HISTORICAL BACKGROUND OF THE FLORA OF THE PACIFIC NORTHWEST by LEROY E. DETLING Bulletin No. 13 Museum of Natural History University of Oregon Eugene, Oregon July 1968 ., . ' Funds for publication of Bulletin No. 13 provided by the National Science Foundation, grant GB 3670. 1 HISTORICAL BACKGROUND OF THE FLORA OF THE PACIFIC NOR THWES'T by LEROY E. DETLING Museum of Natural History University of Oregon ABSTRACT The modern flora of the Pacific Northwest is characterized by associations which show affinities to floras now occupying widely separated areas (Eurasia, South and Central America) and to floras shown by paleo- botanical evidence to have occupied all these areas, but particularly the American West. Distinct distribu- tion patterns, both in time and space, manifest themselves. These patterns are and have been influenced by topographic and climatic changes from the Cretaceous to the present. Three principal sources of associations are evident: evolution in situ; northern regions as shown in the Arcto-Tertiary Geoflora; western Mexico and the southwestern United States as shown in the Madro-Tertiary Geoflora. FOREWORD (This work has extended over a period of several years, and was unfinished at the time of the author's death in September of 1967. The following is essentially a report of the state of the work at that time. The sections of the rough draft have been rearranged, some repetitions clarified, references checked, an appendix add- ed and maps made. Many aspects of the prob- lem, particularly those dealing with Mexico, are obviously incomplete. The available data have been presented in hope that they will have some value in this unfinished form. Gratitude is here expressed to Stanton Cook of the De- partment of Biology, Jane Gray and J. Arnold Shotwell of the Museum of Natural History, all of the University of Oregon, and Orlin Ireland, for invaluable advice and assistance in the preparation of this manuscript for publication. July 1968 Mildred R. Detling) The scope of a work such as this, with its wide and varied ramifications, makes it neces- sary to rely heavily upon the published work of investigators in many fields. Especially have standard works on the geologic history of west- ern North America and the numerous publica- tions of outstanding paleobotanists, including palynologists, been an invaluable source of in- formation in the writing of this paper. The author has drawn upon his own field observa- tions and research in modern floristics in an at- tempt to explain the characteristics and more recent history of the vegetation of the Pacific Northwest. Field work in Mexico was partially financed by a grant-in-aid in 1960-61 from the Penrose Fund of the American Philosophical Society and by a grant from the National Science Foun- dation in 1965 ( GB-3670). The scientific names of plants have been used throughout this paper. A checklist of plants and their more common English names appears at the end of the work. Citation of the authority for specific names has been omitted. These can readily be found in the manuals of Peck (1941), Munz (1959), or Hitchcock et al. ( 1955) . Subspecific categories, whether they were published as varieties or subspecies, are all referred to as simple trinomials. 2 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON t' // ···~:: ,;:;,,,.,,, z u 0 u Ml . u ,.,,,,~ :::_'\I """' ~ . ' . (. ,, _t·: ;, ;.,, ,,,, __ I '·,, I t(.I I I I I I ,,. eK amloo~ ,, ,;. ,, No. 13 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 3 INTRODUCTION The region commonly designated as the Pa- cific Northwest is a natural one floristically, in spite of the great diversity of its habitats and the attendant diversity of life forms, vegetation types, and local plant communities. This is largely because throughout their geologic his- tory the various sectors of the region have in general undergone the same major climatic shifts and been subjected to the same floral in- vasions and evolutionary influences. And al- though its flora has received significant contri- butions in relatively recent times from adja- cent major floras, the resulting composite com- munities have become stabilized in ecological niches of varying extent. As delimited for this study the region ex- tends from the 42nd to the 52nd parallel of latitude north, and from the Pacific Ocean to the 116th meridian west. Thus it comprises the whole of the states of Oregon and Washington, a narrow strip of western Idaho extending to the base of the Bitterroot and Salmon River Mountains, and that part of southern British Columbia lying between the Selkirk Range and the Pacific Ocean, including Vancouver Island. It includes the northern fringe of the Klamath Mountains located in Oregon, the Klamath basin, and the northernmost part of the Creal Basin but excludes the mountain massif of central Idaho (Fig. 1). (Material on vegetation areas in northern and central Mexico was apparently intended by the author to provide background on the Madro-Tertiary elements in the Northwest flora. Such information as written in rough form by the author has been incorporated.) GEOLOGIC HISTORY The oldest land masses in western North America are members of a series of massive granitic batholiths forming the Nevadan oro- genic belt which includes ( 1) the Coast Moun- Figure 1. The Pacific Northwest. tains and Cascades of British Columbia, ( 2) the Okanagan Highlands, ( 3) the Idaho batho- lith of central Idaho, ( 4) the Blue Mountains, ( 5) the Klamath Mountains, ( 6) the Sierra Nevada and (7) the mountains of southern California and Baja California (King 1959, Clark and Stearn 1960). The first five of these form an arc, open to the west, of volcanic mass- es which were intruded from middle Jurassic to middle Cretaceous times along what was then the western margin of the North American con- tinent (Fig. 2). None of them has ever been completely submerged in the sea since the Cre- taceous. Mesozoic land plants have left re- mains in Jurassic rocks of southwest Oregon ( Chaney 1956). During early Cenozoic times the great gulf bounded by the orogenic arc was largely filled in by sedimentation and some uplifting. Fossil leaf deposits assigned to the Oligocene have been found as far west as the Willamette valley (Sanborn 1935, Lakhanpal 1958), giving evi- dence of a terrestrial habitat there by that time. The central and southern portions of the Cascade Range were formed by uplift and vul- canism beginning in the Eocene and continuing through the Tertiary. This range formed direct- ly across the recess of the orogenic arc. Vol- canic activity beginning in the Pliocene, but ap- parently taking place largely in the Pleisto- cene, formed the high ridges and volcanic cones from Glacier Peak and Mount Baker in Wash- ington to Mount Lassen in California. Mean- while, during the last half of the Cenozoic, the Oregon and Washington sectors of the Coast Range, including the Olympic Mountains, were formed by the upfolding of the sediments of the early Cenozoic coastal plain. During the Miocene and long before the Cas- cade Range had reached its final elevation ' there occurred the great outpouring of the ba- salts that form the Columbia Plateau. It is be- lieved that the center of the flows may have been in southeastern Washington and northeastern Oregon. The molten mass did not issue from volcanoes but rather from fissures, and eventu- ally formed beds thousands of feet in thickness. 4 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 Figure 2. The Nevadan orogenic arc, formed by batholiths composed of igneou,.s rock in Jurassic and Cretaceou,.s times ( after McKee, et al., 1956, and Smith and Stevenson, 1956 ). Such in brief were the geologic events that produced the modern physiography of the Pa- cific Northwest whose major features have de- termined the climates of the region and marked the pathways of its plant migrations. PHYSIOGRAPHIC FEATURES The modern physiography of this region is dominated by a series of mountain ranges with intervening systems of river basins, in general running north and south, and an extensive pla- teau region (Fig. 1). The westernmost moun- tains, the Coast Range, parallel the coast throughout the whole of the region. It joins the Klamath Mountains in southwestern Oregon, includes the Olympic Mountains of northwest Washington, and continues north of the Strait of Juan de Fuca as the Vancouver Island Moun- tains. It is separated from the ocean by a nar- row strip of flat land which tapers to nothing in the south where it borders the Klamath Moun- tains and in the north along the western edge of Vancouver Island. Farther inland a higher range extends north- ward from the eastern edge of the Klamath Mountains; at the northern end of Puget Sound it swings northwestward along the coast of the British Columbia mainland. In British Colum- bia it is usually called the Coast Range, al- though frequently known as the Cascade Range. To minimize confusion, in this paper the entire range shall be referred to as the Cascades. Between these ranges lies a series of river basins and troughs. North of the Klamath Mountains and draining into the sea are the Rogue and Umpqua River basins. The Willam- ette and Puget troughs are separated by the gash cut through the mountains by the Colum- bia River. The northward extension of the Pu- get trough is partly covered by the marine waters of Puget Sound and the Georgia, John- stone and Queen Charlotte Straits. East of the Cascade Range is a broad ex- panse of tableland. The portion occupying south central and southeastern Oregon consti- tutes the northernmost extension of the Great Basin of Nevada and Utah. The northern Ore- gon and eastern Washington part form the Co- lumbia Plateau. This tableland is bounded on the east by the foothills of the Bitterroots in northern Idaho, the Blue Mountains in Oregon and Washington, and by way of the Snake River Plains of southern Idaho, by the Rocky Moun- tains. Topographically the Columbia Plateau merges at the north through the north-south ori- ented valleys of the Okanagan Highlands with the plateau of south central British Columbia, dissected by the upper drainage basins of the Fraser, Okanogan and Columbia Rivers. The importance of the north-south trend of our major physiographic features from the standpoint of vegetation migrations and evolu- tion cannot be overestimated. The effect of the mountain ranges in intercepting moisture from 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 5 the prevailing westerly winds is to produce climatic belts with extreme differences in pre- cipitation and temperatures. At the same time the mountain and basin axes provide nearly un- broken pathways for the north-south move- ments of plants. So far as topography is con- cerned the flora of the Mexican Meseta Central (Fig. 6) has had free access across southern and western Arizona, Nevada, eastern Oregon and Washington, and well into the plateau country of British Columbia. Also plant spe- cies moving northward from the Great Valley of California, once they have negotiated the warm valleys and canyons of the Klamath Mountains, have had only a few low transverse ranges to impede their progress up the series of basins ending in the lowlands of Vancouver Island and adjacent mainland British Colum- bia. On the other hand, the two major mountain axes of the Pacific Northwest reach well into the subarctic regions of northern Canada and Alaska. South of the 52nd parallel the only places where the boreal life zones of the Cas- cades have been broken are where the Fraser, Columbia and Klamath Rivers have cut their gorges. In like manner the boreal zones of the Blue Mountains are nearly continuous with those of the Rockies, separated only in the vici- nity of the Snake River canyon. CLIMATE The mild oceanic climate west of the Cas- cades is created by latitude and proximity to the sea. Winter temperatures are highest and summer temperatures in general lowest along the narrow coastal lowland strip. In the series of basins lying east of the Coast Range and Vancouver Island Mountains temperatures are still moderate but the extremes are greater. East of the Cascades the climate is continental in nature, with extreme winter minima and summer maxima. The Cascade axis and Blue Mountains, because of their elevation, experi- ence very low temperatures during the winter summer temperatures tend to be higher in the south and lower in the north. A series of rain shadows occurs in the path of the prevailing westerly winds. Most of the precipitation throughout the Pacific Northwest occurs during the winter months, from Novem- ber to April.West of the Cascades summer rain- fall is negligible. East of the range the ratio of summer to winter precipitation is appreciably higher. This is particularly true of the interior plateau of British Columbia. Farther south, es- pecially in the mountains, the ratio is increased by frequent local thunderstorms. Various combinations of winter and summer temperatures along with varying precipitation patterns have given rise to 19 distinct climatic areas in the Pacific Northwest, which corres- pond quite closely to the major vegetation areas of the region. The writer (Detling 1948a, b) has devised a vegetation map ( Fig. 3) for Ore- and have moderately low summer maxima. Figure 3. Vegetation areas of the Pacific North- Within any north-south belt both winter and west. 6 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 gon and Washington based upon these temper- ature and rainfall data along with length of growing season. Since these areas have both climatic and floristic bases, it is apropros to in- troduce them here in order to present in detail some of the climatic differences which appar- ently have been most instrumental in establish- ing a distinct vegetation type in each area. For purposes of this study the vegetation map has been extended to include southern British Co· lumbia. One change in area name has been made. The Rogue and Umpqua basins, although physiographically distinct, are here treated as a single vegetation area, as they were in the original paper. This will simplify later refer- ences to the floristic characteristics of the area. MAJOR VEGETATION TYPES The vegetation areas referred to in the fore- going paragraphs represent the end results of floristic evolution in relatively restricted physi· ographic and climatic units. In the discussion of the long-time evolutionary processes with which this study is dealing we must make use of more general and widespread vegetation types, such as can be readily and obviously cor- related with identical or similar types in re· gions sometimes well beyond the limits of the Pacific Northwest. For this purpose we shall use the following classification of our major vegetation types: A. Boreal types I. Mesic Conifer Forest II. Boreal Forest Ill. Alpine Fell-fields B. Austral types IV. Pine-oak Forest V. Juniper-sagebrush Woodland VI. Grassland VII. Chaparral The accompanying map (Fig. 4) is designed to show in a general way the relative positions and extent of these types, except the Alpine Fell-fields. In our region these latter tend to occur in small areas above timber line. This classification has been found particu- larly useful in this study because (a) the units are easily recognizable, (b) they are rela- tively homogeneous throughout, ( c) they have similar distribution changes, ( d) most of them are already recognized under some name by plant ecologists, ( e) the names selected are self-explanatory, ( f) as will be more evident later, the component species and genera of each type have apparently originated in the same geographic regions of the world and have fol- lowed the same migration routes throughout their history. For each type there is a list of characteristic species, tabulated to show its ap- parent relationships to floras beyond the Paci- fic Northwest. The species selected are all at least moderately abundant in their type, and insofar as possible are widespread within it. The lists are sufficiently extensive and care- fully selected to furnish valid basic communi- ties for the following discussion. MESIC CONIFER FOREST Of the vegetation types in the Pacific North- west the most basic from all points of view is the Mesic Conifer Forest. It is the most wide- spread, and generally speaking the most near- ly continuous throughout its range. In the north- west this forest, largely coincident with the Douglas fir forest, roughly forms an arc open to the south (Fig. 4) ( Reusser 1960). One arm of the arc extends from western British Colum- bia southward west of the Cascade Range. An eastern arm follows the Rocky Mountains southward from eastern British Columbia. The central curve of the arc connects these two arms across the valleys and low ranges of central British Columbia. Along the coast the type ex- tends from near sea level to an elevation of 4000 feet or more. In the Rocky Mountains both its lower and upper limits are higher. In the northern part of its range it forms a rela- tively narrow belt altitudinally. Excluded from the mesic forest are a narrow strip along the Oregon, Washington, and Vancouver Island coast occupied by Boreal Forest, and the lower 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 7 D MH i,• f.o n ife r t'oreo l Figure 4. Vegetation types of the Pacific North- west. elevations of the Rogue River basin occupied by Pine-oak Forest or Chaparral. It occurs at middle elevations in the Blue Mountains and the Okanagan Highlands. The arc thus formed partially encircles an area, to be discussed later, occupied by a fringe of Pine-oak Forest and by extensive areas of Grassland and Great Basin Juniper-sagebrush Woodland. The dominant species in this type is most commonly Pseudotsuga menziesii. This is es- pecially true west of the Cascade axis. East of the mountains Pseudotsuga is usually present in greater or lesser abundance, but dominance may be shared by or lost to other species, such as Tsuga heterophylla, Abies concolor or A. grandis. Local variations in the environment, such as streamside conditions, account for vari- ations in the associations within the framework of the major type. The species most commonly associated with Pseudotsuga west of the Cascades and compris- ing a significant part of the lower stratum of the forest are Acer circinatum, Gaultheria shal- lon, Mahonia nervosa, Corylus cornuta califor- nica and Polystichum munitum. Where the type occurs in eastern British Columbia and northern Idaho the subdominant association varies somewhat from the foregoing-Acer circinatum is replaced by A. glabrum, Mahonia nervosa by M. repens, Gaultheria shallon and Polystichum munitum are absent, while Popu- lus tremuloides and Betula papyri/era occiden- talis are added to the community. The altitudi- nal narrowing of the mesic forest belt in central British Columbia emphasizes the infiltration in that area of species which normally are charac- teristic of adjacent zones, particularly of the Boreal Forest above it, accounting for the abundance not only of the species of Betula, Acer and Populus already mentioned but also of Arctostaphylos uva-ursi, Vaccinium ovali- folium, Cornus canadensis and Pachystima myrsinites. However, a large number of other species characteristic of the forest west of the Cascades do persist east of the range, so that the general aspect of the forest is not essential- ly altered. The Mesic Conifer Forest has a relatively wide tolerance of climatic extremes. In general it occupies an area of abundant rainfall or snow accumulation, with mild or at least not excessively cold winters, and cool to moderate- ly warm summers. It is best developed and most distinctive in the oceanic climate west of the Cascades of Oregon and Washington. In the Douglas fir forests of this region annual preci- pitation ranges from 25 to 140 inches. In the continental climate of the interior the range of precipitation is approximately from 25 to 40 inches annually. The 2.5 inches which consti- tutes the lower limit for this type is apparently critical on both sides of the Cascades; below this figure the Douglas fir forest is replaced by a woodland of yellow pine and oak. East of the Cascades the mesic forest is probably controlled at its upper limits, both altitudinally and lati- . I ' : ' . ' ' ' ' ' 8 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 tudinally, by temperature rather than precipi- tation. This seems to be true also along the coast where the type is bordered by a modified Boreal Forest, and where temperatures, espe- cially those of the summer months, are moder- ately cool. Temperature conditions within the range of the Mesic Conifer Forest vary considerably. Near the southern Oregon coast January means are about 45°F., but at middle elevations east of the Cascade crest they may be as low as 25°F. July means range from 60° along the coast to about 65 ° at some places in eastern Oregon. The basic community of the Mesic Conifer Forest is comprised of the thirty-nine species listed in Table 1. The significance of the distri- butional units used will become evident with further discussion. In this and following basic communities all taxonomic units considered subspecific have been included under the species listed. The sub- specific breakdown of these species will be dis- cussed later in this work where such informa- tion is pertinent. The distributional data summarized in Table 1 show some features that become of interest when we look for the origins of the Mesic Coni- fer Forest. The type as presently constituted is centered in the temperate part of western North Ameri- ca. None of the arboreal species and only two shrubs occur farther north than the Alaskan Panhandle or northern British Columbia. No trees and three shrub species reach central or eastern North America. The herbaceous spe· cies are in general more widespread toward the north and east. Of twenty-one species nine are either circumpolar or of northern Alaskan and Canadian distribution. Seven of these reach eastern North America. Four others occurring in the eastern part of the continent, while not now reported from the far north, probably did flourish there at one time. In the other direc- tion, while all thirty-nine species representing the basic association of the type extend into California or northern Baja California, only three, two trees and one shrub, continue into Mexico. Further light is shed on the relationships and possible history of this type by the supplemen- tary data on the general distribution of the genera represented in the basic community (Table2). The foregoing data show that the Mesic Con- ifer Forest not only is definitely northern in the present distribution of the species that comprise it, but also the general distribution of the gen- era represented gives evidence that it has had a northern origin in the past. The close relation- ship of the Mesic Conifer Forest of western North America with the flora of eastern Asia is emphasized by the fact that of the thirty-four genera in our list thirty-two occur today in the latter region. Only twenty are reported to occur in Europe. This distribution pattern has been cited by botanists, including many paleobotan- ists, as evidence of a former land bridge be- tween Alaska and Siberia, allowing a free movement of plant populations between the two continents (Hulten 1937). The tree and shrub members of the association, while they were undoubtedly originally continuous with populations in other parts of the world, particu- larly with those of eastern Asia, have been iso- lated for so long a period by extreme frigid conditions to the north that they have followed an independent evolutionary course. This seems to be true to a lesser extent of the herbaceous members of this type, since at least seven out of the twenty-one herbaceous species are still com- mon to both continents. Possible additions to Mesic Conifer Forest basic community: Achlys triphylla Aconitum spp. Actaea spicata arguta Allotropa virgata Anemone deltoidea Aplopanax horridum Aquilegia formosa Aruncus sylvester acuminatus Asarum caudatum , . ·- '· j . L, ' ,. [: . 