VITA 
NAME OF AUTHOR: James Craig Hickman 
PIACE OF BIRTH: Ottumwa, Iowa 
DATE OF BIRTH: April 20, 1941 
MARRIED: to Carole Jean Stentz, June 8, 1964 
UNDERGRADUATE AND GRADUATE SCHOOLS ATTENDED: 
Oberlin College 
University of Oregon 
DEGREES AWARDED: 
Bachelor of Arts, with Honors in Biology, 1963, Oberlin College 
AREAS OF SPECIAL INTEREST: 
Plant Geography, Ecology, Systematics, and Evolution 
General Biology 
PROFESSIONAL EXPERIENCE: 
Undergraduate Teaching Assistant, Department of Biology, 
Oberlin College, Oberlin, Ohio, 1961-1963 
Research Assistant, Woods Hole Marine Biological Laboratory, 
Woods Hole, Massachusetts, 1962-1963 
Research Assistant, Oregon Institute of Marine Biology, 
Charleston, 1964 
Teaching Assistant, Department of Biology, University of 
Oregon, Eugene, 1963-1967 
Instructor, University of Oregon Extension Division, 
Eugene, 1966-1967 
i 
AWARDS AND HONORS: 
NSF Summer Traineeship, University of Oregon, 1967 
Research Grant from Mazamas College Thesis Assistance 
Program, 1967 
NDEA Fellowship, University of Oregon, 1967-1968 
PUBLICATIONS: 
Scott, G. T., R. L. Clark, and J. C, Hickman. 1962. Mechanisms 
of chrornatophore control in the common sand flounder 
Scophthalamus aguosus. Biol. Bull. 123:486-487. 
1962. Drugs causing lightening and darkening of 
the common sand dab, Scophthalarnus aguosus. Biol. Bull. 
123:511. 
Hickman, J.C., and M. P. Johnson. 1968. An analysis of 
geographical variation in western North American Menziesia 
(Ericaceae). Madrano (In press). 
ii 
DISJUNCTION AND ENDEMISM IN THE FLORA OF THE CENTRAL WESTERN 
CASCADES OF OREGON: AN HISTORICAL AND ECOLOGICAL 
APPROACH TO PLANT DISTRIBUTIONS 
by 
JAMES CRAIG HICKMAN 
A DISSERTATION 
Presented to the Department of Biology 
and the Graduate School of the University of Oregon 
in partial fulfillment 
of the requirements for the degree of 
Doctor of Philosophy 
December 1968 
APPROVED: ~~~ 
Stanton A. Cook 
iv 
ACKNOWLEDGMENTS 
A study of this scope cannot be the work of a single individual. 
I am indebted to more persons than I can name here for their contribu-
tions of time, interest, and ideas which have considerably improved the 
quality of the work. The following persons made important contributions 
to the study. 
S. A. Cook suggested the Western Cascades as an area worthy of 
further study, and has, through his continuing interest in the problem 
and his ideas for new approaches, been a primary source of inspiration. 
I thank him here especially for his patient and careful reading of the 
entire manuscript. 
I also express my appreciation for many long discussions concerning 
plant distributions and systematics with the late L. E. Detling, who 
first noted the number of disjunct species in the Western Cascades. 
R, H. Waring unstintingly tutored me in special problems of 
physiological plant ecology, and cooperated in the manufacture of a 
sap-tension pressure-bomb. R. W. Becking, in addition to helping focus 
the study, first suggested the use of such an instrument in the Western 
Cascades, and the Mazamas College Thesis Assistance Program and the 
Department of Biology, University of Oregon, provided funds to have it 
constructed, 
Certain herbarium searches were undertaken by J. H. Thomas, Curator 
of the Dudley Herbarium, Stanford University, and Marilyn Lenox Ballard 
V 
of the University of Washington. Both contributed much time to this 
project, for which I am especially grateful. 
Several individuals kindly contributed expert taxonomic assist-
ance with difficult genera. I have been greatly aided by the opinions 
of Marion Ownbey (Allium, Castilleja), Noel Holmgren (Castilleja), and 
Rimo Bacigalupi (Castilleja). 
G. T. Benson, S. S. Tepfer, R.H. Waring, and D. E. Wimber read 
portions of the manuscript and offered many helpful suggestions . J . F . 
Franklin kindly provided information from his own studies, as well as 
stimulating discussion. John Lincoln, A. R. Kruckeberg, and many others 
have provided encouragement and insight at various points in the study, 
and I have been rescued from difficult situations in the field by C. D. 
White. 
Finally, to my wife, Carole Jean Stentz Hickman, I am indebted 
for many months of field assistance. She has been an inspiration, 
comfort, and aid through all portions of the work, and by observations 
and critical comment has supplied countless new ideas and methods and 
has helped place them in proper perspective. In addition she had 
critically read and typed the entire manuscript. To her I gratefully 
dedicate this dissertation. 
vi 
TABLE OF CONTENTS 
INTRODUCTION 1 
REVIEW OF THE LITERATURE 4 
Geology and Physiography 4 
Phytogeography 8 
OBJECTIVES OF THE STUDY 15 
INVESTIGATIONS UNDERTAKEN 16 
PHYSICAL ENVIRONMENTS 17 
Introduction 17 
Physiography and Geology 18 
Classification of Peaks 25 
VEGETATION DESCRIPTION 31 
.,; 
Mesic Conifer Forest 31 
Dry Mixed Conifer Forest 32 
Xeric Conifer Forest 32 
Lowland Xeric Meadow 33 
Snowbed 34 
Peaty Melt Seep 35 
Rocky Melt Seep 35 
Wet Meadow 36 
Mesic Meadow 36 
Subalpine Xeric Meadow 37 
Fine Gravel Scree 38 
Boulder Creep Slope and Outcrop Ridge 38 
Blocky Talus 39 
Vertical Outcrop 40 
INVESTIGATIONS ON SPECIES WITH UNUSUAL DISTRIBUTIONS 41 
Systematics 41 
Geographical Ranges 41 
Ecological Ranges 42 
Dispersibility, Pollinators, and Breeding Systems 48 
Moisture Regimes and Phenology 53 
Laboratory Germination Experiments 82 
Germination and Establishment in the Field 83 
vii 
DISCUSSION 95 
Plant Groups of Similar Geographical Affinity 95 
Speciation in Western Cascade Disjunct Populations 108 
Geological History of the Area and Its Effects on 
Plant Distributions 111 
An Evaluation of the Xeric Island Concept and 
Theories of Migration 118 
SYNTHETIC HYPOTHESIS 126 
LITERATURE CITED 128 
APPENDIX A 139 
Introduction 139 
Checklist of Western Cascade Species by Locality 141 
APPENDIX B 160 
Introduction 160 
Systematics, Ecology, and Geographical Distri-
bution of Disjunct Species 162 
viii 
LIST OF TABLES 
Table 
I, Erosional Features and Classification of Western 
Cascade Peaks 26 
II. Aspects of the Distribution and Ecology of Western 
Cascade Disjunct and Endemic Species 43 
III. Morphological Adaptations for Dispersal and Pollin-
ators, Observed Pollinators, Probable Breeding 
Systems, and Number of Western Cascade Local-
ities for Disjunct and Endemic Species 48 
IV . Numbers of Seeds Sown, Germinated, and Established 
in Cleared and Uncleared Plots in One 
Habitat on Sardine Butte 88 
V. Numbers of Seeds Sown, Germinated, and Established 
in Cleared and Uncleared Plots in Three 
Habitats on Rebel Rock 89 
VI . Numbers of Seeds Sown, Germinated, and Established 
in Cleared and Uncleared Plots in Three 
Habitats on Iron Mountain 90 
VII. Numerical Summary of Species Germinating or Estab-
lishing in Various Circumstances 91 
ix 
LIST OF FIGURES 
Figure 
1. Variation in Sap Tension with Time after 
Cutting in Four Species 58 
2. Diurnal Variation in Sap Tension in Three 
Species 62 
3. Variation in Sap Tension within Individual 
Plants 65 
4. Variation in Sap Tension among Parts of a 
Plant 66 
5. Variation in Sap Tension within a Species 
Population 68 
6. Population Variation in Sap Tension Correlated 
with Plant Size 69 
7. Population Variation at Minimum and Maximum 
Sap Tension 71 
8. Variation in Sap Tension with Habitat 73 
9. Variation in Sap Tension with Habitat and 
Phenological Stage 76 
10. Variation in Sap Tension with Phenological Stage 80 
11. Relationships of Species of Polygonum, Subgenus 
Duravia 179 
X 
INTRODUCTION 
The Western Cascades of central Oregon comprise a unique area 
of grea t floristic diversity. Included in the flora of the area are 
numerous species with disjunct distributions. About 95 percent of this 
region supports some phase of the broadly-defined mesic conifer forest 
(Detling, 1968), a low diversity floristic unit dominated by Pseudotsuga 
menziesii and species of Abies and Tsuga. Occurring in clearings, 
meadows, bogs, and on scree slopes and outcrops are numerous additional 
species many of which are typical of this region of the Cascades. How-
ever, approximately 20 percent of the flora of the region consists of 
species characteristic of other regions. These species a re disjunct, 
often markedly so, in their Western Cascade occurrences. They a re con-
centrated in, but not r~stricted to, the non-forest habitats. Disjunct 
elements in the Western Cascade flora include boreal, high alpine, 
Siskiyou-Klamath, Great Basin, and lowland valley species. Three spe-
cies are endemic to the study region and its immediate environs. The 
study of the distributions of these disjunct and endemic species is the 
subject of this work. 
Two distinct approaches have been taken in previous work on dis-
junctions, One, exemplified by the studies of Baker (1955, 1959a, 
1959b, 1961) and Carlquist (1966a, 1966b), emphasizes the biological 
and evolutionary mechanisms which adapt plants for long-distance dis-
persal. Thus the role of invasion of new areas by p}ants is stress ed, 
2 
and disjunct segments are likely to be considered outliers of expanding 
populations. At the other extreme, the studies of Gleason (1922), 
Fernald (1925), Marie-Victorin (1938), and Detling (1953, 1954, 1958, 
1961, 1966, 1968) are concerned with the historical aspects of disjunc-
tions. These workers have tended to focus exclusively on the relictual 
concept (Fryxell, 1962) and to hypothesize retreating migration or 
incipient extinction to explain disjunct populations. Such studies 
normally center around the climatic changes of the Pleistocene and 
Recent epochs. A special case involves the study of nunataks, or iso-
lated mountain peaks which were ice-free during the Pleistocene but 
surrounded by glacial ice. Such mountain peaks often support relics 
from the Wisconsin and previous glaciations, often far to the south of 
thei r normal ranges (Marie-Victorin, 1938; etc.). 
In studies with both biological and historical hypotheses, sev-
eral assumptions are made concerning the nature of disjunctions. Bio-
logical studies often consider geographical disjunction only, completely 
disregarding the time factor; while histo~ical studies typically concen-
trate on large changes through time and often fail to consider important 
biological differences between species of similar geographic affinity. 
In addition, the historical approach normally assumes that at some time 
plants of a relict species occupied all areas between each locality of 
present occurrence. Inherent in both approaches is the assumption that 
the habitats of disjunct . species, wher·ever they occur, are generally 
similar. 
The Western Cascade Range bears significantly on the general 
problem of disjunctions in several ways. First, a large number of more 
3 
or less discrete islands of disjunct species occur in the area. Second, 
the boreal, austral, cold desert, and high and low altitude species are 
found together, often in the s ame a ssociation type and occasionally so 
physically close that they are probably in competition with one another 
during certain seasons. Thus it seems that the general assumption that 
disjunct organisms are found in habitats generally similar to those of 
their parental populations is here violated. Such considerations show 
the Western Cascades to be an ideal area for studying the history and 
biology of disjunct species . 
REVIEW OF THE LITERATURE 
There is a paucity of published work that pertains directly to 
the geology or floristics of the Western Cascade Range. A summary of 
the existing literature in these two areas is given below, 
Geology and Physiography 
While Dicken (1955) and Baldwin (1964) present brief discussions 
of the physiogr~phy and geology of the Western and High Cascades, neither 
treats them in detail and both rely primarily on the findings of other 
workers. 
Williams (1942, 1944, 1953) presents, in semi-popular form, the 
geologic history of . these ranges together with considerations of the 
origins, products, and forms of volcanic eruptions. According to 
Williams (1953), the Western Cascades are the result primarily of fis-
sure eruptions in a broad downwarping area, Small cones may have 
developed locally, but no evidence of their original forms remains. 
Eruptions began to the west, perhaps in what is now the Willamette 
Valley, during the Eocene and migrated eastward to the present crest 
of the High Cascade Range by the middle of the Pliocene. There were 
several periods of intense volcanism during the Tertiary, the most ex-
treme of which included middle and late Miocene times. Between these 
periods of volcanism, erosion degraded the mountains and subsequent 
eruptions tended to fill the topographic lows. At the end of the 
4 
5 
Miocene the older flows were uplifted and thrown into broad gentle folds, 
Although there was still some activity in Western Cascade vents by the 
beginning of the Pliocene, most volcanism by this time was localized in 
the present High Cascade Range, a north-south trending line of impressive 
composite volcanoes. Volcanic activity in the High Cascades has con-
tinued at lower rates into the present century. 
The work of Thayer (1936, 1939) has elucidated the geology of 
the North Santiam River drainage, which extends from the Willamette 
River Valley to the crest of the High Cascades. Thayer describes the 
stratigraphy, structure, and glacial and erosional history of this di-
verse region, which forms the northern border of the area under consid-
eration here. In 'his stratigraphic work, Thayer distinguishes as dis-
crete formations any flows separated by a recognizable unconformity and 
groups the formations into "series" such as the Breitenbush, Mehama, 
and Sardine, Of particular interest are Thayer's ideas concerning the 
putative presence of the "Cascade fault scarp," which is considered 
(Thayer, 1936) to be a structural boundary between the Western and High 
Cascades. Thayer and previous workers had noted the sharp line physio-
graphically delimiting the two ranges essentially throughout their 
length in Oregon. To the west the older range rises steeply to dis-
sected ridges often over 1000 m above the floor of the line of valleys 
marking the boundary. To the east the High Cascade plateau slopes 
gently up to Recent cinder cones and finally to the high peaks which 
range from 2400 to 3500 min elevation, After a reconsideration of some 
as yet unresolved stratigraphic problems, Thayer (1939) modifies his 
original opinion by observing that, at least in the North Santiam 
6 
region, a fault need not be invoked to explain the observed strati-
graphic relations. Of critical concern in this decision is the presence 
of a topographic and stratigraphic bridge between the two ranges which 
includes Outerson Mountain and related peaks. Such bridges are excep-
tional in the central Cascades. 
Thayer (1939) calls attention to glacial moraines and reworked 
tills in the valley of the North Santiam River . The large alpine valley 
glaciers which deposited this debris had their origin high on the slopes 
of Mt . Jefferson, as much as 58 air km to the east. The Wisconsin gla-
ciers, however , came barely into the Western Cascades from the higher 
mountains if Thayer's interpretations are correct. He also notes the 
presence of many small cirques and associated features in the more com-
pletely dissected portions of the Western Cascades. It is likely that 
some of these lower glaciers acted as tributaries to the larger valley 
glaciers. Hopson (1946) notes morainal features in the McKenzie River 
Valley, extending as far west as Blue River, or to an approximate eleva-
tion of 300 m. 
Maps of numerous mining claims and petrographic and economic aspects 
of propylitically altered regions scattered through the Western Cascades 
are presented by Callaghan and Buddington (1938). Such areas of alter-
ation are closely associated with small dioritic intrusive bodies, al -
though the mineralization which makes them of economic value occurred 
later than the intrusions themselves--perhaps in the late Miocene . This 
is evidenced by mineral veins which cut through both altered country 
rock and the intrusive bodies, but are thought to be genetically related 
to the latter. Due to the extreme chemical complexity of the substrate 
7 
in altered areas and possible subsequent effects on the vegetation 
patterns, such regions were excluded from the present study area. 
The most recent and comprehensive studies of Western Cascade 
geology have been made by Peck (1960), Wells and Peck (1961), and Peck 
and others (1964). The last paper contains data on the stratigraphy, 
structure, and petrology of the region, Peck and others also discuss 
the intrusions and the concurrent and subsequent alterations of the 
country rock. A very large area is covered in this work, and a great 
deal of new information is presented, However, it is a pilot study 
needing extensive further documentation. Peck lumps the Mehama and 
Breitenbush Series of Thayer (1936) into the Little Butte Series of 
Wells (1956) and Wells and Peck (1961). He also treats the Sardine 
Series as a single formation. This conservative treatment is more in 
keeping with present lack of knowledge of the correlations of the beds 
in these groups of flows. The Little Butte is an important Western 
Cascade series to the west and south of the study area, but within the 
area it comprises only about 17 percent of the total surface (Peck and 
others, 1964). This dominantly tuffaceous series weathers rapidly and 
except for recent roadcuts and occasional brecciated vents has not been 
found to crop out in the study area, Peck dismisses Thayer's (1936) 
proposal of a series of faults between the Western and High Cascades 
with the observation that a linear scarp is not topographically evi-
denced at all localities. 
Three geologic maps are available for the portion of the Western 
Cascades considered here, The first (Williams, 1957) is a reconnaissance 
map of the central portion of the High Cascades but includes the proposed 
8 
western limit of the Plio-Pleistocene volcanic rocks of that range, 
which occur irregularly well into the physiographically-defined Western 
Cascades. According to the most recent map, Plio-Pleistocene flows 
cover about 44 percent of the present study area. However, proposed 
contacts between Miocene and Plio-Pleistocene flows were visited by 
the mappers in only a small number of instances, and these were confined 
to the most heavily forested areas of the Western Cascades, where good 
outcrops are extremely rare. Peck does demonstrate overlapping but 
probably statistically significant differences between the chemistries 
of Miocene and Plio-Pleistocene volcanics of the Cascade Ranges, but it 
appears that none of the reputed Pliocene rocks from the Western Cas-
cades were included in these samples. Thus, the extent of Plio-Pleisto-
cene rocks in the Western Cascades and the locations of contacts between 
them and older flows are still open to considerable question. 
In summary, the basic history, structure, and stratigraphy of 
both Western and High Cascade Ranges is relatively well known and docu-
mented, but detailed information on any portion of the study area is 
lacking. 
Phytogeography 
The first published note of extra-limital species in western 
Oregon was made by J.C. Nelson in 1918. Using terms striking in their 
immediacy, Nelson (p. 23) states that 
•.• species which have been regarded as distinctively 
Californian are pressing steadily northward •••• In the 
same way, species that have been considered as belonging to 
the flora of the interior are continually being transported 
down the Columbia, and even travelling over the low summits 
of the Cas cades. 
9 
Although Nelson discusses only tw species which are here considered 
important Western Cascade disjuncts (Gilia aggregata and Silene 
campanulata glandulosa) , hi s early recognition of the major pathways 
by which disjunct species have arrived in western Oregon is noteworthy, 
Two papers by Detling deal directly with floristics and phyto-
geography in the study area (1953, 1968). In a paper calling attention 
to the existence of dry-adapted species in "xeric islands" in Oregon 
west of the Cascade crest, Detling (1953) describes the general nature 
of the islands and discus ses the present and postulated past di s tribu-
tions'  of the various "xer ic " plant species. "Xeric" literally means 
"dry." Detling evidently uses the term only in a general sense, to 
denote plants found under conditions of low moisture availability. Sea-
sonal changes in moisture i n the Western Cascades render such a gen-
eral definition of little use for more detailed ecological or physio-
logical studies. 
Detlin,g (1953) plo ts the ranges of 32 "xe r i c species" on 8 moun-
tain summits of the Coast and Western Cascade Ranges. Three of thes e 
islands (Castle Rock, Horsepasture Mountain, and Rebel Rock) are in t he 
present study area and now have been sampled more completely. Of 
Detling's 32 xeric species , only 13 are considered of critical impor -
tance here. These are All i um crenulatum, Arabis holboellii r etrofracta, 
Arenaria capillaris ameri cana, Arnica par ryi, Artemisia t r identata, 
Collomia linearis, Erigeron folio sus confinis, Gilia aggregata, Linum 
perenne lewisii, Lupinus arbustu s neolaxiflorus, Phacelia linearis, 
Sedum stenopetalum, and Sil ene douglasii. Of the remaining, four were 
misidentified (Delphinium de pauperatum = D. menziesii pyramidale; 
10 
Hackelia diffusa = H. jessicae; Hieraci~m cynoglossoides nudicale = 
H. scouleri scouleri; Madia minima=!· 1exigua); three are not xeric 
species, but are annuals or perennials of short growing season adapted 
to either snowbed or seep-slope environments (Claytonia lanceolata, 
Crocidium multicaule, Erythronium grandiflorum pallidum); and five are 
not found within the study area (Lupinus lepidus medius, Poa scabrella, 
Sidalcea asprella, Sisyrinchium douglasii, Viola sheltonii). Finally, 
seven of Detling's species, while basically "xeric" and restricted to 
rather specialized environments within the area, have proved quite wide-
spread and common in surrounding regions and cannot be considered dis-
junct in the Western Cascades, although their presence there is notable 
and supports conclusions about more disjunct dryland species. The seven 
species are Bromus polyanthus, Eriogonum compositum, Eriogonum umbellatum, 
Microsteris gracilis, Polygonum douglasii, Prunus emarginata, and 
Sanicula graveolens. 
Concerning the islands themselves, Detling notes rapid erosion 
as being responsible for the shallow soil depth and subsequent rapid 
drainage and desiccation by wind and sun. He suggests that the dark 
color of the igneous rock and associated higher heat-holding capacity 
result in a longer growing season in these habitats, 
Detling uses the bog-pollen data of Hansen (1947) in discussing 
the warm dry climatic maximum of about 6000 years ago and concludes 
that his xeric species 
are relicts of a once widespread xeric flora which originated 
first, in the Rogue River Valley, and secondly, on the plateaus 
of east-central Oregon , They have persisted on the mountain 
summits because of the arid and relatively warm conditions of 
11 
the shallow soil and exposed dark rocks, and the consequent 
freedom from competition with the surrounding mesic forest 
types. (p. 46-47) 
He makes no note in this paper of the boreal and arctic-alpine species 
which occur with the xeric species. 
In the larger context of the history of the vegetation of the 
Pac ific Northwest, Detling (1968) reaffirms, without major change, 
portions of the Xeric Island Hypothesis. Here he summarizes the paleo-
botanical work of Axelrod, Hansen, and Heusser, discusses the migra-
tional hi s tory of · western Nor th American angiosperms from their first 
appearances in the Cretaceous to the present, and describes the major 
extant vegetation units. 
An unpublished dissertation reporting comprehensive n~tur al 
history studies of the McKenzie River drainage has been presented by 
Hopson (1947). Her lists of the major plants and vertebrates and some 
of her comment's on the geology of this region are helpful in under-
standing the general nature of the Cascade Ranges. Hopson reports sev-
eral disjunct plant and animal species, mostly from the Western Cascades. 
The area encompassed by her work extends from the mouth of the McKenzie 
River near Coburg to the summits of the Three Sisters and covers an 
elevational range of over 3000 m. 
Two workers have completed floristi c studies in other portions 
of the Western Cascades. Baker (1949a, 1951) studied the flora of 
Fairview Mountain in southern Lane County where he noted a number of 
disjunct xeric and a few disjunct boreal species. Unfortunately, a 
significant number of Baker's species were misidentified, and vou her 
specimens of many others have not been located. In addition, Baker 
12 
(1951) claimed numerous "range extensions," although many of the species 
concerned had been collected previously either from the Fairview-Bohemia 
region or from localities still more distant from the presumed parental 
populations . Despite these limitations, however, his work was the pio-
neer flo r isti c study in the Western Cascades and is of continuing value 
as the most comprehensive study of this highly diverse mountain peak. 
In a more recent study of the flora of Monument Peak, Aller 
(1956) defines the major communi t y types, hypothesizes on their origins 
and eco logical requirements, and lists the plant species found on the 
mounta in. Of parti cular interest to the present study are his comments 
on t he we t Ac e r -Alnus community, which is a widespread feature of the 
north and east-facing slopes of the Western Cascades. Aller concludes 
that the fac t ors responsible for its existence on Monument Peak inc lude 
some type of disturbance of the dominant mesic forest coupled with only 
moderately steep slopes and abundan-t soil moisture. Aller emphasizes 
the dynamism of t he various Monument Peak plant communities. 
Several publi shed and unpublished papers concern the floristics 
or ecology of the High Cascade Range adjacent to the area of present 
interest . Roach (1952) analyzes the phytosociology of the Nash Crater 
lava flows abutting the eastern side of the study area, naming and 
desc ribing various a s sociations according to the European system (Braun-
Blanquet, 1932). An interes ting outcome of Roach's work was documenta-
tion of the much greater diversity of the plant species of mesic and 
hydric habitats than in drier sites in the lava flows, some of which, 
accordi ng to Roach, may be as young as 400 years. The contrary is true 
in the We stern Cas cade flora, in that more species are adapted to 
13 
habitats that are at least seasonally dry than to all other habitats 
combined. 
In addition, three botanical studies have been completed in the 
region of the Three Sisters. Van Vechten (1960), in an unpublished 
dissertation, describes the timberline and alpine vegetation of the 
area. His work is of value in relating plant communities to the gla-
cial processes which are so marked at high elevation in this region. 
Pechanec (1961) sampled the mosses of the High and Western Cascades in 
the latitude of the Three Sisters and found correlations between the 
moss communities and the trees and other plant species of the sample 
areas. Ireland (1968) published an illustrated flora with keys to the 
vascular plants he found within the region . 
Several papers, while not concerned directly with the study area 
or its immediate environs, report studies of floristically allied areas. 
Publications by Detling (1954, 1958, 1961) discuss the floras of Saddle 
Mountain, Clatsop County, and the Columbia Gorge; and the ecology and 
history of the chaparral formation of southwestern Oregon. In each of 
these studies, s pecies are divided into groups of geographic affinity, 
which are then discussed as migrational units. The areas discussed are 
among the most interesting regions in the Pacific Northwest, and numer-
ous species occur in them which are found again, as disjuncts, in the 
Western Cas cades. 
Of minor importance to floristic considerations within the study 
area are the treatments of broad vegetational units or biotic zones such 
as those given in various manuals and short papers (Peck, 1925a, 1925b
1 
14 
1961; Larrison , 1946) . A large number of species which Peck (1925a, 
1925b) suggests are characteri s tic of the valleys of southwestern 
Oregon or of the diverse areas of Eastern Oregon also occur in the 
Western Cascades. 
OBJECTIVES OF THE STUDY 
The objectives of the present work can be briefly stated as 
follows: 
1. to elucidate the ranges of disjunct species occurring in 
the Western Cascades by analyzing the geographical, geological, his-
torical, ecological, and biological patterns of the species and areas; 
2. to examine the processes leading to and maintaining these 
disjunctions; 
3. to test experimentally Detling's "Xeric Island Hypothesis" 
concerning the origin of Western Cascade disjunctions; 
4. to investigate the role of disjunctions in the evolutionary 
history of the larger groups which have disjunct species in the Western 
Cascades; 
5. to arrive at a better understanding of the nature of dis-
junctions in general. 
15 
INVESTIGATIONS UNDERTAKEN 
Observations were made and experiments were conducted on several 
of the variables that might influence the distribution of Western Cas-
cade disjunct species. An attempt was made to follow a more synthetic 
line of investigation than has previously been used. 
Physical environments and vegetation units have been approached 
in a primarily descriptive manner, although some experiments were per-
formed involving both. Other investigations centered on the various 
species showing extraordinary distribution ranges . Aspects studied 
include systematics, geographical and ecological distributions, dis-
persibility, pollinators, breeding systems, moisture regimes, and 
phenology. Experimen t al studies on germination and establishment were 
performed in the field and in the laboratory and included studies of 
inhibition of germination and growth by soils of differing parent 
materials. 
