as most abundant in adjoining regions; none are, however, peculiar to either of these northern regions.

     The line 5--6 is used to bound approximately the Canadian Region on the southeast, though from the nature of the country there is no well marked division there, the valleys belonging to one region and the mountains to the other.

     The numbers 5--6--7 indicate the general limits of D, the Alleghany Region comprising the eastern slopes, and towards the south some of the highest peaks and ridges of this range; 24 trees are considered characteristic, and 6 peculiar to it. Towards the west and south its limits are not well marked from the uneven surface of the country, but to the southeast the line between it and G is the abrupt [[p. 272]] geological boundary, between the metamorphic rocks of the hills and those of later date, corresponding to the change in soil characterizing the sandy flat low country.

     The line 7--8 is the northwestern boundary of G, and a very natural one both geologically and botanically. The range of animals also corresponds to it in some degree. The boundary indicated by 9--10 is not a fixed limit, since the natural history of Florida is still but little known. There is, however, a geological line near this point, which, with the rapid increase of temperature southward, will probably be found to limit the greater number of species characteristic of Florida. The line connecting this with 8 is also only arbitrary or geographical, as it separates the waters of the Atlantic and of the Gulf. There seem to be several trees, however, limited in range near it, and probably none can be made more natural. 8--12 is a continuation of the line separating the low, sandy country from the granitic and limestone formations of the hilly country. From 12 it extends southwest, bordering the hilly table-land west of the Mississippi valley.

     The broken line 6--12 is one of the most unsettled of all. Though a number of trees are limited in this vicinity, there is no geological or geographical line by which they may be bounded, and an attempt is therefore made to make an average limit, depending on climate and elevation as well as on the actual facts of distribution.

     An eminent geologist has suggested to me that perhaps the southern limit of the drift may be found the true boundary here and on the line of 15--16.

     The lines G--6--15--16 may represent the summer isothermal of about 75°, while 7--6--15--16 are near that of 35° for winter and 55° for the year.

     The numbers 11--15--13 indicate the eastern border of prairies, and the limit of the great number of plants and animals which belong exclusively to them. It is one of the best marked of all the lines, though, since the settlement of the country, it has become obscured both by the extension of forests westward and their destruction eastward of it. The line represents nearly its natural position. The figures 13--23 show the northeastern border of the prairies, which is considered pretty well established, as far as authorities examined give any data for it. Richardson mentions their northern extremity near Great Slave lake.

     The line 13--14--22 is a good geographical boundary, and also nearly coincides with the isothermal of summer 70°, though winter seems to differ much at its two extremities. 14--18 is the line of about 2,000 feet general elevation, and west of which scarcely any of the eastern trees extend; in all not over six or eight species, while, as already stated, most of M and N is entirely woodless.

     The numbers 17--18 show approximately the boundary of the tropical group of animals and plants which are found on the lower Rio Grande. On a larger map the direction of this line would be more irregular in detail.

     The numbers 18--19--16--21--22--23 mark the line limiting the forests and other products of the western mountains. It will be found [[p. 273]] a very well marked, though irregular, boundary, and of trees only two or three western species extend east of it, and as many eastern ones westward. The general elevation of the country on this line is from 3,000 to 4,500 feet, and the sudden change from plains to mountains causes a corresponding change in climate, productions, and general features.

     The line 19--20--21 is a geographical boundary, which, from the great elevation of the dividing ridge, and the different climates of its opposite sides, causes a more marked limitation here than on most of the other western mountains.

     The line 24--26--25, provisionally bounding the Arizonian region, is founded chiefly on the isothermal lines. The great summer heat of 80° and upwards is connected with the growth of the Cactaceae and other peculiar plants, and the mountains between the Colorado and Gila also have many peculiar forms even on their cool, high summits; and 26--20, bounding the upper Colorado valley, is the geographical boundary of the Wasatch region on the west.

     26--27 and 20--27 form the geographical boundaries of the Utah basin; and the latter appears also to be near the summer isotherm of 75°. The whole of this region is more than 4,500 feet above the sea.

     20--31--27 forms the northern boundary of the great Columbian plains or Shoshonee region, and is also a well marked limit of the continuous forests northward of it. It coincides also pretty well with the summer temperature of 70° and 35° for winter, or 45° annually. The sudden change in the surface by forests causes these isothermals to be more abrupt than in a country uniformly bare or wooded. This is probably the effect of increased rains, which, of course, affect other products also. For this reason it is believed that the summer temperature of about 65° will be found to run along the line of 13--4--23, instead of being parallel to 70° on 13--14--22.