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 9 Boykinia major Cimicifuga elata Circaea pacifica Coptis laciniata Corydalis scouleri Cynoglossum grande Dicentra formosa Epilobium angustifolium Fragaria bracteata Galium aparine Galium boreale Galium trifidum Galium triflorum Geum macrophyllum Heuchera micrantha pacifica Hydrophyllum tenuipes Lathyrus polyphyllus Lonicera ciliosa Lysichitum americanum M enziesia f erruginea Mertensia platyphylla subcordata M itella caulescens Monotropa uniflora N emophila parviflora Physocarpus capitatus Pleuricospora fimbriolata Populus trichocarpa Prunus emarginata Pyrolaceae genera and species Rhamnus purshiana Rosa gymnocarpa Rosa pisocarpa Rubus spectabilis Sambucus glauca Satureja douglasii Stachys spp. Symphoricarpos albus Synthyris reniformis T ellima grandiflora Thalictrum hexandra Tiarella unifoliata T olmiea menziesii Trautvetteria grandis Trientalis latifolia Urtica holosericea Urtica lyallii V icia americana BOREAL FOREST In this paper the term Boreal Forest is ap- plied to all Pacific Northwest forest communi- ties at altitudes above the Mesic Conifer For- est, or at lower altitudes when occurring under conditions of cooler summer temperatures where this is apparently the factor giving rise to such a community ( Fig. 4). In the northern- most part of its range it is composed of the for- ests extending from the northern limit of trees in Alaska and the Yukon to the Douglas fir for- ests of the mesic conifer belt in central or southern British Columbia, reaching sea level in southern Alaska and northwest British Co- lumbia. It extends southward in its typical form along the Cascade Range and the Sierra Neva- da, and down the Rocky Mountains as far as Arizona and New Mexico, including the Blue Mountains of Oregon, occupying a belt pro- gressively higher in elevation as it goes south- ward. A narrow strip of Boreal Forest extends from the main body in British Columbia south along the coast as far as northern California. This strip, rarely more than ten miles wide south of Vancouver Island, has been infiltrated by many species from the adjacent Mesic Coni- fer Forest, but contains enough characteristic species to maintain its distinctiveness and show its close relationship to the rest of the type as it occurs at middle or higher altitudes in the Cascade Range. The most reliable indicators of the Boreal Forest are Pinus contorta in its two subspecies, Picea engelmannii or P. sitchensis, Tsuga mer- tensiana, and one or more species of Abies, e.g., A. lasiocarpa, A. procera or A. amabilis. In central and eastern British Columbia Betula papyri/era is a common indicator. Forming the shrub stratum under these dominants are most commonly found species of Vaccinium and Ledum, and Acer glabrum douglasii, while the most characteristic herbaceous species are Polygonum bistortoides, Pedicularis spp., Ane- mone occidentalis, Rubus lasiococcus, R. peda- tus and Xerophyllum tenax. Official climatic data from stations located in the Boreal Forest zone are scarce, except for 10 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 TABLE 1. DISTRIBUTION OF THE BASIC COMMUNITY OF THE MESIC CONIFER FOREST OUTSIDE OF THE PACIFIC NORTHWEST Arctic & Northern Cent. & S. subarctic Rocky Cent. & Rocky Great N. & Cent. sw.u.s. N.A. Mts. NE.N.A. Mts. Basin Calif. & Mexico A bies concolor x x x x A bies granilis x x Acer circinatum x Acer macrophyllum x Adenocaulon bicolor x x x x Alnus oregona x Calypso bulbosa x x x x x Chimaphila umbell,at,a x x x x x Clintonia uniflora x x Corallorhiza striata x x x x Cornus nuttallii x x Camus stolonifera x x x x x Corylus cornuta x x x x Disporum oreganum x x Gaultheria shallon x Goodyera oblongif olia x x x x Hieracium albiflorum x x x H olodiscus discolor x x x Hypopitys latisquama x x x x x Linnaea borealis x x x x x Listera convallarioides x x x x Listera cordata x x x x x Mahonia nervosa x x M ontia sibirica x x x x Osmaronia cerasif ormis x Osmorhiza chilensis x x x x x Oxalis oregana x Polystichum munitum x x x Pseudotsuga menziesii x x x x Rhododendron macrophyllum x Rubus parviflorus x x x x x Smilacina racemosa x x x x Smilacina stellata x x x x x Spiranthes romanzoffianum x x x x x Thuja plicat,a x x Trillium ovatum x x x Tsuga heterophylla x x V accinium parvif olium x x Viola glabella x x x 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 11 TABLE 2. DISTRIBUTION OF THE GENERA REPRESENTED IN THE BASIC COMMUNITY OF THE MESIC CONIFER FOREST < z rn 0 < < 0 .... . ... ti z z ~ < ... 0 as ... Q) Q) .... en ..0 .... ::, .... as '"O as ... Q) rn & i ~ .... o(j 0 as as ~ ... .... .~ 8 & rn as .... Q) Q) .... ~ ~ ... 0 .... ::, ... ~ ~ Q) ~ < z E-t Abies x x x x Acer x x x Adenocaulon x x x Alnus x x x x x Calypso x x x Chimaphila x x x Clintonw x x x Corallorhiza x x x Cornus x x x x Corylus x x x x Disporum x x x Gaultheri,a x x x x Goodyera x x x Hieracium x x x x Holodiscus x x Hypapitys x x x Linnaea x x x x Listera x x x x Mahonw x x Montw x x x Osmaronw x Osmorhiza x x x x Oxalis x x x x x Polystichum x x x x x Pseudotsuga x x Rhododendron x x x Rub us x x x x Smilacina x x x Spiranthes x x x x Thuja x x x Trillium x x x Tsuga x x x x Vaccinium x x x x x Viola x x x ' 12 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 the coastal strip. The following account of its climatic features is based upon the official rec- ords of the few stations available, and from ex- trapolations made from these. These sources probably give a reasonably accurate picture of the climatic environment of the type. In central British Columbia and the adjacent Rocky Mountains the middle and higher eleva- tions, at which the Boreal Forest occurs, receive approximately from 25 to 40 inches of precipi- tation annually. Much of this falls as snow. In general, precipitation in this region is evenly distributed throughout the year. Southward from British Columbia in the mountains of northern Idaho and the Blue Mountains of Ore- gon and Washington the annual preciptation remains about the same, but as was pointed out earlier, a smaller percentage of it falls during the warm season. Throughout the length of the Cascades and the coast strip, annual precipita- tion in the Boreal Forest is much higher than it is farther inland, ranging from nearly 60 inch- es in the south ( Crater Lake) to 100 inches or more in many places in British Columbia and on the coast. The Boreal Forest is tolerant of much lower winter temperatures than is the Mesic Conifer Forest, and requires significantly lower sum- mer maxima. In the mountain ranges through- out the Northwest and in the interior plateau of British Columbia the mean January tempera- ture in the type probably varies from about 10° to 26°F. Along the littoral belt this mean varies from 38° to 46°F. Insofar as we can determine, the July mean temperatures vary much less than do those in January. The July mean in the mountains and interior probably ranges from 52° to 58°F., while along the coast it runs from 57° to 61 °F. The following list of thirty-seven species (Table 3) is representative of the Boreal For- est, and will serve as a basic association like that of the Mesic Conifer Forest. The list has been selected to include the dominant or other- wise common species that characterize the type in as many of its areas as possible, i.e., the lit- toral belt, the Cascades, and the Blue and Rocky Mountains. The distribution of these species as tabulated shows the following pertinent features: Only three species are reported from as far south as Mexico; on the other hand thirty-one are either circumboreal or at least occur in Alaska and northern Canada; all but two occur in the area comprised of southeastern British Columbia and northern Idaho. These two are largely restricted to the littoral belt. The foregoing data point to the fact that the Boreal Forest is distinctly northern in its dis- tribution and in all probability was northern in its origin, as was the Mesic Conifer Forest. Fur- ther evidence in support of this theory is sup- plied by the present distribution of the other species in the genera represented in the type (Table 4). Seven genera only are repeated from the mesic forest of Table 2; the remain- ing thirty-one are new. A small but interesting group of species now closely associated with the coastal strip of Bor- eal Forest in Oregon appears to have a history relating them to the Madro-Tertiary Geoflora. The group is made up of Arctostaphylos colum- biana, Baccharis pilularis consanguineus, Garrya elliptica, Myrica californica, Rhodo- dendron occidentale and U mbellularia cal if or- nica. They all follow the coast northward as far as Coos Bay, constituting an important part of the flora in this strip. Baccharis, Garrya and Myrica continue more or less sporadically along the Oregon coast, and Arctostaphylos as far as southern British Columbia. Only rarely are any of them found beyond the Klamath Mountains in the inland basins: Rhododendron occidentale in the Rogue valley at Myrtle Creek and on Nickel Mountain in the Umpqua basin, Umbellularia in the Umpqua basin, and Arcto- satphylos columbiana occasionally in the Cas- cades of central western Oregon. With the exception of Rhododendron, all the genera cited above are reported by Axelrod ( 1958) to occur among the fossils associated with the Madro-Tertiary Geoflora. Further- more, as a group they occur today in central and southern California, sometimes as far in- land as the Sierra Nevada, in formations rang- r. TABLE 3. DISTRIBUTION OF THE BASIC COMMUNITY OF THE BOREAL FOREST 13 en ..... ~ >, cd ~ 0 '"'O ~ < ..... :i ~ :i en < ..... «I o?J i::: ~ :i rn ..... <1) ~ - <1) i::: cJ u 0 .~ - «I '"'O ,.. o?J ..... rn «I ,.. «I 0 i:Q o?J p.. ,.. «I <1) 0 0 ~ <1) ~ Ei <1) ,.. ,.. ::::: rn ca ~ ::, i::: ~ ...=: ::, 0 ..... «I ~ <1) 0 ~ i:Q ...:l u u [f) u 0 A bies lasiocarpa x x x x x Acer glabrum douglasii x x x x x x Alnus tenuifolia x x x x x Anemone occidentalis x x x x x Arctostaphylos uva-ursi x x x x x x x x x Betula papyri/ era x x x x Caltha leptosepala x x x x x Cornus canadensis x x x x x x x x x Empetrum nigrum x x x x x Mt. Rainier Eriophorum chamissonis x x x x Eriophorum gracile x x x x x x x Gaultheria ovatif olia x x x x x H abenaria dilatata x x x x x x x x Heuchera glabra x x x x !uniperus communis x x x x x x x x Kalmia polif olia x x x x x x Ledum palustre x x x x x Lonicera involucrata x x x x x x Chihuahua Luetkea pectinata x x x x x M aianthemum dilatata x x x x x x Col. Gorge, Saddle Mt. Pachystima myrsinites x x x x x Coast Range Pedicularis groenlandica x x x x x x x Picea engelmannii x x x x Picea sitchensis x Pinus contorta x x x x x x Polygonum bistortoides x x x x x x x x Saddle Mt., Marys Pk. Populus tremuloides x x x x x x x Chihuahua Pyrola secunda x x x x x x x x Mt. Orizaba, Tillamook Rhododendron albiflorum x x x x Rubus pedatus x x x x Saddle Mt., Tillamook Shepherdia canadensis x x x x x Trientalis arctica x x x T suga mertensiana x x x x x x Vaccinium ovalifolium x x x x x x x V accinium ovatum x x V accinium scoparium x x x x V eratrum viride x x x x x Xerophyllum tenax x x x x x x 14 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 TABLE 4. DISTRIBUTION OF THE GENERA REPRESENTED IN THE BASIC COMMUNITY OF THE BOREAL FOREST (.) ..... - (.) 11) ... ~ < .a i:: o:s 0 "§ ..... ... :i ... N 11) 11) [f) ~ i:: t i:: 11) ~ 8.. rtJ ... c., <:.) 11) -~ ..i:: - 11) 0 ·a rtJ o:s < E-< - ... 11) ·.;; <:.) ::::, ... 0 ~ :i :i ... 0 ::::, ... < < [f) ~ E-< Abies x x x x x Acer x x x Alnus x x x x x Anemone (PulsatiUa) x x A rctostaphylos x Betula x x x x x Caltha x x Cornus x x Empetrum x x x x x x Eriophorum x x Gaultheria x x x x Habenaria (Limnorchis) x Heuchera x !uniperus x x Kalmia x Led um x x Lonicera x x x x x Luetkea x x M aianthemum x x x Pachystima x P edicularis x x x x Pie ea x Pi nus x x x x x Polygonum x x x x x x Populus x x x x x Pyrola x x x x x Rhododendron x x x Rub us x x x Shepherdia x Trientalis x x x x Tsuga x x x x Vaccinium x x x x x x Veratrum x x Xerophyllum x x 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 15 ing from chaparral and woodland to mixed conifer forests. All the evidence points to these species, or at least their immediate ancestors, having migrat- ed northward through California with the rest of the Madro-Tertiary Geoflora, reaching the Klamath Mountains and southwestern Oregon by early and middle Pliocene. There, in re- sponse to increasing cold and humidity, they became sufficiently adapted to this type of cli- mate and edaphic conditions to invade the mild, wet coastal region of the Del Norte Area and the more northern coastal strip, where they have persisted until the present and mingled with the members of the Boreal Forest. The sporadic occurrences of Baccharis pilularis and Garrya elliptica northward suggest that this group at one time was common as far up the coast as western Washington, probably paralleling the advance of other Madro-Tertiary species through the western basins during the xero- thermic maximum. It was possibly in conjunction with the coast- ward and northward movement of these species, and this also perhaps at the xerothermic maxi- mum, that Quercus garryana and Rhus diversi- loba, dominant features of the interior oak woodlands, advanced up the coastal strip of the Del Norte Area (Fig. 3)-Quercus as far as Pistol River, Rhus as far as Cape Sebastian. It is conceivable that an environment which at first glance appears to be moist and cool may in actual fact have an effect which is quite the reverse. A soil which is sandy and very well drained in an area where summer rainfall is negligible creates a xeric situation as far as plants growing in it are concerned. 1£, in addi- tion, salinity of the environment is increased by sea water in the form of spray in the winter, the total effect is xeric ( Reusser 1960). Sum- mer temperatures, while tempered by proximi- ty to the ocean, probably exert more than aver- age influence if moisture is insufficient. The net result is a xerothermic condition. When this condition is extreme, as on the west coast of northern Mexico, cactus plants extend from the desert down into the upper margin of the beach- es. In the Northwest Arctostaphylos columbia- na, Baccharis pilularis and Garrya elliptica mix with those members of the Boreal Forest which tolerate this situation, and in the Del Norte Area Quercus garryana and Rhus diver- siloba of the oak woodlands appear in the coast- al strip. ALPINE FELL-FIELDS The Alpine Fell-fields comprise the treeless vegetation type that occupies the zone above the Boreal Forest, where the absence of trees and any but prostrate shrubs is the result of the ex- tremely rigorous conditions resulting from low winter temperatures, short growing season, deep snow cover and high wind velocities. In the southern Oregon Cascades, as at Cra- ter Lake or on Mount McLoughlin, timber line, or the lower limit of fell-fields vegetation, is at an altitude of about 7500 feet. This line is low- er in the north, and at the 52nd degree of lati- tude it varies from 4500 feet in the Canadian Cascades to 6500 or 7500 feet in the Selkirk Range ( Kendrew and Kerr 1955) . Still farther north it merges with the arctic tundra which continues on to sea level at the Arctic coast. In southern British Columbia, fell-fields occur in extensive areas in the Cascades and in the Selkirks. In Washington and Oregon the areas are smaller in size and more widely scattered southward through the Cascades the Blue Mountains, and in the W allowas. They are not found in the Coast Range south of the Oympic Mountains. Accurate climatic data are even more scarce for the fell-fields than for the Boreal Forest. Annual precipitation in most cases is about the same as in the Boreal Forest immediately be- low it, as is also its seasonal distribution. It is certain that January means are much lower than the 26°F. we have given as the upper limit in the Boreal Forest of the mountains, and that the July average is also lower, with night tem- peratures frequently falling below the freezing point throughout the summer. The number of plant species above timber line is relatively small compared to those of other vegetation types. The fourteen species 16 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 TABLES. DISTRIBUTION OF THE BASIC COMMUNITY OF THE ALPINE FELL FIELDS "' "O "' c "' ca u Cl) :i ..... 0 "' ..0 o(j u s ..... "' ~ ::s ~u "' "' >, ..... ..... < 0 u Cardamine bellulifolia x x Cassiope mertensiana x Collomia debilis x Hulsea algida Hulsea nana Lupinus arcticus x Oxyria digyna x x x Phyllodoce empetriformis x x Phyllodoce glanduliflora x Polemonium elegans Polemonium viscosum P olygonum phytolaccae f olium Saxifraga bronchialis x x Saxifraga tolmiei x x (Table 5) are closely restricted to these areas and together constitute a characteristic and well-defined community. It is possible that the present geographical distribution of the species, as well as the gener- al distribution of the genera represented (Table 6), indicates the northern origin of the alpine flora of the Pacific Northwest. It is inter- esting to note, however, that of the fourteen species representing it here, only three are cir- cumboreal and only seven occur in northern Canada or Alaska. This is in contrast to the relatively large percentage of Boreal Forest species which extend far northward, and sug- gests that the alpine species either have been isolated for a longer time and have undergone rigorous selection, or that the species of Collo- mia, Hulsea, Polemonium and Polygonum arose in the western mountains. PINE-OAK FOREST The Pine-oak Forest has two major phases in the Pacific Northwest. A western phase occu- en ~ cd ~ "O Cl) ...... ::s :i ~ ~ o(j o(j "' u cj "O Cl) ..... ~ "' ~ ~ c u ~ u "' "' ca "' 0 ,;::: ~ u u i::i::: < x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x pies a considerable area at lower elevations in the Rogue River basin and occurs in scattered areas of variable extent in the succession of river basins reaching to the south end of Van- couver Island. The dominant species are com- monly Pinus pondersoa, P. jefjreyi, one of sev- eral species of Quercus, and Arbutus menziesii. This mixed forest continues southward through California, across Arizona, and into the Sierra Madre Occidental of Mexico. In the Sierra Madre it typically occurs as a mixed communi- ty, although under extreme conditions of cli- mate it may appear as pure stands of oak or pine. In the Pacific Northwest, while there is still considerable mixing of the dominants, there is a strong tendency for the oaks and the pines to segregate into altitudinal belts with the former concentrated at lower levels while the pines occupy a belt above them. Arbutus menziesii mixes freely with both. The eastern phase of the type is the yellow pine forests east of the Cascade Mountains. It forms an arc within and roughly paralleling that of the Mesic Conifer Forest (Fig. 4), ex- 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 17 TABLE 6. DISTRIBUTION OF THE GENERA REPRESENTED IN THE BASIC COMMUNITY OF THE ALPINE FELL FIELDS < -; z ... Ill = = rj ..... Ill ..... -~ < en ~ en z < Cardamine Cassiope x x Collomui x Hulsea x Lupinus x Oxyrui x x Phyllodoce x x Polemonium Polygonum Saxifraga tending from northeastern California through the Klamath basin and eastern Oregon, Wash- ington and British Columbia as far as the lati- tude of Kamloops and Shuswap Lakes, and from there south and east into northern Idaho and through the Blue Mountains. The dominant species in the eastern phase is Pinus pondero- sa; Arbutus is entirely absent and Quercus oc- curs only in a limited area in the vicinity of the Columbia Gorge. Further, a large percentage of the shrubby and herbaceous associates of the western phase is absent from the eastern. In spite of this there are reasons for treating the two phases as a single unit. They are not dis- junct populations, being confluent in a broad area in northeastern California where the grad- ient of occurrence and dominance of the constit- uent species can be observed. A sufficiently sig- nificant number of species is common to both phases; besides Pinus ponderosa the most obvi- ous are probably Ceanothus velutinus and Arc- tostaphylos patula. Finally the eastern and western forms of Pinus ponderosa are appar- ently identical throughout the Pacific North- west, and although the eastern population does extend southward to Arizona east of the Sierra Nevada, it is distinct from though closely re- .;:; ti "' Ill ..... = "' u 0 -§ ......... N Ill rJ) s ci. ~ < en ~ & u u ...c:: .... .... ..... 