16 
PHYSICAL ENVIRONMENTS 
Introduction 
The area of the Western Cascades in which this s~udy has con-
centrated is shown in Appendix B, p. 162. It is irregular in shape, 
and the north-south axis is elongated. The latitudinal boundaries ap-
proximately coincide with N 43° 58' and N 44° 40'; while longitudinal 
boundaries are W 122° 0' and W 122° 22'. The region is characterized 
geologically by primarily horizontal flows of andesites and basalts 
which have been locally intruded by dikes and plugs. Oligocene and 
Miocene flows have been gently warped, but more recent volcanic rocks 
have retained their original dips (Thayer, 1939). Erosion has produced 
a maturely dissected topography with numerous valleys and steep ridges, 
the peaks of which stand at a remarkably uniform height of 1500 to 
1700 m. Average annual precipitation ranges from 1525 mm to 1900 mm. 
This is comparable to the rainfall on the immediate coast and is ex-
c eeded in Oregon only in isolated spots in the high Coast Range or the 
Cascades (Detling, 1948a, 1948b). Spring rainfall may be considerable, 
but little or no precipitation falls during the summer months. In 
winter the snowpack in the area may become quite deep (up to 4 m), but 
on south-facing slopes and ridgetops the snow frequently melts or is 
blown away. Winter air temperatures average well below freezing, al-
though temperatures under or within the snowpack are consistently 
near O0  C. 
17 
18 
Physiography and Geology 
Several type s of obser vations were made on the peaks themselves. 
Ro ck types were observed, atid fresh samples were cotlectrd wher ever 
possible. Field r elations between flows, attitudes of flows, and evi-
dences of faulting were noted but not quantified. Erosional features 
were noted together with physiographic effects on the composition of 
plant communities. The most significant of these finding s are reported 
in the following paragraphs . 
Relationships of Vegetational Units to Physiography 
Several vegetational units were found to be close ly assoc iated 
with repeated physiographic features of Weste r n Cas cade peaks . Many of 
the south or west-facing slopes have thin, easily erodable soils. In 
addition, they are swept relatively c lear of snow during winte r and are 
the first habitats to melt free of snow in spring. These slopes support 
meadows which vary in cover and species composition with the amount of 
runoff available through the growing season and the depth of the soil. 
Such meadows were later used to help deduce erosional pat t erns and pat-
terns of snow accumulation. 
Snowpocket areas were f ound frequently on t he north or east sides 
of ridges and in gentle depres sions on the sou t h-fac ing slopes. Such 
areas support a character i stic flora regardl ess of exposure direction. 
The presenc e of species such a s Orogenia fusiformis and Di. cen tra uni.flora 
indicate heavy accumulations of snow. 
Certain species were found only in a ssociation wi th glacial 
19 
phys iographic features. Ivesia gordonii, Polemonium pulcherrimum, 
and Douglasia laevigata, all highly r estricted spec ies in the area, are 
found on ver ti cal walls or t n the derived fine scree around the head-
walls of old cirques. These are all Bpecies of arctic or high alpine 
derivation. 
Er os ional Features 
Various aspec ts of eros ional processes were investigated. Iso-
lation of various peaks was considered as a possible factor influencing 
the number of disjunct species supported. Problems both of dispersal 
and of reduced habitat size are greater on isolated peaks. The dominant 
ridge and valley structure of the Western Cascades results in few truly 
isolated mountain masses, but regions where several ridges come together 
can be compared with the few peaks which do not connect with any other 
by a high ridge. Three Pyramids, Cres cent Peak, and Carpenter Mountain, 
isolated peaks of markedly differing aspect, support 19, 14, and 13 dis-
junct spec ies respec tively. On the other hand, Iron Mountain, Horse-
pasture Mountain, Lowder Mountain, and Rebel Rock, all at or near the 
j unc tions of major ridges, support 42, 30, 25, and 41 disjunct species 
respec tively. Although other factors are also important, this evidence 
poin t s toward isolati on as one fa c tor i n fluencing the concentration of 
distributionally interesting spec i es . 
Numerous small Pleistocene glaciers were an important factor in 
the erosion of the We s tern Cascades. St eep headwalls of cirques and 
associated tarns occur for example on Three Pyramids, North Peak, Echo 
Mountain, Cone Peak, Browder Ridge, Lowder Mountain, and Indian Ridge. 
20 
The largest cirques are confined to the northern or eastern slopes of 
these mountains. 
Certain other typical glacial features have not been observed in 
the Western Cascades. Many of the glacial valleys have been deepened 
by stream action since retreat of the ice, and V-shaped valleys are pre-
dominant in the Western Cascades except for the major river valleys, 
which contained valley glaciers from the High Cascades. In addition, 
no primary evidence of abrasion such as glacial striations have been 
found in the Western Cascades. 
These findings may support Thayer's (1939) proposal that the 
Wisconsin ice did not advance as far in this region as did glaciers of 
earlier maxima. Hopson (1946) invoked a sudden melting of glacial ice 
to sluice away the striated rocks in the U-shaped White Branch Valley. 
Such sluicing would, however, produce a V-shaped valley, and this hy-
pothesis does not appear tenable. In this instance, striations (if 
persistent) may well be buried beneath morainal deposits and talus. 
Another alternative to Thayer's proposal is that at these low eleva-
tions the Wisconsin ice would have melted much earlier than ice in higher 
alpine regions to the east, allowing more time and better conditions for 
weathering and erosion of the less durable glacial features. 
Since many of the slopes supporting disjunct species seem to be 
in rapid movement, experiments were conducted on the rates of erosion 
• 
of various types of slopes on Iron Mountain. In October, 1966, irides-
cent yellow spray paint was applied in horizontal and vertical lines at 
a height of about 1.2 m to two vertical scoriaceous outcrops of differ-
ent freshness and to two slopes of unconsolidated material. One of the 
21 
latter was a creep-slope of fragments averaging over 3 cm in diameter, 
and the other was composed of fine scoria averaging less than 5 mm in 
diameter. All slopes were examined after nine months and af t er one 
year. After nine months, only occasional pieces of paint-flecked gravel 
and sand were evident on the fine scoria slope. Some of these fragmen t s 
had migrated 3 m or more downslope from the original line. At the end 
of one year, a thorough search recovered no painted rocks in the a rea. 
Other surfaces did not erode so rapidly. The line on the s lope 
of larger unconsolidated material was still evident after one year; t he 
most distant pieces were found about 1 m downslope. After 21 mon t hs , 
painted fragments were aligned parallel to the slope in a drainage fur-
row. This spot was found later to be covered continually with snow from 
early winter until July. Lines on the vertical outcrops were still 
quite fresh, except that the paint had flaked off the rock in s everal 
spots, perhaps indicating that the rock was not covered by snow dur ing 
the winter and was alterna t ely heated and cooled. The more wea t he r ed 
outcrop had lost several small chunks of rock, amounting to approxi-
mately five percent of the total line · area. Evidently all loss of 
paint from the less weathered face was due to flaking. These r esul ts 
demonstrate a wide range in stability of closely pr oximate habita t s on 
the same slope. Similar r esults were obtained using al l ied technique s 
in British Columbia by Br ink (1964) . 
Stratigraphic and Structural Geology 
Struc tural and s t ratigraphic investigations in t he Western Ca s-
cade s are diffi cult for the following r easons: (1) the percen t age of 
22 
unweathered outcrops is very small; (2) the individual flows seem to be 
of limited extent and cannot ordinarily be followed from peak to peak; 
(3) the lithology of the volcanic rocks can vary considerably, even 
within a single flow, while the chemistry may remain essentially the 
same; (4) the orientation of flows is difficult to determine; (5) abso-
lute dates for the rocks are not obtainable with any degree of accuracy. 
Each of these points merits further consideration. 
In only one instance was a single flow observed in two sepa~ate 
localities. An outstanding bright red band of scoriaceous basalt of 
approximately 2 m thickness on Iron Mountain was also found, much re-
duced in thickness, on the lower slopes of Browder Ridge about 1.6 km 
distant. The andesitic basalts in the region of Iron Mountain are 
strongly flow-banded. According to three-point determinations they dip 
0 
to the east-southeast at 5-8 • Peck and others (1964) map these flows 
as a vent in the Plio-Pleistocene High Cascade volcanic rocks, but flows 
throughout this area are nearly parallel and show no brecciation. Fur-
thermore, according to both Peck and others (1964) and Thayer (1939), 
Plio-Pleistoc ene flows are not deformed but retain their original dips . 
No Plio-Pleistocene volcano of the magnitude required to have produc ed 
all these parallel flows can be postulated for this deeply dissec ted 
area; and the attitudes correspond to those reported for the eastern 
limb of the Breitenbush Anticline (Peck and o t hers, 1964; Thayer, 1939) . 
Thus it seems likely that much or all of this region is ac tually com-
posed of the youngest flows of the Miocene Sardine Formation, which are 
presumably the most recent rocks included in the general period of de-
formation which produced the Breitenbush Anticline. It is also possible 
23 
that contrary to the opinions of Peck and Thayer, some of the early 
Pliocene flows may have been warped together with the Miocene and older 
0 
rocks. An angular unconformity of less than 10 between flows exposed 
on the "nigger baby," an erosional remnant on the southeast face of Iron 
Mountain, may suggest that these flows are contemporaneous with the 
uplift and folding of the Western Cascades. The origin of the flows is 
unknown, but the latter hypothesis indicates a source area to the south-
west. 
Volcanic rocks of the Western Cascades range from olivine basalts 
to rhyodacites, but the few samples analyzed from the more recent flows 
are primarily basaltic andesites (Peck and others, 1964). The appear-
ance of these rocks is highly variable, especially regarding color, tex-• 
ture, and the presence of phenocrysts, while the chemical compositions 
seem, from the little work yet completed, to remain rather constant. 
Hand specimens are thus difficult to identify to type. 
Flows where banding is not obvious are most common and are typ-
ically exposed for only short distances, making determinations of their 
orientations extremely difficult. 
Some dating of Western Cascade rocks has been attempted using 
fossil floras and lead radioactivity determinations (Peck and others, 
1964). Fossil leaves are known from a number of localities in the 
Western Cascades, but dating by this method is highly restricted and 
imprecise. Lead radioactivity dates are possible only from the dioriti c 
intrusives. These methods have indicated ages of more than 35 ! 10 
million years for portions of the Little Butte Series and of more than 
25 ! 10 million years for the Sardine Formation (Peck and others, 1964, 
24 
p. 40). Extrapolations from the sites where dates have been obtained 
to other areas should be made only with great care. It is unlikely that 
either of these techniques will aid in discovering the contact between 
Miocene and Pliocene volcanic rocks in the Western Cascades, or the con-
tact between Western Cascade flows and those of the High Cascades. In-
tensive field work in the area will be necessary before these problems 
can be solved. 
A striking physiographic feature noted in the course of this 
study is the "valley-in-valley" effect found along the western margin 
of the area, espe~ially in the valleys of Blue River, Tidbits Creek, 
Canyon Creek, and Squaw Creek. In these areas the higher mountainsides 
show a uniform gentle slope with a break where uniformly steeper lower 
slopes begin. This effect has been caused by more rapid erosion of the 
lower parts of the valleys. This rapid erosion could in turn have sev-
eral causes. Uplift, either regional or through faulting along the 
eastern edge of the range as proposed by Thayer (1936), would re sult in 
increased stream gradients and more rapidly incised valleys. Al so, the 
region where this effect is most striking is near the axis of the 
Breitenbush Anticline, where the Sardine Formation and the Little Butte 
Series are in contact over large exposed areas. The rapid erosion of 
the tuffaceous Little Butte pyroclastics indicates that the break in the 
steepness of these slopes may approximate the contact between the Little 
Butte and the more resistant overlying flows. The last hypothesis seems 
most lik' ely. 
25 
Classification of Peaks 
Methods 
Forty-two peaks were sampled, of which 28, ranging from maximum 
elevations of 1150 m to 1800 m, fall within the boundaries of the study 
area. Five other Western Cascade peaks to the north, south, and west 
were sampled for comparative purposes, as were seven lower peaks of the 
High Cascade Range (1250 m to 1900 m), the Three Sister region, and 
Crater Lake National Park. 
For ease of presentation, these peaks have been grouped into 
classes according to a set of four physiographic characteristics which 
are important influences on floristic composition and vegetational 
patterns (see Table I). The criteria chosen include the following: 
(1) texture of parent rock; (2) proportion of peak on which outcrops 
are found; (3) direction of exposure of outcrops; and (4) slope steep-
ness, with outcrop slope emphasized. 
Non-rigorous manipulations of the data have shown that any given 
character state is highly correlated with particular states of the other 
characters, and in large part, division of the peaks into two large 
classes was easy. Howeve r , since this is an artifi c ial classification 
system using only a small number of charac ters, certain peaks had to be 
considered intermediate. The system was also used for peaks outside the 
Western Cascades, although under such conditions it occasionally proved 
to be misleading. 
26 
Table I. Physiographic Features, Classification, and Number of Dis-
junct Species on 42 Western and High Cascade Mountain Peaks. 
(Peaks are arranged from north to south within groups.) 
Key.!£. Symbols: 
Rock Texture 
0 pumice or scoria gravel 
1 highly scoriaceous andesite or basalt 
2 scoriaceous andesite or basalt 
3 slightly scoriaceous andesite or basalt 
4 moderately dense blocky andesite 
5 dense blocky andesite 
Outcrop Size 
s small 
m medium 
1 large 
Outcrop Exposure 
0 outcrops facing various directions 
1 outcrops facing mostly north 
2 outcrops facing mostly east 
3 outcrops facing mostly south and/or west 
Outcrop Steepness 
1 slopes relatively gentle 
2 slopes moderately steep 
3 slopes precipitous 
27 
(/) 
Q) (/) µ 
Q) 0. 0. H 0. Q) <.) (/) 
H 0 0 :::, 0 p p Q) 
:::, H H rtl H 0. (/) :::, •M 
,.!G..J <.) Q) 0 <.) Q) (/) . ,....., <.) <.) 
<J X µ N µ 0.. µ Q) ~ (/) Q) 
0 Q) ;J•M :::, X :::,..J ..... "M 0. 
i:G E-< QC/) Q~ QC/) u Cl Cl) 
Peak 
Bachelor Mt. 3 m 3 2 int. 26 
Three Pyramids 2 1 3 3 1 19 
Crescent Mt. 5 s 0 1 2 14 
North Peak 1 m 3 3 1 17 
Echo Mt. 2 m 3 2 1 33 
South Peak 1 m 3 2 1 30 
Cone Peak 1 m 3 2 1 37 
Iron Mt. 1 1 3 3 1 42 
Browder Ridge 2 1 3 3 1 34 
Jumpoff Joe 3 1 1 2 int. 5 
Squaw Peak 5 s 0 3 2 (3) 
Twin Buttes 3 s 0 2 2 15 
Carpenter Mt. 4 s 0 3 2 13 
Tidbits Mt. 5 m 0 3 2 15 
Lookout Mt. 5 s 0 2 2 (9) 
Frissell Point 5 s 0 2 2 17 
Castle Rock 4 s 0 3 2 12 
O'Leary Mt. 3 m 3 3 1 23 
Horsepasture Mt. 3 s 0 2 2 30 
Lamb Butte 5 s 0 3 2 4 
English Mt. 3 s 0 2 2 14 
Lowder Mt, 3 m 0 3 int. 25 
Yankee Mt. 4 m 3 2 2 (9) 
Tipsoo Butte 5 m 3 2 2 17 
Olallie Mt. 5 s 0 2 2 18 
Indian Ridge 5 m 2 2 2 12 
Sardine Butte 5 s 0 2 2 5 
Rebel Rock 3 1 3 3 1 41 
Monument Peak 4 s 0 2 2 10 
Huckleberry Mt. 5 s 2 2 2 (5) 
Fairview Peak 3 1 3 2 1 33 
Bohemia Mt. 2 1 0 3 1 37 
Hershberger Mt. 3 m 3 3 1 23 
Grizzly Peak 3 m 0 2 2 7 
Maxwe 11 Butte 1 m 0 1 0 10 
Little Nash Cr. 0 1 0 2 0 1 
Hogg Ro ck 5 1 3 3 1 1 
Hoodoo Bu tte 0 m 3 1 0 1 
Sand Mt. 0 1 0 1 0 1 
Step toes 0 1 3 1 0 5 
Three Sisters 2 1 0 3 1 28 
Crater Lake 1 1 0 2 0 32 
28 
Results 
Class 1 mountains include those composed of moderately to highly 
scoriaceous basalt or andesite flows, and with relatively large, south 
or west-facing, precipitous outcrop areas. The upper portions of these 
peaks are dominated by open meadow or scree habitats rather than forest. 
Class 2 mountains, on the other hand, are defined as those composed of 
dense vo lcani c material with small, moderately steep or gently sloping 
ou t crops which face in a variety of direc tions. Class 2 peaks are usu-
ally forested, except for summit and ridgetop outcrops . 
Good examples of Class 1 mountains are Three Pyramids, Iron 
Moun t ain, Browder Ridge, and Rebel Rock. Also included in this class 
are peaks which differed in one c riterion by one or two of the relative 
units, or in two criteria by only one unit. Such peaks are North Peak, 
Echo Mountain, South Peak, Cone Peak, O'Leary Mountain, Fairview Peak, 
Bohemia Mountain, and Hershberger Mountain. 
Exemplary Class 2 mountains include Crescent Mountain, Twin 
Bu t tes, Lookout Mountain, Frissell Point, Horsepasture Mountain, English 
Mountain, Olallie Mountain, Sardine Butte, Monument Peak, and Huckle-
berry Mountain o Peaks included in this class whi ch differ slightly in 
one or two charac ters are Squaw Peak, Carpenter Mountain, Tidbits Moun-
tain, Castle Ro ck, Lamb Butte, Yankee Mountain, Tipsoo Butte, and Indian 
Ridge. 
Thr ee peaks had to be considered intermediate in this classifi-
cation . Bachelor Mountain, while having south and west-facing outcrops 
of mode r a t ely scoriaceous texture, has only an average outcrop area of 
29 
moderate steepness. Lowder Mountain, a flat-topped peak perhaps repre-
senting an old flow surface, has precipitous outcrops of moderately 
scoriaceous texture, but the cliffs encircle essentially the entire top 
of the mountain and are of only moderate extent. Jumpoff Joe is a low 
elevation vent in the Little Butte Series near the axis of the Breiten-
bush Anticline. It is composed of a friable breccia which weathers in 
similar fashion to more scoriaceous rocks. The vent has been exposed 
on the north by stream or glacial action in the otherwise rapidly 
weathering Little Butte pyroclastics, accounting for the large expanse 
of nearly vertical north-facing outcrop. Jumpoff Joe, with a summit at 
only 1300 m, is the only peak in the study area composed entirely of 
Oligo-Miocene Little Butte Series (Peck and others, 1964), which per-
haps explains its unusual erosional behavior. 
Most of the lower mountains of the High Cascade Range are cinder 
I 
cones composed of pumice and clinkery volcanic ash rather than flows 
of consolidated material. Such peaks are considered ''Class 0" in this 
work. Included in this category is the region of Mount Mazama surround-
ing Crater Lake, since , the denser flows of the ancient mountain have 
been deeply covered with a mantle of pumice. 
Table I also lists the number of distributionally interesting 
species found on each mountain. Those numbers appearing in parentheses 
indicate peaks which were inadequately sampled. Numbers of disjunct 
species are highly correlated with peak classes. The mean number of 
disjunct species on Class 1, intermediate, Class 2, and Class 0 peaks 
are 29, 19, 12, and 8 respectively. With the present data Class 1 moun-
tains are separated from those of Class 2 at the level of 17 disjunct 
30 
species. 
Horsepasture Mountain, a Class 2 peak, is the only exception 
to this statement. It supports at least 30 disjunct species and is 
one of the most floristically diverse areas in the Western Cascades. 
Unlike other such centers of diversity, Horsepasture Mountain has nu-
merous, small, widely-scattered outcrop and meadow areas offering a 
variety of habitats. Although it must be placed with much less diverse 
peaks according to the system used here, these unique aspects perhaps 
explain why it is floristi cally more closely related to Class 1 moun-
tains. 
VEGETATION DESCRIPTION 
All species of vascular plants except members of the Gramineae 
and Cyperaceae were recorded in the field from each major habitat on 
each peak. A checklist of the species found on each of the sampled 
peaks, including both those within the study area and from adjacent 
regions is presented as Appendix A. 
Associated groups of plants became obvious with extensive obser-
vations, but no quantitative sampling was undertaken. The vegetation 
units described below consist of groups of species which recur wherever 
certain physical conformations are found. The environment is thus of 
great importance in defining the units, which are not "communities" in 
the standard phytosociological sense. The associations intergrade, 
often to a marked degree, but are the result of careful study in many 
different localities within the study area. 
Mesic Conifer Forest 
The mesic conifer forest is most important. It covers a s much 
as 85 percent of the land area in the Western Cascades and is respon-
sible for most of Oregon' s timber production. Although generally found 
under 1350 m elevation, under certain conditions it may extend much 
higher. It is dominated by Pseudotsuga menziesii. Important associated 
species include Abies grandis, Tsuga heterophylla, Cornus canadensis, 
Pedicularis racemosa, Chimaphila umbellata, Pyrola asarifolia, Pyrola 
31 
32 
picta, Listera caurina, Anemone deltoidea, Montia sibirica, Senecio 
harfordii, Viola glabella, Synthyris reniformis, Vaccinum parvifolium, 
Streptopus curvipes, Sedum oreganum, Galium oreganum, Polystichum munitum, 
Berberis nervosa, Gaultheria shallon, Rosa gymnocarpa, Tiarella unifoli-
ata, Clintonia uniflora, Smilacina sessilifolia, Arenaria macrophylla, 
Dicentra formosa, and Ribe s lacustre. 
Dry Mixed Conifer Forest 
Dominant tree species in this broadly defined vegetational unit 
vary with elevation. Abies amabilis and Pinus monticola are co-dominants 
in all stands. At lower elevations (about 900 m) the unit is apparently 
subclimax to re-establishing Mesic Conifer Forest following fire. Here 
other important tree species are Pseudotsuga menziesii and occas iona lly 
Tsuga heterophylla. On forested ridgetops at about 1500 m elevation, 
Abies procera and Tsuga mertensiana occur with Abies amabilis and Pinus 
monticola. This facies of t he unit appears to be climax. 
Both subdivisions contain a strikingly similar understory flora, 
which includes Xerophyllum tenax, Rubus lasiococcus, Coptis laciniata, 
Hypopitys monotropa, Allotropa virgata, Rhododendron macrophyllum, 
Gaultheria ovatifolia, and Eburophyton austinae. It is the similarity 
in non-tree species that warrants the fusion of these otherwise distinct 
units. 
Xeric Conifer Forest 
This uni t intergrades comple tely with the last . I t is found 
primarily on south-facing slopes where extensive ground fires have 
33 
burned within the last century. Living trees show a bimodal age distri-
bution, with large old trees (mostly Pseudotsuga menziesii) having thick 
fire-blackened bark on the lower portion of the trunk, Trees of inter-
mediate ages were evidently killed by the fire. Younger post-fire vege-
tation also includes douglas fir together with many species which are 
characteristic of the dry hillsides within the major valleys of western 
Oregon or are characteristic fire-following species. They include 
Libocedrus decurrens, Abies grandis, Castanopsis chrysophylla, Arbutus 
menziesii, Ceanothus velutinus, Ceanothus integerrimus, Quercus garryana, 
Rhus diversiloba, Rhododendron macrophyllum, Arctostaphylos columbiana, 
Rubus ursinus, Penstemon cardwellii, and Hieracium albiflorum, 
Lowland Xeric Meadow 
Especially in the region of the South Fork of the McKenzie River, 
clearings and meadows associated with the Xeric Conifer Forest contain 
a large number of species characteristic of the dry hills and fields 
within the Willamette Valley and other major valleys in western Oregon. 
Although their occurrences in the Western Cascades are disjunct and note-
worthy, most of the components of this association do not occur with 
those disjunct species found at higher elevations and have not been 
treated in detail in this study. This grassy meadow association is not 
found above 600 m, except on Castle Rock, where it reaches 1000 m and is 
confined to dry south-facing slopes . Such areas often support scattered 
trees of Quercus garryana, Arbutus menziesii, and Philadelphus gordonianus 
lewisii; and shrubs such as Rhus diversiloba, Ceanothus sanguineus, and 
Ribes sanguineum. The dominant perennial herbs include Brodiaea 
34 
hyacinthina, Brodiaea pul chell a , Berberis a9uifol i um, Psoralea physodes, 
Convolvulus nyctag ineus , and Apocynum androsaemi.fol i um. Annuals are 
abundant, including many spec ies of grasses, Polygonum spergulariaeforme, 
Clarkia rhomboidea, Clarkia amoena, Linanthus bicolor, Orthocarpus 
attenuatus, Githopsis specularioides, Conyza canadensis, Madia gracilis, 
and Madia exigua. 
Boreal Forest 
The above name is borrowed from Detling (1968) and is here de-
fined as that phase of the for est vegetation tha t is typically found at 
high altitudes (over 1500 m) on north-facing slopes, where dense stands 
of small trees are often encountered. Dominant tree species include 
Tsuga mertensiana, Abie s lasiocarpa, and Chamaecyparis nootkatens i s , 
wi t h a sparse herbaceous flora of Arnica la t ifolia, Pedicularis bracteosa 
flavi.da, Valer i ana sitchensi s , and Viola orbi culata. Snow cover in such 
communitie s i s always heavy and lasts until mi ds ummer or la t er. 
Snowbed 
Beneath north-fac i ng ou t c rops or on s t eep op en s lopes, as well 
-as in certai n wind-protec ted s i te s on south or east-fac ing slopes, snow 
accumulates to grea t dep t h s i n the winte r , of ten excee ding 8 m. Numer-
ous s pec ies, mostly of h igh alpine or boreal a ff inity, flower and rapidl y 
set seed and de s i ccate at t he edge of these snowbanks as t hey melt. 
Such s pec ies include Orogen i a fu siformis , Dicentra un i.flora, Claytonia 
lanceolata, Erythr onium gr andiflorum pall i dum, Lue t kea pec tinata, 
Mer t ens ia bella, and occasionally wi der-ranging spec ies s uch as Trillium 
35 
ovatum, Senecio triangularis, and Hydrophyllum occidentale. 
Peaty Melt Seep 
A continually wet habitat of gentle or level slope where snow-
melt or spring water trickles throughout most of the supper supports 
several species characteristic of true sphagnum bogs, which are rare in 
the Western Cascades. A similar environment has been found on the sedge 
mat surrounding small tarns at elevations above 1600 m. Characteristic 
species of this habitat include Dodecatheon jeffreyi, Pedicularis 
groenlandica, Trifolium longipes, Boykinia major, Caltha biflora, 
Habenaria dilatata, Stenanthium occidentale, Ranunculus alismaefolius 
alismellus, Drosera rotundifolia, Kalmia polifolia, and Tofieldia 
glutinosa. 
Rocky Melt Seep 
Occasionally snowmelt trickles over outcrops, especially on 
south-facing slopes where the thin soil characteristic of non-forest 
habitats in the Western Cascades has eroded away in response to fires 
and frost action. Unlike Peaty Melt Seeps, these habitats typically 
desiccate shortly after midsummer, and plants found here tend to be 
ephemeral annuals or stoloniferous perennials with ephemeral above-
ground parts. Such plants can survive the dry season in dormant con-
dition. Prominent species include Dodecatheon jeffreyi, Romanzoffia 
sitchensis, Lewisia triphylla, Mimulus breweri, Mimulus guttatus, 
Polygonum kelloggii, Linanthus harknessii, Allium amplec tens, Saxifraga 
integrifolia, Saxifraga occ identalis rufidula, and Gayophytum humile. 