     26--27--31 is the eastern limit of the continuous forests of the Sierra Nevada and Cascade ranges, and a very good natural boundary. 27--28 is the approximate limit between W and X, and though in reality very irregular from the rough character of the country, its general direction is sufficiently shown. It limits the range of many trees and other products.

     A boundary probably exists between Upper and Lower California, but cannot yet be determined naturally.

     The broken lines 31--29, 21--30, 23--2, are of course merely conjectural, but are made to correspond to the general features of the country as well as the direction of the isothermal lines. On the coast at 30° the annual temperature is as high as 43° 5', corresponding to that of the coast of Maine; while what little is known of the interior shows that though exceedingly rugged and often inaccessible the Yukon region possesses a good climate and a great variety of natural products, probably comparable with Sweden and Norway. Its winters are milder near the coast than those of Maine, though fifteen degrees farther north.

     These few notes, hastily thrown together, will, it is hoped, make the design intelligible, and lead others to investigate and increase our very imperfect knowledge of animal and vegetable distribution. The [[p. 274]] writer has collected a large amount of material relating to it, which he hopes at some future time to publish in the form of physical charts, and intends the present paper merely as a circular requesting further information, for which due acknowledgment will always be made, and the favor reciprocated as far as possible.

GENERAL CONCLUSIONS.

     Looking now at the regions as combined in groups, we may derive the following conclusions from the examination of the number of species, abundance and arrangement of their trees and forests.

     1st. Coming southward from the treeless Esquimaux region and passing a certain well marked line, we suddenly find a continuous forest, broken only where the poverty of the soil or other local causes prevent the growth of trees. About fifteen species compose this forest, but the prevailing forms are the Coniferae. For much interesting detail on the influence of climate and other causes on this northern line, as well as for evidence of its encroachment southward, I must refer the reader to Dr. Richardson's interesting Arctic explorations.

     Passing into the Canadian region we find seventeen species of trees suddenly added as characteristic, and towards its southern border many of those of the Apalachian Province begin to appear. Yet the Lacustrian Province is a well marked one, and in the grouping of its trees differs materially from those farther south.

     Too little is known of the climate, in its interior especially, as to the amount of rain at different seasons, to form conclusions as to its influence on this peculiar grouping. It appears more probable that other physical agents were the chief ones in this respect, particularly the geographical and geological barriers which those interested may perceive on good maps. Another affecting the western part of the boundary will be hereafter referred to.

     2d. Coming next to the Apalachian Province, and remembering that the forests of a great part of the Alleghenies are only a southward extension of those farther north, we find a vast increase in the variety of our forest trees. In fact, looking at all its natural products collectively, one of the most striking, as compared with the rest of the world between the 30th and 45th degrees of north latitude, is its richness in trees, which will compare favorably with almost any part of the tropics.

     It contains more than twenty species which have no representatives in the temperate climates of the Old World, and a far greater number of species of the forms found there. In fact, those trees which have been considered characteristic of the Lacustrian Province, with perhaps the addition of those of the Allegheny region, will fully represent all the common trees of Europe in number of species, though a few other genera must be added from other regions to give them all.

     Yet with all this variety the bulk of the Apalachian forests is of forms characteristic of the temperate zone. Only about twenty-three out of one hundred and thirty-six species are broad-leaved evergreens, and these nearly all in the two southern regions. We will hereafter see a marked difference in another part of the continent in this respect, [[p. 275]] and I will merely allude to its connexion with, though not dependence on, the peculiarities of our climate. We have with a tropical summer a tropical variety of trees, but chiefly of northern forms. Again, with our arctic winters we have a group of trees, which, though of tropical forms, are so adapted to the climate as to lose their leaves, like the northern forms, in winter. But here, it must be distinctly understood, is no alteration produced by climate. The trees were made for and not by the climate, and they keep their characteristics throughout their whole range, which with some extends through a great variety of climate. I will not stop to discuss the relations of specific distinctions and external physical influences, assuming that the reader understands the scientific views of the subject. Besides the connexion between the winter climate and the deciduous trees, we find a remarkable instance of a similar character in the great forests of pines which appear in our subtropical southern borders, and with them a cypress, chiefly a subtropical form of the Coniferae, which, unlike all others, loses its leaves in winter like northern trees. Thus, as our climate shows great extremes, so do the forms of vegetation adapted for it; and the very wide distribution of many of these forms may be considered by some a proof of the great past duration of the same climate, on the supposition that each species has spread slowly from a narrow centre of creation.