0 0.. ti :::, ..... 0 z ..... 0 :::, ..... < rJ) ~ E-< x x x x x x x x x x x x x x x lated to the more southern segregates that have sometimes been made of the species, viz., Pinus arizonica, P. latifolia and P. brachyptera, thus minimizing the possibility that the eastern phase of the type is more closely related to a Great Basin or Rocky Mountain flora than to the Californian (Martinez 1948, Mirov 1967). The Pine-oak Forest tolerates a more xeric climate than does the Mesic Conifer Forest. East of the Cascades it occurs where mean an- nual precipitation is from 15 to 25 inches. In the Rogue Area these figures are a little higher, from about 20 to 30 inches. January mean tem- peratures range from 25° to 30°F. east of the mountains and from 34° to 39°F. in the Rogue Area. There is less difference in the summer temperatures, the July means ranging from 61 ° to 71 °F. and from 67° to 71 °F. in the east and west respectively. Because of the greater diversity of the sub- dominants on the two sides of the Cascade axis it is more difficult to select a basic association in the Pine-oak Forest than in the preceding types. The species list presented in Table 7 has been selected in such a way as to include (a) the major dominants and subdominants that occur commonly on both sides of the axis, (b) 18 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 TABLE 7. DISTRIBUTION OF THE BASIC COMMUNITY OF THE PINE-OAK FOREST "' ., Q,) ,_. i u < ~ p:i d ui ., .,; - .. ., ~ ~ "O ~ ..... ,_. 0 - < < 0 Q.) "" Cl) ::s E-< x x x x x x x x x x x x x All Asian exc. 1 sp. x x Temp. to frigid x x x to here as the Interior Southwest). b) Thirteen species extend northward on the west side only of the Sierra-Cascade axis. These include, however, four species which have obviously migrated through the Columbia Gorge and now occur in a restricted area near its upper or eastern end. Only two of these thir- teen species are represented in the Interior Southwest region named above. c) Two species extend northward on the east side only of the Sierra-Cascade axis. Both of these occur in the Interior Southwest. d) Of fifteen species whose distribution ex- 20 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 tends north on both sides of the axis seven oc- cur in the Interior Southwest. e) All species occurring in the Rogue Area are found also in California west of the Sierra Nevada. f) No species extends north of the Pacific Northwest region, i.e., into northern Canada or Alaska. g) Only two species range into eastern North America; both of these are connected with the western portion of their populations through Mexico, New Mexico or Arizona. The foregoing brief analysis, along with con- sideration of the present distribution of the genera involved (Table 8), indicates that the Pine-oak Forest, in contrast with the preceding three types, is of southern origin. The center of diffusion of the species comprising the type was in all likelihood the region included in northwestern and central Mexico and Arizona ( Fig. 6). This idea will be developed later. Whenever the original northern migrations may have occurred, the eastern and western phases of the type were at least eventually sep- arated by the Sierra-Cascade axis, and signi- ficant divergent evolution has taken place on the two sides of the mountain ranges. JUNIPER-SAGEBRUSH WOODLAND The Juniper-sagebrush Woodland is the northward extension of the Great Basin pin- yon-juniper woodland, from which association, at our latitude, the pinyon pines have dropped out. As indicated by the name the type is domi- nated by one of two species of Juniperus (!. occidentalis in the south and J. scopulorum in the north) and/ or Artemisia tridentata. It is most extensive in southeastern Oregon where it occupies the tableland that constitutes the northern end of the Great Basin, continuing southward into northeastern California and Nevada and eastward across southern Idaho. It is almost cut off by the Ochoco Mountains in central Oregon, but after passing through the gap of the Deschutes River between these moun- tains and the Cascades, the type continues northward as a more or less narrow strip inside the arc of the yellow pine forest, itself sur- rounding the grasslands of the Columbia Area of north central Oregon and southeast Wash- ington ( Fig. 4). In southeast British Columbia the Juniper-sagebrush Woodland extends northward to approximately the latitude of Kamloops. An interesting shift in the dominant species of this type takes place along its north- south axis. On the central plateau of northern Mexico the type is represented by a woodland of junipers and pinyon pines; sagebrush is ab- sent. The latter species enters the community in Arizona, and the three are closely associated as far north as central or northern Nevada. Here the pinyon pines are eliminated and juni- per and sagebrush dominate the type until the northern extremity of its range is reached. In southeastern British Columbia the junipers largely disappear and the sagebrush and its associates merge with the yellow pines, so that in this area it is frequently difficult or impossi- ble to determine which type we are dealing with. Climatically the Juniper-sagebrush Wood- land is more xeric than the Pine-oak Woodland. Mean annual precipitation ranges from 7 to 13 ( rarely up to 16) inches, most of which falls during the winter, although there are signifi- cantly heavy thunder storms during the sum- mer months. Most of the plant species bloom during a short period in May and June, with only a few deep-rooted shrubs ( e.g. Artemisia and Chrysothamnus) delaying their flowering season until late summer or early fall. Winter temperatures in this community in the Pacific Northwest are moderately low, the January means ranging from 22 ° to 32°F., the lower averages tending to occur in central Brit- ish Columbia, the northern extremity of the region, and the higher ones in the Deschutes Area ( Fig. 3), the area in which we find prob- ably the best development of this association in the Pacific Northwest. These temperatures are not greatly different from those in the Pine-oak Forest east of the Cascades, although January means average from five to seven degrees 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 21 TABLE 9. DISTRIBUTION OF THE BASIC COMMUNITY OF THE JUNIPER·SAGEBRUSH UJ "' Q) .... ~ < 1 d .; "' ..;,i "' "O .; .... .lo: ..... .... UJ q u Q) in :i .ci .... UJ .... Cl) rn I .s 0 I .s 1 I ..... UJ "' 0 ......: "' {-? u ~ 1l:: "; ~ "' Q) ~ ,..c; u .... ~ .... .... t >< ::::, ~ C) ~ "' u "; ...:l l:>O UJ ...:l Q) ..c Q) 0 .... Q) 0 .... .... ;::.J ~ u z i:i:: c.., Q rn i:i:: c.., 0 Artemisia tri.dentata x x x x x x x x x Astragalus lentiginosus x x x x x x x Astragalus purshii x x x x x x Calochortus macrocarpus x x x x Calochortus nuttallii x x x x x Cercocarpus ledifolius x x x x x x Chaenactis douglasii x x x x x x x Chrysothamnus nauseosus x x x x x x x Descurainia pinnata x x x x x x x x x SE. U.S. Erigeron filifolius x x x x Eriogonum umbellatum x x x x x x x J uniperus occi.dentalis x x x x !uniperus scopulorum x x x x Replaces/. o northward Leptodactylon pungens x x x x x x Lupinus argenteus x x x x x x notn. of Ochocos, N. Cal. via NE. Cal. Mentzelia albicaulis x x x x x Mimulus nanus x x x Penstemon speciosus x x x x Phacelia linearis x x x x x Phlox hoodii x x x x x n. to Yukon & Alaska Purshia tri.dentata x x x x x Rhus glabra x x x x x E.N.A. Ribes cereum x x x x x x x Ribes viscosissimum x x x x T ownsendia fiorif er x x 22 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 TABLE 10. DISTRIBUTION OF THE GENERA REPRESENTED IN THE RASIC COMMUNITY OF THE JUNIPER-SAGEBRUSH ll ..... v ..!>fl i:: t.) en 0 0 i::i::: N ;:i t .... 0 E Q) v v "fl < E-< :i ~ z Artemisia x Astragalus x x Calochortus x Cercocarpus x Chaenactis x Chrysothamnus x Descurainia x Erigeron x Eriogonum x ]uniperus (Sabina ) x Leptodactylon x Lupinus x Mentzelia x Mimulus x Penstemon x x Phacelia x x Phlox x Purshia x Rhus (sensu strictu ) x Ribes x Townsendia x warmer in the Rogue Area Pine-oak Forest. July mean temperatures range from 65° to 73°F. at stations in the juniper-sagebrush belt, running from two to four degrees higher than in the surrounding pine forests. A group of twenty-five species can be readily selected which is clearly characteristic of the Juniper-sagebrush Woodland. This basic asso- ciation is listed in Table 9 where the distribu- tion of the various species is tabulated. Analy- sis of the tabulation shows that the community is compact and homogeneous, restricted in gen- eral to the intermontane region but with the same southward extension into the arid South- west that was evident in the Pine-oak Forest. The following distributional details may be ctl ..... (/J < IA x x s 5 < < ctl bl) t.) u :i "' ..... ... t.) v o,J - ..... s ctl 0.. :,< v 0 < v ... ... en ~ 0 E-< i:Q x x x x x 1 sp. in Mex. x x x x x x x x x esp. W. U.S. x esp. W. U.S. x x esp. W. N.A. x esp. W. N.A. x x x esp. W. N.A. x worth noting here: a) Of the twenty-five species included in the basic association, all occur in the Great Basin and the Columbia Plateau and seventeen ex- tend northward into southeastern British Col- umbia. b) Twenty-one occupy the westward exten- sion of the Great Basin in northeastern Califor- nia. One of these, Phacelia linearis, crosses over into the Rogue Area while two, Chaen- actis douglasii and Lupinus argenteus, occur in n01thern California west of the Sierra Neva- da with no present-day connection with the type through southern California. c) Sixteen species occur in the Interior Southwest area comprised of north central ------ -- 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 23 Mexico, New Mexico and Arizona. d) Only three species range to any appre- ciable distance from the Great Basin. One sub- species of Descurainia pinnata follows the southern coastal plain as far as North Caro- lina; Rhus glabra is common in central and eastern North America, while Phlox hoodii ranges northward along the western margin of the Great Plains to Yukon and central Alaska, the only one of our species to extend into boreal America. The features of the present distribution of its constituent species, together with the general distribution of the genera involved (Table 10), indicate a southern origin for this community, with its center possibly on the Meseta Central of Mexico and in Arizona and western New Mexico ( Fig. 6). Although in general the spread of this vegetation type and that of the Pine-oak Forest may have paralleled each oth- er, the greater tolerance of the juniper wood- land species for xeric conditions probably kept the two distinct in area and in time and stages at which they developed in any given area. It is noteworthy that of the genera represented in our selection of the basic flora of the Juniper- sagebrush Woodland, more than half are either restricted to western North America or are es- pecially well developed there. It would seem that with the rigorous conditions under which this type has existed evolution has progressed rapidly at all levels. CHAPARRAL FORMATION As defined here chaparral is restricted in the Pacific Northwest to lower elevations in the Rogue Area, occurring extensively in the Rogue watershed, including that of its tributary, the Illinois River, and in relatively small areas in the Umpqua valley (Fig. 4). It is a formation dominated by shrubs with extensive root sys- tems, usually growing in dense stands. The spe- cies are characterized by relatively small, thick, heavily cutinized leaves, usually but not always evergreen, borne on rigid twigs and branches. A recent paper by the author (Detling 1961) treats in detail the floristic constitution, cli- matic requirements, and probable postglacial history of this vegetation type in southwestern Oregon and northern California. In our region the chaparral is usually domi- nated by Ceanothus cuneatus, although this spe- cies is sometimes largely replaced by Arcto- staphylos viscida or A. canescens. The chaparral species are more tolerant of drought conditions than are those of the Pine- oak Forest which usually occupies the slopes above them. Mean annual rainfall ranges from 12 to 20 inches, with less than 20 per cent oc- curring in the six driest months of the year. January mean temperatures are relatively high, running from 34° to 38°F. July means are from 69° to 73°F. These also are in general higher than those in the Pine-oak Forest, and more nearly approximate those occurring in the Juniper-sagebrush Woodland. The species making up the Chaparral For- mation are even more restricted in their range than are those of the juniper-sagebrush type. The basic community with its distribution is shown in Table 11. They all occur in the Cali- fornia Central Valley, Klamath Mountains and the Rogue Area, with about half of them extend- ing southward to southern California or Baja California. Relatively few are found outside this general region. Of the four species listed for the Great Basin three occur in western Ne- vada only in that area. Only two go as far north as the Columbia River. The genera to which these species belong are mostly either restricted to western North America or are at least best developed there. The present-day chaparral may have developed from vegetation of a more southern origin (cf. Table 12) , but many chap- arral species probably developed in the areas in which they now occur. This view is support- ed by Stebbins ( 1952) who postulates that greater numbers of species in geologic time evolved in a dry habitat than in a mesic one because of a greater turnover in species, and that "environments limiting or deficient in one all important factor, moisture, have often pro- moted evolution." 24 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 TABLE 11. DISTRIBUTION OF THE BASIC COMMUNITY OF THE CHAPARRAL ~ 1 oj .;; '"' '"' u Q.) .~ ..... en ifJ '"' -1 I 0 ~ cd ~ u Q.) .;; ~ i:ci u Amdanchier pallida x x Arctostaphylos canescens x Arctostaphylos viscida x Asclepias cordif olia x Ceanothus cuneatus x x Cercocarpus betuloides x x Chlorogalum pomeridianum x x Eriodictyon calif ornicum x FritiUaria recurva x Garrya fremontii x x Lonicera interrupta x x x M onardella villosa x Quercus vaccinifolia x Rhamnus calif ornica x x x Rhamnus crocea x x x Rhus trilobata x x x Thysanocarpus radians x Trichostema lanceolatum x x GRASSLAND FORMATION Studies of post-Pleistocene pollen of the grasslands in eastern Washington show that the vegetation immediately following the Wiscon- sin period was similar to that found today ( Hansen 1939). The grassland and desert zones increased in area for a time during the xerothermic interval following this time, and then the forest returned to approximately its present extent (Daubenmire 1942). Overgraz- ing and other man-induced factors may have accelerated the invasion of this type by second- ary species often to the extent of eliminating the grasses that were originally dominant. Much of the plateau of southern Idaho ( Crad- dock and Forsling 1939) and the Great Basin (Pickford 1932) that is now occupied by sage- brush was grassland before livestock grazed out the associated plants. ..i.: rn iii I .;; u ~ z x x x x x x x x x x x x x x x x x x .s rn rn oj s::: s::: ~ ·;;; ·; ...c: oj a'.l 0::: - ~ Q.) oj ::, s ..... ..... .... -bO oj ~ oj Q.) oj Q.) -.) Q.) bO 0 E2 '"' 0 '"' ::, ~ 0 ~ 0 ~ x x x x x x x W. Nev. only x x x x x W. Nev. only x x x x x x x x W. Nev. only x x x x x x x The most typical grassland left today is to be found in southeastern and south central Washington, particularly on the rolling hills of the Palouse and Snake River valleys (Fig. 4). There is evidence from relict areas that much of the Columbia Plateau north of the Ochoco Mountains in Oregon was once climax grassland. In British Columbia natural grass- lands extend far north of our area on the floors and lower slopes of the interior valleys, where they are often interspersed with forest areas. In south central British Columbia much of the grassland has been seriously invaded by sage- brush and related vegetation (Tisdale 194 7). The Grassland Formation is dominated largely by bunchgrasses such as Agropyron. Streambank and frequently rimrock associa- tions may contain shrub species, usually the same ones that occur in similar situations in the - ----· -- 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 25 TABLE 12. DISTRIBUTION OF THE GENERA REPRESENTED IN THE BASIC COMMUNITY OF THE CHAPARRAL "' Q.l Q.l ~ = s I) 0 cr.i 0 i:: N ;:i < i::i::: Q.l "' < ci. cj ..... bO .;:: :i s 0 "' ..... 