36 
Wet Meadow 
Open areas, frequen t ly on east-facing slopes, which have rela-
tively constant sources of moisture and sufficiently gentle slopes to 
build up a relatively deep organic soil support a characteristic wet 
meadow flo ra. Dominated by Ve ratrum viride, Senecio triangularis, and 
Vale r iana sitchensis, this association also includes Ribes bracteosum, 
Rubus s pectabi lis , Mitella brewe ri , Ligusticum gr ayi, Mertensia bella, 
Hydrophyllum fendleri albifrons, and Hydrophyll um tenuipes. Much of 
this association is repeated around t he bor de rs of larger boggy areas, 
such as Quaking Aspen Swamp or the Potho l es . This phase grades into the 
Peaty Melt Seep community dis cussed above. Ano t her phase of this asso-
ciation is dominated by Acer c irc inatum and Alnus sinuata, whi ch fre-
quently form dense thickets of small tangled trunks and bran ches on the 
steeper wet slopes throughout the Western Cascades. Most of the species 
mentioned above can be found around t he edges of a maple or alder thicket 
or under its canopy. It is notable t hat these t wo tree species are also 
found in much less abundance on open slopes supporting Mes i c or Xeric 
Meadow communities. 
Mesic Meadow 
The factors res ponsible for t he maintenance of the remarkable 
open meadow s lope s in the Western Cascades are not c ear but doub tl es s 
include fi re, heavy snowpack , occasional snow- and landslides, r apid 
downslope creep of the light loamy soil, and the churning and cutting 
action of rodents, especially Ap lodontia rufa. Oc as i onally saplings 
seem to be establi shing in t hese habitats, but not in sufficient numbers 
37 
to allow the conclusion t hat succession is returning such areas to 
forest. This community, while normally encountered on south or west-
facing slopes, has an adequate supply of moisture until shortly past 
midsummer when the drier patches begin to desiccate. Most of the asso-
ciated species are herbaceou s perennials whi ch have adequate time to 
set see d during the warm moist part of the early summer. The stems of 
many of them are brown and withered by August. Mesic Meadows are dom-
inated by Rubus parviflorus, Pteridium aquilinum, and Rudbeckia occi-
dentalis but also include such herbaceous perennials as Aquilegia 
formosa, Erigeron aliceae, Lupinus latifolius columbianus, Ribes binom-
inatum, Ribes viscosissimum, Polygonum phytolaccaefolium, Cirsium 
centaurea, Mertensia paniculata, Dicentra formosa, Vicia americana 
truncata, and Epilobium angustifolium, together with occasional ephem-
eral annua ls such as Gal ium bifolium and Gayophytum humile, which grow 
in the shade of the taller perennials. 
Subalpine Xeric Meadow 
Between the Mesic Meadows and dry, rocky, surrounding areas, 
especially on Class 1 peaks, is found a loose association of species 
many of which have di sjunct ranges. In this habitat soils are much 
thinner and rockier t han those of the Mesi c Meadow, and moisture stress 
increases earlier in the season. Few of t hese species are ephemeral; 
rather they are capable of sustaining high moisture tensions in their 
stems until very late in the summer. Other species, mostly herbaceous 
perennials, die back earlier in the season. Representative s pecies 
are Gilia aggregata, Collomia linearis, Gayophytum giffusum parviflorum, 
38 
Orthocarpu s imbricatus, Luina stricta, Polygonum minimum, Polygonum 
douglasii, Polygonum cascadense, Navarretia divaricata, Lupinus arbustus 
neolaxiflorus, Linum perenne lewisii, Eriogonum nudum, Microsteris 
gracilis, Collinsia parviflora, Potentilla glandulosa, Cerastium arvense, 
Artemisia ludoviciana latiloba, Calochortus lobbii, Rumex acetosella, 
Pachystima myrsinites, Amelanchier alnifolia semiintegrifolia, and 
Phacelia heterophylla. 
Fine Gravel Scree 
This community is continuous with the Xeric Meadow on Class 1 
mountains and suppor ts, in lower densities, many of the species that are 
common in that habitat, as well as others such as Lotus nevadensis 
douglasii, Chrysothamnus nauseosus albicaulis, Allium crenulatum, Ivesia 
gordonii, Trifolium productum, Aster gormanii, Crepis occidentalis, 
Sedum oregonense, and Sanicula graveolens o In this community plants are 
charac ter istically widely dispers ed with much intervening bare substrate 
which is in rapid movement downslope o The community, consisting pri-
marily of disjunct species, is conf ined to ridges of rapidly weathering 
s coriac eous material which are oriented nearly perpendicular to the 
prevailing northwesterly winds. 
Boulder Creep Slope and Outcrop Ridge 
Al though these two habitats are physi cally different, they sup-
port many of the same species and so are considered together in this 
treatmen t. The Boulder Creep Slope is normally south or west-fac ing 
and consis ts of weathered rock f r agments in a fine loamy matrix, t he 
39 
whole moving downslope at moderate speed. Drainage is good, and the 
soils are typically quite dry through the first 10 cm but may be moister 
below that level. The Outcrop Ridge habitat is also found on south or 
west-facing slopes, where mass wasting of small fragments results in 
cropping out of small patches of parent rock which are barely exposed 
and eroded parallel to the general slope of the area. Many species are 
able to root in the weathered cracks of the outcrops or in occasional 
small pockets of finer material. Included are Delphinium menziesii 
pyramidale, Castilleja hispida, Penstemon procerus brachyanthus, Sedum 
stenopetalum, Sedum divergens, Eriophyllum lanatum, Arctostaphylos 
nevadensis, Haplopappus hallii, Silene douglasii, Cheilanthes siliguosa, 
Cheilanthes gracillima, Comandra umbellata, Lomatium martindalei, 
Sanicula graveolens, Eriogonum umbellatum, Eriogonum compositum, 
Junipe r us communis saxatilis, Erigeron foliosus confinis, Penstemon 
deustus, Arenar ia capillaris americana, Erysimum asperum, Antennaria 
rosea, Phacelia heterophylla, Anaphalis margaritacea, and Penstemon 
cardwellii. Occasionally these plants are found on ridgetop deflation 
armor flats where frequent high winds carry away finer soil particles, 
leaving an impervious pile of weathered rock fragments. 
Blocky Talus 
Below cliffs on high north-facing slopes of Class 2 mountains, 
Western Cascade andesites weather into large rec tangular blocks which 
form extensive talus piles. This phenomenon has been found only on the 
north sides of peaks composed of relatively dense parent rock. Vegeta-
tion is sparse in these areas, but it is quite constant. Plant species 
40 
include Sambucus racemosa pubens arborescens, Acer circinatum, Thalictrum 
occidentale, Aquilegia formosa, Cardamine integrifolia sinuata , and 
Campanula rotundifolia. 
Vertical Outcrop 
On dikes, cirque headwalls, or other er0sional surfaces that are 
steep or vertical, slow weathering produces crevices and pockets that 
provide some protection from excessive wind and heat. Adapted to these 
exposed environments are Saxifraga bronchialis vespertina, Penstemon 
rupicola, Selaginella wallacei, Erigeron cascadensis, Polemonium .P.!:!1:.-
cherrimum, Douglasia laevigata, Castilleja rupicola, Saxifraga cespitosa, 
Heuchera micrantha, and Polypodium hesperium. Certain of these species, 
such as Saxifraga bronchialis verpertina, Douglasia laevigata, and 
Castilleja rupicola, are found only on the windward sides of the out-
crops. If the outcrop is oriented 60 degrees or more to the prevailing 
winds, they are not likely to be found at all. This association is 
restricted to Class 1 peaks. 
INVESTIGATIONS ON SPECIES WITH UNUSUAL DISTRIBUTIONS 
Systematics 
Since determination of the range of a taxon requires an uncom-
promisingly distinct impression of the range of variation allowable 
within that taxon, detailed studies of herbarium specimens and compar-
isons with the available taxonomic literature were undertaken in a 
number of instances . In even the most ambitious of these stcd ies, the 
results are not monographic in scope but represent better synthese s of 
the taxonomic and evolutionary knowledge of these species than is avail-
able in present literature. These studies have shown that many of the 
Western Cascade disjuncts are characterized by complex and poorly under-
stood intra- and interspecific relationships. More deta iled information 
on materials and methods and taxonomic treatments of the species are 
given in Appendix B. 
Geographical Ranges 
With the discovery of a number of species new to the Western 
Cascades, it was considered desirable to document thoroughly the known 
ranges of all these species and to attempt to locate all populations of 
these plants within the study area. Nine West Coast herbaria were 
sampled, and the study area was sufficiently traversed to locate by far 
the largest portion of the interesting populations. In the herbarium 
sampling label information was copied from all Oregon Cascade specimens 
41 
42 
and others representing the complete range of documented localities. 
Voucher specimens from new and previously known localities collected 
during the course of this study have been deposited in the Herbarium 
of the University of Oregon . They were taken for all species showing 
unusual distribution patterns from all their major habitats on all the 
peaks sampled. In addition, range information was gathered from mono-
graphic treatments and from local and regional floras. The herbaria 
sampled and the completed dot distribution maps are found in Appendix 
B. Table II contains a s ummary of the geographical information found 
in Appendix B. 
Also given in Table II for reference is the number of localities 
for each of the disjunct species now known from the entire Western Cas-
cade Range and the number of these localities visited during the present 
study. These numbers show that the area has been poorly botanized over 
the last 100 years : 87 percent of the known disjunct species localities 
are ones discovered or visited during this study. 
Ecological Ranges 
Also presented in Appendix Bare qualitative observations on the 
ecology of the distributionally interesting species both in the Western 
Cascades and, insofar as could be determined, in their more typical as-
sociations. Field observations were the most important source of infor-
mation, but studies of floristic literature and herbarium labels were of 
some help, especially in determining the nature of typical habitats for 
the species in other regions. In the study area observations were made 
on aspects of the environment and the relationships of interesting 
43 
Table II. Aspects of the Distribution and Ecology of Western Cascade 
Disjunct and Endemic Species. 
Key!£_ Symbols : 
Associations (listed in order of importance for each species) 
MF Mesic Forest 
CF Dry Mixed Conifer Forest 
LXM Low Elevation Xeric Meadow 
WM Wet Meadow 
MM Mesic Meadow 
SXM Subalpine Xeric Meadow 
GS Fine Gravel Scree 
OR Outcrop Ridge 
BC Boulder Creep Slope 
RMS Ro cky Melt Seep 
S Snowbed 
BT Blocky Talus 
VO Vertical Outcrop 
(seep. 31-40 for more complete descriptions) 
Growth Form 
EA ephemeral annual 
PA persistent annual 
LA late-blooming annual 
EP ephemeral perennial 
pp persistent herbaceous perennial 
s shrub 
T tree 
44 
'O 
QJ QJ 
'O :> 
CO I-< 
(J QJ 
Cl) Cl) 
co ..0 
uo 
Cl) 
p p Cl) 
.,0..,  I-< p QJ 0 
;.J ;.J .,.., 
.,c..o,  Cl) ;.J QJ co 
(.) !3: ,-l 
0 ;:J 
Cl) • 0.. 
Cl) 0 0 
zi:i... 
Southern and Eastern Element <t: 
Selaginella scopulorum OR EP 13 13 
Pinus ponderosa GS T 2 2 
Arabis platysperma howellii S,OR pp 3 3 
Linum perenne lewisii SXM , OR pp 6 7 
Gayophytum humile RMS,MM EA 27 28 
Linanthus harknessii GS,SXM,RMS EA 8 9 
Navarretia divaricata GS,SXM , RMS PA 20 21 
Monardella odoratissima BC,SXM,OR pp 4 4 
Penstemon deustus BC,OR pp 7 8 
Mimulus breweri RMS,GS,MM , SXM EA 30 32 
Castilleja pruinosa OR,BC pp 5 11 
Galium bifolium MM EA 22 22 
Lonicera conjugialis CF,SXM,BC s 2 2 
Chrysothamnus nauseosus albicaulis GS,OR s 3 4 
Nothocalais alpestris MM,SXM pp 1 1 
Southern Element 
Cheilanthes siliquosa BC,OR,SXM pp 20 22 
Allium c renulatum GS EP 10 12 
Arenaria pumicola OR , BC pp 2 5 
Silene campanulata glandulosa OR pp 10 13 
Cardamine integrifolia sinuata BT pp 3 6 
Ribes binominatum MM,MF s 25 26 
Ribes erythrocarpum MF,S s 4 6 
Trifolium productum GS,OR pp 5 6 
Trifolium howellii MF,WM pp 4 6 
Mimulus pulsiferae RMS,GS,OR EA 4 5 
Erigeron foliosus confinis OR,BC pp 19 21 
Crepis occidentalis GS pp 5 5 
Eastern Element 
Populus tremuloides MM T 1 1 
Polygonum kelloggii RMS,OR EA 5 5 
Lewisia triphylla RMS , S EP 13 13 
Arabis holboellii retrofracta OR,BC pp 24 26 
Horkelia fus ca SXM , S pp 1 1 
45 
Table II - continued 
'O 
Q) Q) p (fJ 
'O :> H p 
CU H Q) 0 
c.) Q) .j.J •H 
(f) (f) (f).W....-... cu ..0 Q) cu  uo ::S:.--l.--l 
(f) ;:J cu 
p 
~ p (fJ 0 H p § g-u 
•H 0 Q) 0 0 A-, 0 
.j.J ~ .W •H p .j.J 
cu (fJ .j.J ~ Q) 
•.-l ,..c:: Q) ..cu(.) .j.J :,: ...  . ..... 'Ocu  ,..c:: 
0 ~ ;:J cu (_) (fJ 
(f) 0 • 0.. .µ (fJ cu 
(f) H 0 0 0 cu ::S: 
<t! 0 ZP... E-' u-
Lathyrus lanszwertii aridus GS,OR,CF pp 4 4 
Gayophytum diffusum parviflorum SXM LA 10 11 
Gilia aggrega ta SXM,GS,OR pp 18 19 
Collomia linearis SXM PA 7 9 
Gentiana calycosa n. subsp. MF,WM pp 4 4 
Pterospora andromedea MF,CF pp 1 2 
Cryptantha affinis SXM,MM PA 1 1 
Helianthus cusickii OR,MM,CF pp 1 1 
Artemisia ludoviciana latiloba SXM pp 16 17 
Artemisia tridentata SXM,CF,OR,BC s 2 2 
Arnica parryi SXM,CF pp 8 9 
Microseris nutans SXM,GS pp 14 14 
Crepis acuminata OR pp 2 2 
Northern Element 
Polystichum andersonii WM pp 2 2 
Chamaecyparis nootkatensis OR,MF T 23 26 
Arenaria capillaris america·na OR,BC pp 17 17 
Sedum divergens BC,OR,GS pp 14 16 
Hydrophyllum fendleri albifrons WM,MM pp 13 14 
Orogenia fusiformis s EP 15 18 
Rhododendron albiflorum MF,WM s 3 4 
Menziesia ferruginea MF,MM s 2 3 
Douglasia laevigata VO,OR pp 5 5 
Mertensia bella WM EP 19 20 
Castilleja rupicola VO,OR pp 5 5 
Lonicera utahensis RMS,S s 2 3 
Haplopappus hallii OR,BC pp 19 21 
Luina stricta MM pp 12 12 
Alpine Element 
Polygonum newberryi S,GS pp 8 9 
Spraguea umbellata GS,S pp 3 7 
Arenaria rubella VO pp 6 6 
Luetkea pectinata S,GS pp 2 3 
46 
Table II - continued 
'tl 
(I) (I) p rJJ 
'tl > H p 
CO H (I) 0 
l) (I) .µ •H 
rJJ CfJ CfJ .µ,,...._ 
co ..0 (I) co . 
uo ::;: .-1 .-1 
rJJ ::, co 
p 
~ p CfJ 0 H p ~ g-u 
•H 0 (I) 0 
.µ 0 '1-1 0 ii. .IJ•H p .µ 
co CfJ .µ ::.::: (I) 
•H ..c: (I) co 
.µ 'tl 
. 
u :;3:.-1 .-1 co ..c: 
0 ~ ::, co l) CfJ 
Cf) 0 • 0.. .µ CfJ co 
Cf) H 
z0  0 c., 0 co :3: <t:: '1-1 1:-"'U'--' 
Ivesia gordonii GS , OR pp 2 2 
Polemonium pulcherrimum VO pp l 1 
Linanthastrum nuttallii OR,VO,GS pp 3 3 
Erigeron compositus , OR,GS,BC pp 1 2 
Valley Element 
Populus trichocarpa BT T 4 4 
Quercus garryana LXM , OR T 4 4 
Polygonum s pergulariaeforme LXM LA 3 4 
Convolvulus nyctagineus LXM , GS , BC pp 3 7 
Pl agiobothrys scouleri GS, SXM PA 1 1 
Wide spread Element 
Polygonum douglasii SXM , GS PA 28 30 
Polygonum minimum SXM 1 GS PA 30 36 
Silene douglasii OR , BC pp 22 22 
Sedum stenopetalum BC , OR,GS ,SXM pp 16 18 
Lupinus arbustus neolaxiflorus SXM , GS pp 27 30 
Lotu s nevadensis dougla s ii GS , OR , BC pp 16 22 
Phacelia linear is OR , GS , SXM PA 2 3 
Endemic Element 
Polygonum cascadense SXM , GS PA 17 19 
Aster gormani i GS , OR pp 1 1 
Erigeron cascadensis VO , OR GS pp 9 9 10 
47 
species to the rest of the vegetation. Some of these findings are 
presented in the preceding section on vegetation as well as in Appendix 
B. Table II presents growth forms of the species and the associations 
in which they have been found in the Western Cascades. 
Dispersibility, Pollinators, and Breeding Systems 
Since breeding systems have been shown to be important in estab-
lishing (Baker, 1956) and maintaining (Carlquist, 1966a) species popu-
l ations after long-distance dispersal, observations were made in the 
field on various mechanical adaptations for dispersal, on pollinators 
noted visiting the various species, and on floral morphology, size, 
color , and display to possible pollinators. The results are presented 
in Table III, together with the kind of breeding system suggested by 
the behavior of each species. It should be stressed that the proposed 
breeding systems are not supported by experimental evidence but are 
dedu c tions from the trends compiled by numerou s other workers. In 
general, small-flowered species which always set seed are presumed to 
be self-fertilizing, while large-flowered forms which are regularly 
visited by animal pollinators are concluded to be at least partially 
I 
outcrossing. Because modal apomixis ~ often a difficult reproductive 
method to deduce, it is suggested in Table III only for forms which 
produce little or no pollen. 
Among the dispersal adaptations noted were vegetative reproduc-
tion by fragmenting stems which root at the nodes; prickly, bur-like 
leaves ; tumble-weed-like, disarticulating flower heads that are blown 
by the wind, scattering seed; winged, disarticulating flowers that 
48 
Tab l e III. Dispersal Adaptations, Pollinators, Seed Set, Floral Mor-
phology and Exposure, Probable Breeding Systems, and Number of 
Western Cascade Localities of Disjunct and Endemic Species. 
Key to Symbols 
Dispersal Adaptations Pollinators 
1 vegetative r eproduction: 0 none 
fragmenting stems W wind 
2 prickly, bu r -like leaves D Dip te:ra 
3 tumbleweed inflorescences L Lepidoptera 
4 winged , disarticulating flowers H Hymenoptera 
5 spiny, bur-like calyces C Coleoptera 
6 exp l osive fruits B hummingbirds 
7 s piny berries 
8 bird-eaten berries 
9 f ruits with hooked spines Flower Exposure 
10 pappus or coma: wind-blown fruits 
11 wind-blown spores CC completely concealed 
12 spiny-tuberculate seeds PC partly concea led 
13 winged seeds EX fully exposed 
14 light, minute seeds 
15 large, heavy seeds 
16 glutinous seeds Flower Color 
G green 
Spopophyll Exsertion GW greenish-white 
GP greenish-pink 
0 no perianth w white 
CL flowers cleistogamous p pink 
IN stamens and style completely included L lavender 
SI stamens and style slightly included V violet 
EX stamens and style exserted BL blue 
y yellow 
s salmon 
Proposed Breeding Systems R red 
M maroon 
A apomictic BR brown 
SF probably self-fertilizing 
SC probably self-compatible, but 
partially out-crossing Flower Size 
OC probably mostly out-crossing 
SI probably self-incompatible d diameter 
h diameter of head 
L length 
(all measurements in mm) 
49 
M 
,,.-.._ i:::: 
H M •..-1 
0 ._~__ , i:::: '"O 4-4 •..-1 (lJ 
H .I,.) (lJ 
V) V) 4-4 0 (lJ .I,.) H 
i:::: H 0 .--I .--I N (lJ ~ 
Q.--1 0 .--I i:::: 0 •..-1 C/) 
•..-1 ell .I,.) (lJ >, 0 u C/) ,,.-.._ (lJ 
.I,.) V) ell H rJJ ..c •..-1 VJ~ .--I 
ell H i:::: ::J H 0.. .I,.) H H H'--' ..0 s 
.I,.) (lJ •..-1 rJ) (lJ 0 H (lJ (lJ (lJ ell (lJ 
0.. 0.. .--I 0 ~ H <U ~ ~ ~'"O ..0 .I,.) 
ell rJ) .--I 0..0 0 V) 0 0 0 (lJ 0 V) 
'"O •..-1 0 )< .--I 0.. )< .--I .--I .--I (lJ H >, 
<I! Cl 0.. ~Cz. C/)~ Cz. Cz. i:z. C/) 0.. C/) 
Southern &. Eastern Element 
Selaginella scopulorum 1,2 SC 
Pinus ponderosa 13 w oc 
Arabi s platysperma 
howellii 13 0 EX EX V Sd 80- SC 
Linum perenne lewisii 15 D,L,H EX EX BL 35d 9o+ SI 
Ga yophytum humile 0 0 EX EX w 2d 100 SF 
Linanthus harknessii 0 0 EX IN w ld Bo+ SF 
Navarretia divaricata 16 0 EX EX w ld Bo+ SF 
Monardella odoratissima 0 H EX EX L 14h ? oc 
Pens temon deustus 0 0 EX IN L 121 90+ oc 
Mimulus breweri 14 0 EX IN V 3d 100 SF 
Ca s tilleja pruinosa 13 0 EX IN R 231 ? oc 
Galium bifolium 9 0 cc EX G ld 100 SF 
Lonicera conjugialis 8 0 PC IN M 71 Bo+ SC 
Chrysothamnus nauseosus 
albicaulis 10 L EX EX y 91 60+ SI 
Nothocalais alpestris 10 0 EX EX y 30h 90+ oc 
Southern Element 
Cheilanthes siliquosa 11 SC 
Allium crenulatum 3 0 EX EX p 12h 9o+ SC 
Arenaria pumicola 0 0 EX EX w 8d ? oc 
Silene campanulata 
glandulosa 12 H,C EX IN GW lld ? oc 
Cardamine integrifolia 
s inuata 6 0 EX EX L 10d 80- SC 
Ribes binominatum 7 0 PC SI GW 6d Bo+ SC 
Ribes ~rythrocarpum 8 0 EX EX s Sd 30+ oc 
Trifolium productum 4 0 EX IN L 101 so SC 
Trifolium howellii 4 0 EX IN w 101 ? SC 
Mimulus pulsiferae 0 0 EX IN y Sd ? SC 
Erigeron foliosus 
confinis 10 L,H EX EX V 20h Bo+ OC? 
Crepis occidentalis 10 L EX EX y 22h 7o+ A 
so 
Table Ill - continued 
,,...., bl) 
,_. p bl) •.-1 
0 ...~_, , p 'O 4-< ,_. ·.-I (I) _µ (,_I).  
Cl) C,_l).  4-< 0 (I) _µ p 0 .--l .--l N (I) il'.'.l 
0 .--l 0 .--l p 0 •.-1 U) 
•.-1 cl! _µ (I) ::,.,o u U) ,,...., (I) 
_µ Cl) cl! ,_. Cl) ..c:: •.-1 Cl) ~ .--l 
cl! ,_. p ::i ,_. O,.-IJ ,_. ,_. ,_......,, .D E 
_µ Q) •.-1 Cl) (I) 0 ,_. Q) (I) (I) cl! Q) 
0.. 0.. .--l 0 ~ ,_. (I) ~ ~ 'O .D _µ 
cl! Cl) .--l 0..0 0 Cl) 0 ~ 0 (I) 0 Cl) 
'O •.-1 0 >< .--l o..>< .--l .--l .--l (I) ,_. >, ~o i:i... w~ UlW ~ ~ ~U) i:i... U) 
Eastern Element 
Populus tremuloides 10 w PC 0 G 7L O? oc 
Polygonum kelloggii 0 0 PC EX p 2d 100 SF 
Lewisia triphylla 3 D,C EX EX p 6d 100 SC 
Arabis holboellii 
retrofracta 6 C EX EX L Sd 9o+ SC 
Horkelia fusca 0 D,C EX EX w 8d ? oc 
La thyrus lanszwertii 
aridus 0 0 PC IN L lOL ? ? 
Gayophytum diffusum 
parviflorum 0 0 EX EX w 3d 100 SC 
Gi lia aggrega ta 16 B EX EX R 20L 9o+ SI 
Collomia linear is 0 0 EX SI p 12L ? SC 
Gentiana calycosa n. subsp. 0 0 EX SI BL 50d 9o+ oc 
Pterospora andromedea 13 0 EX IN BR 6d 100 SF 
Cryptantha affinis 5 0 PC IN w ld 100 SF 
Helianthus cusickii 0 L EX EX y 65h ? SI? 
Artemisia ludoviciana 
la tiloba 0 W? EX EX y 4h ? SI? 
Artemis ia tridenta ta 0 W? EX EX y 4h ? SI 
Arnica parryi 10 0 EX EX y 10h 100 A? 
Microseris nutans 10 L EX EX y 20h Bo+ oc 
Crepis acuminata 10 L EX EX y 16h ? SI 
Northern Element 
Polystichum andersonii 11 SC 
Chamaecyparis nootkatensis 0 w oc 
Arenaria capillaris 
americana 0 D,C EX EX w 10d ? oc 
Sedum divergens 1, 14 L,H EX EX y 9d 9o+ oc 
Hydrophyllum fendleri 
albifrons 15 0 PC EX GW 9d 2o+ SC 
Orogenia fusiformis 0 C PC EX w 12h 8o+ SC 
Rhododendron albiflorum 0 0 EX EX w 25d ? oc 
51 
Table III - continued 
,,......_ M p
µ oD .,..