     Another marked climatic connexion is that of the moisture annually deposited on this province. I need not prove to those who have studied the subject that this is one of the most essential elements for the growth of forests of all kinds of trees. The dense growth of the ever-rainy tropics in some parts of the world, with the bare plains or deserts of other equally tropical countries which have little or no rain, prove that heat alone has little influence. On the other hand, our Lacustrian Province is densely wooded almost to the limits of perpetual frost, while the Steppes of Siberia, though of similar temperature, have little or no wood, being supplied with little moisture. Other proofs will be referred to hereafter. But besides the total amount, the equal distribution of the rains is important, as will be seen when we speak of regions having a dry and a wet season alternately.

     Without going further into this interesting subject here, I will say that I believe the reader, upon close study of it, will come to the conclusion that so intimate a relation exists between the trees and the climate of this Apalachian Province that their peculiarities always have and always must exist together. The disadvantages we feel in the climate are in this way compensated for, and will in time be looked upon as among the greatest natural advantages of the country. The midday heat, by rarefaction and evaporation, brings northward from the Gulf its abundant moisture. After several hours or days the cold westerly or northerly wind condenses this in rain over all that part of the continent included in this province, and in a less amount for some distance west and north of it. Thus in summer a continual supply is provided, while in winter the same winds which condense the summer rains become our most dreaded arctic blasts.

     The Lacustrian Province does not derive much moisture from the Gulf winds, but while its colder climate makes evaporation slower, it receives a share both from its own waters and probably also from the [[p. 276]] Pacific, as is shown by Lieutenant Maury in his Physical Geography of the Sea, though he makes the rains come down entirely too far to the south in regions known to receive but little.

     3. A few words on Florida will show its peculiarities in relation to trees and climate. The trees of Georgia extend for a long distance, some of them continuing as prevailing forms almost to its extremity. But they are gradually replaced by more tropical species, although where the most marked line of distinction exists is not well known. Most of those in the special list have been yet found only on Key West, but the examination of the almost unexplored interior, especially the Everglades, will doubtless extend their range materially. Forty-eight out of seventy-eight species found in it are evergreen, and all but four of these the broad-leaved tropical forms.

     Towards the middle of the State are found extensive prairies and treeless tracts, which are evidently connected with the alternation of wet and dry seasons, generally well marked in its climate. Though the wet season is in summer, yet the little interruption of growth by cold at other seasons makes their dryness influential. Its effects will be hereafter more fully alluded to.

     4. Now coming to the CAMPESTRIAN PROVINCE we find, as already stated, that no new forms of trees appear, while those found rapidly diminish and disappear towards the west. Thirteen species have not been traced west of its eastern border; about ninety extend pretty far into the Texan and Illinois regions, but only five or six get across the eastern limit of the Camanche and Dacotah regions, which, however, receive nine or ten more from the west and north.

     The Saskatchewan region, bordering close upon the well-wooded Lacustrian Province, may have a few more eastern species, and possibly more from the west, as there is evidence that it is better watered and approaches in character to the Illinois region.

     It will be observed that the southeast and northeast borders of this province form nearly a right angle with each other, and extending east into Michigan cause a wide separation of the Lacustrian and Apalachian provinces. This is one of the most well defined facts in the distribution of trees. A careful examination of the minute land office surveys has shown that the line is exceedingly distinct in Wisconsin and Minnesota, prairies prevailing to the south of it interspersed with oak-openings and groves of deciduous trees along the streams, while to the north pine and spruce forests with tamarack swamps cover the whole country, having the other Canadian trees with them. This is doubtless in great part due to the change in the character of soil and of the underlying rocks, which retain the moisture, while it is completely drained off to the south. Thus we have here a distinct division of the two eastern forest provinces, assisting to determine where it would be eastward were it not disguised by local irregularities of surface.