5 ~ (1) "' I) Q.l cc Q.l ..... s 'lil >< "' ·s. E-t < < < Q.l Q.l Q.l 8 ..... :i ~ ~ :i ~ cr.i 0 E-t i:i::i Ame/,anchier x A rctostaphylos x x Asclepias x x x x Ceanothus x x x Cercocarpus x x Chlorogalum x Eriodictyon x x Fritillaria x Garry a x x Lonicera x x Monardella x x Quercus x x Rhamnus x x x Rhus x x x x T hysanocarpus x x Trichostema x x x Mostly Pac. Coast juniper-sagebrush type. In general, the climatic requirements of grassland in the Pacific Northwest are not greatly different from those of the Juniper- sagebrush Woodland. Annual precipitation ranges from 7 to 18 inches, January mean tem- peratures from 29° to 33°F., and July mean temperatures from 65° to 76° F. What may be a significant factor in maintaining grassland rather than sagebrush and juniper is a fairly consistent tendency for relatively high amounts of rainfall to continue well through the late spring months (May and June) in the grass- land areas. CONSTITUENTS OF THE GRASSLAND Agropyron inerme Agropyron spicatum Balsamorhiza sagittata Clarkia pulchella Clematis hirsutissima Elymus glaucus F estuca idahoensis F estuca scabrella Fritillaria pudica Gaillardia aristata Geranium viscosissimum H elianthella uni/fora Iris missouriensis Koeleria cristata Lupinus sericeus M ertensia longiflora Poa secunda Ranunculus glaberrimus Sporobolus cryptandrus Sporobolus columbianus Stipa comata Trillium petiolatum W yethia amplexicaulis 26 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 CRETACEOUS The history of modern vegetation goes back largely to the Cretaceous. By this time the gym- nosperms had undergone a long period of evo- ution. Many of their primitive lines were ex- tinct, and of those remaining most were dwin- dling in importance. One of the last of their or- ders to evolve, the Coniferales, had originated at least by the beginning of the Mesozoic, and by late Cretaceous it was still the most import- ant line of gymnosperms, and was widely dis- tributed in both the Old and New Worlds. By the end of the Cretaceous the flowering plants far surpassed the gymnosperms in num- bers of families and genera and dominated the face of the earth. Their primitive representa- tives may have originated in the tropics and moved northward to undergo this great devel- opment (Axelrod 1952, 1959), or, as some botanists believe, they may have originated in the Holarctic regions ( Just 1946). Most of the angiosperm families as we know them now, and many of our modern genera, are known from the Cretaceous of the Northern Hemisphere. Whatever may have been the time and place of origin and early development of the angio- sperms, by the end of the Cretaceous there was a well developed flora of modern aspect in Alaska, northern Canada, and Greenland, ex- tending southward into eastern North America, the Rocky Mountain region and California. The constitution of this flora has been determined or otherwise called attention to by the work of Holli ck ( 1930) in Alaska, Dorf ( 1942) in the central Rocky Mountains, Chaney ( 1946) and others. Let us recall that the only land masses in the Northwest not covered by the Cretaceous sea were the great batholiths forming the Ne- vadan orogenic arc (Fig. 2). Emergent land then existed in the mountain masses that have persisted as the Vancouver Island Mountains, the Canadian Cascades and the highlands form- ing the dissected plateau of southern interior British Columbia and north central Washing- ton, the Selkirk and Bitteroot Ranges and the Blue and Klamath Mountains. During the long period of time encompassed by the Cretaceous the elevation of the land masses varied with stages in uplift and erosion, but they were never areas of low relief. These land masses were closely associated with those of Alaska, north- ern Canada and the Rocky Mountain axis, and from what we know of their Cretaceous floras and the evidence we have from the Vancouver Island Cretaceous ( Chaney 1946) we can prob- ably draw some very accurate conclusions re- garding the flora of our region at the close of that period. The vegetation of the uplands bordering the embayment of the northwestern Cretaceous sea apparently consisted of a forest cover made up largely of hardwood genera ( Chaney 1946), Represented among them was a rather large number that still occur commonly in our flora. These genera are mostly associates of the cool- temperate forest, although some are found more regularly in other types. The more im- portant members of the group include: Abies Acer Alnus Betula Cornus Corylus Crataegus Fraxinus Gaultheria ]uglans Larix Lithocarpus Mahonia Malus Myrica Philadelphus Picea Pinus Populus P seudotsuga Quercus Rhamnus Salix Sambucus Sequoia Smilax 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 27 Sorbus Stipa Torreya Tsuga U mbellularia Viburnum Vitis These are genera which have persisted in west- ern North America down to modem times and are referred to by paleobotanists as the West American Element of the early flora. A second group of genera that was wide- spread in the Cretaceous forest occurs com- monly today in the eastern part of North Amer- ica but has been eliminated from the region west of the Rocky Mountains. This East Ameri- can Element includes the following: Aristolochia Asimina Carpinus Cary a Cassia Castanea Celastrus Comptonia Cotinus Diospyros Fagus /lex Lindera Liquidambar Liriodendron Magnolia Myrsine Nelumbo Nyssa Os try a Piper Sapindus Sassafras Taxodium Tilia Ulmus Still a third group of genera, now restricted either to Asia or to tropical or subtropical re- gions of the New World, includes: Ailanthus Cercidiphyllum Cinnamomum Ficus Ginkgo Glyptostrobus Grewia Hedera Keteleeria Laurus M etasequoia Paulinia Per sea Pistachia Pterocarya Sabal Sterculia Zelkova Zizyphus The present distribution of the genera rep- resented in this whole Arcto-Cretaceous Geo- flora indicates for that period a temperate to warm-temperate and humid climate, though not to the point of being subtropical, even in the subpolar region above the Arctic Circle. Abun- dant rainfall apparently was distributed fairly evenly throughout the year, as evidenced by the presence of genera which today are character- istic of climates that have abundant summer rainfall. Proximity to the sea doubtless mini- mized seasonal temperature variations. Stebbins and Major ( 1965) and Axelrod ( 1959) postulate that the Madro-Tertiary Geo- flora had it origins in an ancient xeric or semi- xeric floristic element which migrated from one continent to another. They point out that many relict species in California have close relatives in the other arid regions of the world -southwest Africa, the Mediterranean region and Chile. This great similarity suggests that they have evolved very little since they became separated. Climatological evidence shows (Stebbins and Major 1965) that the Madro-Tertiary Geoflora could not have migrated from one continent to another in the Tertiary. The northern limits of the evidence is far south of any Tertiary cli- 28 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 mate capable of supporting such a flora. Further, Stebbins and Major postulate the existence of pockets of xerophytic floras in late Cretaceous and Tertiary times when the cli- mate of the American Southwest was tropical to warm-temperate. They cite similar forma- tions in South America. EOCENE Certain major geologic and climatic changes occurred at about the close of the Cretaceous period that had a far-reaching effect upon the flora of the Northwest. The first of these was the elimination, through combined sedimentation and uplift, of the embayment of the sea bor- dered by the Nevadan orogenic arc. The result of this was to provide a relatively level low- land area extending from the Blue Mountains and Okanagan Highlands eventually almost to the present coast line of Washington and Oregon. The Cascade and Coast Ranges were formed much later, and in the absence of moun- tain barriers on the west, an oceanic influence prevailed in the climate of this entire plain. A second significant change for which we have evidence from the Eocene was an increase in temperatures over those prevailing during the Cretaceous. This warming trend was gen- eral, at least throughout the Northern Hemi- sphere, so its effect on vegetation was not re- stricted to the Pacific Northwest (Chaney 1949, Axelrod 1958) . A few fossil floras have been uncovered in this region which have been referred to the Eocene, and which reveal the nature of the flora during the epoch. The best known of these are the Clarno (Knowlton 1902, Chaney 1946) in the John Day basin of eastern Oregon, and the Goshen (Chaney and Sanborn 1933) and Comstock (Sanborn 1935) floras of the Wil- lamette valley in western Oregon. These floras indicate that the lowland plain was occupied by a broad-leafed forest whose genera came from two distinct sources. A relatively small number of genera had been represented in the temper- ate Cretaceous floras of Alaska and the central Rocky Mountains (Rollick 1930), and we may assume that these genera had been present in the mountains of British Columbia and in the Blue Mountains. This group was made up of species of Aralia, Aristolochia, Celastrus, Cin- namomum, Diospyros, Ficus, Magnolia, Per- sea, Platanus, Quercus, Rhamnus, Smilax and Viburnum. They are genera which are well represented today in warm temperate North America. We may suppose that as the Pacific shoreline advanced westward during upper Cretaceous and lower Eocene times this frag- ment of flora of boreal origin invaded the new- ly formed lowlands, and persisted there in spite of the warm climate, probably as newly evolved species adapted to the changed condi- tions. The majority of the genera making up the lowland forest, however, were apparently new to this region and had no previous relation to the temperate boreal forest. Such were: Anona, Aporosa, Astroninm, Calyptranthes, Cordia, Cupania, Inga, Lucuma, M allotus, M eliosma, Nectrandra, Ocotea, Sapium, Siparuna, Strych- nos, Symplocos, and Tetracera. These are gen- era which today are found largely in tropical or subtropical countries, many of them restricted to Central and South America. The fossil rec- ord shows that a closely related community was already in Central America in the Eocene, and persists there under subtropical conditions in such places as the central plateau of Costa Rica. Chaney and Sanborn ( 1933), in discussing the Goshen flora, emphasize the point that the Eocene climate of the Pacific Northwest low- lands was probably not tropical but rather sub- tropical, similar to the present climate of up- land Costa Rica. They postulate a mean annual temperature of about 68°F. and an annual rainfall of about 70 inches. While the subtropical forest dominated the warm lowlands, the mountains that bordered them on the north, east and south-the Cana- dian Cascades, Blue and Klamath Mountains (Fig. 1)-undoubtedly served as refugia for the boreal flora of the region. Direct evidence of the constitution of this upland flora is scarce, 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 29 although one layer of leaves in the Clarno for- mation in Oregon contains representatives of such typical temperate genera as Metasequoia, Alnus, Lithocarpus and Ulmus, mixed with the subtropical genera of the typical flora (Knowlton 1902) . Chaney ( 1938) believes these represent the Eocene vegetation of the up- lands of the Blue Mountain region. Later de- velopments in the eastern Oregon and Wash- ington flora seem to substantiate this view. OLIGOCENE Subtropical conditions persisted in the low- lands through the Eocene and into the early part of the Oligocene, but during the latter epoch there was initiated a return to a more temperate climate-a cooling and drying ten- dency that continued progressively through the remainder of the Tertiary Period, with only brief recessions. Plant megafossils are not numerous in the Oligocene of the Pacific Northwest, but from those available, along with the evidence from such floras as the Weaverville in northern Cali- fornia (MacGinitie 1937), the Lower Cedar- ville in northwestern Nevada, and the Floris- sant in Colorado (MacGinitie 1953), and from the recent pollen studies by Gray (1964), we get what is probably an accurate general pic- ture of the floristic changes that took place dur- ing the epoch. With cooling temperatures the subtropical forest of the low plain was gradually replaced by a temperate forest very similar to that which dominated the upland region during the Cre- taceous. Most of its species probably migrated down the slopes from the highlands bordering the plain, where temperatures had been cool enough for them to persist through the Eocene. Other species and genera may have come south with the changing climate from northern Brit- ish Columbia and Alaska where the original temperate boreal forest still flourished. It was still largely a broad-leafed decidu- ous forest. Many of its genera are familiar in more mesic situations in the Northwest today: Acer, Alnus, Amelanchier, Betula, Corylus, Crataegus, Fraxinus, Mahonia, Malus, Phila- delphus, Populus, Quercus, Ribes, Rosa, Salix, Sambucus, Viburnum and Vitis. Others that were common at that time are now restricted to eastern North America: Carpinus, Carya, Cas- tanea, Fagus, !lex, Liquidambar, Nyssa, Os- trya, Tilia and Ulmus. The conifers apparently were represented largely by members of the Taxodiaceae. Leaves of Sequoia, M etasequoia and T axodium are common in the Oligocene sediments, and Gray ( 1964) reports that " ... Oligocene strata at present are perhaps best characterized by the abundance of coniferous grains of presumed taxodiaceous affinities .... " Thuja and Libo- cedrus were probably present, especially in the uplands. Pollen of Cupressaceae is reported from both western and eastern Oregon and from Idaho, and both of these genera were present in the Miocene. While Abies, Picea, Pinus, Pseudotsuga and Tsuga were present and most widespread as shown by megafossils and pollen, they were not abundant, and the great coniferous forests made up largely of Pinaceae that today occupy so much of the Pacific Northwest at all elevations were still to develop. MIOCENE The Miocene and Pliocene floras of the Pa- cific Northwest are more abundant and better known than those of the preceding epochs. Our best known Miocene floras are the Eagle Creek ( Chaney 1920) , Molalla and Ashland in west- ern Oregon, and the Mascall ( Chaney 1925), Blue Mountains (Oliver 1934) , Stinking W a- ter ( Chaney and Axelrod 1959), Sucker Creek (H. V. Smith 1938, 1939), Trout Creek (Mac- Ginitie 1933), Lower Ellensburg, and Latah (Knowlton 1926) in eastern Oregon and Wash- ington, and the Upper Cedarville (LaMotte 1936) in northwestern Nevada. The tendency toward a cooler and drier climate that began in the Oligocene and con- tinued through the Tertiary was a general one that was felt at least throughout the Northern 30 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 Hemisphere. A more local factor that eventual- ly had a profound effect upon the climate, and thus upon the vegetation of the Pacific North- west was the volcanic activity which was re- sponsible for the building of the Cascade Range ( Clark and Stearn 1960). During the lower and probably middle Miocene the mountains were low with only a moderate influence on the cli- mate to the east. At that time M etasequoia was the dominant tree in the temperate forest of eastern Oregon and Washington. But by upper Miocene time, as revealed in the Mascall flora at the base of the Blue Mountains, while Meta- sequoia was still present it was greatly reduced in abundance. The Cascades were now inter- cepting some of the moisture from the Pacific Ocean, and seasonal temperature fluctuations and extremes were becoming more pronounced. Axelrod (1950), however, postulates still an average annual rainfall of 35 to 50 inches, dis- tributed more or less evenly throughout the year, and moderate temperature ranges for the lowland areas of the Great Basin and Columbia Plateau, where the Arcto-Tertiary Geoflora was dominant. The Pinaceae increased markedly during the Miocene. At the beginning of the Miocene mem- bers of the Taxodiaceae were predominant among the conifers, represented by Sequoia, M etasequoia and T axodium. But by the end of the epoch several of our familiar genera among the Pinaceae, e.g., Abies, Picea and Pinus, were often abundant, while Pseudotsuga and Tsuga are known to have been present, as well as Thuja and Libocedrus of the Cupressaceae. Another Miocene feature which was probab- ly the result of changing climatic conditions was the appearance of an increasing number of herbaceous types, including a few predomi- nantly herbaceous families such as the Malva- ceae, Onagraceae, Polemoniaceae, Umbellife- rae, Gramineae, Cyperaceae and Compositae. Herbaceous types were much more highly de- veloped at this time in California and the rest of the Pacific Southwest (Gray 1964), prob- ably as a result of adverse climatic conditions for most trees and shrubs ( in this case extreme drought), and it is to be expected that colder winters in the Northwest would be conducive as well to the herbaceous habit. PLIOCENE The culmination of the xeric trend was reached by middle Pliocene time. The Cascades and Coast Range had been elevated to approxi- mately their present elevation and the modern pattern of ranges and intervening basins was well established. Rainfall east of the Cascades was reduced to the minimum reached at any time previous to the Pleistocene Ice Age. Axel- rod ( 1950) postulates a middle Pliocene an- nual rainfall for the northern Great Basin and Columbia Plateau of about 15 to 17 inches, with somewhat less farther south. The contin- ued presence of the East American Element in the forest, however, indicates that there was still considerable summer rainfall. During the lower Pliocene the northern in- termontane region in Oregon, with a rainfall of 25 to 30 inches, was dominated by a modi- fied Arcto-Tertiary Geoflora which included Abies, Acer, Amelanchier, Mahonia, Picea, Pinus, Populus, Pseudotsuga and Sorbns ( Ax- elrod 1944, Chaney 1944) . As early as the middle Eocene (Brown 1934) in the Southwest, where their evolution has been traced through abundant fossil records, there had been developing a series of vegeta- tion types new to the Pacific slope. At that time they probably grew on some of the lowlands while the Arcto-Tertiary Geoflora was still dominant in the uplands. By Oligocene times differentiation of the generalized flora into spe- cialized types adapted to varying degrees of aridity and temperature had reached the stage where Axelrod (1958) distinguishes: (a) a woodland savanna comprised of Arbutus, Cel- tis, Dodonaea, M orus, Quercus ( live oak type), Platanus, Rhus, Sapindus, Stipa and Vauqueli- nia; (b) a chaparral type with Ceanothus, Cer- cocarpus, Colubrina, Mahonia, Quercus (scrub oak type) and Rhus; and ( c) a thorn scrub of Bursera, Caesalpinia, Colubrina, Euphorbia, Prosopis, Tephrosia, Thouinia and Zizyphus. 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 31 For the most part they appeared in the Pacific Northwest in the Pliocene, and from the time of their first appearance they have played an important role in Pacific Northwest vegetation- al history. Middle Pliocene floras indicate that the climate throughout western United States at that time was semiarid, and milder and warmer than at present. It was in response to these climatic changes that this new flora evolved in northern Mexico and southwestern United States from a warm- temperate flora that already existed there un- der humid conditions. The fact that the most closely related descendants of the original flora are found today in the Sierra Madre Occidental of northwestern Mexico or adjacent southwest- ern United States has resulted in the applica- tion to the whole flora of the name Madro- Tertiary Geoflora. The details of its evolution and migrations have been summarized by Ax- elrod (1958). Axelrod lists almost 100 genera in which fossils have been described from the Madro- Tertiary Geoflora. The following selection lists some of the more important members of this group that are found today in northern Mexico and southwestern United States. The genera preceded by an asterisk extend northward to- day into the Pacific Northwest. Acacia Adenostoma Agave *Amelanchier *Arbutus * Arctostaphylos *Artemisia *Baccharis Bursera Caesalpinia Cassia *Ceanothus *Celtis Cercis *Cercocarpus *Chamaebatiaria Clethra Condalia Dendromecon Dodonaea *Ephedra *Euphorbia Eysenhardtia Ficus *Fraxinus Fremontia *Garrya * H olodiscus *Juniperus Karwinskia Leucaena Ly silo ma *Mahonia Mimosa *Myrica * P hiladelphus Photinia *Pinus Pithecolobium Platanus *Populus Prosopis *Prunus *Purshia *Quercus (live and scrub oaks) Randia *Rhamnus *Rhus *Ribes Sabal Salvia * Symphoricarpos * U mbellularia Central California was dominated by a live oak woodland during the middle Pliocene. As- sociated with Quercus were such genera as Cel- tis, Lyonothamnus, Platanus, Populus, Robi- nia, Salix and Sapindus. Some parts of the re- gion were given over to chaparral, as evidenced by aggregations of Arctostaphylos, Ceanothus, Cercocarpus, Dendromecon, Fremontia, Ma- honia, Photinia, Quercus (scrub oak) and Rhus. Bordering upland areas were forested with such genera as Alnus, Arbutus, Comus, Fraxinus, Populus and Prunus. It is estimated ! J 32 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 that the central California middle Pliocene rainfall varied from 15 to 17 inches over the lowlands to 23 inches in the forested uplands ( Axelrod 1948) . Southern California also was dominated by woodland, not greatly different from that far- ther north. Here, however, more arid condi- tions are indicated by desert-border communi- ties including Baccharis, Cercidium, Chilopsis, Condalia, Ephedra and Prunus (Emplectocla- dus), and an arid subtropical scrub with Do- donaea, Eysenhardtia and Ficus. Rainfall must have been from 12 to 15 inches on the lowlands, distributed as summer showers and wint_er rams. The northern Great Basin by middle Plio- cene was apparently largely grassland ( Axel- rod 1948, 1958), with Acer, Populus, Prunus and Salix along the streams. The uplands were probably forested, a situation similar to that today in the Deschutes area with yellow pine and fir on the slopes of the Cascades. The central Great Basin, with a rainfall of about 12 inches according to Axelrod ( 1948), had semiarid woodland over what is now des- ert, with Juglans, Edwinia, Purshia, Rhus and a live oak. Chaparral communities included Arctostaphylos, Cercocarpus, Rhus, Ceanothus, M ahonia and Prunus ( Emplectocladus). Only in the still more xeric southern inter- montane areas did the flora include Pinus ( the pinyon pines), Robinia and Acacia; these ap- parently never reached the Pacific Northwest. The Madro-Tertiary Geoflora reached south- western Oregon by early Pliocene, and un- doubtedly reached its greatest development in this area by the middle of the epoch, just as it did in other areas throughout the west ( Axel- rod 1958). Its further development and spread in the Pacific Northwest was arrested after the middle Pliocene by the widespread climatic change that initiated a return to the moister and cooler conditions leading up to the glacia- tions of the Pleistocene. Whether it was ever entirely eliminated from the Rogue basin after it once arrived there is problematical. Yellow pine apparently was abundant in the upper Willamette valley at the beginning of the last glacial retreat, and if it persisted there through the cold maximum it is entirely possible that other xeric species of the Madro-Tertiary Geo- flora persisted as far north as the Rogue basin. However, the greatest importance for us of the Madro-Tertiary Geoflora at that stage lies in its continuing proximity to our region, to whose flora it made significant contributions in later times as climatic conditions allowed. During late Pliocene times there began a trend toward cooler temperatures, which cul- minated in the ice ages of the Pleistocene. And while the Cascades still acted as a barrier to moisture from the Pacific Ocean, the lower tem- peratures and consequent lower evaporation rates may have resulted in a greater effective precipitation than that which obtained in the preceding middle Pliocene. By late Pliocene a major secular change