 , 
0 .._~,,  .,p.. , "O 44 Q) 
µ µ Q) 
<J) <J) 44 0 Q) µ µ 
p µ 0 ..-I ..-I 
0..-1 0 ..-Ip 0 .,N..,  
Q) i:Q 
Cl) 
.,.., (IS µ Q) 
µ >.o u Cl) ,,......_ Q) <J) (IS µ <J) ,..C:•,-, U)~ ..-I 
(ISµ .,p.. , :i µ o.µ µ µ µ.._,, µ Q) ..0 s <J) Q) 01-l Q) Q) Q) (IS Q) 
0.. 0.. ..-I 0 !3: µ Q) ..O..J 
(IS !3: !3:-0 <J) ..-I 0..0 O<Jl 0 Q) 
"O .,.., ~ 0 0 <J) 0 ><: ..-I 0.><: ..-I ..-I ..-I Q) µ >, <O A... ~i:,:.. Cl)~ i:,:.. i:,:.. i:,:.. Cl) A... Cl) 
Menziesia ferruginea 0 0 PC IN GP 6d 9o+ SC 
Douglasia laeviga ta 0 0 EX IN V 16d 60+ oc 
Mertensia bella 12 H PC IN BL 6d 8o+ SC 
Castilleja rupicola 13 0 EX IN R 30L ? oc 
Lonicera utahensis 0 C PC IN GW lOL ? SC 
Haplopappus hallii 10 0 EX EX y 14h 60+ oc 
Luina stricta 10 L, C EX EX y lBL Bo+ SI 
Alpine Element 
Polygonum newberryi 0 0 PC IN GW 3d ? SF 
Spraguea umbellata 4 0 EX IN p 30h ? SF 
Arenaria rubella 0 D EX EX w 4d ? SC 
Luetkea pectinata 0 0 EX EX w 7d ? oc 
Ivesia gordonii 3 L EX EX y 6d 50- oc 
Polemonium pulcherrimum 0 H,C EX EX BL 15d 20- oc 
Linanthastrum nuttallii 0 0 EX IN w 12d ? oc 
Erigeron compositus 10 0 EX EX w 14h Bo+ OC? 
Valley Element 
Populus trichocarpa 10 w PC 0 G 6L ? oc 
Quercus garryana 0 w cc 0 G (5) ? oc 
Polygonum spergulariaeforme 0 D PC EX p 4d ? SC 
Convolvulus nyctagineus 0 D,C EX SI w 40d ? oc 
Plagiobothrys scouleri 5 0 EX IN w ld ? SF 
Widespread Element 
Polygonum douglasii 0 0 PC CL GP 4d 100 SF 
Polygonum minimum 0 0 PC EX p 2d 100 SF 
Silene douglasii 12 L,H,C EX IN w 14d 90+ oc 
Sedum stenopetalum 1 L, H EX EX y lld 0 A 
Lupinus arbustus 
neolaxiflorus 0 H EX IN BL llL 90+ oc 
Lotus nevadensis douglasii 0 H EX IN y lOL 90+ SC 
Phacelia linear is 0 0 EX EX w 10d ? oc 
Endemic Element 
Polygonum cascadense 0 0 PC EX p 2d 100 SF 
Aster gormanii 10 0 EX EX w 20h 4o+ ? 
Erigeron cascadensis 10 L EX EX w 14h Bo+ OC? 
52 
might be wind dispersed; spiny, bur-like calyces; somewhat explosive 
si liques; spiny , bur-like berries; red , bird-eaten berries; bur-like 
nutlets with hooked spines; fruits or seeds with plumed appendages 
such as pappus or coma for wind dispersal; windblown spores or light 
minute seeds; and spiny-tuberculate, winged, or glutinous seeds. Sev-
eral species showed the evidently negative dispersal adaptation of 
large, heavy, unpalatable seeds. This characteristic is also noted 
in Table III. 
Some species of quite limited distribution could not be observed 
long enough in the field to see visits by pollinators. Such species 
include Arabis platyspema, Luetkea pectinata , Rhododendron albiflorum, 
Douglasia laevigata, Gentiana calycosa n. subsp., Linanthastrum 
nuttallii, Phacelia linearis , Castilleja pruinosa , Castilleja rupicola , 
and Nothocalais alpestris . I ns ect visitors are reported here only to 
order although more specific identifications were made for some Lep i dop -
tera , Hymenoptera, and Diptera. All major groups of temperate pol li-
nators were encountered, but the most abundant and frequently seen were 
unidentified small beetles, which spend considerable time within a 
single flower and are often found in rather large numbers. Th e only 
evidently obligate relationship between a disjunct plant and pollinator 
i nvolv ed Cilia aggregata, a polemoniaceous species with scarlet tubular 
flowers, and the rufous hummingbird , Selasphorous rufus . Female s of 
other hummingbird species may also visit Q. aggregata in the Western 
Cascades , but they could not be differentiated from~- rufus in t he 
f ield , Q, aggregata ranges throughout most of the Western United States 
and undoubtedly is po l linated by other species of hummingbirds in 
53 
other parts of its range. It has not been shown conclusively in this 
study that any of the observed visitors actually effect either cross 
or self-fertilization, although at least in the case of the larger 
more attractive visitors a considerable amount of cross pollination 
can be assumed. 
The most striking feature of Table III is the evident range in 
breeding systems represented by Western Cascade disjunct species. The 
range in adaptations for dispersal , pollinators, and breeding systems 
for disjunct species seems to be quite comparable with that of the 
Western Cascade flora as a whole. The presently available evidence 
thus indicates that type of breeding system has been of no general 
significance either in establishing or maintaining Western Cascade 
disjunct populations. This may in turn argue for relatively easy, 
repeated dispersal to the area. Sixty percent of the species have 
specific adaptation for such dispersal. 
Moisture Regimes and Phenology 
The present hypotheses concerning the presence of disjunct 
s pecies in the Western Cascades implicate the moisture tolerances of 
t h e disjunct plants or the cycle of moisture availability in the areas 
where these species are concentrated. An attempt was made to gather 
bio l ogically significant data, using a sap-tension pressure-chamber, 
from a number of species that might help test these idea s, especially 
Detling ' s "Xeric Island Hypothesis. " Iron Mountain was chosen to 
s ample intens i vely, because of its diversity of habitats and disjunct 
s pecies and its accessibility. Some additional corroborative 
54 
information was collected on Sardine Butte and in Crater Lake National 
Park. A great deal of information on variability within species, among 
species, among habitats, and with time was obtained from these studies. 
The results have elucidated the range of stresses encountered under 
various conditions and have indicated different kinds of adaptation to 
seasonally increasing moisture stress. In addition , it has been shown 
that there is no reasonable a priori way of defining xeric and mesic 
l ocalities or xeric and mesic species. Such definitions must always 
refer to the functional environment of the individual plant or at most 
to members of a species population within a given habitat. 
Literature Review and Methods 
Moisture stress data were obtained using the sap-tension 
pressure-bomb devised by Scholander and others ( 1965 ) to measure in 
an almost direct fashion the water tension in the xylem cells of 
plants. Scholander's simple but elegant techn i que involves cutting 
the stem to be measured, sealing it in a pressure -chamber with the 
cut end exposed to the atmosphere, and opposing the original tension 
in the stem with a positive gas pressure app l ied through the stomata 
of the leaves. When the applied pressure is equal to the original 
tension, an equilibrium state is produced , and sap reappears at the 
cut surface of the stem. Scho l ander and his colleagues performed 
studies on the sap tension of various plants as related to habitat, 
height, and parasitism, finding l ow tensions in all herbs and ferns 
(those measured were confined to damp forest or fresh water environ-
ments) and high tensions in the trees and shrubs of desert and 
55 
seashore environments. Tensions in trees increase with height, and 
parasites of the genera Phoradendron and Phrygilanthus were found 
without exception to have higher tensions than those found in their 
hos ts. In addition, Scholander presents a theoretical model showing 
t hat the equilibration pressure is, in fact, equal and opposite to 
t h e hydrostatic pressure in the xylem of the intact stem. 
The data of Boyer (1967) support Scholander's model ~ Boyer 
compares the water potential of leaves as measured by a thermocouple 
psychrometer with the water potential of the xylem cells as measured 
with a pressure-chamber. He used greenhouse cultures of Taxus 
cuspidatus, Helianthus annuus, and Rhododendron roseum. Although 
Boyer found that configuration of the tissues in the stem may be 
responsible for considerable error in some cases, as in Rhododendron, 
he repor ts accuracy of the pressure - bomb technique within about two 
atmospheres. 
Boyer states that the pressure-chamber technique actually 
measures leaf water potential and not tension in the xylem. However, 
if the various tissues of the leaf are in equilibrium with regard to 
diffusion pressure deficits, the two measures will be identical. 
There is no theoretical reason to assume that mesophyll cells and 
vascular tissue in the leaf would not be in moisture equilibrium, and 
Boyer found no differences, even in rapidly transpiring material. 
Thus, it can be assumed reasonably t hat pressure-bomb readings are 
measurements of negative hydros ta t ic pressure within the water con-
ducting tissues. 
56 
Scholander and others (1965 ) have shown that initial xylem exu-
dates from a wide variety of species, including extreme halophytes, are 
almost pure water , leaving hydrostatis tension as the only component 
of the xylem diffusion pressure deficit. Boyer (1967) found somewhat 
h igher osmotic potential in the xylem sap, but he used exudate from 
t i ssue s which had prev iously been pressurized. It is possible that 
s ome damage was incurred by mesophyll cells during the first pressur-
ization and that later exudate a c tually represents a mixture of xylem 
and cell saps. 
Waring and Cleary ( 196 7) found good correlation between water 
stress in twigs of douglas fir as measured by the pressure-bomb and 
the water potential of associated leaves determined by a vapor equi-
libration technique. These findings add experimental support to 
Scholander's model. Waring and Cleary also report on field studies 
of coniferous species in southwestern Oregon. A moisture gradient 
respons ive to exposure and soil type is described. Reported differ-
ences among the various coniferous tree species in the same stand are 
small once root systems are well established , 
A portable .model of Scholander ' s pressure-bomb was used in this 
study. The model was designed by B. D. Cleary at the Oregon State 
Forest Research Laboratory, Corvallis. 
Over 1000 readings were made with t h e instrument during the 
summer of 1967. No rainfall was recorded in the study area during 
t h i s time , and almost every day wa s c lear and sunny, resulting in un-
usually high and continued water stresses and accompanying extreme 
f i r e danger. Since a large number of readings was desired , specimens 
57 
were cut and sealed immediately in plastic _bags, kept out of direct 
sunlight, and returned to a convenient location to be measured using 
standard large cylinders of compressed nitrogen. After the initial 
cutt ing, stems were not recut, but phloem was trimmed (except for 
smal l annua ls), the stems were fitted into slits in rubber stoppers 
and sealed into the pressure-chamber with one cm or less of the stem 
protruding into a normal atmosphere. Pressure in the chamber was 
slowl y raised, at a rate of about 0.6 atmospheres per second . Gas 
flow wa s stopped and the gauge pressure read when the cut xylem dark-
ened a s the tracheids or vessels filled with sap. The pressure was 
then released and the measured stem was discarded. A broad ecological 
and taxonomic spectrum of plant species was studied using the pressure-
bomb method, and several more or less distinct sources of variation in 
readings were encountered, They are described and exemplified below, 
Variation with Time after Cutting 
Since the above method required occasional long intervals be-
tween cutting the stems and measuring the sap tension, the error intro-
duced by this delay was determined for several species of different 
growth habit (Figure 1). In the first such experiment, 20 stems from 
one large tree of Pseudotsuga menziesii were cut, divided into four 
groups, and sealed into one~pint plastic bags with most of the air 
expressed. Each group of stems was measured at approximate half-hour 
intervals, the first being read about 15 minutes after cutting. F 
tests between t h e means demonstrate no significant differences in 
readings taken immediately and two hours after cutting. 
15 8 o STEMS SHADED 0 8 
0 
• STEMS EXPOSED TO SUNLIGHT 
10 I 8 I 8 
Cl) 5 PSEUDOTSUGA MENZIESII 
IJJ 
0:: 
IJJ 
:c 
0. 
Cl) 20 i 0 
0 
:E § 0 0 0 
t- 15 0 
<t 8 0 
0 
10 
z GILIA AGGREGATA 
0 0 
en 
z 25 
IJJ 
t- • 
0 
20 0 
0. I s0  8 <t §i 0 
en 15 0 0 0 
0 0 
• 
RIBES BINOMINATUM 
2 3 4 5 6 7 
TIME AFTER CUTTING: HOURS 
Figure l. Variation in Maximum Sap Tension with Time after Cutting: Stems Stored in 
Plastic Bags. 
' 
' ' 
70 0 
0 0 
(J) • 
w 60 
a: 
w 
:c I 
a_ 
(J) 
0 50 
~ 
;- 0 
<t 
0 
40 0 • 
z 
0 • 
0 
(J) 8 
z 0 0 
w 0 0 30 0 0 8 0 0 0 0 t- 0 0 0 
~ 0 
a_ 0 0 
<t 0 
(J) 
2:~ POLYGONUM DOUGLASII --
I 
2 3 4 5 6 7 8 
TIME AFTER CUTTING: HOURS 
Vt 
\!) 
Figure 1 (cont) . Variation in Maximum Sap Tension with Time after Cutting: Stems Sto red in 
Plastic Bags. 
60 
After discovering that the bagging method would often entail 
delays of longer than two hours, additional experiments were performed , 
Stems representing populations of Ribes binominatum , a perennial shrub; 
Gilia aggregata, a biennial herb; and Polygonum douglasii , a small 
annual , were collected in a similar manner in the early afternoon of 
a sunny day. Four bags of stems were kept shaded at ambient temper-
ature, and on e was left on open ground in full sun until sunset. 
Shaded bags were measured at irregular intervals averaging longer than 
one hour, and the last shaded bag and that left in the sun were meas-
ure.d together about seven hours after they were collected--a much 
l onger delay than was necessary in other experiments. 
In groups of stems left in the shade, differences between ini-
tia l and final samples were not significant as determined by an F test 
between the means. Bags of stems left in the sun were coated internal ly 
with condensed moisture, b~t for Ribes binominatum and Gilia aggregata 
pres sure bomb readings fell well within the range of the shaded samples. 
Only in Polygonum douglasii did treatment of the cut and sealed stems 
result in obviously different readings. Although the highest readings 
for the experiment came from shaded stems measured early in the experi -
ment, the moisture tension in sun- treated stems averaged much higher, 
indicating that they had lost considerable water. It seems likely 
t ha t the difference between the two perennials and the annual resulted 
from the relative masses of the stem pieces, indicating that Polygonum 
doug l a si i lost relatively more moisture to the air which remained in 
the p l astic bag. Also pertinent is the fact that the annual showed 
greater s tress at the beginning of the experiment, indicating a lower 
61 
water to dry weight ratio than in the other species. This series of 
experiments demonstrates clearly that with reasonably careful treat-
ment, sap tensions, even in small annuals, remain remarkably constant 
for long periods after cutting. The plastic bag storage technique 
was used for all subsequent measurements. 
An interesting result of this large-scale sampling of popula-
tions was the range in variance from one population subsample to an-
other. This is particularly marked in Gilia aggregata and Polygonum 
douglasii. 
Diurnal Variation 
Scholander and others (1965) report marked diurnal variation 
in sap tension in various plants. Waring and Cleary (1967) note that 
equilibration between plant and soil occurs by early evening but pres-
ent measurements indicate that it continues at lesser rates throughout 
the dark hours. Due to this fact and to field sampling problems after 
dark, minima in this study are represented by cuttings taken before 
full light in the morning. Maxima were found to occur between about 
1300 and 1700 hours on cloudless days with little haze. 
In Figure 2 diurnal curves are compared for three species. 
Each point in this graph represents a single measurement. Later 
studies showed that sap tensions in phenologically similar populations 
of Polygonum cascadense ranged over about 8 atmospheres at minimum 
tension and 11 atmospheres at maximum tension. Lotus nevadensis doug-
lasii and Luina stricta are herbaceous perennials and show considerably 
less variability than the annual knotweed. In the two perennials 
50 62 
POLYGONUM CASCAOENSE 
40 
~ 
Cl) 
w 
0:: 
w 
:I: 
a. 30 
Cl) 
0 6 G----------
~ 
t- / 0 
<( 0 LOTUS NEVADENSIS 
OOUGLASII 
z Cl 
0 
Cl) 20 
z 
w ~0 
t- 0 0 0 
a. 
<( 
Cl) LUINA STRICTA 
10 
0' 
0 
' . 
0 
7 9 II 13 15 17 19 , 21 
TIME OF DAY 
Figure 2. Diurnal Variation in Sap Tension in Three Species on Iron 
Mountain, ·July 23, 1967. 
63 
population variation was wi thin 3 atmospheres at minimum tension and 
within 5 atmospheres at maximum. 
The most symmetrical curve is that for Lotus nevadensis douglasii 
taken from a south-facing scree slope. Luina stricta, confined to more 
westerly meadow slopes, reaches maximum tension later in the evening, 
due evidently to its direction of exposure. Polygonum cascadense shows 
both the highest minimum tension and the greatest diurnal variation. 
It is dependent on moisture near the soil surface, unlike the deeper-
rooted perennials . The Polygonum population grew in dry rocky loam on 
a southwest-facing slope. In these studies actual minimum tensions 
probably were not obtained, since the sun had risen (although not on 
the slopes sampled) when the stems were cut. 
Variation within Individual Plants 
It might be expected from the anatomical and molecular arrange-
ment of water in plant stems that equilibration of diffusion pressure 
deficits would be rapid and that differences in sap tension measure-
ments within a single plant would be due to experimental error. How-
ever, Vit~ and Rudinsky (1959) have demonstrated that in conifers, even 
without the more continuous water columns provided by angiosperm ves -
sels, sap migrates up the trunk in definite spiral patterns and does 
not diffuse through the entire stem. Water from one sector of roots 
will therefore supply only a portion of thi branches, but all roots 
supply the crown tip. Using injected dyes, Vit~ and Rudinsky charac-
terized a variety of patterns of water movement, the most restricted 
of which were characteristic of Pseudotsuga menziesii and Chamaecyparis 
64 
lawsoniana. These findings indicate that sap tensions in portions of a 
single plant occupying different micro-environments might be markedly 
different. This phenomenon has been demonstrated in Pseudotsuga 
menziesii (Figure 1) and Vaccinium membranaceum (Figure 3). 
In general, however , when measurements were taken from several 
stems of a single individual , the readings were tightly clustered and 
s panned less than one atmosphere (Figure 3). These measurements demon-
strate the repeatibility and precision of pressure-bomb measurements. 
Variation among Parts of~ Pl ant 
An attempt was made to determine the reliability of extrapo-
l ating from one portion of a single plant to another, as well as to 
determine how small a p l ant part might be used to produce a reliable 
measurement. A large plant of Ivesia gordonii, a cespitose perennia l 
herb, was dug; and individua l leaves, flowering s t ems having a single 
bract, and portions of the branching rootlike caudex were sealed in 
plastic bags and measured. Th e results are shown in Figure 4. Meas-
urements for all three plant parts overlap, and differences between 
the means are insignificant at the 5 percent level as determined by 
the F test. Flowering stem and caudex measures showed low variance , 
but that of the leaves was considerably greater. Since the leaves are 
smal l and had to be carried in a knapsack for most of a hot afternoon , 
the greater variance may in part be the result of loss of water to 
air in the plastic bag. These f indings do indicate that vi rtually any 
measurable part of the plant can be used with considerable reliability . 
11 
~ RIBES BINOMINATUM Cl) ..J ct 
::> 
>-
0 
 ~ ' CHAMAECYPARI S NOOTKA liENSI S 0 z 
LL 
0 'f I 
~ CHRYSOTHAMNUS NAUSEOSUS ALBICAULI S 0::: w m 
~ 10 20 30 40 50 60 70 
::> ua 
z ■ 
ff VACCINIUM MEMBRANACEUM 
SAP TENSION 1 ATMOSPHERES 
Figure 3. Variation in Minimum Sap Tension within Single Plants on Iron Mountain, 
August 1967. 
' ' 
°V',  
' ' 
FLOWE RING STEM S 
CJ) 
_J 
<t I 
:::, ~I 
-0 > 
0 
z 
LEAVES 
~ 
0 I ■ Pf.\1 I 
a: 
w 
m 
~ 
:::, 
.z CAUDEX BRANCHES 
rn l1 
10 20 30 40· 50 60 70 
SAP TENS ION• ATM OSPHERE S 
Figure 4. Variation in Maximum Sap Tension among Plant Parts in Ivesia gordonii (Cone 
Peak, 3 August 1967). 
67 
Variation within a Population 
Figure 5 shows that minimum sap tensions are quite constant 
within populations growing in more mesic habitats (Chamaecyparis 
0
nootkatensis, Ribes binominatum, Lupinus arbustus neolaxiflorus), but 
that in areas where moisture stress is generally higher, even minimum 
tens ions show large variances within small populations (Chrysothamnus 
nau s eosus albicaulis, Gilia aggregata). 
The Chamaecyparis nootkatensis population measured is actually 
a c lon e of individuals beneath the Mesic Forest canopy. Many new 
trunks have been added by self-layering of branches. These trunks are 
still interconnected, but each has a well-established root system of 
its own. The remarkable clustering of readings, as well as showing 
the reproducibility of results with this technique, indicates that 
such clones should be considered single individuals for physiological 
purposes. 
Population Variation Correlated with Plant Size 
It was noted early in the sap tension determinations that in 
a variety of small annuals growing in drier sites, high measurements 
were strongly correlated with small plant size. This phenomenon has 
been demonstrated in trees by Waring and Cleary (1967) but is best 
exemplified in annuals such as Polygonum douglasii, as shown in Fig-
ure 6. A sample of nine plants showing the range of sizes present 
within 2500 square cm was ordered by size of plant and measured. 
Without exception sap tension increased with decreasing plant size. 
' ' 
LUPI NUS ARBUSTUS NEOLAXI FLORUS 
(/) 
....J 
ct ~ V RISES BINOMINATUM ~ 0 
> 
C 
z 
~ J CHAMAECYPARIS NOOTKATENSIS LL 0 
a: Id. I 
l1J 
m ~ CHRYSOTHAMNUS NAUSEOSUS ALBICAULIS ~ ~ z 
10 20 30 40 50 60 70 
i ii a I I 
H GILIA AGGREGATA 
SAP TENSION: ATMOSPHERES °00'  
Figure 5. Variation in Minimum Sap Tension within Species Populations on Iron Mountain , 
August 1967 •· 
70 
0 
65 
(/) 60 
w 
0::: 
w 
J: 55 
a. 
(/) 
0 
~ 50 .... 
<l 
45 
z 
0 
40 . 
(/) 
z 
.w...  35 
a. 
<l 
(/) 30 
25 
0 '----LA-R-G'E-S-T -----'-----'------l-----.1-.-----L----i-----L..S.M-A-L-L-E-LS T 
.. 
RANK BY · SIZE 
Figure 6. Maximum Sap Tension a s Related to Plant Size in Polygonum douglasii (Iron 
Mountain , 2 Augus't 196 7). 
70 
These annuals have shallow root systems and almost certainly compete 
for water near the surface of the soil. Digging numerous plants has 
demonstrated that depth of root penetration is correlated with plant 
heigh t . It is thus likely that high sap tension is to some extent 
both cause and effect of small plant size, since small plants are poor 
c ompetitors for soil mois t ure , which is probably limiting, and there -
fo r e remain of lesser sta tu re. 
Popu l ation Variation a t Minimum and Maximum Tens ion 
Wa r ing and Cleary ( 1967 ) note that samp l e variation i s greater 
at maximum tens ion than a t min imum tension. This has been confirmed 
i n a number of Western Cascade species , exemp l ified in Figure 7 by 
Collomia l ineari s , an annua l . While minimum tensions span on ly 2.5 
atmo sph eres, maximum tensions range over at l eas t 43 atmosphere s , t h e 
highest reading being over 68 atmosphere s . I n this, as in other annu-
als , variation in depth of root penetration is probably of major sig -
nificance to variation in maximum tension. Even at thes e high maximum 
tensions, individuals of C. linearis seem capable of completely equili -
brating with the soil dur ing the night . 
Variation with Habitat 
A number of We s te rn Cascade di s junc t species a r e found i n mo r e 
than one habitat type. Th ey are of ma j or importance in comparing the 
moi s ture regime s in the va rious sites where they occur . Chamaecypari s 
nootkatens is is found a s a fore s t t r ee among similar - sized specimens 
of Ts uga me r tens iana and Abies l a s iocarpa on north - facing slopes near 
' ' 
en 
...J 
<t 
:::> 
0 8 
> ~6 
0 :!: . 
z x4 ~ 2 
~ 
0 
u. :!: 2 10 30 40 50 60 70 
0 ::> 
~4 
z 
er: i 6 
lJ.J 
m 8 
~ 
:::> SAP TENSION: ATMOSPHERES 
z 
Figure 7. Population Varia tion at Ninimum and Maximum Sap Tension in Collomia linearis 
(Iron Mountain, 23 August 1967). 
-......J.  
72 
the summits of some Western Cascade peaks and as occasional understory 
clones beneath less dense but more mature Pseudotsuga menziesii - Abies 
amabilis forests on lower slopes. It also occurs in very shallow 
soils, sometimes rooted in rock crevices on exposed ridges of various 
exposures. Four such habitats were compared using pressure-bomb tech-
niques, but only two are reported here since the two forest localities 
were found to be nearly identical , as were the two ridge localities. 
Figure 8 shows that the forest popu l ations had considerably lower ten-
sions, both at maximum and minimum tension , than the ridge populations. 
Specimens of Ribes binominatum from the same shaded forest lo-
cality, however, had significantly higher minimum tensions than compar-
able populations in open , southwest-facing meadows. Maxima were nearly 
the same. This indicates that the hot open meadows actually comprise 
a less stressful environment for Ribes binominatum than does the shaded 
forest. This is probably due to the heavy dew in the meadows, resulting 
in less nighttime transpiration and consequently more rapid equilibra-
tion with the soil. Nightly dew may also add a considerable amount of 
moisture to the surface soil during a pro l onged drought. 
Stems of Pachystima myrsinites were taken from three different 
habitats on Sardine Butte. The first was a dense forest with much con-
iferous litter and little ground-cover, the second a moderately steep 
east-facing rockfall slope, and the third the rocky exposed summit of 
the east spur of the peak. Figure 8 shows the expected relationships 
between the moisture regimes of t h e three hab i tats for this species: 
increasingly high tensions were recorded from the forest, rockfall 
slope, and exposed summit habitats. 
CHAMAECYPARIS NOOTKATENSIS (FOREST) 
' ' 
(/) 
_J 
<( CHA MAECYPARIS NOOTKATENSIS (RIDGE) 
:::> 
-0 > 
0 
z 
u. r 
0 RISES S INO M INATUM (FOREST) 
a: 
w I bUD □fJD 
co . 
~ 
:::> 
z 
~ 6 
:x:!1: 4'  RISES BINOMINATUM (MEADOW) 
~ 2 
0 l-....-r.=-r--..l.--......1.::.&........1w.-------'-----......1.___ ___. ,__ ____. ..._ _____  
10 
2 20 30 40 50 ::!1: 60 70 
:> 
~ 4 
~ 6 -..J 
::& w 
SAP TENSION: AT MO SP HERES 
Figure 8 . Var i ation in Sap Tension with Habitat ( Iron Mountain, 2 Augus t 1967). 
PACHYSTIMA MYRSINITES (FOREST) 
' ' 
a ·o 
CJ) 
_j 
<( PACHYSTI MA MYRSINITES (OPEN E AST SLOPE) 
::> 
0 
> 
0 
z 
PACHYSTIMA MYRSINITES (ROCKY SU M MIT} 
LL 
0 
i ~'-· DOUGLAS I A LA EVI GATA (SHADED} ---...L---U.Ja.a.nL .aU, -..B.1..------.JL-----..1------'----..-.1.---------
DOUGLAS IA LAEVIGATA ( EXPOSED) 
n 
10 20 30 40 50 60 70 
SAP TENSION: ATMOSPHERES 
Figur e 8 (cont .). Variation i n Sap Tension wi t h Habi t a t (Sard ine Butte, 27 J uly 1967; 
and Cone Peak , 3 Augus t 1967). 