     The cause of the disappearance of trees in the Campestrian Province is, in a word, the deficient and irregular supply of moisture. I need not enter into the proofs of this, but refer to the records of meteorologists. It is true that this does not materially affect agriculture in the more eastern regions; in fact, most crops will succeed better with [[p. 277]] less rain than is necessary for most trees to thrive, and in some years there is even a greater supply of rain in the Texan and Illinois regions than eastward. But there are years and series of years of drought, when in their natural condition the forests take fire from the slightest cause and burn over large tracts. This was made even more general by the Indians, but since the white settlement has in great degree ceased and forests have been re-established. In the Apalachian region droughts have never been sufficient to keep trees from extending themselves as soon as a forest might be partially destroyed by fire, and thus the formation of prairies has been prevented. A consideration of the source of the rains will explain why the limit of prairies has its present direction. Coming north from the Gulf they are continually carried more and more eastward by the westerly winds; and as the greater part of the moisture is precipitated before reaching the Ohio river, the Illinois region is deprived for many years of its due share of rains.

     The Texan region lying quite west of the line of travel of those Gulf streams has to depend on less abundant sources for its rains. Now, as we go westward the supply rapidly diminishes until in the Camanche and Dacotah regions it is entirely inadequate to the growth of trees as well as of most cultivated products; and in some parts even grass and other herbage entirely disappear over vast tracts. From the great bend of the Missouri north, however, there seems to be an improvement in the country. On the banks of that river, above Fort Union, there is no long interval without trees as there is farther south on nearly all the streams, and on the Saskatchewan there is even less.

     The very porous character of the soil and underlying rocks assists much in this aridity of the country, and we therefore find that the line marking the junction of the carboniferous rocks of the Illinois region with the cretaceous and tertiary is a distinct limitation of many trees.

     When better known the geological character will help much in defining the physical geography of the surface of this province. In Texas the border of the Llano Estacado coincides with that of the Camanche region for a long distance. It is evidently more the retentiveness of the soil than its mineral composition that affects the growth of trees, for all soils contain more or less of their essential ingredients.

     Even the saline substances, which are supposed by some to make deserts of portions of the Great Plains, are rather the secondary effects of the climate; for if rains were abundant these salts would become diffused, and in their proper proportions enter into the structure of trees and other plants.

     It is certain, however, that even if the fires cease very few trees will ever be made to grow in these two arid regions.

     5. Coming now to what I have called the Rocky Mountain province, we find that the relations of climate and forest characteristic both of the Campestrian and Apalachian provinces are repeated, but combined in an entirely new manner.

     The high mountain ranges resemble the latter in their regular supply of rain, while near their summits the vegetation of the Athabascan region appears either in identical or allied forms, and still higher, [[p. 278]] near the limits of perpetual snow, the Esquimaux vegetation is almost precisely copied. But, on the other hand, the lower plains present every shade of succession, from the continuous forests of the Apalachian, through the rich prairies of the Illinois, to the barren deserts of the Camanche region. All these characteristics occur, however, in comparatively narrow belts surrounding isolated peaks or ranges, and the species of trees met with are nearly all distinct from those of the eastern provinces.

     Another distinctive character is in the fact that this province receives its rains from the west, (except, perhaps, some of the most eastern mountain slopes and those of Arizona,) and the supply of moisture is in direct proportion to the vicinity of any region to the Pacific, and the obstacles between it and that reservoir. Thus the Sierra Nevada cuts off almost all the rain from Utah, the little that reaches its eastern part being from local evaporation and what is intercepted by the lofty central ranges from the higher currents of the atmosphere.

     It thus happens that no constant elevation and no similar exposure has always the same amount of forest or other vegetation; local circumstances make every range of mountains and every valley differ somewhat from those around it.

     But, as a general fact, we find that those regions towards the north are the best supplied with moisture, and therefore best wooded--exactly contrary to the character of those regions which receive their moisture from the Gulf of Mexico.

     Near the Mexican boundary we enter the belt of rainless regions described by Lieutenant Maury; and here the supply is indeed precarious, though apparently more adequate to vegetation than in the Utah region.

     Although I have enumerated a long list of trees as first appearing on this line, it is in reality, for the most part, a treeless belt. Scattered individuals of numerous species occur, often limited to one narrow locality, as if merely outlines of more extensive forests in Mexico, or of what were once more extensive here, and have been destroyed by drought. Lieutenant Ives found great tracts of some of the more common trees thus standing dry and dead, as if killed within a recent period; but this is not the place to discuss these apparent changes in the climate of the country.

     The higher San Francisco and other ranges seem, however, to receive a better supply of moisture from the upper strata of the air, while their more impervious rocks probably retain it, and their cool summits condense around them enough moisture for the leaves of trees. I may remark here, that it would seem as if trees, rising high above the surface of the ground and expanding a vast evaporating surface of leaves to the air, require a greater degree of moisture in the air than herbaceous plants. They cannot, like the herbs of all arid regions, dry up and die down to their roots, to spring again with the wet season; they must retain vitality throughout or die. This constitutes a real physiological distinction between trees and herbs. The shrubs which live in those arid regions, presenting less evaporating surface, and having larger rootstocks in proportion, withstand droughts.