had occurred in the climate on both sides of the Cascades. The amount of summer rainfall pro- portionate to the annual total had dropped sharply ( Chaney, Condit and Axelrod 1944). This resulted in the elimination from our flora of both the East American and East Asian Ele- ments, species and genera which were depend- ent upon wet summers for their continued ex- istence. PLEISTOCENE By the beginning of the Pleistocene and be- fore the advance of the glaciers the modern pattern of vegetation in the Pacific Northwest must have been well established even to spe- cies, an association directly derived from the West American Element of the Arcto-Cretace- ous and Arcto-Tertiary Geofloras. This forest covered the valleys and coastal strip and con- sisted of Pseudotsuga, Tsuga, Larix, Thuja, Alnus, Acer macrophyllum, Cornus nuttallii and Populus, with an ,rnderstory of Corylus, Acer circinatum, Mahonia nervosa, Philadel- phus, Rhamnus purshiana, Sambucus glauca and V accinium parviflorum. A number of disjunct distributions among the Boreal Forest species, where their range is ,, 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 33 broken at or near the southern limit of the con- tinental ice sheet, demonstrates the influence of the glaciers. Figure 5 shows the maximum ex- tent of continental glaciation in our region (Flint 1945). This maximum occurred late in the series of ice advances, in most places during the last or Wisconsin stage, and its southern limit constitutes a critical line from the stand- point of the distribution of many northwest species. Glaciologists have shown that during the ice age there was not just one advance of an ice sheet, but at least four, and that these advances presumably were separated by relatively warm periods. If this is so, we must assume that south of the glaciers the cold wet maxima alternated with periods during which the boreal flora took refuge in the non-glaciated areas in the high mountains, the far north and in the south. Evi- dence for these climatic fluctuations is found in the complicated vegetational pattern in the Pacific Northwest, demonstrated in modern times by peculiar species and subspecies dis- tributions and by the presence of islands of boreal flora. Rand (1948) describes birds hav- ing continuous distributions which show bio- logical discontinuity, and postulates temporary barriers in the form of glacial ice which per- mitted the fixing of genetic entities. We may assume that the pattern of vegeta- tional adjustment to conditions caused by either advancing or retreating glaciers would be much the same throughout the whole Pleis- tocene epoch since the temperatures of all the interglacial periods were about the same (Flint 1957). With increasing cold, mesic species would be eliminated from higher elevations of this region, and from lower elevations in the vicinity of the margin of the continental ice sheet. They would be replaced by more boreal types already occupying the highest mountain tops or the lower elevations to northward. Judg- ing by present distributions, the climate adja- cent to the ocean, at least as far south as north- ern California, was cool enough to permit the invasion of the coastal lowlands by this same boreal flora. We have used relict islands of boreal vegeta- tion to show past distribution patterns of bor- eal associations. Another type of vegetation which left its pattern in disjunct distributions was more xeric in its requirements and invaded the warmer and drier areas as they appeared in the wake of receding glaciers. These xeric vegetation types, the Pine-oak Forest, Juniper- sagebrush Woodland, Chaparral and Grass- land, moved north from the Great Basin or developed approximately in the areas they now occupy (Axelrod 1950). This evidence is particularly striking in the Pacific Northwest where there is great physio- graphic and climatic variation. Even here, however, the evidence of retreat and advance is not clear-cut. As a result of subsequent ad- vances and retreats the locus of an island might or might not be repopulated with the same com- bination of species as before. We have no way of knowing which glacial advance or which in- terglacial maximum brought elements of the flora of a given present-day relict island into place. And it is probable that the present islands are not all of the same age. This increases the difficulty of determining the chronology of suc- cessive waves of floral migration, and seems to be particularly true of our boreal flora. That part of the continental ice sheet which affected the Pacific Northwest was centered in the Canadian Rockies. At its maximum stage it extended southward well into Washington and northern Idaho ( Fig. 5). East of the Cascade Range it reached its southern limit about where the Columbia River forms the southern boun- dary of Okanogan County in Washington. It extended beyond the present site of Spokane, and in Idaho extended south of Lake Pend d'Oreille. West of the Cascades a lobe of the same glacier covered the Puget Sound basin as far south as the valley of the Chehalis River. Apparently all of southern British Columbia was ice-covered during the maximum glacial advances. Besides the lowlands and plateaus covered by the main ice sheet, extensive areas of some of the mountain ranges were covered by gla- 34 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 I I I I (J : I ----------- ---- - r- - ----- _ _ ___ L __ ~ 0 i ' I I I Figure 5. Maximum continental glaciation in the Pacific Northwest during the Pleisto- cene epoch (after Flint 1945). ciers which originated at high elevations and moved downward in response to decreasing temperatures. Such glaciers were almost con- tinuous in the Cascades as far south as the Klamath Gap. Widespread glaciation occurred in the Wallowas and the Olympic Mountains (Flint 1957). Two relict islands are of major importance in our subsequent discussion of the former ex- tent of the boreal flora. The most extensive in area of these islands is the Columbia River Gorge (Fig. 1). This local flora was discussed in some detail by the writer in a previous pub- lication ( Detling 1958). The gorge has been carved through the Cascade Range by the Co- lumbia River, its cutting having kept pace with the gradual rise of the range since its beginning. In some place precipitous walls rise to a height of 2500 feet. The boreal flora of the Columbia Gorge oc- curs at numerous places along the bottom of the canyon at elevations ranging from 50 to 1500 feet, mostly along the south wall in the al- most constant shade of the steep bluffs. Many of the species thrive in the cool misty situations about the numerous waterfalls and in the nar- row lateral canyons. The nearest related flora occurs above the 4000-foot level on the north facing slopes that rise toward Larch Mountain and Mount Hood, but the two populations are separated by at least 2500 vertical feet of Mesic Conifer Forest. A second boreal island, less extensive in area but equally as significant in its implications, is on Saddle Mountain (Fig. 1). This flora also has been discussed in detail by the writer in a previous paper (Detling 1954). Located in the Coast Range of Clatsop County, Oregon, this peak rises to a height of 3250 feet. The sheer cliffs and interspersed hanging meadows near its summit harbor about forty species that are distinctly boreal in their distribution. Many of them occur again no nearer than the Olympic Mountains, the Columbia Gorge, or at other stations in the Cascade Mountains of Washing- ton or Oregon. The most ambitious project aimed at deter- mining the succession of vegetation types in the Pacific Northwest since the last glacial re· treat has been the study of fossil bog pollen carried on for the last thirty years by Henry P. Hansen. By 1947 Hansen had investigated some sixty-seven bogs in this region and has since added several more to the list. Most bogs suitable for study occur west of the Cascade Mountains, but a sufficiently significant num- ber were found in eastern Washington and Ore- gon to furnish valuable clues to climatic shifts in that area as well. His work shows that most of our bogs were initiated at about the begin- ning of the last retreat of the continental ice. Pollen profiles from the glaciated region show that Pinus contorta was in all cases the first arboreal invader of the newly uncovered ground. The species is well adapted to cool temperatures and disturbed edaphic condi- 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 35 tions, and probably established itself close to the retreating glacier. South of the ice sheet P. contorta was present in greater or lesser abundance at the time of glacial maximum. It was especially abundant west of the Cascades as at Lake Labish (Fig. 1) and Silverton 'bogs in the central Willamette valley, but in lesser degree was present east of the range as far south as the northern Great Basin, where it probably was limited to upland situations from which its pollen was carried to adjacent bogs. Pollen of this species is usually most abun- dant at the lowest levels of the bog profiles, where it is dominant. It soon decreases upward and is gradually replaced west of the Cascades by Pinus monticola, Picea sitchensis and Pseu- dotsuga menziesii, the last two of which eventu- ally become dominant in certain areas. East of the Cascades Pinus contorta was replaced by P. monticola, Abies spp., or over much of the area, finally by Pinus ponderosa. Reusser ( 1960) interprets the pollen record in the Pacific Northwest in general as follows: The maximum advance of the last Wisconsin glaciation, which occurred about 15,000 years ago, was followed by a cool moist period of about 10,000 years duration. This was the per- iod of dominance of the more boreal forest spe- cies such as Pinus contorta, P. monticola and Abies spp. The period from 13,000 to 11,000 years ago was one of increasing warming and drying, represented in the vegetation by an ex- pansion of the more mesic species, viz., Pseu- dotsuga menziessi, Picea sitchensis and T suga heterophylla west of the Cascades, and Pinus ponderosa to the east. The next postglacial pe- riod extended from 11,000 to 4000 years ago. During this time average temperatures were slightly higher and average precipitation less than at present. Vegetation was marked by the dominance of Quercus garryana in the valleys west of the Cascades, and a maximum of grass- es, composites and chenopods in the mid-Co- lumbia and Great Basin areas. Picea sitchensis apparently was dominant along the coast. The period from 4000 years ago to the present has been one of cooler and moister climate marked by an increase in Pseudotsuga menziesii and Pinus ponderosa at the expense of Quercus garryana and the grasses, composites and chen- opods. By utilizing the results of pollen research and what we know of present relict floras we can piece together with fair accuracy the gener- al story of Pacific Northwest vegetation since the early Postglacial. At the glacial maximum a boreal forest dom- inated by Pinus contorta and P. monticola ex- tended along the ice front from the Washing- ton coast to the Rocky Mountains of northern Idaho (Reusser 1960). It was probably similar to the present Boreal Forest occurring at eleva- tions of 4000 to 6000 feet in the Cascade Range of Oregon and at lower elevations northward. This forest extended southward along the coast as far as northwestern California (Han- sen 194 7), although at its southern limits it probably was infiltrated by austral genera and species. It occupied the western Washington and Oregon valleys as far as the Calapooya Mountains between the Willamette and Ump- qua drainages. A relict flora in a bog 22 miles west of Eugene today contains Ledum glandu- losum columbianum, the coastal counterpart of the montane L. glandulosum glandulosum, Ra- nunculus reptans strigulosus, H ypericum ana- galloides and Drosera rotundifolia, all boreal species or subspecies. In the same bog Hansen found pollen of Pinus contorta and P. monti- cola, the former of which was present almost to recent times. An abundance of Pinus pon- derosa pollen in the lower half of the profile confirms our belief that the southern end of the Willamette valley was near the margin of the Boreal Forest so far as the interior basins are concerned. Picea sitchensis was present at other points in the central Willamette valley during this time, at Onion Flats and Lake Labish (Hansen 1947). Betula glandulosa was col- lected at Lake Labish by Thomas Howell in 1893, and Picea sitchensis is now growing on the Nisqually River in the Cascades of western Washington. 36 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 The disjunct occurrence of a number of bor- eal species, mostly on mountain tops through- out the range, indicates that the Boreal Forest covered the Coast Range southward through most of Oregon. The following representative species occur as members of typical boreal associations at several points. The letters in parentheses following the species names indi- cate the localities (Fig. 1) (S=Saddle Moun- tain, Clatsop County; H=Mt. Hebo, Tillamook County; M=Marys Peak, Benton-Lincoln Counties; R=Roman Nose, Douglas County; T=high Coast Mountains, Tillamook County). Abies procera (S-M) Acer glabrum doublasii (S-M) Cornus canadensis (H) Cryptogramma acrostichoides (S-R) Heuchera glabra (M) Lomatium martindalei augustatum (M) Lupinus lepidus lyallii (M-R) Penstemon davidsonii menziesii (S-H) Phleum alpinum (S-M) Polystichum munitum imbricans (R) Pyrola secunda (T) Rubus lasiococcus (H) Rubus pedatus (S-T) Senecio triagularis (H-M) Vaccinium membranaceum (M) Vaccinium ovalifolium (H) Climatic conditions in the Cascades were se- vere enough to push the boreal forest to valley level in the Columbia Gorge area. About fif - teen per cent of the species listed by the writer (Detling 1958) as occurring in the Columbia Gorge are constituents of the boreal flora being considered here. Following are a few of the more striking representatives of the group. Species preceded by an asterisk occur also on Saddle Mountain. * Acer glabrum douglasii * Alnus sinuata Antennaria racemosa *Campanula rotundifolia Cornus canadensis *Cryptogramma acrostichoides Dodecatheon dentatum H abernaria unalaschensis H aplopappus hallii Lewisia columbiana Lomatium martindalei martindalei M itella trifida *Penstemon nemorosus Penstemon rupicola *Phlox diffusa longistylis *Polypodium vulgare columbianum * Saxifraga bronchialis vespertina *Saxifraga caespitosa * Saxifraga rufidula Suksdorfia violacea V accinium membranaceum All of these species occur in other boreal situa- tions more or less distant from the Columbia Gorge. Two other members of the Columbia Gorge boreal flora, Bolandra oregana and Sullivantia oregana, otherwise endemic in the area, are of interest here because they have been collected also at Elk Rock, a sheer northeast-facing bluff on the Willamette River at Oswego, about twen- ty miles from the west portal of the gorge. They furnish additional evidence that the gorge bor- eal flora formerly extended out over the Wil- lamette valley floor at least as far as Oswego. To summarize for this early Postglacial time we may assume for southwestern Washington and the Willamette valley, as well as for the valleys and ridges of the Coast Range and for the Columbia Gorge, a forest dominated by Pinus contorta and P. monticola, and in some areas by Abies procera and Picea sitchensis, with an understory composed of Acer glabrum douglasii, V accinium membranaceum, V. oval- ifolium, V. scoparium, Rhododendron albifior- um and Cladothamnus pyrolaefiorus. Forming the ground cover among the shrubs were the more or less woody trailing Rubus nivalis, R. pedatus, R. lasiococcus and Cornus canadensis, as well as Pyrola secunda and patches of Xero- phyllum tenax. On steep slopes and rocky ledg- es grew several species of Penstemon ( david- sonii, fruticosus , rattanii), Lomatium martin- dalei, Saxifraga f erruginea and other species of Saxifraga, Montia parvifolia fiagellaris, 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 37 V aleriana sitchensis scouleri, H euchera glabra, and the fern species Polystichum munitum im- bricans and Cryptogramma acrostichoides. Where conditions favored the persistence for any length of time of open meadows, these were dominated by Carex spp., Koelaria cris- tata and Phleum alpinum; but interspersed among the sedges and grasses were Lupinus Lepidus lyallii, Viola adunca, Erythronium grandiflorum, Senecio triangularis and Anem- one lyallii. Swampy places were occupied by various species of Salix, Ledum glandulosum columbianum, and such small herbaceous spe- cies as Drosera rotundifolia, H ypericum ana- gallidoides, Trientalis arctica and Ranuncu- lus reptans strigulosus. This is the assemblage of species which to- day occupies the slopes of the Cascade Range between the elevations of approximately 4500 and 5500 feet, and at lower elevations is essen- tially the forest of north central British Colum- bia. The littoral belt of the Boreal Forest may have differed to some degree from that in the Coast Range and the leeward valleys. Its cli- mate was certainly tempered by its proximity to the ocean. Picea sitchensis was dominant from the glacial maximum and Pinus contorta was present in considerable quantity as they are now. The coastal bog plants today are the same ones which were present in the Willam- ette valley during the period we are consider- ing and were undoubtedly common along the coast at that time. Except for Ledum we have no direct evidence that the shrubs list- ed above as forming the understory of the val- ley and Coast Range forest were present in the coast strip forest. However, several of the her- baceous species of the foregoing list were pres- ent in at least some parts of the coastal strip, e.g. , Xerophyllum tenax, Saxifraga spp., Val- eriana sitchensis and M ontia parvif olia flagel- laris. The Boreal Forest may have occupied the plateau south of the continental ice sheet east of the Cascade Range at least as far as the northern Great Basin if we can interpret the distribution of a number of typical boreal spe- cies ordinarily occurring today either in the Cascades or the Blue-Rocky massif, or fre- quently in both, as evidence of such an occupa- tion. They are also found on one or more of the higher mountains of the northern Great Basin, the widely dispersed peaks of the Steens, Hart, Warner, Pueblo and Mahogany Mountains, Gearhart Butte, Drake's Peak and Abert Rim, all separated by many miles of arid plateau in southeastern Oregon. Occurring at elevations of 6000-9000 feet on these peaks we find: Agoseris aurantiaca Arctostaphylos nevadensis Caltha leptosepala Cryptogramma acrostichoides Epilobium alpinum Habenaria dilatata leucostachys Hulsea nana Kalmia polifolia microphylla Lupinus Lepidus lyalli M itella pentandra Oxyria digyna Penstemon davidsonii Phleum alpinum Phyllodoce empetriformis Pinus albicaulis Polemonium elegans Polygonum bistortoides Populus tremuloides Salix barclayi Sibbaldia procumbens Sorbus scopulina Thermopsis montana montana V eratrum viride This distribution pattern gives evidence of a former more or less continuous range of the Boreal Forest at plateau level at least as far south as southeastern Oregon. The bog pollen profiles from eastern Wash- ington and Oregon and northern Idaho show that while Pinus contorta was the dominant ar- boreal species, P. monticola and P. ponderosa were significantly present from the beginning within a short distance of the face of the ice. The same is true of the grasses, composites and 38 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 chenopods. These more xeric species and gen- era were the ones which replaced Pinus con- torta as the glaciers retreated, and not Pseudo- tsuga or T suga as was the case west of the Cas- cades. As the Boreal Forest retreated northward and to higher elevations, it was succeeeded first by the Mesic Conifer Forest, and this in turn by a series of xeric austral vegetation types which included the Pine-oak Forest, and the woodlan and chaparral types. We have already noted the origin of these communities as a part of the Madro-Tertiary Geoflora early in the Tertiary and their culmi- nation during the middle Pliocene. Although some elements of this flora probably persisted in the Rogue basin through the cooler and moister period of the Pleistocene, as shown by the abundance of yellow pine at the beginning of the glacial retreat (Hansen 1947), it is doubtful that the more extremely xeric ele- ments such as the chaparral and pure oak wood- land would have withstood the conditions north of the Klamath highlands during that period. However, they were present throughout this time in northern and central California (Axel- rod 1958). Evidences of the northward advance of the more extremely xeric types following the re- treat of the continental glaciers, however, are to be found not only in the fossil pollen record but also, at least west of the Cascades, in a number of relict islands of this xeric vegetation. During the last few years the writer has made detailed studies of the distribution, floral con- stitution and ecological relations of some of these relicts ( Detling 1953, 1954). The is- lands that have been studied occupy special habitats consisting of exposed mountain tops or ridges, with shallow soil in which the more deeply-rooted trees or shrubs have