75 
Douglasia laevigata is a highly restricted species in the 
Western Cascades, being found only on certain west or north-facing 
vertical basaltic outcrops exposed to the prevailing wind s. On one 
such rock wall on Cone Peak, specimens were s ampled from a rather 
deep protected crevice and from a more expo s ed crev ice less than a 
meter away. The more protected plants were etiolated, had very large 
leaves for the species, and did not flower abundantly compared with 
close neighbors but exhibited sap tens ions lower by more than 16 at-
mospheres (Figure 8). Such major differences in moisture regimes over 
small distances illustrate the difficulty in attempting to describe a 
species or a region, no matter how small , as "xeric" or "mesic." 
Variation with Habitat and Ph enological Stage 
Especially in annua ls, it is difficu l t or impossible to sepa -
rate habitat differences from phenolog ical diffe r ences, since as the 
plant matures and dies during a sing le s eason 9 it exhibits a distinc t 
progression of sap tens ions. A moister habitat may simply prolong 
this characteristic progress ion . Such problems are il l ustrated well 
by populations of Polygonum cascadense on Iron Mountain. On 2 August 
plants were measured from three habitats of diff ering insolation, 
exposure, and soil type ( Figure 9). The highest popu l ation was found 
on a s teep s outh-facing scree slope; t he population of intermediate 
elevation on part ly-shaded, more gently -sloping fine gravel scree near 
xeric meadow species; and the lowest popu l ation in the shade of robust 
meadow species. The apparen t age of the individuals became l ess with 
decrease in altitude, and s ap tensions decreased correspondingly. On 
2 AUGUST 
SHADED - YOUNGEST 
B 
Cl) LOW SCREE - INTERMEDIATE AGES 
_J 
<t 
::> 
Cl " 0 afle H p " n 
> 
Cl 
z HIGH SCREE - OLDEST 
u. Q n I ll n n 
0 
a: 
w 23 AUGUST 
m a· 
~ 
::> ~ 6 SHADED LOW SCREE 
.z ~4 X 
; 2 
0 1---n----'-'-Ll-t'~--......- -'--------i.---,,,-n--....__ __~ __.__ ____. .._ ___- ir. ..............- 
~ 2 20 30 40 50 60 
=> 
~ 4 
z 
i 6 
8.  
SAP TENSION: ATMOSPHERES 
Figur e 9~ Variation in Sap Tension with Habitat and Phenological Stage in Polygonum --..J 
cas-cadense ( Iron Mountain ). °' 
u 
~ POLYGONUM MINIMUM (SHADED - YOUNGER) 
C/) 
..J 
<( u 'Y u 0 
:::> ~ POLYGONU M MINIMUM (EXPOSED - OLDER) 0- > 
0 
z 
LL 
0 
0:: GAYOPHYTU M DIFFUSUM PARVIFLORU M (G ENTLE SLOPE - YOUNGER) 
IJ.1 
CD 
:E 10 20 30 40 50 60 70 
:::> 0 
z 2 
4 GAYOPHYTUM DIFFUSUM PARVIFLORUM (STEEPER SLOPE - OLDER ) 
6 
8 
SAP TENSION 1 ATMOSPH ERES 
' ' 
Figure 9 (cont.). Variation in Sap Tension with Habitat and Phenological Stage in 
Polygonum minimum (Iron Mountain, 2 August 1967) and Gayophytum 
-..J 
di ffusum parviflorum (Iron Mountain, 23 August 1967 ). -..J 
78 
2 August plants at the highest locality (separated by 60 vertical m 
from the lowest locality) were still blooming freely, although some 
specimens registered tensions beyond the capacity of the instrument 
( above 68 atmospheres). By 23 August the highest population had com-
pletely disappeared, but the lower two were again measured. The inter-
mediate population then supported tensions comparable with those of 
the highest population on 2 August. The lowest population was in turn 
comparable to the intermed i ate population on the first date of measure -
ment, although the range exhibited by individual plants was not as 
great, as might be expected from the habitat configuration. I n years 
of higher summer rainfall than 1967, plants oft· cascadense have been 
observed blooming into October, indicating that available moisture 
closely controls their phenology. Other variables in the habitat are 
evidently much less important for the distribution of the species, as 
long as some direct sunlight is available. Gayophytum diffusum parvi-
florum and Polygonum minimum show patterns which are similar but docu-
mented by fewer measurements, 
Variations with Phenological Stage 
Occasionally a direc t relationship between sap tension and 
phenological stage can be demonstrated. This is possible where within 
a small area (less than 100 square cm) a population of ephemeral an-
nuals occurs in which the individuals are of similar sizes but are at 
different developmental stages. Two such situations were observed at 
Crater Lake National Park, where the season was considerably delayed 
over that on Iron Mountain, and earlier phenological stages of many 
79 
species were available for study. 
Specimens of Polygonum kelloggii exhibited uniform minimum 
tensions, even though a complete progression of phenological stages 
was represented in the population (Figure 10). When measured at 
maximum tension, the phenologically oldest specimens (not necessarily 
the smallest individuals) had sap tensions almost 30 atmospheres 
higher than the youngest. It is obvious that moisture was beginning 
to be limiting for some members of the population and that within a 
few days at most the population would be entirely gone. 
An even more striking example of rapid desiccation was illus-
trated by a population of Mimulus breweri from the same meadow, al-
though age differences among the individuals were less obvious than in 
Polygonum kelloggii. Minimum tensions were consistently low except in 
one individual which represented a slightly more advanced phenological 
stage. This specimen registered above 68 atmospheres on the pressure-
bomb apparatus (Figure 10). Maximum tensions were only slightly higher 
than minima except, again, for the oldest individual, which measured 
18 atmospheres higher than any other. Within the population there are 
never more than a few plants with intermediate sap tensions. The sap 
tens ion measurements corroborate the morphological evidence that when 
water begins to become limiting for these plants, they desiccate rap-
idly and die. Unfortunately, it is impossible to take repeated meas-
urements on a single individual, and this rapid progression cannot be 
completely followed with the pressure-bomb apparatus. 
8 
:!: 
:::, 6 
~ 
CJ) x 4 POLYGONU M KELLOGG II 
...J <( 
<( :!: 2 
:::> 0 
0 ::E 
:::, 2 ,,, 
> :!: 
z 4 
0 
~ 6 z 8 
u. 
0 8 ~ 
:::, 6 
::E 
a: X 4 MIMULU S BREWER I 
LL.I <( 
m ~ 2 
~ 0 30 40 50 60. 70 
:::> 20 ::E 
z 2 ::, ::E 
z 4 
~ 6 
8 
SAP T ENS IO N : ATMOSPHERES 
Figure 10. Variation i n Sap Tension with Phenologica l Stage (Crater Lake, 13 August 1967~ 
s ee text fo r explanation . 
' ' 
00 
0 
81 
Xeric Indicators 
Sap tensions were measured in seven of the species termed "xeric 
indicators" by Detling (1953): Lupinus arbustus neolaxiflorus, Collomia 
linearis, Gilia aggregata, Linum perenne lewisii, Polygonum douglasii, 
Sedum stenopetalum, and Silene douglasii. Sap tensions ranged from the 
lowest registered by any species (Lupinus arbustus neolaxiflorus: 2 
atmospheres) to readings above 68 atmospheres ( Collomia linearis, Linum 
perenne lewisii, Polygonum douglasii, and Sedum stenopetalum). No gen-
eral conclusions are possible from these observations except that the se 
species encompass a wide range of adaptations to a progressively de-
creasing water supply. It is clear, as discussed by Thoday (1933), 
that many meanings might be attached to such a term a s "xeric species," 
and that investigators must take care to define usage of such terms in 
as strict a sense as possible. 
Maximum Recorded Stresses 
Twelve Western Cascade species were found to have sap tensions 
above 68 atmospheres before seeds were entirely ripe. Most of these 
species were annuals, such as Polygonum douglasii, Polygonum minimum~ 
Polygonum cascadense, Gayophytum diffusum parviflorum , Linanthus hark-
nessii, Collomia linearis , Orthocarpus imbricatus, Mimulus breweri , 
and Galium bifolium. The last two species are spring ephemerals, and 
it is likely that their seeds would have matured rapidly even if de-
tached from the desiccating plants . Linanthus harknessii exhibits 
modified ephemeral tendencies even at these high sap tensions. The 
82 
highest measured tension, exceeding 71.5 atmospheres, was recorded in 
Polygonum cascadense. 
Sedum stenopetalum and Linum perenne lewisii, herbaceous peren-
nials, and Arctostaphylos nevadensis, a woody perennial, also exhibited 
tensions above 68 atmospheres. This is particularly noteworthy in 
Arctostaphylos, a typically high-montane species, which showed no die -
back of stems sustaining these extreme tensions. 
Tensions in an annual, Navarretia divaricata, and two woody 
perennials, Juniperus communis saxatilis and Pachystima myrs inites, 
measured greater than 50 atmospheres. The perennials again showed no 
ill effects from the moisture stress. 
A number of species (including only one annual) characteristic 
of diverse geographic regions were found to have sap tensions above 40 
atmospheres, They are Polygonum kelloggii, Ivesia gordonii, Lotus 
nevadensis douglasii, Gilia aggregata, Penstemon deustus , Haplopappus 
hallii, Chrysothamnus nauseosus albicaulis, and Artemisia ludoviciana 
latiloba. 
Laboratory Germination Experiments 
Germination experiments in sub-irrigated peat pots using a 
variety of volcanic Western Cascade soils , serpentine soil from the 
Illinois Valley, Josephine County, and greenhouse rotted leaf mold 
indicated no re s triction of germination or establishment among those 
species tested. The volcanic soi ls us ed were derived from all three 
major volcanic units in the Western Cascadeso The Little Butte Series 
wa s represented by soil from Jumpoff J oe, the Sardine Formation by 
83 
soil from the lower slopes of Iron Mountain , and Plio-Pleistocene vol -
canics by s oi ls from Rebe l Rock and Sand Mountain. Species used had 
prev iously been shown to germinate on wet s terilized filter paper in 
l i gh t without previous stratification. Replicate pots were autoclaved 
to destroy the original mic roflora. The autoclaved soils , howeve r , 
broke down physically and became sufficiently hydrophobic that capil-
lary water would not rise fiv e cm. No species germinated on such 
soils, but al l did readily on unautocla v ed counterparts. Difference 
in growth rate was the on l y notable outcome of this experiment . All 
species pers isted but grew slowl y on serpentine soil. The Western 
Cascade soils produced a range of growth rates, correlated with the 
texture and amount of organ ic matter in the particular soil. All spe -
cies were most robust on greenhou s e soil. These results indicate that 
such factors as secondary organic nutrients and water availability are 
more important to germination and growth in these species than the 
chemical compo s ition of t h e parent rock from which the soi ls were de -
rived. 
Germination and Establi shment in the Field 
Plots were sown with s eed of disjunct species at several local -
ities in the Western Cascades in an attempt to establish them in sites 
where they had not previou s ly been found. Experiments of this type, 
which are similar to those repor ted by Cav ers and Harper (1967a, 1967b), 
test the a cceptibility of t he var ious habitats to the propagules of 
the disjunct spec ies. With the addition of proper controls , informa -
tion can be gathered on t h e degree of past dispersal of the species a s 
84 
well as on competitive exclusion by existing vegetation. 
Methods 
Seeds of 22 disjunct or endemic Western Cascade specie~ were 
collected during the late summer and fall of 1966. They were cleaned , 
counted, and bagged in equal numbers depending on the availability of 
good seed. Field sites were chosen on Iron Mountain, Rebel Rock, and 
Sardine Butte. The first peak, because of its easy access and great 
number of disjunct species, supplied much of the seed used in the 
experiments. Since some seeds were resown at or near their sites of 
collection, Iron Mountain acted as a partial control in the establish-
ment studies. Rebel Rock also supports a large number of disjunct 
species, but the disjunct flora is of a different composition. Six of 
the species tested in these experiments do not occur naturally on 
Rebel Rock, and three of them have not been found on Iron Mountain. 
Sardine Butte, which lies farther to the west and has only a small 
outcrop area at the summit, supports five of the disjunct and endemic 
species including only two of those utilized in the experiments. 
Representative habitats were chosen on each of these peaks. 
Three were selected from Iron Mountain including a mesic meadow dom-
inated by Rubus parviflorus and Pteridium aquilinum, with frequent 
occurrences of Ribes binominatum, Luina stricta, Gilia aggregata, 
Artemisia ludoviciana latiloba, Lupinus latifolius, Vicia americana, 
Eriophyllum lanatum, and Cirsium centaurea. This site is on a moder-
ately steep, southwest-facing slope, partially shaded to the west by 
several large, free-standing douglas fir trees. The soil is a deep 
85 
light loam, rich in organic matter, which, judging from spring slump-
ing and observable creep of cleared areas, undergoes considerable 
movement downslope during those seasons when it is not covered by snow 
or dense herbaceous vegetation. The second habitat is a considerably 
less steep, southwest-facing slope of fine scoriaceous material which 
presumably creeps at a rate at least comparable with that of the meadow 
soil. There is little or no organic matter in the "soil" of this 
slope; the only cover consists of widely dispersed small annuals. The 
third and most severe habitat is a flat but rough area of scoriaceous 
rock which crops out at the brink of the south-facing precipice on 
Iron Mountain. Fine mineral soil collects only in small pockets of 
the scoria, and the site supports only mosses and mats of Selaginella 
wallacei, both of which desiccate during the dry portion of the sununer. 
Habitats on Rebel Rock include forest, dry meadow, and cliff-
brink sites. The most protected site is under a moderately dense 
canopy of mature Abies lasiocarpa and Pseudotsuga menziesii. The dom-
inant species growing in the damp, rich, loamy, black soil are Valeriana 
sitchensis, Galium oreganum, Viola orbiculata, Hieracium albiflorum, 
Aquilegia formosa, and Aster ledophyllus. A small patch of Mertensia 
bella occurs in a slightly more exposed area only a few meters away. 
The second habitat is a dry ridgetop meadow dominated by an unidenti-
fied species of Carex. Also abundant are Viola nuttallii bakeri, 
Calochortus lobbii, Lupinus sericeus, and Cirsium centaurea. The 
ridge at this point slopes slightly to the southwest, and the soil is 
a well-drained peaty loam with numerous included fragments of andesite. 
The soil is not as deep as that in the forest habitat or in the meadow 
86 
site on Iron Mountain; the numerous rocks preclude accurate measure-
ments of depth. Two parallel but slightly different sites were chosen 
at the top of the south-facing precipitous slope that runs the length 
of the Rebel Rock ridge. One site, in a pocket of outcropping scori-
aceous rock, contains an accumulation of reddish rocky loam and sup-
ports -a moderately dense growth of Arc tos taphylos nevadensis, Erio-· 
phyllum lanatum, Eriogonum umbellatum, Antennaria ~' and such 
annuals as Polygonum douglasii, Polygonum minimum, and Navarretia 
divaricata. The neighboring plot has considerably less soil develop-
ment and supports mosses, Selaginella wallacei, Calochortus lobbii, 
Eriophyllum lanatum, and Polygonum minimum. The soil-pocket plot was 
cleared following the method described below, and the plot having less 
soil was left undisturbed. The area is flat, and exposure to the sun 
in both plots is complete. 
Only one site was chosen on Sardine Butte because of the paucity 
of habitats available. This site is located at the summit of the east 
spur of the mountain in a flat area of blocky andesite. Light mineral 
soil has accumulated between the large blocks of parent material, giv-
ing the entire area a well-drained surface of uneven depth. Drought-
tolerant species such as Arctostaphylos nevadensis , Juniperus communis 
saxatilis, Comandra umbellata, Hieracium scouleri, and Pachystima 
myrsinites are mixed with a few "ephemeral" perennials, especially 
Selaginella wallacei and All ium ampiec tens. 
Sites were visited in early October, 1966, and pairs of square 
meter plot~ were chosen in each habitat. Plot pairs were as uniform 
in soil and vegetative cover as possible. One of the plots was cleared 
87 
of vegetation and the soil cleaned of most of the roots and rhizomes. 
Both plots were divided into 400 square cm subplots onto which the 
s eeds were scattered. They were then pres sed gently into the soil to 
minimize downslope movemen t o Such variables as inequalities in the 
vegetation within the undisturbed plots and edge effects in the cleared 
plo ts were ignored. 
All plots were visited at least twice during the growing season 
of 1967 - -in the early s pring and early fa ll . Ob s ervations were made 
on the number, size , and phenological stage of seedlings of each of the 
species a s well as the ra tes of recolonization of the cleared p l ots by 
native vegetation. 
Resu l t s 
The information gained from the seed plots is summarized in 
Tables IV , V, VI , and VII , Several interesting features of these 
tables are discussed below . 
In the Rebel Rock p lo t s only one species showed no germination 
at all--Chamaecyparis nootkatensis (Table V). It did not germinate 
either in the field or in s ub s equent laboratory tests, and the seed 
can probably be considered inviable. Four species germinated but 
failed to liv e in any of the plots throughout a single summer. These 
include Allium crenulatum , Lewisia triphylla, Sedum divergens , and 
Castilleja rupico l a. All 17 other species were still living in the 
establishment plots by late Augu st, 1967. Sixteen species established 
themselves in the cleared forest plot, far more than in any other hab-
itat, although it is al~ost certain that most of these will be 
Table IV. Numbers of Seeds Sown, Germinated, and Established in 
Cleared and Uncleared Plots in One Habitat on Sardine Butte. 
Outcrop 
Species A Cleared Uncleared 
1 2 3 1 2 3 
Chamaecyparis nootka-
tensis 10 0 0 0 0 0 0 
Allium crenulatum 25 0 0 0 5 0 0 
Polygonym douglasii 8 7 0 0 0 0 0 
Polygonum minimum 15 2 0 0 0 0 0 
Polygonum cascadense 25 1? 0 0 0 0 0 
Polygonum kelloggii 50 1 1 1 0 0 0 
Lewisia triphylla 50 0 0 0 0 0 0 Explanation: 
Arabis holboellii 
retrofracta 50 0 0 0 0 0 0 A: Number of seeds sown per plot 
Sedum ~ivergens 50 0 0 0 0 0 0 1: Number of germinated seeds 
Ivesia gordonii 13 1? 0 0 4 0 0 2: Number of seedlings surviving 
Trifolium productum 16 2 0 0 2 0 0 one year 
Linum perenne lewis ii 50 27 18 0 17 0 0 3: Number of plants setting some seed 
,,. 
Douglasia laeviga ta 13 0 0 0 0 0 0 ... Species which were abundant both 
Gil ia aggrega ta 25 0 0 0 11 7 0 in and around the sown plots 
Linanthus harknessii 50 26 9 9 11 4 4 
Navarretia divaricata 50 9 4 4 3 1 1 
Mertensia bella 25 0 0 0 0 0 0 
Penstemon deustus 30 0 0 0 0 0 0 
Castilleja rupicola 25 0 0 0 0 0 0 
Castilleja pruinosa 50 0 0 0 0 0 0 
Galium bifolium 13 2 2 2 4 4 4 
Luina stricta 10 0 0 0 0 0 0 00 
00 
Table V. Numbers of Seeds Sown, Germinated, and Established in Cleared and Uncleared Plots 
in Three Habitats on Rebel Rock. (See Explanation, Table IV . ) 
Forest Meadow Outcrop 
A Cleared Uncleared Cleared Uncleared Cleared 
Uncleared 
Species 
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 
Chamaecyparis nootka-
tensis 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Allium crenulatum 25 11 0 0 9 0 0 7 0 0 0 0 0 3 0 0 0 0 0 
Polygonum douglasii 8 2 2 1 0 0 0 0 0 0 0 0 0 14>'( 141: 14'>': l'>': 0 0 
Polygonum minimum 15 2 2 2 1 0 0 10 4 4 7 0 0 14>': 14>': 14>': 71>': 5()-k 50>': 
Polygonum cascadense 25 8 8 8 0 0 0 7 7 7 2 0 0 6 5 5 0 0 0 
Polygonum kelloggii 50 0 0 0 0 0 0 0 0 0 0 0 0 19 19 19 0 0 0 
Lewisia tr i phylla 50 8 0 0 0 0 0 0 0 0 0 0 0 8 0 0 0 0 0 
Arabis holboellii 
re trofrac ta 50 13 13 0 5 0 0 0 0 0 7 0 0 4 0 0 0 0 0 
Sedum divergens 50 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 0 
Ivesia gordonii 13 7 7 0 5 0 0 3 2 0 1 0 0 1 0 0 0 0 0 
Trifolium productum 16 6 6 0 7 7 0 0 0 0 0 0 0 0 0 0 0 0 0 
Linum perenne lewisii 50 32 32 0 12 9 0 10 3 0 31 0 0 13 0 0 0 0 0 
Douglasia laevigata 13 2? 2? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Gil ia aggrega ta 25 21 21 0 5 5 0 9 4 0 4 0 0 0 0 0 0 0 0 
Linanthus harknessii 50 30 29 29 11 10 10 13 8 8 4 0 0 40 26 26 0 0 0 
Navarretia divaricata 50 12 12 12 0 0 0 18 12 12 0 0 0 50>": 501: 50>": 7>': 7>'( 7'>': 
Mertensia bella 25 9>': 9>': 0 161: 16* 0 0 0 0 0 0 0 0 0 0 0 0 0 
Penstemon deustus 30 4? 4? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Castilleja rupicola 25 1 0 0 0 0 0 0 0 0 0 , 0 0 0 0 0 0 0 0 
Castilleja pruinosa 50 10 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Galium bifolium 13 6 6 6 4 4 4 4 4 4 4 4 4 10 10 10 0 0 0 
Lu ina s tric ta 10 5 5 0 l . 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
I 00 
'° 
Table VI. Numbers of Seeds Sown, Germinated, and Established in Cleared and Uncleared Plots 
in Three Habitats on Iron Mountain. (See Explanation, Table IV.) 
Meadow Scree Slope Outcrop 
Species A Cleared Uncleared Cleared Uncleared Cleared Uncleared 
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 
Chamaecyparis nootka-
tensis 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Allium crenulatum 25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Polygonum douglasii 8 9;': 5;': 51: 181: 18;': 181: 1 0 0 0 0 0 0 0 0 0 0 0 
Polygonum minimum 15 5 5 5 4 4 4 0 0 0 0 0 0 0 0 0 0 0 0 
Polygonum cascadense 25 2 0 0 0 0 0 U: 11: 1,•: 8;': 81: 81: 0 0 0 2 0 0 
Polygonum kelloggii so 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Lewisia triphylla so 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Arabis holboelli 
retrofracta so 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Sedum divergens so 0 0 0 0 0 0 0 0 0 S>'<' 51: 0 0 0 0 0 0 0 
Ivesia gordonii 13 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Trifolium productum 16 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 
Linum perenne lewis ii so 3 3 0 15 15 0 0 0 0 0 0 0 2 2 0 0 0 0 
Douglasia laeviga ta 13 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Cilia aggregata 25 20;': 20>'<' 0 301: 301: 0 1 0 0 0 0 0 0 0 0 0 0 0 
Linanthus harknessii so 6'·k 61: 61: 7>'<' 7* 71: 11: l;': 11: 121: 121: 121: 0 0 0 0 0 0 
Navarretia divaricata so 11 11 11 0 0 0 0 0 0 15>'<' 151: 15;': 0 0 0 0 0 0 
Mertensia bella 25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Penstemon deustus 30 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Castilleja rupicola 25 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Castilleja pruinosa so 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
Galium bifolium 13 14;': 8;': 81: 8 ;': 8·.'.·  8;': 0 0 0 0 0 0 0 0 0 0 0 0 
Lu ina s tric ta 10 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 
'0°  
t""" 0 0 0 '"CJ ::a:::z t""" 0 0 t""" ,-,3 H Cl) :i>t"""'"CJ'"CJ'"CJ'"CJ;i> 0 ,-,3 
.C..:. . ill ill ill 
(1) (1) ll> .......... 0 ..... ti < (1) ti (1) 0 0 I-' 
.I..-..'  en 
0 0 ::r 
en ;:l ti < ;:l I-' C: ;:l ..... (1) a. ill ti ll> ~ I-' I-' I-' I-' I-' rt ill 
;:l .r..t..  rt en rt ill ill 
CT 
1-'•CJ'Q C: Mi en C: (1) o" I-'•'<'<'<'< ..... (1) a t--' 
ll> Ca: 
..... rt (1) 
 ti ;:l ll> I-' a 0 ..... a rt I-'• en CJ'Q CJ'Q CJ'Q CJ'Q C: ;:l ill (/) (1) I-' I-' (1) ;:l ti rt ill I-' ill ti en ..... 0 0 0 0 a en (1) 'O 
(/l I-' I-' a en (1) ::r ll> (/l 'O ..... a. 0 ill ;:l ;:l ;:l ;:l ..... () (1) 
rtcJ"(1)(1)0 ..... rt C: (J'Q ..... (1) C: (J'Q ..... 1-11 ::r <: C: C: C: C: () en '< () H 
(/) ti ..... L.,. L.,. ;:l ll> ..... en CJ'Q ll> ti a 0 < ti 0 rt a a a a ti 'O ..... 
'O ..... 1-11 ill H ll> ill ti (1) ::r ti (1) ill I-' ti a (1) a. ill (1) (1) () 0 CT (1) I-' ;:l 'O a. ti () CT I-'• ;,:;' n a. ;:l ti en 
.(.).. . rt I-' 'O ti (1) (1) a. ll> (J'Q ill ;:l ti 0 (J'Q rt O 'O (1) ll> ..... 0 C: ..... ll> ..... ti C: C: I-' ..... ti ill (1) (1) 0 ;:l (1) ll> (1) ::r I-' (/l ;:l C: I-' en ..... 
(1) 
aC: 
z 
 C: 'O en I-' < ;,:;' rt < a. ..... ;:l I-''< I-' () 1-'•CJ'Q ll> en .......... rt a ;:l C: ll> ll> ;:l ill ..... I-' C: ..... (/l I-' I-' 0 ill I-' rt ;:l 
;:l () C: ti (1) (J'Q (1) a () ..... I-' (J'Q 
0 0 0 en ..... 
a. C: 
en ill C: 0 <: (1) ll> 
en ~.... . rt ..... ll> (J.'.Q.. . (1) a .(./.l..  a 0 en ll> ti () I-' () 
() ill ill ll> ..... rt C: ;:l rt .... . ll> (/l a ..... en ..... ti ;,:;' ..... () 
C: rt ..... ........ (1) ll> 0 ill ti ill .. I C: I-' 
en 
;:l (/) 
ill I-' t--' Number of cleared and uncleared plots . 0 C: 
rt ~\Ot--'t--' t--'NOOt--'00 1-'0\Jl-..JN~ NN\0-.0-..J\JIO 
C: •-v ( n=l4) in which germination occurred ~- § 
ti () ll> 
ill Ca:  ti I-' I-' -< Number of cleared and uncleared en 
'< rt 0 
..... plots (n=l4) in which species ll> 1-11 t--' ;:l 
;:l t--' '° t--' Ot--'NOOOa- I-' -..J N N t--' t--' ON\Jla-~oo established for one year () (/) •-v (1) 'O 
g. , .en  (1) () 
(1) ..... 
< (1) .....  en Number of cleared plots (n=7) 
.(.).. . a, a, 0 \J1 N 0-, 0-, 0-, · 0-, N t--' W t--' \J1 W \J1 \Jl\Jlt--'NN~-..J in which no evidence of (1) 
.;.:.l. . germination was ti seen 
rt .a....  
'< ;:l 
Number ll> of cleared plots (n=7) .r..t..  
in which species germinated ;:l 
t--'O O ,t--' 00001-'0t--'t--'Nt--'t--' N O N t--' N W 0 (J'Q 
but died within one year 
0 
ti 
tr! 