     6. The Californian region stands alone, unless combined with the [[p. 279]] Peninsula by natural affinities. Its mountains essentially resemble in climate and forest growth the Lacustrian province. Its valleys are like the Illinois and Texan regions, with this difference: that they have periodic dry and wet seasons, occurring at seasons opposite to those of Florida. Connected with this and with the mildness of its winters we again find a large proportion of broad-leaved evergreens, several genera, as the oaks and chestnut, which are deciduous in the east, being nearly all evergreen there. I should have mentioned that the same is the case in Arizona, though there a climate of tropical heat requires no interruption of vegetation.

     7. The Oregon region, and those north of it, as far as known, have very peculiar characters as well as points of resemblance to the eastern region. The climate is mild and equable, without excessive heat or cold, the rains abundant, and towards the coast excessive, with but short intervals, scarcely amounting to a dry season. The forests, while mostly composed of the northern forms of Coniferae, have also several broad-leaved evergreen trees and shrubs, which give them almost a tropical aspect. Both in climate and vegetation this western coast resembles much the coast of Europe, species and genera of trees being almost identically represented, and in about the same numbers. On the other hand, a similar analogy exists between the Apalachian forests and those of China.

     The marked differences in the character of the various mountains and valleys in these western regions may in future lead to more minute division than I have adopted, but present information does not warrant it now. Species formerly supposed to be limited to a narrow district have unexpectedly been found in others far distant, but of similar natural character, the intervening wide tracts being entirely destitute of them. Many facts go to show that the distribution of trees and forests was once very different, and is even now constantly changing, together with the climate; but as even the possibility of this change is doubted by some, except with such geological convulsions as can upheave mountains and sink continents, we must not be hasty in deciding the question.

     It may be objected to what I have said of the connexion between the cold winds and the constant rains of the Apalachian province that they do not coexist in the northwestern regions, which, together, form the CAURINE PROVINCE. But there are other causes which produce the precipitation of rain there. One is the cold northwest sea-breeze, which in winter precipitates the moisture brought by the southwest winds; the other is the cold air around the peaks of perpetual snow, which in summer produces at night a downward cold current with the same condensing effect.

     Thus the snowy mountains assist to improve the climate, and intercept much of the rain which, more to the north, seems to pass over the lower mountain ranges and to reach the Lacustrian province.

     Much more might be said respecting the connexions of forests and climate, but the general and best known facts are presented so as to lay the way for more complete observations. The provinces and regions may be classified in the following manner as to these connexions:

     [[p. 280]] A.--Completely wooded; rains equally distributed and abundant.--The Lacustrian and Apalachian.

     B.--Partially wooded; rains sometimes deficient.--Florida, Texas, and Illinois regions.

     C.--Almost woodless; rains always deficient.--Dacotah, Camanche, and Utah regions.

     D.--Plains and valleys unwooded; mountains wooded in proportion to their moisture, which is irregularly distributed, or periodical.--Rocky mountain province, except Utah?

     E.--Partially wooded; rains periodical.--Californian region and Mexican province?

     F.--Nearly all densely wooded; rains somewhat periodic, increasing in amount to the north, and with elevation.--Caurian province.

     From what examination I have been able to give the subject, I conclude that at least fifteen inches of rain during the growing season is essential to the vegetation of trees of all kinds. This, however, must vary with the retentiveness of the soil, the rapidity of evaporation, and the species of tree, some requiring much more than this. We have seen that, with its abundant moisture at all seasons, the Apalachian province has far the greatest number of species of trees, while the Caurian, though with perhaps more rain, unequally distributed, has much fewer. This is, however, connected also with its cooler summers, and, as before remarked, we have in the east a tropical forest with our tropical summers, in spite of arctic winters.

     From an accurate determination of the range of trees much interesting information on both climatic and other physical influences is expected to be derived. At the same time the distribution of all other plants and of animals must be studied in order to arrive at a knowledge of that harmonious system which undoubtedly prevails throughout the organic world, however obscured by the accidents of time and of external influences. Among them all we believe that forests will be found one of the most important, and respectfully invite the attention of the reader to its investigation.