probably never gained a foothold and thus have not en- tered into serious competition with the low- growing xeric species. They occur at elevations of from 2000 to 6000 feet, mostly along the west slopes of the Cascades or rising from the floor of the Puget trough. The only one seen and studied in the Coast Range is on Marys Peak in Benton-Lincoln Counties, Oregon. Those studied by the writer were all in Oregon, al- though others occur as far north as Puget Sound. In most cases the islands are bounded below by the Mesic Conifer Forest; a few of the higher ones are bounded by a belt of Boreal Forest. The most striking of the islands from the standpoint of numbers of xeric species sup- ported are those on Bohemia-Fairview Moun- tain, Rebel Peak and Horsepasture Mountain, all in eastern Lane County. Others of signifi- cant importance are on Spencer Butte and the Coburg Hills, both near Eugene. In one of the studies the writer analyzed the distribution patterns of the species associated specifically with chaparral, while in the other he was concerned with the relict distribution of Rogue basin xeric species in general. In the present discussion of the post-Pleistocene mi- grations of Madro-Tertiary elements in our flora, no advantage would be gained by disting- uishing among the various associations of the xeric flora. Many chaparral associates occur also in the oak and pine woodlands of south- western Oregon, and while chaparral is a dis- tinct vegetation type, it responds to climatic changes in the same way as do the other types comprising the xeric flora. As the writer pointed out in the earlier paper on xeric islands, the most significant feature of a vegetation area is not its boundaries but rath- er its environmental center, the point about which its environmental ( in this case climatic) extremes are concentrated, which point func- tions as a center of influence for the whole area. In most cases, and certainly in the Pacific Northwest, the positions of the environmental centers are determined by the topography of the region. Since our topography has not been ap- preciably altered since the close of the Pleisto- cene we may reasonably assume that the centers of climatic extremes, and therefore of vegeta- tion areas, have also remained unchanged. What changes is the intensity of influence of the combined climatic extremes at the environmen- 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 39 tal center. If the climate of the Pacific North- west were to undergo a trend toward increased warmth and dryness the influence of the warm and dry extremes in such areas as the Rogue and the Deschutes would be intensified, and the floras of those areas would spread progres- sively away from the centers at the expense of the floras of areas such as the South Cascade and Siskiyou which have had a cool-humid combination of extremes. In the case of the Rogue Area, the xeric yellow pine-oak type un- der such conditions would migrate northward and to higher elevations around the Rogue basin, followed by the more xeric chaparral and finally, if the climatic trend continued, by grassland types. It is difficult or impossible to trace through the xeric islands the former distribution of the shrubby species so characteristic of the chapar- ral. Their absence from the xeric sites probably is due largely to the shallowness of the soil, al- though other limiting factors may also play a part. Rhus diversiloba, while abundant in the chaparral, is so widespread in the Pine-oak Forest west of the Cascades and even in the drier phases of the Mesic Conifer Forest, that its value in determining the former range of its chaparral associates is negligible. Relict stands of Ceanothus cuneatus occur in the deeper soils of the Puget Area as far north as the Columbia River, especially on well drained sites such as those underlain by old gravel bars. Garrya fremontii, a common chaparral species in the Rogue Area, occurs occasionally as far north in the Puget Area as the McKenzie River, and again in the Columbia Gorge. Chrysothamnus nauseosus albicaulis, less typical as a chapar- ral species but still a member of the associa- tion, occurs in several of the Puget xeric is- lands. These four are the only chaparral shrubs known to the writer to extend northward from the Rogue basin. All other chaparral associates found relict north of this area are herbaceous. While a considerable number of Madro- Tertiary genera are found on both sides of the Cascade axis, the number of species that are generally distributed at present both east and west of the range is not large. More numerous are those species which have a fairly wide dis- tribution, but on one side only of the Cascades. Examples of such species west of the moun- tains are Quercus garryana, Arbutus menziesii, Corylus cornnta, Phacelia heterophylla, Cean- othus sanguineus, Erysimum capitatum and Godetia amoena. Widespread east of the moun- tains are Purshia tridentata, Artemisia triden- tata, Cercocarpus ledifolius, Ceanothus veluti- nus, Descurainia pinnata and Lupinus laxi- florus. Another group of xeric species, however, is much more closely restricted to one or some- times two areas today, and we find no evidence of their ever having spread beyond the confines of their present limits. Such are: Quercus chry- solepis, Arctostaphylos viscida, A. canescens, Eriodict'j"On californicum, Chlorogalum pom- eridianum and Fritillaria recurva, restricted to the Rogue Area and California west of the Sierra Nevada; Asclepias cordifolia, Cerco- carpus betuloides, Convolvulus polymorphus, Lomatium californicum, M ahonia piperiana, Micropus californicus and Rhus trilobata, oc- curring typically in the Rogue Area but ex- tending at least into the Klamath basin portion of the Washoe Area ( Fig. 3). Between these extremes of wide diffusion and restricted distribution is a large group of species which are much more pertinent to our study because of their present-day distribution and their occurrence in the relict islands of xeric species. The distributions of most of these involve the Rogue Area, and when they do they usually continue southward through the warm canyons of the Klamath Mountains and the foothills of the interior valleys of central Cali- fornia. Many of them are chaparral associates, either typically or more or less incidentally. The simplest distribution pattern of this group, and a very common one, consists of the Rogue Area with outlying stations in the xeric islands or other isolated xeric sites of the Puget Area. Examples of this pattern are seen in Ceanothus cuneatus, Erigeron foliosus confinis, Hieracium cynoglossoides nudicaulis and Si- 40 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 dalcea asprella. A relatively small number of species with this basic distribution occurs also in the Klamath basin, but apparently no far- ther north east of the Cascades. Castilleja pru- inosa and Viola sheltonii are examples. A significant variation of this pattern is one in which typical Rogue species appear in the xeric islands ( rarely more extensively) of the Puget Area and also in that portion of the Co- lumbia Area contiguous with the east end of the Columbia Gorge. A few of these Rogue spe- cies may occur in the Klamath basin but are not found farther north on the east side of the Cascades. This distributional variant is repre- sented by such species as Eriogonum composi- tum pilicaule, Sisyrinchium douglasii, Quer- cus garryana, Dentaria tenella pulcherrima, Lupinus bicolor, L. micranthus, Tonella tenel- la, Linanthus bolanderi, Orthocarpus attenu- atus, Pectocarya pusilla and Plagiobothrys no- thofulvus. One of the most interesting of these examples is afforded by Eriogonum composi- tum. The typical variety of this species ( no- menclatorially), with glabrous scapes and pe- duncles, occurs commonly east of the Cascades except in the Columbia Area ( Fig. 3). The var- iety pilicaule, characterized by pubescent scapes and peduncles, replaces this in the Rogue and Siskiyou Areas and in the Columbia Area at the head of the Columbia Gorge, and is the form which is always found on the xeric islands of the Puget Area where it occurs with considerable frequency. A lesser number of xeric island species are typical of arid vegetation areas east of the Cas- cades, namely one or more of the Washoe, Des- chutes or Columbia Areas (Fig. 3). They may or may not occur also in the Rogue Area. They include such species as Chrysothamnus nauseo- sus albicaulis, Phacelia linearis, Artemisia tri- dentata, Gila aggregata, Collomia linearis, Pen- stemon deustus, Lupinus lepidus medius, Aren- aria formosa, Bromus polyanthus, Amica par- ryi and Sanicula graveolens. The most interest- ing refugium of this group of species is prob- ably the rather extensive colony of Artemisia tridentata which occurs on Rebel Peak in east- em Lane County at an altitude of 5384 feet, as- sociated with about twenty other xeric species normally found in the Rogue or Deschutes Areas, or both. This is the only occurrence of sagebrush so far reported from west of the Cascade crest. From the evidence of the relict distributions summarized in the foregoing paragraphs we can infer the movements of the xeric flora in the Pacific Northwest with the return of the warmer and drier climate following the retreat of the continental glaciers. The migration was merely the resumption and continuation of the northward spread of the Madro-Tertiary Geoflora that was tempo- rarily interrupted by the cold-humid period of the late Pliocene and Pleistocene. In this region it involved any species that might have sur- vived in the Rogue basin during the generally changed conditions, and those species that had reached and remained in northern and central California and the central areas of the Great Basin. The high elevation of the Sierra-Cascade axis prevented any general east-west inter- change of the flora until late in the xerothermic trend. Exceptions may have been the Klamath River Gap between the Central Valley of Cali- fornia and the Klamath basin, the Columbia Gorge, and most probably the Fraser River Gap in southern British Columbia. West of the Cascades the first Madro-Ter- tiary species to migrate northward with the ad- vent of warmer and drier conditions were un- doubtedly the pine-oak associates that had per- sisted in the Rogue basin. With increasing warmth and aridity the low cross ranges such as the Wolf Creek and Calapooya Mountains be- came less effective as barriers, and the xeric flora moved north and attained dominance through the Puget trough as far as Vancouver Island and the lowlands of mainland south- western British Columbia. The continuing xerothermic trend brought new genera and species over the Klamath Mountains from northern California, and the chaparral type became established in the 1968 DETLING : HISTORICAL BACKGROUND NORTHWEST FLORA 41 Rogue basin. This moved northward at lower elevations as the earlier migrants into the Puget trough moved to higher elevations in the Cas· cades and the Coast Range. Xeric species occupied extensive warm open slopes and ridges in the Cascades to an eleva- tion of 6000 feet, and at this stage must have mixed with an ascending xeric flora from the east slopes of the range to form such communi- ties as that on Rebel Peak. Boreal conditions in the strip along the Cascade summit were least effective as a barrier at this stage. Meanwhile the xeric flora that had reached the lower Willamette valley, made up of Quer- cus garryana and its associates, was migrating eastward through the Columbia Gorge, and hav- ing reached the upper end, spread out over the warm valleys of the Deschutes, John Day and Yakima Rivers and such moderate uplands as the Simcoe Mountains. Later migrations through the Columbia Gorge included some of the herbaceous typical chaparral associates. On the east slope of the Coast Range, border- ing the Willamette valley, the early migrants of the xeric flora ascended at their maximum to an elevation of 4000 feet. On a few of the higher peaks they mingled with but failed to sup- plant a relief boreal flora persisting from the glacial maximum. This presumably explains the occurrence today on Marys Peak of the xeric Allium crenulatum, Eriogonum umbella- tum and Silene douglasii alongside the boreal Abies procera, Erythronium grandifiorum pal- lidum, Lomatium angustatum fiavum, Lupinus Lepidus lyallii, Phleum alpinum and Polygo- num bistortoides, and on Saddle Mountain, Al- lium crenulatum, Festuca howellii, F. pacifica, Poa secunda, Senecio fastigiatus and Silene douglasii associated with such boreal species as Abies procera, Acer glabmm douglasii, An- emone globosa, Cladothamnus pyrolaeflorus, Lewisia columbiana rupicola, Rudbeckia oc- cidentalis, Saxifraga bronchialis vespertina and V accinium scoparium. On the basis of precipitation and tempera- ture data selected from recording stations in the Shasta River valley and Rogue basin chaparral belt, and from others in the transverse ranges and areas where chaparral relicts occur today, as well as from Klamath Falls and The Dalles, the writer has postulated that for the more xeric phases of the Madro-Tertiary Flora to cross the ecologic barriers interposed by the higher ridges of the Siskiyou Mountains and the transverse ranges to the north it would re- quire that the mean annual rainfall west of the Cascades be decreased by about 230 mm below that prevailing at present with not more than twenty-seven per cent occurring during the six driest months of the year. At the same time a general increase in the July mean maximum temperatures of 5.0° Cover that of today would be required. An increase of only 1.0° C in the January mean minimum temperatures would have favored the invasion of the Klamath basin and the Columbia Area above the Columbia Gorge by chaparral of its present associates (Detling 1961). The general aridity just cited as part of the requirement for the elimination of the ecologic barriers to the northward progress of chaparral undoubtedly decreased rainfall in the broader valleys to a figure below the 250 mm which Cooper ( 1922) found to be the minimum re- quired by chaparral in California. As a conse- quence, during the xerothermic maximum chaparral probably was replaced by grassland in the lowlands of the Rogue, Umpqua and Willamette valleys and in the area about The Dalles. As pointed out earlier, bog pollen profiles indicate that the xerothermic trend was initi- ated about 13,000 years ago (Reusser 1960), at which time the postglacial northward migra- tion of the Rogue River Madro-Tertiary Geo- flora got well under way. By the same source of evidence the xerothermic maximum was reached about 6000 years ago, the time at which the xeric vegetation reached its greatest northward advance and greatest altitudinal dis- tribution, and had spread into the area east of the Columbia Gorge and into the Klamath basin. Much of the floor of the basins west of the Cascades was probably covered by grass- 42 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 land, and near-desert conditions may have ob- tained along the Columbia River from near the head of the gorge eastward to its confluence with the Snake River. The trend toward a cooler and more humid climate since the xerothermic maximum has witnessed the reversal of the vegetational mi- grations. The increasingly more mesic types, pine and pine-oak woodlands and Mesic Coni- fer Forest, moving southward in the basins and downward from the mountains replaced the grassland and chaparral. Only in especially favorable sites have a few of the xeric species persisted, and some of these colonies may have been reduced to actual "islands" only within the last one or two thousand years. East of the Cascades the postglacial migra- tions of the Great Basin flora followed the same general pattern as that of the Rogue. While we have nothing from xeric refugia upon which to base this statement, Hansen's pollen studies on bogs east of the mountains bear out such a conclusion. We have already pointed out in this connection that at the end of the Pleistocene the Boreal Forest occupied at least the uplands as far south as the northern Great Basin, with yel- low pine occupying the lower slopes and grasses and chenopods on the plateau. We may recall also that Axelrod has shown that by middle Pliocene, when the Madro-Tertiary Geoflora reached its farthest north pre-Pleistocene ex- tension, the northern Great Basin was occupied largely by grassland and by arboreal genera of northern rather than southern origin. Judging then by these sources of evidence it seems extremely likely that a major north- ward migration of Great Basin Madro-Tertiary Geoflora was initiated at the end of the Pleisto- cene and kept pace with the advance of the Californian species of the same flora from the Rogue basin. Yellow pine woodland as a phase of the more southern and western Pine-oak Forest ex- tends at present as far north as Kamloops Lake in central British Columbia, and is well repre- sented along the Thompson River between Lyt- ton and Kamloops, and in adjacent valleys such as that of the Nicola River. Mingled with the woodland in several of the valleys are areas of well developed sagebrush type. Within these mingled types occurs an association of species of austral origin common in the pine or Juni- per-sagebrush Woodland to the south. These are the species that moved north with the xero- thermic trend, reaching the southern margin of the Fraser Plateau by the time of its maxi- mum. Included in the association are: Amelanchier alnifolia cusickii Artemisia tridentata Astragalus beckwithii Astragalus purshii Astragalus stenophyllus Balsamorhiza careyana Balsamorhiza sagittata Castilleja thompsonii Chaenactis douglasii Chrysopsis villosa hispida Chrysothamnus nauseosus albicaulis Chrysothamnus viscidiflorus Cirsium undulatum Clematis ligusticifolia Delphinium bicolor Descurainia pinnata filipes Erigeron filifolius Erigeron pumilus Erigeron speciosus Eriogonum heracleoides Eriogonum niveum Fragaria vesca bracteata Gaillardia aristata Cilia aggregata H euchera cylindrica Juniperus scopulorum Lithospermum ruderale Lomatium dissectum multifidum M ahonia re pens M entzelia laevicaulis parviflora Opuntia fragilis Orobanche f asciculata Phacelia hastata leucophylla Phacelia linearis Prunus virginiana melanocarpa Purshia tridentata Rhus radicans 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 43 Ribes cereum Senecio canus Sieversia ciliata Symphoricarpos albus laevigatus Zygadenus venenosus RECENT As the climate became warmer and drier in the wake of the receding continental ice, the Boreal Forest flora retreated northward and upward in the mountain ranges. This move- ment was more clear-cut in areas removed from oceanic influence. Along the narrow coastal strip it was modified to the extent that while there is evidence that many boreal spe- cies did move northward, at the same time many persisted at sea level. In the Coast Range south of the Olympic Mountains the mild cli- mate of the low, more inland mountains re- sulted in the elimination of the boreal flora except in a few restricted localities such as the summits of Saddle Mountain, Mount Hebo and Marys Peak (Fig. 1). The final result of the migration was the establishment of a broad zone of boreal flora extending northward from central British Co- lumbia and reaching from the coast to the Rocky Mountains, with three fingerlike exten- sions of the same flora extending southward down the coastal strip, the Cascades and the Blue-Rocky Mountain massifs (Fig. 4). It is not known to what extent the British Columbia boreal zone was repopulated by the flora that moved north from below the ice limit, and to what extent from refugia that may have existed within the glaciated region. Various authors have pointed out that ice-free areas did exist during the Wisconsin phase of the ice age in the Canadian Rockies as well as in northern British Columbia and the Yukon (Hulten 1937, Hansen 1947, 1950, 1953, 1955). Be that as it may, it is certain that all vegetation was elim- inated from a broad strip for many miles be- hind the ice front, extending from the Puget Sound area to the Rockies and beyond, and that this strip was apparently repopulated largely by the boreal species that occupied the terrain contiguous to the ice front at the beginning of its recession. With the amelioration of the climate many of the species have persisted in more than one of the southerly extensions of the boreal flora in Canada ( Fig. 4). The following list is rep re· sentative of this group of species. The occur- rence of each species in the various fingers south of glaciation is shown by the symbols C ( coastal strip), Cas ( Cascade Range), B ( Blue Mountains) and R (Rocky Mountains). An asterisk following a symbol indicates that the species occurs only a very short distance south of the limit of glaciation. Abies amabilis C-Cas Abies lasiocarpa Cas-B-R Acer glabrum douglasii Cas-B-R Agoseris aurantiaca Cas-B-R Alnus sinuata C-Cas-B-R Anemone occidentalis C-Cas-B-R Antennaria racemosa C-Cas-B-R Arctostaphylos uva-ursi C-Cas-B-R Amica latifolia Cas-B-R Betula glandulosa Cas-B-R Betula papyri/era occidentalis Cas*-R* Caltha leptosepala Cas-B-R Cardamine bellidifolia Cas-R Cassiope mertensiana Cas-R* Cornus canadensis C-Cas-R Crepis nana Cas-B-R Cryptogramma acrostichoides C-Cas-R Empetrum nigrum G-Cas* Epilobium alpinum Cas-B-R Epilobium latifolium