Number of uncleared plots (n=7) in en 
which no establishment occurred , rt ill 
t--'t--'1-'000WNI-' 1-'WON 0 I-' ONWNNOO 
•-v while es ta bl ishmen t occurred in the CT t--' 
corresponding cleared plot ..... en 
.:.:..r.  
Number of uncleared plots (n=7) in ;:l 
(J'Q 
0 I-' I-' N which species established better than Ot--' 00 ,:- ,:- ,:- NOt--' t--' 0 t,:---'O 0 0 I,-:-' t,-:--' t--' 00 't°--'  in the corresponding cleared plot 
92 
eliminated by competition as the forest vegetation reinvades the plot. 
Three species ( Ivesia gordonii, Trifolium productum, and Luina stricta), 
all herbaceou s perenn ial , were establish ed on Rebel Ro ck ev en though 
they are not native to the region. 
The Iron Mountain p lots, near where mot of the seed originated, 
showed remarkably litt le germination and establi hment ( Table VI) . 
Two of the chosen habitats are s evere, being expo s ed to alternate 
freezing and thawing in winter and having oil s urface avai l able for 
germinating s eeds fo r only a short time in the spring. The meadow 
plots were sufficient ly steep that many of the seeds may have been 
washed completely out of the cleared plot and were never found in the 
surrounding dense vegetation, It is l ikely that competition for light 
and water from previously established perennial s pecies limited th e 
establishment of s eedlings in t h e undistu r bed meadow p lot in the early 
stages of germination. If s eeds germinated and died soon thereafter , 
there would be little likelihood of witnes ing the phenomenon in the 
field. The failure to obs erve germination , especially in the cleared 
meadow plot on Iron Mountain, indicates that higher rates of germina -
tion and establishment should be expected given more ideal conditions. 
The numbers reported here thus represent minima of species actually 
able to establish within a given region . These considerations may 
explain many of the zeros in Tables IV 9 V, and VI. 
The plots on Sardine Butte a r e especially interesting since 
they show estab lishment of six disj unct s pecies in an area where none 
existed prev iously . Of these s pecies the f our annual s (Polygonum 
kelloggii, Linan thus harkne ssii, Navarretia d ivarica ta, and Gali um 
bifolium) all set some seed which appeared viable. It is not yet 
93 
known whether they will su cceed in estab l ishing again in later years. 
Conclusions 
The tentative conclusions based on this information and pre-
sented below are equivocal, given the small number of habitats actually 
tested, the small number of seeds frequently us ed, and the short dura-
tion of the experiment. On the basis of my field experience with the 
establishment plots, I believe that few, if any , of the established 
species will remain for any length of time in their tenuously held new 
habitats . Nevertheless, the present information does point to impor-
tant conclus i ons regarding the establishment potentialities of many of 
t he disjunct species. 
If species prev iously found in or around the germination plots 
are disregarded , the number of established species in uncleared plots 
is only 40 percent of the number established in cleared sites. However , 
as indicated in Table VI, some species, especial l y those which are 
often found as "understory" vegetation in mesic meadows, were occa-
sionally better able to establish in uncleared than in cleared plots. 
If species germinated and then died (or failed to germinate 
completely) in a cleared habitat , it must be considered that their 
elimination was due to some combination of physical properties inher-
ent in the habitat or to predation. Cor respondingly, if establishment 
occurred in a cleared plot with none in the adjacent undisturbed plot, 
competition with existing vegetation must be invoked to explain lack 
of establi shment in the uncleared s ite , Table VII shows both phenomena 
to be important in the situations tested, with competition being the 
controlling fa c tor much less frequently than environmental unsuitability. 
94 
Species which germinated in more than half of the cleared plots 
showed the highest rates of establishment in all plots, and have been 
eliminated from some uncleared plots. This evidence suggests that com-
petition is the most important factor in determining the distributions 
of these species, which include Polygonum douglasii, Polygonum min i mum, 
Polygonum cascadense, Ivesia gordonii, Linum perenne lewisii, Navarret ia 
divaricata, and Galium bifolium. All but three of these are widespread 
in the Western Cascades. Polygonum cascadense is endemic to the region, 
Linum perenne lewisii is relatively restricted within the study area , 
and Ivesia gordonii is known from only one locality within the Oregon 
Cascade Ranges. It seems likely that these three species have much 
wider potential ranges than they pr esently inhabit. Historica l disper -
sal factors, together with elimination from many sites by c omp e tition , 
may explain their presently restricted di s tribution in the Western 
Cascades. 
DISCUSSION 
Plant Groups of Similar Geographical Affinity 
It is possible to a ss ign disjunct species to groups or elements 
according to the geographical region or biotic province of their major 
popu l ations. Such e l ements may indicate the migrational path by which 
the species reached the Western Ca s cades and are theoretically similar 
to the groups of species showi ng "equiformal progressive areas" di s-
r 
cussed by Hulten (1937). Equiformal progressive areas are more close ly 
,, 
tied to glacial history in the boreal regions treated by Hulten than in 
the Western Cascades. The present groupings are notable in their diver-
sity. Only rarely do the total ranges of even two disjunct species ap -
proach identity. In addition, diverse ecological patterns are shown by 
species within a single geographical group. Nevertheless , it i s likely 
that geographical affinity reflects similarity in migrational history, 
and qualified use of such groups as Hulttn proposes is both warranted 
and helpful. 
Southern and Eastern Element 
This large group of disjunct spec ies has major centers of dis-
tribution both in the Columbia Basin and the High Lava Plains of eastern 
Washington and Oregon (Freeman and others , 1945) , and in the southern 
Cascade, Siskiyou-Klamath-North Coast Range, and Sierra Nevada regions 
of southern Oregon and northern California. Ranges of the 15 species 
95 
96 
considered to belong in this group overlap with those of the following 
two groups. In general, species whose distributions indicate that they 
have occupied both southern and eastern regions for considerable time 
are included here. 
There is evidence that eight species have reached the Western 
Cascades by migration both across the Cascade crest from the east and 
northward through the Western Cascades from the Siskiyou-southern Cas-
cades region. These includ e Pinus ponderosa, Arabis platysperma 
howellii, Gayophytum humile, Linanthus harknessii, Navarretia divari-
cata, Mimulus breweri, Castilleja pruinosa, and Lonicera conjugialis. 
Their distributions are relatively continuous in the Western Cascades 
south of the study area, and some specimens are known either from the 
High Cascade peaks or from the passes between them. Castilleja pruinosa 
is not common east of the Cascade crest, but Western Cascade forms have 
been implicated in the C. peckiana complex of that region, and so are 
included here. 
This group contains ephemeral annuals restricted to habitats 
where abundant snowmelt is available in early spring, annuals of drier 
situations, an herbaceous perennial of dry habitats, and a large shrub . 
One species, Navarretia divaricata, grows in disturbed sites and is some-
what weedy in behavior, All the species in this group except Arabis 
piatisp~rma, Castilleja pruinosa, and Lonicera conjugialis are rela-
tively widespread in the drier portions of the Pacific States, 
Although Nothocalais alpestris extends almost to the Pacific 
Ocean in the Siskiyou region, only two collections are known from the 
Western Cascades. These are both in the vicinity of Olallie Mountain 
97 
and are evidently westward extensions of the large populations found 
at higher altitudes in the Three Sisters. Its entry is thus entirely 
from the east, although it is not found in the lower elevation areas 
of eastern Washington and Oregon. 
Another group of six species, while most common east of the 
Cascade-Sierra axis, shows no evidence of having reached the Western 
Cascades from the east and has probably migrated northward through the 
Western Cascades. These species , Selaginella scopulorum, Linum perenne 
lewisii, Monardella odoratissima , Penstemon deustus, Galium bifolium 9 
and Chrysothamnus nauseosus albicaulis, are fully as diverse with regard 
to habit, breeding system, moisture relations, and habitats occupied 
as the first group discussed. All of these species except the first 
are widespread in western North America. ~- scopulorum is primarily 
confined to the Klamath Mountains and the Wallowa Mountains, but has 
also been reported from several sites in Washington and Montana. To 
date, Western Cascade populations comprise half of the species popula-
tions known to me. Migrational patterns are thus difficult to interpret. 
Species of this element are known from seven different vegetation 
units but are most abundant in dry rocky areas (Gravel Scree and Outcrop 
Ridge), Xeric Meadow, and Rocky Melt Seep associations • . 
Southern Element 
Twelve species have major populations only to the south of the 
study area and appear to have reached the area by northward migration 
along one or more of several possible routes. Five of these, Cheilanthes 
siliguosa, Allium crenulatum, Silene campanulata glandulosa, Cardamine 
98 
integrifolia sinuata, and Mimulus pulsiferae , also have stations in 
the interior valleys west of the Cascades. In addition, three of them 
occur near the eastern end of the Columbia Gorge. These species prob-
ably reached the Western Cascades by migration up the tributaries of 
the Umpqua and Willamette Rivers. These species are also ecologically 
diverse. The only widespread species jn this group is the fern 
Cheilanthes siliguosa , whose center of distribution nevertheless seems 
to be serpentine areas in southwestern Oregon. 
The other seven species of t h is element have migrated northward 
through the mountains and are not found west of the Cascades except in 
their parental area. Trifolium productum and Crepis occidentalis are 
relatively widespread and have major population centers in the northern 
Sierra Nevada as well as in the Siskiyou region. Trifolium howellii, 
Arenaria pumicola, and Ribes erythrocarpum have restricted ranges. The 
first is found in southern Oregon and northernmost California wes t of 
the Cascade crest , and the last two have been cons idered heretofore to 
be narrowly restricted to the region around Crater Lake. Ribes binom-
inatum and Erigeron foliosus confinis have remarkably similar ranges in 
the southern Cascades, Siskiyou Mountains, and the North Coast Range of 
California. 
Mimulus pulsiferae is the only annual in this group. The herba-
ceous and woody perennial s , however , show a wide range of ecological 
tolerances, pollinators, and breeding systems. Members of the southern 
element inhabit 12 of the described vegetation units--more than any 
other element--but the dry rocky areas are the only units of importance. 
Seven of the units, most of which are occupied by a single southern 
99 
species, comprise mesic environments. 
Eastern Element 
The largest group of disjunct species is centered in the inter-
mountain region of the Western United States. All of these species 
have evidently entered the Western Cascades across the Cascade crest. 
Ten species (Polygonum kelloggii, Lewisia triphylla, Arabis 
holboellii retrofracta, Horkelia fusca, Lathyrus lanszwertii aridus , 
Gayophytum diffusum parviflorum , Gilia aggregata, Gentiana calycosa 
n. subsp., Pterospora andromedea , and Cryptantha affinis) are presently 
established at a few points along the crest, supplying evidence of the 
proposed migration route . 
A second group consisting of Helianthus cusickii, Artemisia 
tridentata, and Arnica parryi is restricted to the study area (and a 
few similar spots in the Western Cascades and eastern Siskiyous ) west 
of the Cascade cres t but shows no further evidence of migration route. 
Two of the species of this group, Helianthus cusickii and 
Artemisia tridentata, are hypothesized to be of very recent origin in 
the Western Cascades. Both are known from only one locality, and the 
present populations are not in equilibrium with the surrounding vege-
tation but are associated with diverse habitats. It is not yet certain 
whether they will be able to persist. 
Populus tremuloides, Collomia linearis, Artemisia ludoviciana 
latiloba, Microseris nutans, and Crepis acuminata typica have few local-
ities in southern Oregon or northern California west of the Cascade 
crest, and I believe these populations to be equally recent in origin 
100 
to populations in the study area. All western Oregon populations are 
hypothesized to have come west across the Cascade crest. Collomia 
linearis has two present localities in the Willamette Valley. It is 
possible that it came west through the Columbia Gorge and reached the 
Western Cascades through the tributary valleys from the west. 
Of the Eastern Element, all the annuals and biennials except 
Collomia linearis are found along the Cascade crest. This may indi-
cate that the annuals and biennials, which have fewer morphological 
adaptations for long-distance dispersal, typically migrate in shorter 
steps than do the wind-dispersed perennials. This group is also ex-
tremely diverse ecologically, including ephemeral annuals and desert 
shrubs. Lewisia triphylla grows around snowbeds and in running snow-
melt, and Crepis acuminata typica is restricted to dry south-facing 
cliffs. The annuals grow in generally moister environments than the 
larger perennials. These species occur in 11 associations, the most 
important of which are dry rocky areas and Xeric Meadow and Xeric 
Forest associations. 
Northern Element 
Another well-represented element in the disjunct flora occurs 
to the north and has reached the study area by migration south through 
the Cascade Ranges. 
One subgroup of the northern element is known from the coastal 
mountains of northwestern North America, from the Cascades, and from 
the northern Rocky Mountains. This group includes Polystichum ander-
sonii, Arenaria capillaris americana, Rhododendron albiflorum, 
101 
Menziesia ferruginea, Hydrophyllum fendleri albifrons, and Lonicera 
utahensis. 
The ranges of several of these species are especially noteworthy. 
Polystichum andersonii is a rarely collected boreal species that is 
presently known from only three populations in Oregon. Two of these 
are in the study area. Arenaria capillaris americana reaches the south-
ernmost points in its range within the study area where it overlaps and 
merges with~- pumicola , a member of the southern element. Lonicera 
utahensis has a uniquely disjunct distribution. It is found commonly 
in northeastern Oregon, Idaho, and northern Washington but is repre-
sented in the Cascades only from the study area south to the region of 
Mt. Shasta. The species has evidently been poorly collected and there 
is little evidence concerning its entry into the southern half of the 
Cascade Ranges. 
A second group is known from the northern coast mountains and 
the Cascades. These species are Chamaecyparis nootkatensis, Sedum 
divergens, and Douglasia laevigata. The first two species occur as 
far south as the California-Oregon border, but Douglasia reaches its 
southern limit at the latitude of the study area. The distribution of 
Chamaecyparis nootkatensis is noteworthy in that, like Menziesia 
ferruginea, it is found in the High Cascades as far south as Mt. 
Jefferson, at which latitude it becomes exclusively a Western Cascade 
species. 
Three species of the northern element, Castilleja rupicola, 
Haplopappus hallii, and Luina stricta, are restricted to the Cascade 
Ranges . Haplopappus hallii is probably a spurious member of this 
102 
element, sinc e it is a member of a basically southern genus. It was 
described from the dry eastern end of the Columbia Gorge, but at present 
most of its known populations occur in the Western Cascades, extending 
as far south as Hershberger Mountain. Its parental area is unknown. 
Mertensia bella and Orogenia fusiformis show remarkably similar 
and striking distribution patterns. Both s pecies were previously known 
primarily from the Siskiyou region, but had also been recorded from 
localities in northeastern Idaho and adjacent Montana. These distri-
bution ranges , together with other evidence , indicate that the species 
are actually boreal in orig in and tha t the disjunct populations fol l owed 
separate mountainous routes during a southward migration. Orogenia 
fusiformis is slightly more widespread than Mertensia bella, but the 
present work has more than doubled the known localities for both species. 
This element is much less diverse ecologically than those pre-
viously considered . Annuals are not represented at all, and more than 
half of the species grow only in the moistest terrestrial habitats in 
the Western Cascades, The exceptions are noteworthy. Chamaecyparis 
nootkatensis typically grows on barren south-facing ridges where other 
tree species cannot establi sh. It has not been found to exhibit sap 
tensions over 19 atmospheres , however, in spite of the general dryness 
of its habitat. This i s presumably due to its deeply penetrating root 
sys tern. Thi s species occurs less fr equently under the mesic forest 
canopy where it reproduces only vegetatively. Trees rarely attain con-
s iderable size, 
Sedum divergens , Arenaria capillaris americana, and Haplopappus 
hallii occur on dry, rocky , south-facing slopes , Haplopappus has 
103 
already been noted as a possible spurious member of this element. Sedum 
divergens has an excellent water-storage organ in its almost globose 
succulent leaves. Members of this genus have been shown to exhibit 
little moisture stress until late in the season. Arenaria capillaris 
americana is often found along dry ridgetops near late-persisting snow-
banks . It blooms early in the season and has usually set seed by the 
time water becomes limiting in its environment. 
Doug l asia laevigata and Castilleja rupicola are confined to 
cliffs which are exposed to the prevailing winds (which are fog- and 
rain-laden throughout most of the year) and partially shaded from direct 
insolation. 
Thus all disjunct species of the northern element are either di-
rectly dependent on an abundant and continuing water s upply or have 
reached adaptive compromises with their seasona l ly dry environments. 
Northern species grow in 11 vegetation units , having particular abun-
dance in the Outcrop Ridge, Wet Meadow 9 and Mesic Forest associations. 
Alpine Element 
Eight species are characteri s tic members of the Arctic-Alpine 
Life Zone (Merriam, 1889), which does not occur in the Western Cascades 
in its typical form. Most of these species (Polygonum newberryi, 
Spraguea umbellata, Arenaria rubella , Luetkea pectinata, Polemonium 
pulcherrimum pu l cherrimum, and Erigeron compositus) are found on the 
uppermost slopes of the High Ca scade peaks 9 especially the Three Sis-
ters . They have presumably reached the study area from these high 
mountains immediately to the east , Arenaria rubella, Polemonium 
104 
pulcherrimum pulcherrimum , and Erigeron compositus have also developed 
lower elevation forms, bu t since only the alpine forms occur in the 
Western Cascades , they ar e included in this element. 
Two species , Ivesia gordonii and Linanthastrum nuttallii, are 
not otherwise known from the Cascade Ranges at this latitude, and their 
modes of entry into the study area are unknown. Major populations of 
both species occur to the north, ea s t , and south. Ivesia is disjunct 
by more than 300 km, the closest populations occurring in the Strawberry 
Mountains of east-central Oregon and on Abert Rim in the south-central 
portion of the state. Northern Cascade populations also supply a pos-
sible point from which immigrants could have reached the Western Cas-
cades. 
With the exception of Spraguea umbellata, all members of the 
Alpine element are deep-rooted herbaceous perennials (Bliss, 1956) 
which live either in crevices in the Outcrop Ridge or Vertical Outcrop 
associations or in fine scree such as that a ssoc iated with morainal 
features at higher elevations (Gravel Scree association). Spraguea 
is a short-lived perennial, and is found (like Luetkea pectinata and 
Polygonum newberryi) in barren areas of gravel scree or fine rock frag-
ments where snow lies deep in winter and persists late into summer 
(Snowbed association). The occurrence of these high alpine species at 
elevations as low as 1200 mis str i k i ng . 
Valley Element 
More than 30 species that are charac teristic of the lower val-
leys west of the Cascades have been found in the study area. All of 
105 
these are found at lower elevations, but only five reach an elevation 
such that they can be found with disjunct species of other elements 
(about 1100 m). These include Populus trichocarpa, Quercus garryana, 
Polygonum spergulariaefome, Convolvulus nyctagineus , and Plagiobothrys 
scouleri . All of these species are corrnnon on the floor of the Will-
amet t e, Umpqua, and Rogue River Valleys in Oregon 1 and several extend 
into the Puget Sound region. They are also all found at the dry east-
ern end of the Columbia Gorge and are comparable with species of the 
Rogue - Columbia element of Detling (1953). Only Quercus garryana has 
been reported from the Deschutes area of eastern Oregon (Ornduff and 
French, 1958). 
These species range from late-blooming annuals to trees, but 
are frequently found together, both in their major populations and in 
the Western Cascades. They are known from eight associations, the 
primary of which are the Lowland Xeric Meadow and Outcrop Ridge a s so-
ciations. 
Widespread Element 
Seven species occur in all major geographical-physiographical 
regions of Oregon in addition to being more or less widespread in the 
Western or northwestern United States. These species are all restricted 
to dry habitats and are known from localities in the ·Willamette Valley, 
Columbia Gorge , and eastern and southern Oregon. They are less disjunct 
in the Western Cascades than many other species treated in detail here, 
but nevertheless comprise an interesting element of the flora. Most of 
these species (Polygonum douglasii, Polygonum minimum, Silene douglasii, 
106 
Lotus nevadensis douglasii, and Lupinus arbustus neolaxiflorus) are 
also known from the High Cascades at this latitude, perhaps indicating 
an eastern entry into the Western Cascades. Sedum stenopetalum probably 
migrated northward into the area, while Phacelia linearis made its way 
up the tributary valleys of the west slope of the range. Members of 
this element appear to be basically eastern in origin and show three 
possible modes of entry into western Oregon: 1) through the Columbia 
Gorge; 2) across the crest of the Cascades in central Oregon; and 3) 
acro s s the low passes of the southern Cascades and north from the 
valley of the Rogue River, either through the Western Cascades or 
through the interior valleys. These species are all annuals or herba-
ceous perennials and occupy similar habitats ( especially Outcrop Ridge 
and Xeric Meadow) in the Western Cascades, although it is likely that 
they have never acted as a single floristic unit. 
Endemic Element 
Only three species are endemic to the central and central West-
ern Cascades: Polygonum cascadense, Aster gomanii, and Erigeron cas-
cadensis. This is a surprisingly small number of endemics relative to 
the floristic complexity of the region. 
Polygon~m cascadense is a small annual derived from the P. 
spergulariaefome complex which grows in the Xeric Meadow and Gravel 
Scree associations. It had the highest recorded sap tension (71.5 
atmos pheres) of any actively growing Western Cascade species. The two 
compos ites are deep-rooted herbaceous perennials occurring in crevices 
(Outcrop Ridge and Vertical Outcrop), Blocky Talus, or Gravel Scree. 
107 
Polygonum ca scadense and Erigeron cascadensis have strikingly similar 
distributions, but Aster gormanii is currently known from only three 
localities, of which two are in the vicinity of Mt. Jefferson in the 
High Cascade Range. If it were not so closely limited to this region, 
it would be considered a member of the Alpine Element. 
Summary 
The geographical elements described are diverse and often over-
lapping. Some correlations exist between e l ements and the habitats 
occupied by member species. The majority of species from the three 
southern and eas tern elements , which correspond in large part with 
Detling's "xer ic s pecies ," occur in dry ro cky habitats. In addition, 
many species of the Eastern , and Southern and Eastern Elements, but 
not of the Southern Element , occur in the Xeric Meadow association. 
All elements conta in some species wh ich grow in the dry rocky areas, 
but only the Northern Element contains a ma j ority of species which 
occur in mes ic or wet habitats. Alpine species are rather closely re -
stricted to Gravel Scree and Vertical Outcrop associations, and all 
endemic species sometimes occur in the Gravel Scree association. 
All large elements conta i n species showing a diversity of mois-
ture regimes. Ephemeral annual s, however , occur only in the three 
s outhern and eastern elements , while other. life forms are more or less 
scattered t h rough all groups. The No r t h ern Element is unique among 
those measu r ed in showing no sap tens ions greater than 19 atmospheres. 
This is partially correlated with the moistness of the habitats occu-
pied by northern pecies 9 but is a lso t r ue where northern species 
108 
inhabit dry environments. 
Breed i ng systems , pollina tion mechanisms, dispersal adaptations , 
and ease of estab l ishment in new env ironments cannot be correlated with 
the geographical elements. Each element shows nearly the complete range 
of variation for these phenomena fo und in the entire group of disjunct 
and endemic species. 
Th e ev idence indica t es that none of the elements, with the pos-
s ible exception of the Valley Ele.men t 9 cons titutes a discrete f l oristic 
unit. Th e like l ihood that the e l emen s hav e migrated a s a ssocia ti.on s 
is extremely small, but the indiv idual pattern s of migration f o llowed 
by member species are generally similar , 
Speciation in Western Cascade Disj unc t Populations 
Theoretical considerations and evidence compiled by numerous 
workers predict rapid rates of evolutionary divergence in such eco log-
ically diverse , s ea sonally arid environments a s the Western Cascades . 
Cain (1944) proposes that a compensation of critical or limiting factors 
iE of major importance in the establishment of extra -limi tal distribu -
tion patterns. Mason (1945) pre fe rs to rely on the ability of species 
to undergo ecotypic variation and readily adapt to new env ironments. 
These proposals are closely related, and bo th call attention to the 
fact that se l ective regimes will di ffer between continuous and disjunct 
popula tions. Wils on ( 1959) and Car l qui s t 1966a) strikingly demonstrate 
this phenomenon on oceanic i slands 9 much mor e isolated situations than 
any terrestrial mountaintop "is l and." Kruckeberg 1967) found numerous 
examp l es of ecotypic di fferentiation in dry regions of serpentine soils. 
109 
Lewis (1962) and Warburg (1965) show in both plants and animals that 
drought, especially if extreme (as was the case in the Western Cascades 
in the summer of 1967), can be a potent selective factor. In the case 
of Clarkia a single exceptionally dry season brought about a major 
shift in the physiology and morphology of an annual population (Lewis, 
1962). Stebbins (1952) demonstrates that aridity also acts as a stim-
ulus for evolution in the opposite direction, i.e. into more mesic hab-
itats. In addition, Stebbins and Major ( 1965) clearly show that "eco-
tonal" regions, in which forms characteristic of two or more provinces 
overlap, are centers of evolutionary divergence and novelty. The number 
and diversity of disjunct elements show that the Western Cascades com-
prise such an "ecotonal" region, and that unique forms should be impor-
tant in the flora. 
In spite of the weight of such considerations , little evidence 
of divergence has been found in the present study. Relative to the 
diversity, both ecological and floristic, of the Western Cascades , 
there are few endemic species. All three of these species are here 
considered to be relatively recently derived, and never to have had a 
greatly wider distribution than at present (neo-endemics of Stebbins 
and Major, 1965). Thus divergence to the level of species within the 
Western Cascades amounts to considerably less than one percent of the 
total number of species known from the area. 
An equally small proportion of subspecific divergence among the 
disjuncts has been noted. No subspecific taxa have yet been described 
from the disjuncts, but some taxonomic recognition might be useful for 
the Western Cascade forms of Selaginella scopulorum, Allium crenulatum, 
110 
Ivesia gordonii, Gentiana calycosa, and Haplopappus hallii. Lonicera 
utahensis and Cheilanthes siliquosa show questionably significant dif-
ferences from parental populations, the former morphological and the 
latter edaphic. Complex, morphologically variable taxa which contain 
unique Western Cascade forms that should not now be recognized taxonomi-
cally are Arenaria capillaris-Arenaria pumicola, Monardella odoratissima, 
Castilleja pruinosa (peckiana), Castilleja rupicola , and Crepis occi-
dentalis. 
There are several possible explanations of this lack of observed 
divergence. The simplest is that the disjunctions are of such recent 
origin that they do not yet show morphological divergence. Evidence 
against this hypothesis is found in the ecological stability of almost 
all of the disjuncts and the typically large number of sites occupied, 
as well as in the rapidity with which divergence has occasionally been 
shown to occur. Perhaps more pertinent is the fact that the multiple -
allelic nature of physiological race formation need not involve large 
morphological differences between sibling races (Clausen, Keck, and 
Hiesey, 1947). In this connection Mason (1946) has noted that classi-
cal taxonomic techniques are generally not sufficiently discriminating 
to isolate small local differences in morphology or physiology. The 
problem here is not nearly of the magnitude of that shown by the many 
amphi-tropical disjunctions, a large number of which show identical 
variation patterns in both northern and southern hemispheres (Raven, 
1963). It does, however, seem likely that further studies with herbar-
ium material and use of transplant and cytological techniques will i s o-
late more differentiation of Western Cascade forms than is presently 
known. 