C-Cas-B-R Eriophorum chamissonis C-R* Eriophorum gracile Cas-B-R Erythronium grandifiorum C-Cas-B-R chry sand rum R grandifiorum Cas-B-R pallidum C-Cas-B-R Gaultheria humifusa C-Cas-B-R Gaultheria ovatifolia Cas-B-R H abenaria dilatata C--Cas- B-R H euchera glabra C-Cas H ieracium gracile Cas-B-R densifioccum B-R detonsum Cas 44 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 Juniperus communis montana C-Cas- B-R Kalmia polifolia C-Cas-B-R microphylla Cas-B-R polifolia C-Cas Ledum glandulosum C-Cas-B-R columbianum C glandulosum Cas-B-R Led um palustre groenlandicum C-R * Leptarrhena pyrolaeflora C-Cas-R * Lonicera involucrata C-Cas-B-R involucrata C-Cas-B-R ledebourii C-Cas-B-R Luetkea pectinata Cas-B-R M aianthemum dilatatum C-Cas-R * M itella breweri C--Cas-R * M itella pentandra C-Cas-B-R Mitella trifida C-Cas-·B-R M yrica gale C Oxyria digyna C-Cas-B-R Penstemon davidsonii C-Cas davidsonii Cas menziesii C-Cas Petasites frigidus Cas Petasites sagittatus R Phleum alpinum C-Cas-B-R Phyllodoce empetriformis C-Cas-B-R Phyllodoce glanduliflora Cas-B-R Picea engelmannii Cas-B-R Picea sitchensis C Pinus albicaulis Cas-B-R Pinus contorta C-Cas-B-R contorta C murrayana Cas- B-R Polygonum bistortoides C-Cas-B-R Pyrola secunda C-Cas-B-R Rhododendron albiflorum C-Cas-B-R Rubus nivalis C-Cas-R * Rubus pedatus C-Cas-R * Saxifraga bronchialis C-Cas-B-R austromontana Cas-B-R vespertina C-Cas Saxifraga caespitosa C-Cas-R emarginata C-Cas * minima R subgemmif era C-Cas Saxifraga f erruginea C-Cas-R * Saxifraga oppositifolia C-Cas*-B-R Saxifraga punctata cascadensis C-Cas Saxifraga tolmiei C-Cas-R * Senecio triangularis C--Cas-B-R angustifolius C triangularis C-Cas-B-R Shepherdia canadensis B-R Sibbaldia procumbens Cas-B-R Sorbus scopulina C-Cas-B-R cascadensis C-Cas scopulina C-Cas-B-R Sorbus sitchensis C--Cas-R grayi C-Cas sitchensis C-Cas-R * Stenanthium occidentale C-Cas-R Trientalis arctica C-Cas-R* T suga mertensiana C-Cas-B-R V accinium caespitosum C-Cas-R Vaccinium ovalifolium C-Cas-B-R V accinium scoparium Cas--B-R V accinium uliginosum C-Cas-R V eratrum viride Cas-B-R Viola sempervirens orbiculata Cas-B-R Xerophyllum tenax C-Cas-R A lesser number of species apparently did not follow in the wake of the retreating ice front but have been stranded instead in one or more of the boreal fingers, and have maintained no connection in the north. Examples of this type of distribution are shown in the following list: Arctostaphylos nevadensis Cas-B Bolandra oregana Cas-B imnahensis B oregana Cas Calochortus lobbii Cas Chamaecyparis nootkatensis C-Cas-B C ladothamnus pyrolaeflorus C Claytonia megarhiza Cas-B-R bellidifolia B megarhiza R nivalis Cas Collomia debilis C-Cas-B-R debilis Cas-B-R larsenii C-Cas Dodecatheon dentatum Cas-R Douglasia laevigata Cas 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 45 ciliolata Cas laevigata Cas ( Columbia Gorge) Draba aureola Cas Erigeron howellii Cas Erigeron oreganus Cas Hieracium longiberbe Cas H ulsea algida B-R Hulsea nana Cas Lewisia columbiana C-Cas-B columbiana Cas rupicola C-Cas wallowensis B Lomatium martindalei C-Cas angustatum C-Cas flavum C martindalei Cas Lupinus lepidus lyallii C-Cas-B-R M icroseris alpestris Cas Penstemon rupicola Cas Polemonium elegans Cas-B Polemonium viscosum B-R Polygonum newberryi C-Cas Polygonum phytolaccaefolium Cas-B-R Rubus lasiococcus C-Cas Salix hookeriana C Saxifraga tolmiei C-Cas-R Suksdorfia violacea Cas-R Sullivantia oregana Cas Synthyris missurica B-R Synthyris schizantha C-Cas Synthyris stellata Cas V accinium deliciosum C-Cas V accinium membranaceum C- Cas- B-R V accinium occidentale Cas-B-R V accinium ovatum C The significance for evolution of the partial or complete isolation of population fragments in the mountain or coastal fingers is obvious. Gene mutations occurring in one fragment have been separated by such great distances from the rest of the population that in numerous cases divergent evolution has proceeded to the point where separate subspecies or even sepa· rate but closely related species now occupy one or more distinct fingers. Such undoubtedly has been the origin of the subspecific differentia- tion noted in the two preceding plant lists. The climatic trend of the last 4000 years, toward cooler and moister conditions ( Hansen 194 7, Reusser 1960), has brought the Pacific Northwest flora to its present status. With pos- sible temporary unrecorded reversals the Mesic Conifer Forest has again become dominant over much of the western portion of the region. It has largely replaced the xeric Rogue flora except for numerous refugia where soil condi- tions have probably hindered its competitors. Many Madro-Tertiary species have been suffi- ciently adaptable to become incorporated into the mesic forest where we find them, especially in its drier phases. On the valley floors the mesic forest has not taken complete possession; the chaparral has retreated southward, but con- siderable areas are occupied by oak wood- land, and occasional stands of ponderosa pine persist. Judging by the vestiges of Madro-Tertiary associations remaining in the high mountains, the Bo real Forest has increased in area along the crest of the Cascades, descending the west- ern slopes by 1000 feet or more, and widen- ing the strip it occupies to the point where the vegetation types on either side are now far separated. The coastal strip of Boreal Forest was prob- ably the least affected by the xerothermic trend of any of our types. No evidence of the Madro- Tertiary invasion is to be seen north of the central Oregon coast, and although the Mesic Conifer Forest encroached greatly upon it from the landward side during the warm period it has maintained itself intact, including the nu- merous bog associations that have persisted since the glacial period. East of the Cascades, even with the relative increase in humidity and lowering of temper- atures in the last 4000 years, the region is still one of actual aridity. Consequently the Madro- Tertiary elements still dominate except at higher elevations, as in the mountains of south- ern British Columbia and the Blue Mountains. The only change over the period has been a migration down slopes and out on to the plateau 46 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 of the yellow pine forest at the expense of sage- brush and grassland. Certain genera and families in the Madro- Tertiary Geoflora have undergone marked evo- lutionary development during their adjustment to the changing Tertiary climate, and have given rise to specially adapted forms appear- ing in different vegetation types. This seems to account for the presence in the more northern Mesic Conifer and Boreal Forests of members of typically drought-adapted genera. The more recently derived northern species are usually taller plants with larger leaves, mesophytic structure and surfaces. Some of the outstand- ing examples of this species evolution and adaptation are the following: a) Ceanothus cuneatus and C. cordulatus in the chaparral, replaced by C. velutinus in the Pine-oak Forest, especially east of the Cas- cades, and by C. thyrsiflorus and C. integerri- mus in the drier phases of the Mesic Conifer Forest. b) Arctostaphylos viscida and A. canescens in the chaparral, represented by A. patula in the Pine-oak Forest and by A. columbiana in the coastal Bo real Forest. c) Rhamnus crocea in the chaparral, and R. purshiana in the Mesic Conifer Forest and coastal Boreal Forest. d) Mahonia repens in the Pine-oak Forest east of the Cascades, replaced by M. aquifolia in dry phases of the Mesic Conifer Forest. M. nervosa apparently is of more ancient origin and is related to East Asian species and has probably come into the Mesic Conifer Forest from the north. e) Holodiscus glabrescens of the arid Pine- oak and Juniper Woodlands, and H. discolor of dry situations in the Mesic Conifer Forest. f) Cercocarpus ledifolius and C. betuloides of the Juniper Woodland and Chaparral For- mations, respectively. g) Garrya fremontii of the chaparral and G. elliptica of the coastal Bo real Forest. h) Rhus trilobata in the chaparral, repre- sented in both chaparral and dry phases of the Mesic Conifer Forest by R. diversiloba and in the Juniper Woodland by R. toxicodendron. It is of interest to note that in several of the genera drought-adapted species in arid forma- tions are replaced by morphologically more mesic species or subspecies in the Mesic Coni- fer or Boreal Forests. They are represented in Mexico by similarly mesic forms ( taller plants with mesophytic foliage, etc.). These features are displayed by such Mexican species as Ce- anothus coeruleus, Arctostaphylos pungens and Mahonia fascicularis. VEGETATION AREAS IN WESTERN MEXICO (The following section consists of brief and incomplete notes dealing with work done in Mexico in connection with the geologic history of the Northwest flora. Most of the plant collec- tions are as yet unidentified.) According to paleobotanical studies ( Axel- rod 1938, l 950a), a large part of the xeric flora of western United States developed in response to warm dry climates in the Tertiary, evolving from a subtropical flora that already existed in northwest Mexico. The closest pres- ent relatives of this Madro-Tertiary Geoflora are centered in the western Sierra Madre of Mexico ( Fig. 6). Since there are several close- ly related vegetation types in that region through which evolutionary lines from sub- tropical to desert can be traced, it is of consid- erable interest and value to determine these pathways from their origins eventually to their contributions to our present western United States flora. We may distinguish six major vegetation types in western Mexico: Subtropical Forest, Xeric Matorral, Thorn Savanna, Desert Scrub, Pinyon-juniper Woodland and Pine-oak For- est. Degrees of relationship between major types may be shown by the relative number of genera ( sometimes families) which they hold in common. Sometimes a genus ( e.g. Bombax, Ceiba) may be represented by several species, all of which have remained typically subtropi- cal in their distribution. On the other hand, 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA I \ \ 47 ---------1 ----- Pin c-oa k Fores t ••••••••••••••• Ju11iper -Sagebru sh I) .... Pue rl o Va I lar I · .... ·· .. ... MJ..seta \ C e 00 n Ir a I oAguascali CJil i's..- " "11 ., ..... , .. ... ahuali ca ::::'~ /,' .:::.I~ Guadal ajara ,\JI/, '?; ;:-- i/ //1'~ "//1' , <;).L. Pii1 zcuaro ,,Ii,,, ~II/,, /',:: //\' //\I J/;_ 111'' Figure 6. Tertwry migration routes of the Ma-dro-Tertwry Geo/fora. I I 48 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 genera such as Acacia have some species ( coch- liacantha, hindsii) which are typically sub- tropical, while other species, through long pe- riods of evolution, have adapted to the dry up- land conditions of the Xeric Matorral (pen- natula), or the even drier Thorn Savanna of the central plateau (farvesiana, tortuosa). Sometimes a whole family has undergone a great burst of adaptive evolution almost exclu- sively in one type of environment. As an exam- ple the Compositae have developed a great many species and even genera in the dry belts of the Xeric Matorral, Thorn Savanna and Desert Scrub, while they are only sparsely rep- resented in the warm-humid Subtropical Forest. The genera that constituted the Madro- Tertiary Geoflora are represented today in three vegetation types occurring in northwest Mexico: a) Thorn Savanna in which grasses and other low herbaceous genera are mingled with well- spaced shrubs and low trees, largely of the Mimosaceae; b) Xeric Matorral made up of a moderately dense stand of shrubs of such genera as Acacia, Mimosa, Salvia, Hyptis, Asterohyptis, Solan- um, Bursera, Dalea, Eysenhardtia, and Te- coma, along with numerous herbaceous to woody species of Papilionaceae and Composi- tae; c) Pine-oak Forest, dominated by numerous species of Pinus and Quercus, and containing species of Arbutus, Arctostaphylos and Ceano- thus. It is largely this last type which is repre- sented in the Pacific Northwest vegetation, and has contributed the most characteristic genera making up three of our austral formations, the Juniper-sagebrush Woodland, Chaparral and Pine-oak Forest. The Subtropical Forest may be considered one of the two basic vegetation types of western Mexico ( the other is the Pine-oak Forest). It is the one which the paleobotanists infer gives rise to the Madro-Tertiary Geoflora and so from the historic standpoint is basic also. Many of the genera are still restricted to tropical or subtropical regions, but some have evolved species adapted to more arid condi- tions, and are important components of some of the other vegetation types. This probably reflects the process which gave rise to the drought-adapted Madro-Tertiary Geoflora. In western Mexico the type occurs on the lowlands bordering the Pacific Ocean, from about latitude 23° at Mazatlan south, and on the foothills and lower west slopes of the Sierra Madre. It also follows the deep barrancas of some of the rivers far inland. A striking exam- ple of this last feature is its occurrence along the bottom and lower slopes of the canyon of the Rio Santiago from its mouth near San Blas, inland as far as Guadalajara, and even up its tributary, the Rio Verde, to near Y ahualica (Fig. 6). The type is characterized by the presence of a large number of species at any given locality, none of which, however, is truly dominant. Herbaceous species are rare; the maximum height of the trees seldom exceeds 10 m., and there are abundant shrubby species of almost impenetrable density, of varying heights and with no apparent stratification. The Rio Verde forest has species invading from the adjacent xeric upland, e.g., Trago- ceros schiedianum, Acacia f arnesiana, A. pen- stula, Dahlia coccinea, Zinnia peruviana. See field note comments for numbers 8953 ff. (Field note: "Rio Verde at the bridge, Elev. 1434 m.-4700 ft. River bank. On previous visits this was listed as subtropical forest. Such is not apparent now, but rather it appears as thorn type.") In J alisco, pure subtropical forest is prob- ably coastal only, as at Manzanillo and Puerto Vallarta, also farther north as at Villa Union and San Blas. This forest has followed up the rivers and barrancas as far as Guadalajara and Rio Verde, but since it occurs in narrow strips cutting through the more xeric upland types it is infiltrated by species from the latter. This explains the situation on the Rio Verde men- tioned above. Following is a list of plants typical of the Subtropical Forest in western Mexico: 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 49 Acacia cochliacantha Acacia hindsii Adiantum princeps Agdestis clematidea Anona longiflora Antigonon flavescens Antigonon leptopus Aristolochia sp. Baukinia longiflora Bravaisia integerrima Bursera kerberi Capparis verrucosa Casearia pringlei Cassia emarginata Cissus sichyoides Clematis dioica Combretum farinosum Conzattia multiflora Corchorus siliquosus Coria eleagnoides Exogonium bracteatum Guazuma tomentosa Hamelia versicolor Laguncularia racemosa Lygodium venustum Mimosa invisa Mimosa pigra Mimosa spirocarpa M omordica charantia Oxybaphus viscosus Passiflora foetida gossypifolia Peperomia campylotropa Pluchea odorata Pouzolzia palmeri Randia armata Rhus barclayi Salix humboldtiana Serjania mexicana Solanum bicolor Solanum umbellatum Tournefortia hirsutissima Xylosma velutinum Under very xeric conditions oak occurs as a pure oak woodland, and then tends to be infiJ. trated by Xeric Matorral species. This is the case with the Q. resinosa woodland 34 km. west of Aguascalientes. Pinus oocarpa is the most xeric of all the pines. Where it occurs as co-dominant with Quercus ( as at 16-17 km. west of Guadala- jara), it is associated with many Xeric Mator- ral species, just as in the pure oak woodland facies. The genus Dalea is best developed in the Oak-Woodland facies or Oak-P. oocarpa wood- land of the Pine-oak. It has two species in the Xeric Matorral of Ixtlahuacan. Probably the genus should be thought of as intermediate be- tween these two xeric types. There are no Dal- ea in Subtropical Forest. Following is a list of the Pine-oak Forest association: Adiantum andicola Adiantum patens Adiantum poiretii Alnus glabrata Alnus jorullensis Arbutus glandulosa Arbutus xalapensis Arctostaphylos pungens Asphodelzts fistulosus Calliandra grandiflora. ( 0-W facies) C eanothus coeruleus Ceanothus durangoinus Centaurium macranthum Cheilanthes angustifolia Cheilanthes kaulf essii Cheilanthes pyramidalis C lethra mexicana Crataegus rosei Cuphea aequipetala ( 0-W facies) Dalea gracilis ( 0-W facies) Dalea pectinata ( w. P. oocarpa) Dalea submontana ( w. P. oocarpa) Dalea tomentosa ( 0-W facies) Dalea tuberculata ( 0-W facies) Dodonaea viscosa ( 0-W facies) Dryopteris patula ( 0-W facies) Erigeron tenellus Eupatorium areolare (0-W facies) H ypoxis rugosperma H yptis oblongif olia ( w. P. oocarpa) lpomoea heterophylla (0-W facies) M ahonia f ascicularis 50 ' . BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 M alaxis aurea Malaxis pringlei Oxalis decaphylla Oxypappus seemannii ( w. P. oocarpa) Phaseolus coccineus ( 0-W facies) Pinaropappus roseus ( 0-W facies) Pinus cooperi Pinus engelmannii Pinus lumholtzii Pinus michoacana Pinus oocarpa Pinus teocote Prunus capuli Pteridium aquilinum ( w. P. oocarpa) Quercus aristata Quercus bolanyosensis ( w. P. oocarpa) Quercus coccolobaef olia Quercus crass if olia Quercus diversifolia (0-W facies) Quercus durif olia Quercus eduardii ( 0-W facies) Quercus grisea Quercus laxa Quercus macrophylla (0-SF) Quercus magnoliaefolia (0-SF) Quercus mexicana Quercus microphylla Quercus omissa Quercus praineana (0-W) Quercus resinosa ( 0-W facies, or w. P. oocarpa) Quercus rugosa Quercus rugulosa ( 0-W facies) Quercus transmontana ( 0-W facies) Ranunculus petiolaris Salix bonplandiana Selaginella pallescens (O~W facies or Oak-SF) Senecio stoechadif or mis Spiranthes velata T agetes lucida Thalictrum tripeltiferum (0-W facies) Triumfetta brevipes ( 0-W facies) Trixis hyposericea ( 0-W facies) Trixis longifolia ( 0-W facies) V accinium geminiflorum Vernonia jaliscana ( 0-W facies) Vernonia mucronata ( w. P. oocarpa) Viola grahamii W igandia caracasana ( 0-W facies) Yucca decipiens (0-W facies) The Desert Scrub Formation is an edaphic phase of the Juniper Woodland dependent up- on extreme saline ( or otherwise basic) condi- tions. It is more common in Nevada and Utah, but in the Oregon portion of the Great Basin it occurs on the playas of ancient lakes or on sur- rounding lower slopes. The genera are austral in origin ( except Artemisia, and this genus is probably of austral origin) and developed in the southwestern deserts during the Madro- Tertiary evolution and migrations. Constituents of the Desert Scrub Formation: Sarcobatus vermiculatus Atriplex confertifolius Grayia spinosa Artemisa spinescens Atriplex canescens T etradymia ? Eurotia All the ferns collected at Lake Patzcuaro were along a (now) dry watercourse. Normal· ly they would belong in the pine forest. Patzcu- aro is originally noted as a mixed pine-oak- thorn forest. See field notes 8481 ff. The Chapalita district of Guadalajara is un- doubtedly of thorn savanna type. This is at- tested by remnants of this type around the Poli- tecnico, in the fields or roadsides just outside of town, Zapopan, etc., where such genera as Prosopis and Pithecolobium are common. How- ever, a number of xeric matorral species are abundant in the vacant lots of Chapalita. The same is true of Zapopan. Does intensive land use in the savanna encourage the infiltration (development) of matorral? Examine carefully the association on the slopes about 19 km. north of Tapalpa (Fig. 6). Now classified as Thorn Savanna, but they seem to be nearer Xeric Matorral because of some genera. Axelrod refers to this as mesquite grassland (1958). 