111 
Geological History of the Area and Its Effects 
on Plant Distributions 
S true ture 
The history of the relationships between the Western and High 
Cascade Ranges is poorly known but may have had profound effects on 
plant migration between the two ranges. Wells and Peck (1961) and Peck 
and others (1964) imply that the sharp physiographic line separating 
the ranges, a valley which varies from 300 to 1100 min depth along the 
eastern border of the study area , has been cut completely by erosion 
since the middle or late Pliocene. This hypothesis seems inadequate 
for several reasons. It seems quantitatively unlikely that at least 
1200 m of Plio-Pleistocene flows were deposited and then completely 
eroded by stream action. In addition, the authors offer no explanation 
for the marked differences in slope steepness on the eastern and western 
sides of this physiographic boundary nor for the typical fault-associ-
ated features of the area, especially the numerous hot springs which 
more or less follow the physiographic boundary. Peck and his associates 
have left unanswered the question of why there should be such a pro-
nounced physiographic boundary if there is no underlying structural one. 
If the flat surfaces of Hogg Rock and Hayrick Butte on the crest 
of the Cascades can be presumed to be remnants of Pliocene flow sur-
faces (which is unlikely), a maximum of 300 m of glacial stripping from 
the various ice sheets occupying t he region during the glacial maxima 
can be postulated. Present river valleys are incised into the continu-
ously gentle western slope of the High Cascades a maximum of 100 m. 
112 
These considerations also indicate that erosion will not account for 
the present boundary. This is seen to be true whether or not the upper-
most flows in the Western Cascades are properly mapped as Plio-Pleisto-
cene, 
It is known that the Western Cascades have been uplifted 
(Williams, 1953), and in addition that small-scale faulting is not un-
common (G. T. Benson, personal communication, 1968). It thus seems to 
me that a fault or series of intermittent faults trending north-south 
along the eastern margin of the Western Cascades best explains the 
existing physiographic boundary, Such faulting would account for the 
early entrenchment of the headwaters of the major river systems along 
this line, for the steepness of the eastern edge of the Western Cas-
cades, and for the broken line of hot springs; but would not preclude 
the formation of bridges between th e two ranges such as are presently 
found at Outerson Mountain and Olallie Mountain. 
A continuous bridge between the two rang es until the Plei stocene 
such as that implied by Wells and Peck (1961) would have allowed for 
easy migration from the high peaks to refugia in the Western Cascades 
with the first advance of the glacial ice, but is not consistent with 
other physiographic evidence, It is most likely that there has been a 
distinct boundary between the Western and High Cascades since the origin 
of the latter. 
Physiography 
The physiography of the Western Cascades affects plant distri-
bution patterns in various ways. Observations have shown that within 
113 
the area the isolation of the various peaks and the number of disjunct 
species they support are inversely correlated. This is considered a 
function of non-forest habitat size per unit area and resulting ease 
of dispersal from a given peak to and from its neighbors. 
The "island" terminology employed by Detling ( 1953, 1968) and 
other workers is not strictly accurate. Patches of non-forest habitat 
are seldom separated by more than 1 km, and recurrent fires have kept 
the forest and non-forest habitats in shifting equilibrium, perhaps re-
sulting in changing patterns of bridges between the islands. Thus long-
distance dispersal as discussed by Baker (1955) and Carlquist (1966a) 
need not be invoked within the area nor, for that matter, between the 
area and surrounding regions in former times. 
Cruden ( 1966) shows "mountain-hopping" dispersal to be an attrac-
tive hypothesis even for explaining amphi-tropical distributions. An 
example of how such a mechanism might operate in the Western Cascades is 
given below. Families of blue grouse , Dendragapus obscurus, are fre-
quently observed in the Western Cascades, and are particularly abundant 
in forest-meadow ecotones where large numbers of disjunct species are 
concentrated. These gallinaceous birds are territorial during the 
breeding season. They feed primarily on fruits and seeds. Yearly dis-
persal of young birds to neighboring peaks, as well as the mobile be-
havior involved in establishing and maintaining territories, l ikely re-
sults in the transfer of numerous seeds from the meadows of almost all 
of the peaks to their neighbors. Though this mechanism may be of some 
importance, it is an isolated example from many possib ilities, and it is 
not intended that it should be overstressed in its particulars. My major 
114 
point is that di spersal of many plant species among the peaks of the 
Western Cascades must be considered easy, natural, and a function of 
the distance between them. 
Glacial history has had a controlling influence on both the 
physiography of the area and the distributions of several of the dis-
junct species . Available ev idence indicates that there was no ice 
sheet, but rather a large number of relatively small alpine valley 
glaciers scattered through the Western Cascades. 
The occurrence of eight basically high alpine disjunct species 
in regions where signs of glacial activity are concentrated suggests a 
comparison of these peaks with nunataks ( Fernald, 1925; Gjaerevoll, 
1963 ) . The classic nunatak, as discussed by Gjaerevoll (1963), is a 
barren peak emerging from an essentially continuous sheet of ice. Al-
though there has been much argument concerning the possibility of p lants 
surviving the rigors of such environments, observations on the Greenland 
ice sheet show that many nunataks are characterized by south-facing 
slopes of loose gravelly material that support a surprising divers ity 
of plant types. This description is striking in its resemblance to 
many of the highly glaciated Class 1 peaks of the Western Cascades. I n 
addition, Gjaerevoll demonstrates, using floristic considerations and 
the discovery of a pre-Wtrm (= Wiscons in ) soil surface, that many high 
arctic species have persisted on Norwegian nunataks through times of 
glacial maxima to t he present. Finally , one of the alpine disjuncts, 
Polemonium pu lcherr imum pulcherrimum has been noted by Van Vechten 
( 1960 ) to occur in the Three Sisters only above all evidences of glacial 
activity. In the Western Cascades Arenaria rubella, Ivesia gordoni i, 
115 
Polemonium pulcherrimum, Linanthastrum nuttallii, and Erigeron compos-
itus are all restricted to sites above former glaciers. 
Substrate texture is a third important factor in the physiography 
of the Western Cascades. This characteristic, together with three re-
lated variables, has been used to divide the mountain peaks into classes. 
Mountains with dense, non-scoriaceous volcanic rocks (Class 2) erode 
into piles of slowly weathering blocks of fallrock. Except near the 
summits, slopes are relatively gentle and forested unless cleared by 
recent fires. Meadow areas are not abundant, and there are still fewer 
open rocky areas available for plant establishment. On the other hand , 
Class 1 peaks, frequently composed of highly scoriaceous volcanic rocks, 
erode more rapidly to a fine gravelly scree which is in continuous 
motion downslope. Near the summits outcrop areas are plentiful, and 
lower on the slopes accumulation of weathered volcanic material leads 
to the formation of extensive subalpine meadows which vary greatly in 
moisture availability. On these peaks the rate of weathering of the 
parent rock approximately equals the rate of its erosion, and large 
areas remain in youthful condition, unable to follow the normal succes-
sional sequence to closed forest. 
Age and Its Effect on Diversity 
Margalef (1958, 1963) has proposed that the maturity of an eco -
system is directly and causally correlated with its structural and tax-
onomic diversity, stability, efficiency, and high longevity and low 
reproductive potential of member organisms. In fact, maturity of a 
system is impossible to def ine objectively other than by use of such 
116 
measures as Margalef suggests . Such vague criteria as stability do 
not, however, add to the force of his arguments. Much of his hypothesis 
is supported by his own work with marine zooplankton, and by others 
working with various aquatic systems. Related to Margalef's proposal 
is the well-known time theory of increasing diversity reviewed by Pianka 
(1966). This theory states that the longer an area is free of major 
disturbance the more diverse it will be. Pianka concludes tha t t he 
time theory is undoubtedly true for small areas but is inadequate as 
a global hypothesis. Comparative observations of the relative diver-
sities of the Western and High Cascades demonstrate the necessity of 
reconsidering these ideas as they apply to the Western Cascades. 
Seven peaks of the lower High Cascades, exhibiting the same 
range of elevation, slope steepness, and parent material were sampled 
floristically for comparison with the Western Cascades. The climatic 
regime at this longitude (about 8 km east of the study area) i s s lightly 
drier and windier, since cloudbanks typically begin to lift at the 
eastern border of the Western Cascades, but the major differences be-
tween the two regions are the age and texture of the parent material. 
Flows of dense volcanic rocks are present but local in distribution , 
while most of the surface is covered with relatively recent ash fall. 
The peaks in this area are primarily cinder cones, some of which are 
eroding at approximately the same rate as Class 1 peaks in the Weste r n 
Cascades. 
These peaks have a markedly less diverse flora than do Western 
Cascade peaks. Sand Mountain and Hoodoo Butte, relatively recent 
cinder cones, support 26 and 63 species respectively, while Maxwe l l 
117 
Butte and Grizzly Peak, older mountains with a greater proportion of 
dense volcanics (the latter is an erosional remnant of the flank of 
Mt. Jefferson), totaled 122 and 138 species respectively. Typical 
Class 1 peaks within the study area support 289 (Iron Mountain), 286 
(Rebel Rock), 228 (Horsepasture Mountain), and 210 (Browder Ridge) 
species. 
It is likely that diversity of substrate texture and substrate 
modification by plants with time are inseparable variables in this 
analysis, since both correlate directly with diversity. Roach's (1952) 
finding that very recent lava flows are floristically more diverse 
where abundant moisture was available (bogs, lakes, and streambanks) 
indicates that substrate modification is important, and perhaps ob-
scures any age factor that might be present in the above data. 
This problem can be approached from another point of view by 
considering habitats within the Western Cascades. When the described 
habitats from fresh outcropping rock to mature mesic forest are placed 
in a general successional scheme, analysis shows that it is the younger, 
dry, open areas which support the greater portion (60 percent) of the 
floristic diversity within the study area. Combined non-forest habitats 
contain 80 percent of all Western Cascade species, while only 35 percent 
are ever found in forest associations. This is quite the opposite of 
the implication of Roach's data and at first glance seems contradictory 
to both Margalef's ideas and to the time theory of diversity. 
It has been demonstrated that the rocky areas and derived dry 
meadows cannot be considered stable--the substrate is moving downslope 
at a considerable rate. It is proposed here that these areas are 
118 
predictably unstable, and that because of the approximate equilibrium 
between weathering and mass wasting on floristically diverse Class l 
peaks they are continually rejuvenated. This creates a system low in 
the successional sequence, but sufficiently predictable that many non-
weedy species have been able to continue existence and to adjust com-
petitively to the addition of an unusually large number of species to 
a basically unproductive habitat. It is likely that these perennially 
young systems have been evolving for at least 10,000 years. 
Thus the Western Cascades are seen to comprise a specialized 
series of terrestrial environments which do not entirely support 
Margalef's hypothesis. Although considerable diversity in succession-
ally young habitats can be accounted for by replacing Margalef's sta-
bility criterion with one of predictability, the relative lack of 
diversity in the presumably ancient, stable, structurally diverse, 
efficient, longevous forest associations contradicts both Margalef and 
the time theory of diversity. 
An Evaluation of the Xeric Island Concept 
and Theories of Migration 
The Hypothesis 
The essence of the Xeric Island Hypothesis (Detling, 195 3) can 
be stated as follows: During the post-glacial xerothermal maximum, 
xeric species from the Rogue River Valley and from east-central Oregon 
became widespread in western Oregon, occupying dry habitats in the 
Western Cascades (and to some extent in the Coast Range). Following a 
climatic reversal, they have persisted as relicts in areas of minimal 
119 
competition with gradually re-establishing mesic forest species. 
The hypothesis is dependent on the palynological studies of 
Hansen (1942, 1947, 1955) and Heusser (1960), who have demonstrated a 
xerothermal maximum at about 6000 years ago. Pollen of Pinus ponderosa, 
composites, grasses, and chenopods reach maxima just above a distinc-
tive layer of Mt. Mazama ash in many bogs. Radiocarbon dating of 
charred stumps has placed the eruption of Mt. Mazama at approximately 
6500 years B.P. (Flint&. Deevey, 1951). Sears ( 1942) has pointed out the 
difficulty in adequately defining and dating xerothermal intervals in 
eastern North America and Europe. The fortuitous timing of the last 
violent eruptions of Mt. Mazama and the wide distribution of the re-
sulting ash fall in western North America obviate many of the diffi-
culties experienced in other regions. 
Problems Raised El. the Hypothesis 
Detling's synthesis of information, as represented by the Xeric 
Island Hypothesis, has been the outstanding contribution to knowledge 
of the vegetational history of the Pacific Northwest. However, incor-
porated into the hypothesis are inaccuracies both in theoretical con-
siderations and in supporting data largely because of the imprecise 
definition of "xeric" in the latter. 
One source of error in the supporting information includes sev-
eral misinterpretations of ecological field evidence. Detling suggests 
(1953) that heat absorption by the exposed dark rocks results in longer 
growing seasons for plants occupying these habitats. This conclusion 
is not warranted since, in any modified Mediterranean climate such as 
120 
that in western Oregon , growing seasons for most species are limited 
by lack of moisture, not by low temperatures. This is especially true 
on steep, bare, south-facing slopes at moderate elevation where growing 
seasons are considerably shorter than elsewhere in the Western Cascades, 
In addition, only 14 of the 32 species discussed as xeric by 
Detling actually occur in the extreme, hot , dry environments he de-
scribes in formulating the hypothesis. The rest are more typically 
found in meadow a s sociations, wh ich have been shown by sap tension 
measurements frequently to offer less stressful environments than even 
the surrounding mesic conifer forest. Species of other habitats, such 
as snowbeds, are also included in Detling's list. 
Like many earlier phytogeographers, Detling relies heavily on 
the geographic origin of species in interpreting not only thei r migra-
tional paths but also, with considerably less validity, their ecological 
relationships to other vegetational elements, Persistence or extinction 
of a disjunct species cannot be predicted from its direction of origin 
alone. It is rather a result of many aspects of the ecology of the 
species, including ranges of tolerance to many fa ctors, opportun ism, 
competitive ability, etc, 
Definition of Terms 
A major result of these cons iderations is that it is impossible 
to determine what Detling and other workers have meant when they have 
referred to "xeric species" and "xeric environments." In the Wes tern 
Cascades all habitats are to some degree seasonally dry and could a t 
those times be classified as xeric. Application of the sap-tension 
121 
pressure-bomb in the Western Cascades has supplied a means of quanti-
fying functional moisture stress. This technique is useful in formu-
lating definitions for individual species in given habitats. -Ranges of 
tolerance can be analyzed and specific habitats can be compared. The 
resulting definitions of plant-environment a ssociations are complex and 
involve many variables. Generalizations from them are difficult to make 
without loss of essential information. Th e term "xeric" is thus too 
general to be applied meaningfully in such definitions or wherever 
moistures regimes are being studied on a specific level. 
Patterns of Migration 
It has often been assumed, as in Detling's work, that advancing 
species migrate in a continuous front and that disjunct populations 
perforce indicate former continuity (at least through time) between 
them. Deevey (1949) points out that the ranges of virtually all species 
are characterized by disjunctions of some magnitude. Work with the 
biota of oceanic islands (Carlquist, 1966a; Wilson, 1959; etc.) has 
emphasized long-distance dispersal, and others (Raven, 1963; Cruden, 
1966) have discussed the likelihood of various types of shorter-dis tance 
disjunct dispersal in terrestrial environments. Detling's recognition 
of the Western Cascade peaks as "island s" in fact requires this kind of 
dispersal to some extent rather than continuous-front migrational ad-
vance. The work of numerous biogeographers has made it clear that in 
every species disjunction of any magnitude, both kinds of migration 
must be carefully considered. 
122 
Relict Terminology 
Failure to recognize the possibility of long-distance dispersal 
as a factor in migration forces the conclusion that all disjuncts are 
relictual in nature. Thus an examination of the meaning and implica-
tions of relictual status is in order. Fryxell (1962) and Holmquist 
(1962) are in agreement that relict concepts have been much confus ed 
in the past. Working independently, they achieved a botanical-zoolog-
ical consensus in defining a relict as a plant or animal living on as 
a fragmentary s urv ival from an earlier period. They term the type of 
relict suggested by the Xeric Island Hypothesis a geographical or eco-
l ogical relict--a locally relictual portion of an otherwise non-relict 
entity. 
Most biogeographical studies have deep roots in the relict con-
cept, although several workers have attempted to de-emphasize it. Cain 
(1944) proposes that relictual status of a di sjunc t population cannot 
be assumed, but must require reasonable evidence. Deevey (1949) goes 
to great lengths to provide alternative hypotheses to those of "per-
sistence" of species in any area during a major climatic change. Such 
an extreme approach stems from a reaction to Fernald ' s original "nuna tak 
hypothesis" (1925), which, while in large part wrong, has been in my 
opinion the greatest single stimulus to phyotogeographic inquiry in 
North America. Cain ' s modest argument is sound, but Deevey strains 
credibility by over-reacting to one-sided presentations such as those 
of Fernald, Braun (1955, etc.) , and Detling. Raup (1941) and Marie -
Victorin (1938) offer the least biased but early reviews of these con -
cepts. 
123 
Ecological Relicts vs. Recent Adventives 
J. C. Nelson made the following statement regarding the ignoral 
by botanists of newly discovered extra-limital species in western Oregon: 
To say that these species are only recent introductions, and do 
not belong ecologically to this district, is only to beg the 
question. How can we show that they have not been here as long 
as the species which are more characteristic? The desert plants 
growing on gravelly prairies about Salem are as integral a part 
of the local flora as any typical west-coast forms. (p. 23) 
Nelson's argument appears valid, but it is occasionally disre-
garded, even today. Conversation with various officers and employees 
of the Willamette National Forest indicates that the predominant opin-
ion concerning Western Cascade disjuncts in that organization sharply 
contradicts the ideas of Detling. According to this view, the disjuncts 
were carried into the area by summer - grazing sheep from eastern Or egon 
during the last 80 years or so and have persisted because of the pro-
found environmental degradation due to overgrazing. The conclu s ion 
reached is that the disjunct species are a product of the activities of 
man, and that they should thus not be considered a unique or espec i a lly 
interesting feature of the region. 
The volume of information gathered in this study about the large 
number of geographically, ecologically, florist i cally, and behaviorally 
diverse disjunct species in the Western Cascades stands as testimony 
against this hypothesis as any sor t of general explanation of Western 
Cascade disjunctions. It is like ly, however, that sheep have had an 
important effect, together with the native fauna, on dispersal of dis-
junct plants within the area. It ts also important to note that pre-
vious overgrazing has probably altered the vegetation aspect of th e 
124 
Western Cascades, especially the non-forest areas. It is unlikely 
that many species have become extinct or that large numbers of new 
species have been added to the area by such mechanisms. However, rel-
ative abundances, especially of nativ e grasses and species such as 
Pteridium aguilinum and Rubus parviflorus which probably replaced them, 
have no doubt been considerably altered by such disturbance. Sheep 
grazing in the Western Cascades was discontinued about 25 years ago. 
Observations over the last 10 years ( John Lincoln , personal communi-
cation, 1968) indicate that gra s e s are currently becoming more abun-
dant in several areas. 
The presence of many of the disjunct species in the pass es of 
the Ca s cade crest does indicate recent and perhaps continuing migration 
of these s pecies across the high mountains to the Western Cascades . I 
seriously doubt that these species can accurate l y be considered re l icts . 
Two species, Artemisia tridentata and Helianthus cusickii 9 are 
known from only a single locality i n t he We s t ern Cas cades. They are 
not in equilibrium with the surrounding communities, a s is evidenced by 
their lack of habitat specificity. It is likely that these two s pec ies , 
at least, represent introductions within the last half century, perhap s 
by sheep. 
Thirteen large trees of Populu s tremuloide s with numerous root 
sprouts have been discovered in Quaking As pen Swamp . Heartwood is 
rotten in most of these specimens, but corings and diameter measure-
ments indicate that the age of t he s e trees is well over 100 years. 
This age significantly decreases the probability of human-related dis-
persal factors for this species of the Eastern Element. 
125 
Thus, recent adventives into the area are considerably more 
important than thought by Detling, but dispersal within the last cen-
tury cannot account for the diversity of disjunct forms in the area, 
Concluding Comments on the Xeric Island Hypothesis 
The basic ideas of the Xeric Island Hypothesis are strong 
enough to stand independent of the inaccurate information and question-
able theory originally used to support them . Detling's general obser-
vations, which are essential to any unders tanding of the vegetational 
history of the Pacific Northwest, are supported by the present, more 
intensive study, The situation is, however, much more complex than 
envisioned by Detling, While further investigations have pared Det-
ling's original list of 32 disjunct species to 13, additional discov-
eries have brought the number to more than 80, These species are found 
in every major habitat type in the Western Cascades, show great diver-
sity in patterns of migration and evolutionary divergence from clos e 
relatives , and comprise a wide range of life histories . The Xeric 
Island Hypothesis must be modified and expanded to ass i mi l ate all the 
evidence now at hand. 
SYNTHETIC HYPOTHESIS 
Results of the present study support the following major con -
clusions: 
1) Western Cascade disjuncts are notable in the diversity of 
their geographical, ecological, physiological , taxonomic, evolutionary , 
and behavioral patterns. Generalized geographical elements do not rep-
resent floristic or vegeta t ional units, and no single migrational mech-
anism can account for the presence in the Western Cascades of all the 
members of any of the elements. 
2) Origin of the disjunct populations is best explained using 
a synthesis of the hypotheses and principles of many workers . Northern 
and high alpine species are true relicts which have survived in the 
area since the glacial maxima of the Pleistocene , Many of these spe-
cies were dispersed over distances of many kilometers. Southern, East-
ern, Southeastern, Valley, and Widespread Elements have migrated to the 
area since the most recent glacial retreat. Such species followed sev-
eral migrational paths from the east, south, and west, and were dis-
persed from varying distanc es to the hotter drier localities within 
the study area and throughout western Oregon. Dispersals of these ele-
ments to the area reached their peak during the xerothermal maximum 
about 6000 years ago, and f or some species have continued in lesser 
numbers to the present, due both to natural causes and to disturbance. 
Annuals and biennials establish in new environments more readily and 
126 
127 
appear to migrate in shorter steps than wind-dispersed perennials. 
3) Endemic species probably evolved between the late Pliocene 
and the xerothermal maximum, within or near the areas presently occu-
pied. One endemic species may be approaching extinction, 
4) Continuance of both disjunct and endemic species popula-
tions has been permitted by the near equivalence of the destructive 
geologic proces s es of weathering and ma ss wasting. Con tinued rejuv-
enation of outcrop areas , and rockfa ll and fine scree slopes has pro-
duced a predictably young envi ronmen t for which both high alpine and 
lower elevation dryland species are suited. Surrounding mesic forest 
communities are incapable of succeeding into the young environments. 
The continued existence of Mesic and Wet Meadow associations, which also 
support numerous disjunct species , is more difficult to interpret but 
is partially due to heavy snow pack 9 rapid movement of subst~ate, abun-
dant moisture, and natural disturbance by fire. The equilibrium between 
the latter communities and t he forest seems t o be much less stab le than 
that between forest and the drier perennia lly young env ironments. 
r 
LITERATURE CITED 
Abrams, LeRoy. 1940. Illustrated flora of the Pacific States. I. 
Ophioglossaceae to Aristolochiaceae. Stanford Univ. Press, 
Stanford. 
1944. Illustrated flora of the Pacific States. II. 
Polygonaceae to Krameriaceae. Stanford Univ. Press, Stanford. 
1951. Illustra ted flora of the Pacific States. III. 
Geraniaceae to Scrophulariaceae. Stanford Univ. Press, 
Stanford. 
Abrams, LeRoy, and R. S. Ferris . 1960. Illustrated flora of the 
Pacific States. IV . Bignoniaceae to Compositae. Stanford 
Univ. Press, Stanford, 
Aller, A. R. 1956. A taxonomic and ecological study of the flora of 
Monument Peak, Oregon. Amer. Midl. Nat. 56:454-472. 
Applegate, E. I. 1939. Plants of Crater Lake National Park. Amer. 
Midl. Nat, 22:225-314. 
Babcock, E. B. 1947. The genus Crepis. Univ. Calif. Publ. Bot. 
21-22: 1-1030 . 
Babcock, E. B., and G. L. Stebbins, Jr . 19 38. The American species 
of Crepis. Their interrelationships and distribution as affected 
by polyploidy and apomixis. Carn. Inst. Wash. Publ. 504. 
Baker, H. G. 1955. Self- compatibility and establishment after "long-
d is tance" dispersal. Evolution 9: 34 7-348. 
1959a. The contribution of autecological and genecologica l 
studies to our knowledge of the past migrations of plants. 
Amer. Nat. 93:255- 272. 
1959b. Reproductive methods as factors in speciation in 
flowering plants. Cold Sp ring Harbor Symp. Quant. Biol. 
24: 1 77 - 19 1. 
1961. Pollination mechanisms and inbreeders. Rapid 
speciation in relation to changes in the breeding systems of 
plants. In Recent advances in botany. p. 881-885. Univ. 
Toronto Press, Toronto. 
128 
129 
Baker, W. H. 1949a. A taxonomic and ecologic comparison of the 
floras of Iron Mountain and Fairview Mountain in Oregon. 
PhD Thesis, Oregon State Univ., Corvallis. 
1949b. A new Polygonum from Oregon. Madrono 10:62-64. 
1951. Plants of Fairview Mountain, Calapooya Range, 
Oregon. Amer. Midl. Nat. 46: 132-173. 
Baldwin, E. M. 1964. Geology of Oregon. 2nd ed. Edwards Bros., 
Ann Arbor. 
Beaman, J. H., and S. W. Andresen. 1966. The vegetation, floristics 
/ 
and phytogeography of the summit of Cerro Potosi, Mexico. 
Amer. Midl. Nat. 75:1-33. 
Bliss, L. C. 1956. A comparison of plant development in micro-
environments of arctic and alpine tundras. Ecol. Monog. 
26:303-337. 
Boyer , J.C. 1967. Leaf water potentials measured with a pressure 
chamber. Plant Phys. 42:133-137. 
Braun, E. L, 1955. The phytogeography of unglaciated eastern United 
States and its interpretation. Bot. Rev. 21:297-375. 
Braun-Blanquet, J. 1932. Plant Sociology. McGraw-Hill Co., New York. 
Brink, V. C. 1964. Plant establishment in the high snowfall and 
subalpine regions of British Columbia. Ecology 45:431-438 . 
Cain , S. A. 1944. Foundations of Plant Geography. Harper Bros., 
New York. 
Callaghan, Eugene, and S. F. Buddington. 1938. Metalliferous mineral 
deposits of the Cascade Range in Oregon. U. S, Geol. Surv. • 
Bull. 893. 
Carlquist, Sherwin. 1966a . The biota of long-distance dispersal. I. 
Principles of dispersal and evolution. Quart. Rev. Biol. 
41:247-270. 
1966b. The biota of long-dis tance dispersal. II. Loss 
of disp ersal in Pacific Compositae. Evolution 20:30-48. 
Cavers, P. B. , and J. L. Harper. 1967a. Studies in the dynamics of 
plant populations. I. The fate of seeds and transplants into 
variable habitats. J. Ecol. 55:59-72. 
1967b. The comparative biology of related species living 
in the same area . IX . Rumex : the nature of adaptation to a 
sea - shore habitat. J , Ecol. 55 :7 3-82. 
130 
Chambers, K. L. 1955. A biosystema tic study of the annual species 
of Microseris. Contr. Dudley Herb. 4:207-312. 
1957. Taxonomic notes on some Compositae of the western 
United States. Contr. Dudley Herb. 5:57-68. 
Clausen, Jens, D. D. Keck, and W. M. Hiesey. 1940. Experimental 
studies on the nature of species. I. Effects of various 
environments on western North American plants. Carn. Inst. 
Wash. Publ. 520. 
1947. Heredity of geographically and ecologically iso-
lated races. Amer. Na t. 31:114-133. 
Constance, Lincoln. 1938. A revision of the genus Douglasia. Amer. 
Midl. Nat. 19:249- 259. 