1968 DETLING : HISTORICAL BACKGROUND NORTHWEST FLORA Constituents of the Thorn Savanna: Acacia farnesiana Acacia tortuosa Anisacanthus thurberi Argemone ochroleuca Asclepias linaria Aster tanacetifolius Baccharis ramulosa Bidens pilosa Buddleia scordioides Buddleia sessiliflora Calliandra oaxacana C Zematis drummondii Cynanchium kunthii Desmondium sericophyllum Ficus mexicana Ficus petiolaris Grindelia oxylepis Haplopappus spinulosus scabrellus Haplopappus venetus Hyptis albida Jatropha dioica Karwinskia latifolia Mimosa monancistra Muhlenbergia rigida Nicotiana glauca Oenothera speciosa Parthenium bipinnatifidium Pellaea sagittata Pigueria trinervia Pithecolobium acatlense Pithecolobium dulce Prosopis juliflora Psilactis brevilingulata Psittocanthus calyculatus Rhynchelytrum roseum Sanvitalia procumbens Senecio heracleifolius Senecio salignus Solanum eleagnifolium Solanum madrense Solanum torvum Stevia jaliscensis Tillandsia recurvata Trixis angustifolia W igandia kunthii Woodsia mollis (Patzcuaro) XERIC MATORRAL Acacia pennatula Asterohyptis mociniana Bursera bipinnata Bursera fagarioides Canavallia villosa Cassia stenocarpa Cheilanthes myriophylla Cosmos sulfureus Crotalaria maypurensis Crotalaria pumila Crotalaria sagittalis Croton morifolius Cuphea llavea Cuphea procumbens Dalea diffusa Dalea nutans Desmodium procumbens Dyssodia cancellata Dyssodia tagetiflora Eupatorium pulchellum Eysenhardtia polystachya Gomphrena decumbens Gomphrena nana Indigo/era jaliscensis lpomoea intrapilosa Lagascea decipiens Leucaena esculenta Loeselia mexicana Lysiloma acapulcensis M andevilla f oliosa M anihot f oetida Mimosa albida M ontanoa myriocephala N otholaena aurea N otholaena sinuata Perezia rigida Phaseolus atropurpureus Phaseolus heterophylla Phorandendron carneum Physalis nicandroides Prunus f erruginea Ptelea trif oliata Rhus radicans Salvia leptophylla Salvia purpurea Salvia tiliaefolia 51 52 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 Schkuhria anthemoidea Solanum rostratum Sprekelia formosisissima Stevia rhombifolia T agetes filif olia T agetes micrantha Tecoma stans Thevetia ovata Tithonia tubaeformis Tragoceros schiefianum V erbisina sphaerocephala Vernonia serratuloides V iguiera pachycephala V iguiera quinqueradiata Zinnia peruviana Zornia diphylla Xeric matorral may sometimes be a succes- sional stage, as in Nayarit (numbers 8523 ff.) where it apparently came in after the clearing of an oak woodland. Axelrod refers to this type as "arid sub- tropic scrub" ( 1958, p. 502). PINON-JUNIPER WOODLAND Cowania mexicana ]uniperus deppeana Pinus cembroides Populus arizonica 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 53 REFERENCES CITED Axelrod, Daniel I., 1944, The Alvord Creek flora . Carnegie Inst. Wash. Pub. 553 :225-262. ----- , 1948, Climate and evolution in western North America during middle Pliocene time. Evo- lution 2 :127-144. ----- , 1950, Studies in late Tertiary paleo- hotany. Carnegie Inst. Wash. Pub. 590:1-323. -----, 1952, A theory of angiosperm evolu- tion. Evolution 6:29-60. -----, 1958, Evolution of the Madro-Tertiary Geoflora. Bot. Rev. 24:433-509. , 1959, Poleward migration of early angiosperm flora . Science 130:203-207. Brown, R. W., 1934, Recognizable species of the Green River flora. U.S. Geol. Survey Prof. Paper 185:45-77. Chaney, Ralph W., 1920, The flora of the Eagle Creek formation. Contr. Walker Mus. , 2(2) :115- 181. ----- , 1925, The Mascall flora-its distribu- tion and climatic relation. Carnegie Inst. Wash. Pub. 349:23-48. ----- , 1938, Ancient forests of Oregon: A study of earth history in western America. (In) Cooperation in research. Carnegie Inst. Wash. Pub. 501 :631-648. -----, 1944, The Dalles flora. Carnegie Inst. Wash. Pub. 553:285-321. ----- , 1946, Tertiary centers and migration routes. Ecol. Monogr. 17: 139-148. ----- , 1949, Early Tertiary ecotones in west- ern North America. Nat. Acad. Sci. 35:356-359. -----, 1956, The ancient forests of Oregon. Condon Lectures, Oregon State System of High- er Education, Eugene, Oregon. 56 pp. ----- and Daniel I. Axelrod, 1959, Miocene floras of the Columbia Plateau. Carnegie Inst. Wash. Pub. 617:1-237. ---- , Carlton Condit and Daniel I. Axel- rod, 1944, Pliocene floras of California and Ore- gon. Carnegie Inst. Wash. Pub. 553: 1-407. ---- and Ethel I. Sanborn, 1933, The Go- shen flora of west central Oregon. Carnegie Inst. Wash. Pub. 439:1-103. Clark, Thomas H. and Colin W. Stearn, 1960, The geological evolution of North America. The Ronald Press Co., New York. Cooper, William S., 1922, The Broad-sclerophyll vegetation of California. Carnegie Inst. Wash. Pub. 319:1-124. Craddock, G. W. and C. L. Forsting, 1939, The influence of climate and grazing on spring-fall sheep range in southern Idaho. U.S. Dept. Agr. Tech. Bul. 600:1-42 Daubenmire, R. F ., 1942, An ecological study of the vegetation of southern Washington and adja- cent Idaho. Ecol. Mono gr. 12 :52-79. Detling, LeRoy E., 1948a, Concentration of en- vironmental extremes as the basis for vegetation areas. Madroiio 9:169-185. ----- , 1948b, Environmental extremes and endemism. Madroiio 9:137-149. -----, 1953, Relict islands of xeric flora west of the Cascade Mountains in Oregon. Madroiio 12:39-47. ----- , 1954, Significant features of the flora of Saddle Mountain, Clatsop County, Oregon. Northwest Science 28:52-60. ----- , 1958, Peculiarities of the Columbia River Gorge flora. Madroiio 14:160-172. , 1961, The chaparral formation of southwestern Oregon, with consideration of its postglacial history. Ecology 42 :348-357. Dorf, Erling, 1942, Upper Cretaceous floras of the Rocky Mountain region. Carnegie Inst. Wash. Pub. 508:1-168. Flint, R. F., 1957, Glacial and Pleistocene geology. Wiley, New York. ----- and others, 1945, Glacial map of North America: Geol. Soc. America Special Paper 60, Pt. I-Glacial map; Pt. II-Explanatory notes, 37pp. Gray, Jane, 1964, Northwest American Tertiary palynology: the emerging picture. (In) Ancient Pacific floras. Tenth Pacific Science Congress Series, Honolulu: 21-30. Hansen, Henry P., 1939, Pollen. analysis of a bog near Spokane, Washington. Bul. Torrey Bot. Club 66:215-220. , 1947, Postglacial forest succession, climate, and chronology in the Pacific Northwest. Trans. Amer. Philos. Soc., n.s., 37 (pt. 1) :1-130. , 1950, Postglacial forests along the Alaska highway in British Columbia. Amer. Phi- los. Soc. Proc. 94:411-421. -----, 1953, Postglacial forests in the Yukon Territory and Alaska. Amer. J our. Sci. 251 :505- 542. -----, 1955, Postglacial forests in south cen- tral and central British Columbia. Amer. Jour. Sci. 253: 640-658. Heusser, Calvin J., 1960, Late-Pleistocene envi- ronments of North Pacific North America. Amer. Geographical Soc. Special Pub. 35: 1-308. 54 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No. 13 Hitchcock, C. Leo, Arthur Cronquist, Marion Ownbey and J. W. Thompson, 1955-1964, Vascular plants of the Pacific Northwest, Parts 2-5, Univ. of Washington Press, Seattle. Hollick, Arthur, 1930, The upper Cretaceous flor- as of Alaska, U.S. Geol. Survey Prof. Paper 159. Holten, Eric, 1937, Outline of the history of arctic and boreal biota during the Quaternary period. Bokforlags Aktiebolaget Thule, Stockholm. Just, Theodor, 1946, Geology and plant distribu- tion. Ecol. Monogr. 17: 159-183. Kendrew, W. G. and D. Kerr, 1955, The climate of British Columbia and the Yukon Territory. Edmond Cloutier, Ottawa. King, Philip B., 1959, The evolution of North America. Princeton Univ. Press, Princeton Knowlton, Frank H., 1902, Fossil flora of the John Day basin of Oregon. U.S. Geol. Survey Bul. 204:1-153. ----- , 1926, Flora of the Latah formation of Spokane, Washington, and Coeur d'Alene, Idaho. U.S. Geol. Surv. Prof. Paper 140-A:17-82. Lakhanpal, Rajendra N., 1958, The Rujada flora of west central Oregon. Univ. of Calif. Pub. in Geol. Sciences 35:1-66. LaMotte, R. A., 1936, The Upper Cedarville flora of northwestern Nevada and adjacent California. Carnegie Inst. Wash. Pub. 455, V :57-142. MacGinitie, Harry D., 1933, The Trout Creek flora of southeastern Oregon. Carnegie Inst. Wash. Pub. 416, II :21-68. , 1937, The flora of the Weaverville beds of Trinity County, California. Carnegie Inst. Wash. Pub. 465:83-151. ----- , 1953, Fossil plants of the Florissant beds, Colorado. Carnegie Inst. Wash. Pub. 599: 1- 188. Martinez, Maximino, 1948, Los pinos mexicanos 2d ed.) . Ediciones Botas, Mexico City. McKee, E. D., et al., 1956, Paleo-Tectonic map of the Jurassic system. U.S. Geol. Survey Misc. In- vestigations Map 1-175. Mirov, N. T., 1967, The genus Pinus. The Ronald Press Co., New York. Munz, Philip A. and David D. Keck, 1959, A California flora. Univ. of Calif. Press, Berkeley and Los Angeles. Oliver, E., 1934, A Miocene flora from the Blue Mountains, Oregon. Carnegie Inst. Wash. Pub. 455, 1:1-27. Peck, Morton E., 1941, A manual of the higher plants of Oregon. Binfor-ds and Mort, Portland, Oregon. Pickford, G. D., 1932, The influence of continued heavy grazing and of promiscuous burning on spring-fall ranges in Utah. Ecology 13: 159-171. Rand, A. L., 1948, Glaciation, an isolating factor in speciation. Evolution 2 :314-321. Sanborn, Ethel I., 1935, The Comstock flora of west central Oregon. Carnegie Inst. Wash. Pub. 465:1-28. Smith, Alan R. and John S. Stevenson, 1956, Deformation and igneous intrusion in southern British Columbia, Canada. Bui. Geol. Society of America 66:811-818. Smith, H. V., 1938, Some new and interesting late Tertiary plants from Sucker Creek, Idaho-Oregon boundary. Bui. Torrey Bot. Club 65:557-564. ----- , 1939, Additions to the fossil flora of Sucker Creek, Oregon. Papers Mich. Acad. Sci., Arts and Letters 24: 107 -120. Stebbins, G. Ledyard, Jr., 1952, Aridity as a stimulus to plant evolution. Amer. Nat. 86:33-48. ----- and Jack Major, 1965, Endemism and speciation in the California flora. Ecol. Monogr. 35:1-35. Tisdale, E. D., 1947, The grasslands of the south- ern interior of British Columbia. Ecology 28: 346-383. 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 55 APPENDIX Following is a list of plants and their common names which are diagnostic of the vegetation areas of the Northwest discussed in this paper. Abies amabalis-Amabalis or lovely fir Abies concolor-White fir A bies grand is-Giant or grand fir Abies lasiocarpa-Alpine fir Abies procera-Noble fir Acer circinatum-Vine maple Acer glabrum douglasii-Douglas maple Acer macrophyllum-Big-leaf or Oregon maple Achlys triphylla-Vanilla leaf Aconitum spp.-Aconite, monkshood Actaea spicata arguta-Baneberry Adenocaulon bicolor-Trail plant Agropyron inerme-Great Basin wheat-grass Agropyron spicatum-Wheat bunchgrass Allium crenulatum-Olympic onion Allotropa virgata,-Sugar stick Alnus oregona-Oregon alder, red alder Alnus sinuata,-Sitka alder Alnus tenuifolia-Mountain alder Amelanchier spp.-Serviceberry Anemone spp.-Anemone, wind flower Anemone occidentalis-Mountain pasque flower Antennaria racemosa-Slender everlasting Aplopanax horridum-Devils club Aquilegia formosa-Columbine Arbutus menziesii-Madrone Arctostaphylos canescens-Hoary manzanita Arctostaphylos columbiana--Hairy manzanita Arctostaphylos nevadensis-Pine mat manzanita Arctostaphylos patula-Green-leaf manzanita A rctostaphylos uva-ur si-Kinnikinnick Arctostaphylos viscida-White-leaf manzanita Arenaria formosa-Sandwort Amica latifolia-Broad-leaf arnica Amica parryi-Parry's arnica A rtemisia tridentata,-Sagebrush Aruncus sylvester acuminatus-Goat's beard Asarum caudatum-Western wild ginger Asclepias cordifolia-Purple milkweed Astragalus spp.-Locoweed Baccharis pilularis consanguineus-Chaparral broom Balsamorhiza spp.-Balsam root Betula glandulosa-Resin birch Betula papyri/era-Paper birch Bolandra oregana-Oregon bolandra Boykinia major-Mountain boykinia Bromus polyanthus-Bromegrass Calochortus lobbii-Lobb's star tulip Calochortus macrocarpus-Mariposa lily Calochortus nuttallii-Sego lily Calochortus tolmiei-Mariposa lily, Cat's ear Caltha leptosepala-Marsh marigold Calypso bulbosa-Lady's slipper Campanula rotundifolia-Harebell, bluebell Cardamine bellidifolia-Alpine bitter cress Carex spp.-Sedges Cassiope mertensiana-White heather Castilleja spp.-Paint brush Ceanothus cordulatus-Snow bush Ceanothus cuneatus-Buck brush Ceanothus integerrimus-Deer brush Ceanothus prostratus-Squaw mat Ceanothus sanguineus-Oregon tea tree Ceonothus thyrsiflorus-Blue brush Ceanothus velutinus-Sticky laurel, snow brush Cercocarpus betuloides-Plume tree Cercocarpus ledifolius-Mountain mahogany Chaenactis douglasii-Hoary chaenactis Chamaecyparis nootkatensis-Alaska cedar Chimaphila umbellata-Prince's pine Chlorogalum pomeridianum-Soap plant Chrysopsis villosa hispida-Golden aster Chrysothamnus spp.-Rabbit brush Cimicifuga elata-Bugbane Circaea pacifica-Enchanter's nightshade Cirsium undulatum-Thistle Cladothamnus pyrolaeflorum-Cladothamnus Clarkia pulchella-Clarkia Claytonia megarhiza-Spring beauty Clematis hirsutissima-Leather flower Clematis ligusticifolia-Clematis, virgin's bower Clintonia uniflora-Queen cup Collomia debilis-Alpine collomia Collomia grandiflora-Large-flowered collomia Convolvulus polymorphus-Morning-glory, bindweed Coptis laciniata-Western gold-thread Corallorhiza striata-Coral root C ornus canadensis-Bunchberry Camus nuttallii-Western dogwood Cornus stolonifera-American or red osier dogwood C orydalis scouleri-Western corydalis Corylus cornuta-Western hazel Crepis nana-Hawksbeard Cryptogramma acrostichoides-Parsley fern Cynoglossum grande-Hound's tongue Delphinium bicolor-Larkspur Dentaria pulcherrima-Toothwort Descurainia pinnata-Tansy-mustard Dicentra formosa-Bleeding heart 56 BULLETIN, MUSEUM OF NATURAL HISTORY, UNIVERSITY OF OREGON No.13 Disporum oreganum-Oregon fairy bells Dodecatheon dentatum-Shooting star, bird's bill Douglasia laevigata-Douglasia Drosera rotund if olia-Sundew Elymus glaucus-Western rye-grass Empetrum nigrum-Crowberry Epilobium spp.-Fireweed Erigeron spp.-Wild daisy Eriodictyon californicum-Yerba santa Eriogonum spp.-Wild buckwheat Eriophorum spp.-Cotton grass Erysimum capitatum-Wallflower Erythronium spp.-Lamb's tongue, fawn lily F estuca spp.-Bunchgrass Fragaria spp.-Wild strawberry Fritillaria lanceolata-Mission bells, rice-root lily Fritillaria pudica-Yellow bells Fritillaria recurva-Scarlet fritillaria Gaillardia aristata-Blanket flower Galium spp.-Bedstraw Garrya elliptica-Silk tassel bush Garrya f remontii-Bear brush Gaultheria humifusa-Alpine wintergreen Gaultheria ovatifolia,-Oregon spicy wintergreen Gaultheria shallon -Salal Geranium viscosissimum-Sticky geranium Geum macrophyllum-A vens Cilia aggregata,-Scarlet gilia Godetia amoena-Farewell-to-spring Goodyera oblongif olia-Rattlesnake plantain Habenaria dilatata,-Boreal bog orchid H aplopappus hallii-Hesperodoria Helianthella uni/Zora-False sunflower Heuchera spp.-Alum root Hieracium spp.-Hawkweed Holodiscus spp.-Ocean spray Horkelia fusca-Dusky horkelia Hulsea algida,-Alpine hulsea Hulsea nana-Dwarf hulsea H ydrophyllum tenuipes-W aterleaf Hypericum anagalloides-Tinker's penny Hypopitys latisquama-Broad-leaved pinesap Iris missouriensis-Western iris !uniperus communis-Dwarf juniper luniperus occidentalis-Western juniper luniperus scopulorum-Rocky mountain juniper Kalmia polifolia-Pale laurel Kelloggia galioides-Kelloggia Koeleria cristata-June grass Ledum spp.-Labrador tea Lathyrus polyphyllus-Leafy pea Leptarrhena pyrolaeflora-False saxifrage Leptodactylon pungens-Granite gilia Lewisia columbiana rupicola,-Columbia lewisia Linanthus bolanderi-Bolander's linanthus Linnaea borealis-Twinflower Listera convallarioides-Broad-Iipped twayblade Listera cordata-Heart-leaved twayblade Lithospermum ruderale-Western gromwell Lomatium spp.-Desert parsley Lonicera ciliosa,-Orange honeysuckle Lonicera interrupta-Chaparral honeysuckle Lonicera involucrata-Black twinberry Luetkea pectinata-Luetkea Lupinus spp.-Lupine Lysichitum americanum-Skunk cabbage Mahonia spp.-Oregon grape Maianthemum dilatatum-False lily-of-the-valley M entzelia albicaulis-White-stemmed mentzelia Mentzelia laevicaulis parviflora-Great mentzelia M enziesia f erruginea-Rustyleaf M ertensia spp.-Lungwort Micropus californicus-Cotton weed M icroseris alpestris-Microseris Mimulus nanus-Purple monkey flower MiteUa spp.-Mitrewort M onardella villosa-Coyote mint Monotropa uniflora-Indian pipe Montia parvifolia flagellaris-Miner's lettuce Montia siberica--Spring beauty Myricacalifornica-Western wax myrtle Myrica gale-Sweet gale N emophila parviflora-N emophila Opuntia fragilis-Prickly pear Orobanche fasciculata-Broom-rape Orthocarpus attenuatus-Owl's clover Osmaronia cerasiformis-lndian plum Osmorhiza chilensis-Sweet cicely Oxalis oregana-Wood sorrel Oxyria digyna,-Mountain sorrel Pachystima myrsinites-Oregon boxwood Paeonia brownii-Western peony Pectocarya pusilla-Pectocarya Pedicularis groenlandica-Elephant's head Penstemon spp.-Beard-tongue Petasites spp.-Coltsfoot Phacelia spp.-Phacelia Phleum alpinum-Mountain timothy Phlox difjusa longistylis-Spreading phlox Phlox hoodii-Gray phlox Phyllodoce empetriformis-Rose heath Phyllodoce glanduliflora-White heath Physocarpus capitatus-Ninebark Picea engelmannii-Engelmann spruce Picea sitchensis-Sitka spruce Pinus albicaulis-White bark pine Pinus contorta-Beach pine, lodgepole pine Pinus jefjreyi-J effrey pine Pinus monticola-Western white pine Pinus ponderosa-Western yellow or ponderosa pine Plagiobothrys nothofulvus-Popcorn flower Pleuricospora fimbriolata-Pinesap 1968 DETLING: HISTORICAL BACKGROUND NORTHWEST FLORA 57 Poa secunda-Biuegrass Polemonium elegans-Slender polemonium Polemonium viscosum-Skunk polemonium Polygonum bistortoides-Mountain meadow knotweed Polygonum newberryi-Newberry's knotweed P olygonum phyto/,accae f olium-Alpine knotweed Polypodium vulgare columbianum-Licorice fern Polystichum munitum-Western sword fern Populus tremuloides-Quaking aspen Populus trichocarpa-Poplar Prunus emarginata-Bitter cherry Prunus virginiana-Choke cherry Pseudotsuga menziesii-Douglas fir Pterospora andromedea-Pinedrops Purshia tridentata-Bitter brush, antelope brush Pyro/,a spp.-Wintergreen Quercus chrysolepis-Canyon oak Quercus garryana--Garry oak, Oregon oak Quercus vaccinifolia-Huckleberry oak Ranunculus spp.-Buttercup Rhamnus calif ornica-Coffee berry Rhamnus crocea-Buckthorn Rhamnus purshiana-Cascara Rhododendron albiflorum-White-flowered rhododendron Rhododendron macrophyllum-Rhododendron, rose bay Rhododendron occidentale-W estern azalea Rhus diversiloba-Poison oak Rhus g/,abra--Westem sumac Rhus radicans-Poison ivy Rhus toxicodendron-Poison ivy Rhus trilobata-Squ\w bush Ribes cereum-Squaw currant Ribes viscosissimum-Sticky currant Rosa spp.-Wild rose Rubus lasiococcus-Dwarf bramble Rubus nivalis-Snow bramble Rubus parviflorus-Thimble berry Rubus pedatus-Strawberry dwarf bramble Rubus spectabilis-Salmon berry Rudbeckia occidentalis-Western rudbeckia Salix hookeriana--Hooker's willow Sambucus glauca-Blue elderberry Sanicu/,a graveolens-Snakeroot Satureja doug/,assii-Yerba buena Saxif raga bronchialis-Matted saxifrage Saxifraga caespitosa-Tufted saxifrage Saxifraga ferruginea--Rusty saxifrage Saxif raga oppositif olia-Purple saxifrage Saxifraga punctata--Brook saxifrage Saxifraga rudifu/,a-Red-woolly saxifrage Saxifraga tolmiei-Tolmie's saxifrage Senecio canus-Gray senecio Senecio fastigiatus-Clustered senecio Senecio triangu/,aris-Spear-headed senecio Shepherdia canadensis-Buffalo berry Sibbaldia procumbens-Sibbaldia Sidakia asprella-Harsh sidalcia Sieversia ciliata-Prairie smoke Silene doug/,asii-Campion Sisyrinchium doug/,asii--Grass widow Smi/,acina spp.-False Solomon's seal Smilax calif ornica--California smilax Spiranthes romanzoffianum-Ladies' tresses Sporobolus spp.-Dropseed Stachys spp.-Hedge nettle Stipa columbianus-Subalpine stipa Stipa comata--Needle-and-thread grass Suksdor fia violacea-Suksdorfia Sullivantia oregana-Oregon sullivantia Symphoricarpos albus-Snowberry Synthyris missurica--Western mountain syntheris Synthyris reniformis-Syntheris Synthyris schizantha-Fringed syntheris Synthyris stellata--Columbia syntheris T ellima grandiflora-Fringe cup Thalictrum hexandra--Meadow rue Thuja plicata--Western red cedar Thysanocarpus radians-Lace pod Tiarella unifoliata-Coolwort Tolmiea menziesii-Bristle-flower T onella tenella-Small-flowered tonella T ownsendia florifer-Showy townsendia Trautvetteria grandis-False bugbane Trichostema /,anceolatum-Vinegar weed Trientalis spp.-Starflower Trillium spp.-Trillium, wake-robin Tsuga heterophyl/,a-W estern hemlock Tsuga mertensiana--Mountain hemlock Umbellu/,aria californica--California laurel Urtica spp.-Nettle Vaccinium caespitosum-Dwarf huckleberry V accinium deliciosum-Blue-leaved huckleberry Vaccinium membranaceum-Thin-leaf huckleberry V accinium occidentale-W es tern huckleberry Vaccinium ovalifolium-Oval-leaved huckleberry V accinium ovatum--Shot huckleberry Vaccinium parvifolium-Red huckleberry V accinium scoparium-Small-leaved huckleberry Vaccinium uliginosum-Blueberry, bog huckleberry Veratrum viride-Green false hellebore V icia americana-California vetch Vio/,a adunca-Western long-spurred violet V io/,a glabella-Y ell ow violet Viola sempervirens orbicu/,ata--Evergreen violet Vio/,a sheltonii-Shelton's violet Vitis californica--Western wild grape Whipplea modesta--Yerba de selva, whipple vine W yethia amplexicaulis-Dwarf sunflower Xerophyllum tenax-Bear grass, squaw grass Zygadenus venenosus-Death camas G5401 PUBLICATIONS Museum of Natural History University of Oregon Eugene, Oregon Bulletins Titles Price No. 1 Cenozoic Stratigraphy of the Owyhee Region, Southeastern Oregon; Kittle- man, L. R. et al., 45 pp., 9 plates 11 figures, (December 1965) $1.50 No. 2 Notes on some Upper Miocene Shrews from Oregon, Hutchison, J. H., 23 pp., 17 figures, (March 1966) $1.25 No. 3 A New Archaic Cetacean from the Oligocene of Northwest Oregon, Emlong, D., 51 pp., 15 figures, (October 1966) $1.50 No. 4 The Archaeology of a Late Prehistoric Village in Northwestern California; Leonhardy, Frank C., 41 pp., 17 figures, (March 1967) $1.00 No. 5 Peromyscus of the Late Tertiary in Oregon; Shotwell, J. Arnold, 35 pp., 11 No.6 No. 7 No.8 No.9 No.10 No.11 No.12 figures, (June 1967) $1.25 Ethnomalocology and Paleoecology of the Round Butte Archaeological Sites, Deschutes River Basin, Oregon; Roscoe, E., 20 pp., 4 figures (July 1967) $ .75 Its Own Story, The Museum of Natural History; 16 pp. no charge Geologic Map of the Owyhee Region, Malheur County, Oregon; Kittleman, L., and others, scale 1: 125,000 ( September 1967) $2.00 Late Tertiary Geomyoid rodents of Oregon; Shotwell, J. Arnold, 28 figures (November 1967) $1.25 Refinements in Computerized Item Seriation; Craytor, W.B., and LeRoy Johnson Jr., (March 1968) $ .75 Fossil Talpidae (lnsectivora, Mammalia) from the Tertiary of Oregon; Hutchison, J. $1.25 Plants of the Three Sisters Region, Oregon Cascade Range; Ireland, 0. 130 pp. $3.75 No. 13 Historical Background of the Flora of the Pacific Northwest; Detling, LeRoy (July 1968) $1.50 Cover Photographs by David Cole a:nd Bernard Freemesser