1942. The genus Hydrophyllum. Amer. Midl. Nat. 27:710-731. 
Coolidge, J. 0. 1964. A revision of the genus Polygonum, section 
Avicularia in the western United States. MS Thesis. Univ. 
Idaho, Moscow. 
Copeland, H. F. 1941. Further studies on Monotropoideae. Madrano 
6:97-99. 
Crampton, B. 1954. Morphological and ecological considerations in 
the classification of Navarretia (Polemoniaceae). Madrano 
12:225-238. 
Cronquist, Arthur. 1947. Revision of the North Amer ican species of 
Erigeron, north of Mexico. Brittonia 6:121-302. 
1955. Vascular plants of the Pacific Northwest. V. 
Compositae. Univ. Wash. Press, Seattle. 
Cruden, R. W. 1966. Birds as agents of long-distance dispersal for 
disjunct plant groups of the temperate western hemisphere. 
Evolution 20:517-532. 
Davidson, J. F. 1950. The genus Polemonium (Tournefort) L. Univ. 
Calif. Publ. Bot. 23:2 09-282. 
Deevey, E. S., Jr. 1949. Biogeography of the Pleistoc ene. Bull. 
Geol. Soc. Amer. 60:1315-1416. 
Detling, L. E. 1935. A new Cardamine from northern Idaho. Madrano 
3:176-178. 
1936. The genus Dentaria in the Pacific States. Amer. 
J. Bot. 23:570-576. 
131 
• 1937. The Pacific Coast species of CardalmI i ne. Amer. J, ---B-o-t. 24:70-76. 
. 1948a. Environmental extremes and endemi m, Madrono 
---9-:-137-149. I 
1948b. Concentration of environmental extremes as the 
basis for vegetation areas. Madrono 9:169-185. 
1953. Relict islands of xeric flora west of the Cas cade 
Mountains in Oregon. Madrono 12:39-47. 
1954. Significant features of the flora f Saddle 
Mountain, Clatsop County, Oregon. I Northwest Sc . 28:52-60. 
1958. Peculiarities of the Columbia Rive1r Gorge flora. 
Madrono 14:160-172. I 
1961, The chaparral formation of southwestern Oregon, with 
consideration of its postglacial history. Ecology 42:348-357. 
1966, The flora of the Columbia River Gorge. Northwest 
Sci. 40:133-137. 
1968. Historical background of the Pacific Northwest 
flora. Bull. Mus. Nat. Hist. Univ. Oregon 13:1-57, 
Dicken, S. N. 1955, Oregon geography. Univ. Oreg. Coop. Store, 
Eugene. 
Dunn, D. B, 1955, A new subspecies and new combinations in Lupinus. 
Leafl. West. Bot. 7:254-255. 
1957. A revision of the Lupinus arbustus complex of the 
Laxiflorae. El Aliso 14:54-73. 
Epling, Carl. 1925. Monograph of the genus Monardella, An. Missour i 
Bot. Gard. 12:1-106. 
Ewan, Joseph. 1942. Linanthastrum, a new West American genus of 
Polemoniaceae, J, Wash. Acad. Sci. 32:138-141. 
Fernald, M. L. 1925. Persistence of plants in unglaciated areas of 
boreal America. Mem. Amer. Acad. 15:241-242. 
Flint, R. F., and E. S. Deevey, Jr. 1951, Radiocarbon dating of 
late-Pleistocene events, Amer. J. Sci. 249:257-300. 
Freeman, O. W., J. D. Forrester, and R. L. Lupher. 1945. Physio-
graphic divisions of the Columbia Intermontane Province. 
Ann. Assoc. Amer. Geogr. 35:53-75. 
132 
Fryxell, P. A. 1962. The "relict species" concept. Acta Biotheoretica 
15:105-118. 
Gillett, G. W. 1962. Evolutionary relationships of Phacelia linearis. 
Brittonia 14:231-236. 
Gjaerevoll, 0, 1963. Survival of plants on nunataks in Norway during 
the Pleistocene glaciation. In North Atlantic biota and thei'r 
history, p. 261-283, Pergamon Press, New York . 
Gleason, H. A, 1922. The vegetational history of the Middle West. 
Ann. Assoc. Amer. Geogr. 12:39-85. 
Grant, Adele. 1924. A monograph of the genus Mimulus. Ann. Missouri 
Bot. Gard. 11:99-388 . 
Grant, Verne. 1949. Pollination systems as isolating mechanisms in 
angiosperms, Evolution 3:82-97. 
1956. A synopsis of Ipomopsis. El Aliso 3:351-362. 
Hall, H. M. 1928, The genus Haplopappus. Carn. Inst, Wash, Publ. 389. 
Hall, H, M., and F. E. Clements. 1923a. The North American species of 
Artemisia, In The phylogenetic method in taxonomy, Carn, Inst. 
Wash. Publ. 326:31-156. 
1923b, The North American species of Chrysothamnus. In 
The phylogenetic method in taxonomy. Carn. Inst. Wash. Publ. 
326: 157-234. 
Hansen, H. P. 1942. Post-Mt, Mazama forest succession on the eastern 
slope of the central Cascades of Oregon. Amer. J, Bot. 
2 9: 214-219. 
1947. Postglacial forest succession, climate, and chron-
ology in the Pacific Northwest. Trans. Amer. Phil. Soc. 37:1-1 30, 
1955. Postglacial forests in south-central and central 
British Columbia. Amer. J. Sci. 253:640-658. 
Heckard, Lawrence. 1968. Chromosome numbers and polyploidy in 
Castilleja (Schrophulariaceae). Brittonia (In press). 
Heusser, Calvin. 1960. Late-Pleistocene environments of North Pacific 
North America. Amer. Geogr. Soc. Spec. Publ. 35. 
Hickman, J.C., and M. P. Johnson. 1968. Analysis of geographical 
variation in western North American Menziesia (Ericaceae). 
Madrono (In press). 
133 
Hitchcock, C. L. 1952. A revision of the North Ameri , an species of 
Lathyrus. Univ. Wash. Puhl. Biol, 15:1-104. 
Hitchcock, C. L., and Arthur Cronquist. 1961. Vascul r plants of the 
Pacific Northwest. II I. Saxifragaceae to EricaI1 eae. Univ. 
Wash. Press, Seattle . 
• 1964. Vascular plants of the Pacific Nor hwest. II.• 
----Sa-licaceae to Saxifragaceae. Univ. Wash. Press! Seattle. 
Hitchcock, C. L., Arthur Cronquist, and Marion Ownbey. 1959, 
Vascular plants of the Pacific Northwest. IV. ricaceae 
through Campanulaceae. Univ . Wash . Press , Seat I le . 
Hitchcock, C. L., and Bassett Maguire. 1947. A revision of the North 
I 
American species of Silene. Univ. Wash. Puhl. giol. 13:1-73. 
Holbo, H. R., and H, N. Mozingo. 1965. The chromatogl aphic character-
ization of Artemis ia, Section Tridentatae. Arnet. J. Bot, 
52:970-978. I 
Holmquist, Charlotte, 196 2. The relict concept--is it a merely 
zoogeographical conception? Oikos 13:262-292. I 
Hopson, R. E. 1946. The study of a valley; the McKenzie River region 
of Oregon with special reference to the educatiJnal significance 
I 
of its natural history. PhD Thesis, Cornell Un v., Ithaca. 
1
Howell, J. T. 1945. Studies in Phacelia: Revision of species related 
to~- douglasii, P. linearis, and P. pringlei. ] Amer. Midl. Nat. 
33:460-494. 
Hulten, Eric, 1937. Outline of the history of arctic and boreal biota 
during the Quaternary period. Their evolution uring and after 
the glacial period as indicated by the equiform41 progressive 
areas of present plant species. Bokf~rlags aktlebolaget Thule, 
Stockholm. 
Ireland, Orlin. 1968. Plants of the Three Sisters re ion, Oregon 
Cascade Range. Bull. Mus. Nat. Hist. Univ. Oregon 12: 1-130. 
I 
Johnston, I. M. 1925. The North American species of l ryptantha, 
Contr, Gray Herb. n.s. 74, 
Keck, D. D. 1938. Revision of Horkelia and Ivesia. loydia 1:75- 142. 
1940. Studies in Penstemon. VII. The subsections 
Gairdneriani, Deusti and Arenarii of the GracilJs, and miscel-
laneous new species. Amer. Midl. Nat. 23:594-616. 
1946. A revision of the Artemisia vul ar Js complex in 
North America. Proc. Calif. Acad. Sci. IV. 25:421-468, 
134 
Kruckeberg, A. R. 1964. Ferns associated with ultramafic rocks in 
the Pacific Northwest. Amer. Fern, J. 54 : 113-1216. 
1967. Ecotypic response to ultramafic soils by some 
plant species of the northwestern U. S. Brittonia 19:133-151. 
Lanjouw, J., and F. A. Stafleu. 1964. Index Herbariorum. I. The 
herbaria of the world. 5th ed. Utrecht. 
Larrison, E. J. 1946. Biotic areas in the Pacific Northwest. The 
Murrelet 27:19-24. 
Lewis, Harlan. 1962. Catastrophic selection as a factor in speciation. 
Evolution 16:257-271. 
Lewis, Harlan, and Jerzy Szweykowski. 1964. The genusl· Gayophytum 
(Onagraceae). Brittonia 16:343-391. 
Macbride, J. F., and E. B. Payson. 1917. A rev1s1on of the Erigerons 
of the series Multifidi. Contr. Gray Herb. 49:72-79. 
Maguire, Bassett. 1941. Great Basin plants III. Car~ophyllaceae. 
Madrono 6:22-27. 
1943. A monograph of the genus Arnica. Brittonia 4: 
386~510. 
1947. Studies in the Caryophyllaceae. III. A synopsis 
·of the North American species of Arenaria, Sect ' on Eremogone 
·· Fenzl. Bull. Torr. Bot. Club 74:38-56. 
1951. Studies in the Caryophyllaceae. V. Arenaria in 
America North of Mexico. A conspectus. Amer. Midl. Nat. 
46:493-511. 
Margalef, R. 1958. Mode of evolution of species in rilation to their 
places in ecological succession. XV Int. Congr .l Zool. 10: 
787-789. 
1963. On certain unifying principles in ecology. Amer. 
Nat. 97:357-374. 
Marie-Victorin, Frere. ' 1938. Phytogeographical problems of eastern 
Canada. Amer. Midl. Nat. 19:489-558. I 
Martin, J. S. 1943. A revision of the native clovers o( the United 
States. PhD Thesis , Univ. Wash . , Seattle. 
Mason, H. L. 1938. Two new species of Linanthus from western North 
America. Madrono 4:157-162. 
135 
1945 . Review of Cain , Foundations of plant geography 
Torreya 45 : 17 - 21. 
McKell, C. M. 1956. Some charac t eristics contributing to the estab-
lishmen t of rabbitbr ush (Chrysothamnus spp.). PhD Thesis, 
Oregon State Univ. , Corvallis. 
Merriam, C.H. 1889. Life zones and crop zones of the United States. 
U. S. Dept. Agric. Div . Biol. Surv. Bull. 10 : 9-79. 
Munz, P.A. 1932. The genus Gayophytum. Amer. J. Bot. 19:768-778 . 
Munz , P.A. , and D. D. Keck. 1959. A California flora. Univ. Calif . 
Press, Berkeley. 
Nelson , J.C. 1918 . Addit i ons to t he flora of western Oregon. 
Torreya 18 : 21-35, 
Ornduff, Robert, and D. H. French. 1958. Notable plants of the Warm 
Springs Indian Reservation, Oregon. Leafl. West. Bot. 8:217- 220 . 
Ottley, A. M. 1923. A revision of the Californian species of Lotus. 
Univ. Calif. Publ. Bot. 10 : 189-305. 
1944. The American Loti with special consideration of a 
proposed new section, Simpeteria. Brittonia 5:81-123. 
Payson, E. B. 1926. Erigeron compositus and its allies in the United 
States. Univ. Wyoming Publ. in Science: Bot. 1:172~186. 
Pechanec , A. M. 1961. Some aspects of the ecol ogy of the Bryophy t es 
of the Three Sisters Primitive Area, PhD Thesis, Oregon State 
Univ., Corvallis. 
Peck, D. L. 1960. Cenozoic vulcanism in the Oregon Cascades. U. S. 
Geol. Surv. Prof . Paper 400-B . 
Peck , D. L., A. B. Griggs, H. G. Schli cker, G. G. We lls, and H. M. Do le. 
1964. Geology of the central and northern parts of the Western 
Cascade Range in Oregon . U. S. Geol. Surv. Prof. Paper 449. 
Peck , M. E. 1925a . A preliminary ske tch of the plant regions of 
Oregon. I . Western Oregon . Amer . J . Bot. 12:33-49. 
1925b. A preliminary sketch of the plant regions of 
Oregon . II. The Cascade Mountains and eastward. Amer. J. Bot. 
12 : 69-81. 
1928. New plants from Oregon. Torreya 28:53-56. 
136 
1961. A manual of the higher plants of Oregon. 2nd ed. 
Binfords and Mort, Portland. 
Phillips, Lyle. 1955. A revision of the perennial species of Lupinus 
of North America. Res. Studies State Col. of Wash. 23:161-201. 
Pianka, E. R. 1966. Latitudinal gradients in species diversity: a 
review of concepts. Amer. Midl. Nat, 100:33-46. 
Piper, C. V. 1918. New plants of the Pacific Northwest. Proc. Biol. 
Soc. Wash. 31:75-78. 
1920, A study of Allocarya. Contr. U. S. Nat. Herb. 
22:79-113. 
Raup, H. M. 1941. Botanical problems in boreal America. Bot. Rev. 
7: 147-248. 
Raven, P.H. 1963. Amphitropical relationships in the floras of North 
and South America. In Amphitropical relationships in the herba-
ceous flora of the Pacific Coast of North and South America: a 
symposium. Quart. Rev. Biol. 38: 151-177. 
Rehder, Alfred, 1903. Synopsis of the genus Lonicera. Ann. Missouri 
Bot. Gard. 14:27-232. 
Roach, A. W, 1952. Phytosociology of the Nash Crater lava flows, Linn 
County, Oregon. Ecol. Monog. 22:169-193. 
Rollins, R. C. 1941. A monographic study of Arabis in Western North 
America. Rhodora 43:289-325, 348-411, 425-481. 
Rydberg, P.A. 1916. Artemisia. In North American flora 34:242-285. 
New York Botanical Garden, New York. 
Scholander, P. F., H. T. Hammel, E. D. Bradstreet and E. A. Hemmingsen. 
Sap pressure in vascular plants. Science 148:339-346. 
Sears, P. B. 1942. Xerothermic theory. Bot. Rev. 8:708-736. 
Stebbins, G. L., Jr. 1952. Aridity as a stimulus to plant evolution. 
Amer. Nat. 86:33-44. 
Stebbins, G. L., Jr., and Jack Major. 1965. Endemism and speciation 
in the California flora. Ecol. Monog. 35:1~35. 
Thayer, T, P. 1936. Structure of the North Santiam River section of 
the central Cascade Mountains in Oregon. J. Geol. 44:701-716. 
1939. Geology of the Salem Hills and the North Santiam 
River basin: Oregon. Oreg. Dept. Geol. and Min. Ind. Bull. 15. 
137 
Thoday, D. 1933. The terminology of xerophytism. J. Ecol. 21:1-6. 
Thompson, J. W. 1931. 1930 Washington fern notes. Amer. Fern J. 
21: 117-124. 
Van Vechten, G. W. 1960. The ecology of the timberline and alpine 
vegetation of the Three Sisters, Oregon. PhD Thesis, Oregon 
State Univ., Corvallis. 
Vitt, J.P., and J. A. Rudinsky. 1959. The water-conducting systems 
in conifers and their importance to the distribution of trunk 
injected chemicals. Contr. Boyce Thompson Inst. 20:27-38. 
Warburg, M. R. 1965. The evolutionary significance of the ecological 
niche. Oikos 16:205-213. 
Ward, G. H. 1953, Artemisia, Section Seriphidium, in North America: 
A cytotaxonomic study. Contr. Dudley Herb. 4:155-205. 
Waring, R.H., and B. D. Cleary. 1967. Plant moisture stress: evalu-
ation by Pressure bomb. Science 155:1248~1254. 
Wells, F. G. 1956. Geology of the Medford Quadrangle, Oregon-California. 
U, S. Geol. Surv. Quadrangle Map GQ-89. 
Wells, F. G., and D. L. Peck. 1961. Geologic map of Oregon west of the 
121st meridian. U. S. Geol. Surv. Misc. Inv. Map 1-325. 
Wheeler, L. C. 1937. Notes on California pteridophytes. Amer. Fern J. 
27: 120-132. 
1938. Polygonum kelloggii and its allies. Rhodora 40: 
309-317. 
Wherry, E.T. 1946. The Gilia aggregata group. Bull. Torr. Bot. Club 
73:194-202. 
1961a, Remarks on the Ipomopsis aggregata group. El Aliso 
5:5-8. 
1961b. Remarks on the genus Linanthus. El Aliso 5:9-10. 
1967. Our temperate tufted polemoniums. El Aliso 6:97-101. 
Williams, Howel. 1942. The geology of Crater Lake National Park, 
Oregon, with a reconnaissance of the Cascade Range southward to 
Mount Shasta. Carn. Inst. Wash. Puhl. 540. 
1944. Volcanoes of the Three Sisters region, Oregon Cas-
cades. Univ. Calif. Puhl. Geol. 27:37-84. 
138 
1953. Ancient volcanoes of Oregon. Condon Lectures: 
Oreg. Syst. Higher Educ., Eugene. 
1957. A geologic map of the Bend Quadrangle, Oregon, and 
a reconnaissance geologic map of the central portion of the High 
Cascade Mountains. Oreg. Dept. Geol. and Min. Ind. (map and 
text). 
Williams, L. 1937. A monograph of the genus Mertensia in North 
America. Ann. Missouri Bot. Gard. 24:17-159, 
Wilson, E. 0. 1959. Adaptive shift and dispersal in a tropical ant 
fauna. Evolution 13:122-144. 
Wynd, F. L. 1936. The flora of Crater Lake. Amer. Midl. Nat. 17: 
881-949. 
APPENDIX A 
Introduction 
Appendix A is a checklist of the species of vascular plants 
identified from the Western Cascades during the course of this study. 
Two families, the Gramineae and the Cyperaceae, have been excluded due 
to the author's limited knowledge of their taxonomies. 
Species are listed in the order they are found in Peck (1961), 
although Peck's nomenclature is not strictly followed. Authorities 
are not given here because of space considerations, but can easily be 
found in Hitchcock and others (1959, 1961, 1964); Cronquist (1955); or 
Peck (1961), The rank of trinomials can be similarly obtained. Unless 
trinomials are given, typical subtaxa are ass4med, 
For the purposes of this checklist, mountains are considered to 
extend all the way to the drainage systems which separate them from one 
another. Elevations covered here range from 400 m to over 3100 m, but 
both limits vary from peak to peak. Several peaks not in the area of 
intensive study are included for comparison. These are Monument Peak , 
in the Western Cascades northwest of the main area; Huckleberry Moun-
tain, the Fairview-Bohemia massif, and Hershberger Mountain, progres-
sively more southern peaks of the Western Cascades; Grizzly Peak, 
Maxwell Butte, Little Nash Crater, Hogg Rock, Hoodoo Butte, and Sand 
Mou~tain of the lower High Cascades; the Three Sisters region down to 
an elevation of 1100 m on the western slope and 1900 m on the eastern 
? u,0-:, ~~ 
J 
slope; and the region of the Crater Lake caldera, Several peaks within 
the area, which were visited only briefly, are excluded, Information 
139 
140 
from these peaks was combined with that from nearby, physiographically 
related peaks which were more completely sampled. Excluded peaks (and 
their included counterparts) are North Peak (Echo Mt .) ; Lamb Butte 
(English Mt.); Yankee Mt. (Tipsoo Butte); and the McKenzie Pass Steptoes 
or Sand Hills (Three Sisters). 
An, "X" on the checklist indicates an observation made and noted 
in writing in the field. For those species treated in Appendix B, doc-
umentation of occurrences by previously existing herbarium specimens is 
included, so that the checklist will reflect the total known range of 
those species throughout the area covered. Referral to the maps in 
Appendix B will separate the author's collections from previous ones. 
Additional information from published work is included for 
Monument Peak, Fairview Peak, the Three Sisters, and Crater Lake. These 
• 
reports of occurrences are respectively denoted "A" (Aller, 1956), "B" 
(Baker, 1951), "I" (Ireland, 1968), and "W" or "E" (Wynd, 1936; Apple-
gate, 1939). None of these mountains is within the study area, but all 
were visited a_t least once during the course of the field work. 
Al 141 
'r 
Monument Peak X X X X . X XXXX X X X 
Bachelor Mt. X X X X X X X X X X 
Three Pyramid$ X X· X X X X X X X X X X X X X 
Crescent Mt. X X X X X X X 
Echo Peak X X X X X X X X 
f' . . Sou th Peak X X X X X X X 
Cone Peak X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X ,X XX X 
Browder Ridge X X X X X X X X X X X X 
Jumpoff Joe X X X X X X 
_ Squaw Peak X X X X 
s Twin Buttes X X X X X X X X X X X X 
~,t~carpenter Mt. X X X X X X X 
. d Tidbits Mt. X X X X X X X X X X X X 
a,,.,Jr.w·-~ Lookout Mt. X X X X X X X 
~,,,.wr......,Frissell Pt. X X X X X X X X X X X X X X X 
Castle Rock X X X X X X X X X X X X 
Qr,.V,P (0 ' Lea r:y Mt. X X X X X X X X 
0 (),iJ,f {Horsepasture Mt. X X X X X X X X X X X X X X X X 
" English Mt. X X X X X X 
"> Lowder Mt. X X X X X X X X X X 
. Tipsoo Butte X X X X X X X X X X 
"> Olallie Mt. X X X X X X X X X X X X 
Rebel Rock X X X XX X x·x X X X 
Indian Ridge X X X XX X. X X X X X X X X 
Sardine Butte X X X X X X X 
Huckleberry Mt. X X X X 
Bohemia Mt.(reg.) X X X X X B X X X B X X X X X X 
Hershberger Mt. X X X X X X X X X X X 
) Grizzly Peak X XX X X X X X X X X 
Maxwell Butte X X X X 
Little Nash Cr. X X X X X X X 
Hogg Rock X X X 
Hoodoo Butte X X 
Sand Mt. X X 
'> Thr:ee Sisters I X X X X I X X X X X X X 
Crater Lake X W XX W W X X E X X W X 
142 
Cf) •0..  
•M I-< 
Cf) Cf) ,-I • 
' •M •M•M \ 0 . • Cf) Cf) .jJ ) 0 
P P P · 11' I P <li 
• <l) <l)>(Cf) Cf)(1' ,-I 
~ ~ .jJ .jJ•~ a ::i.-1 • ) 
(13 \ ) •M Cf) Cf) (13Cfl.-l::, ::l<li -iJ 
,_. .1 . Cf) • P ~ • P a0a • P Cf) 
l-<(13 -' <l) ,.C:: <li -iJCfJ-iJ (13 •M \ Cfl<li::l 
11' :J.-111'~~ CfJ I-< O~P O Xl1' J XO~Cfl 
P \ 8.-IP•MN 11' \ I-< OP<li ·M <lil-< 11'P·M0 
11'11' ~11'PP 0.. ..\ 11' ::, p::,~11'1-< 011' P•MOP 
~C/l(1',.C::~p<l/Cfl l-<l-<0 1 0 e~~<li Cll.-l-iJ(1'<li-iJO<li 
.jJ o-lJ o..Cll • e·M Cf)• o•M • <l) • Cf) e 0.-1 e·M ::,1,1-1 (1'-iJ-iJ ::i o P 
1-<l-<l-<OPE .-l~0.-14-11-<~-iJ~OOO•~CfJ~l-<11' .-1 <li 
<li<liOl-<<li.-1(1'•M~00•M<l) (1'1-<004-1 1-<:J>..OOEaOE> 
..0~-iJ<l)-iJ<l)bl)..0p•M0P0CfJ0(1' •MSl-<4-11-<(1'•M::l::, 
e PP .jJ ,_. a0 ::i ••Cf) p a0 o ::i .... o.. C/l Cf) .jJ ::i • 4-1 • ,...( 0...-1 • .... Cf) 
11'00<li<liPCfJ81-<11'011'1-<l-<.-1~::,::,11'-iJO..<liO..bOCfJ.-l~,.C::::, 
.-1 o..o,.c:: e <l)-iJ • a0.-1 o a o..~ o..o ,_. 1-<.-l•M ~~.jJ P 
0 <l) <l) <l) <l) ,.c:: Cf) C/l ••• ,.c:: ,-I p <li 
C/l C/l Cf) • • • ~ Cf) Cf) C/l (fl Cf) C/l 0 1. •..  ..., • 0.. 0.. • 0 ::, ::, ,-I ,-I ,-I 0.. <li • ~ ::l ::l ::l bl) bl) <li ::, <li <li <li <li <li <li 0 E •.-l •M ,.C:: •M O O ::l ::, ::, 0 •.-l P (13 
ppp::,::,o<li~~~~~~..O::i11'PPO..CfJPPNNNl-<4-l<liaO 
~~~C/lC/l~Cll..O..O..O..O..O..O1~ ,.c::,.c::::,::,~~::,::,::,::,::,<lio-iJ~ Cl. Cl. Cl. f,-< f,-< Cl. Cl. <C -<C <C <C <C <C ...l Jf,-< U ', ', f,-< ...l ', ', ...l ...l ...l >< f,-< Ul N 
Monument Peak X X X X X X X X X X X X X X X X 
Bachelor Mt. X X X X X X X X X X X X X X 
·' Three Pyramids X X X X X X X X X X X 
Crescent Mt. X X X X X X X X X X X X X 
Echo Peak X X X X X X X X X X X X X 
South Peak X X X X X X X X X X X X X 
Cone Peak X X X X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X X 
Jumpoff Joe X X X X X X X X X 
Squaw Peak X X X X X X 
"I  Twin Buttes X X X X X X X X X X X 
Carpenter Mt. X X X X X X X X X X X X 
Tidbits Mt. X X X X X X X X X X X X X 
Lookout Mt. X X X X X X X X X X X X X X 
Frissell Pt, X X X X X X X X X X X X X X X X X 
Castle Rock X X X X X 
I O'Leary Mt. X X X X X X X X X X 
.Horsepasture Mt. X X X X X X X X X X X X X X X X X 
English Mt. X X X X X X X X X X X X X X X X 
\ Lowder Mt. X X X X X X X X X X X X X X X X X 
u " Tipsoo Butte X X X X X X X X X X X X 
\ 0 la 11 ie .M-t. X X X X X X X X X X X X X X 
\ Rebel Rock X X X X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X X X X X X X X 
Sardine Butte X X X X X X X X X X X X X X 
Huckleberry Mt, X X X X X X X X X 
Bohemia Mt,(reg.) X X X X X X X X B X X X X 
Hershbergex Mt, X X X X X X X X X X X X X 
Grizzly Peak X X X X X X X X X X X X X 
Maxwell Butte X X X X X X X X X X X X X X 
Little Nash Cr. X X X X X X X X X X X X 
Hogg Rock X X X X X X X X 
Hoodoo Butte X X X X X X 
Sand Mt. X X X X X X 
\ Three Sisters X X X X X X X X X X X X X X X I 
Crater Lake W X X X X X X W X W X WW X E X W X W 
' \ 
143 
Monumef! t Pe~k X A X X A X X X X A X X A A X 
Bachelor Mt. X X X X X X X X X X X X X X X X X 
\ Three Pyramids X X X X X X X X X X X X X X X 
Crescent Mt. X X X X X X X X X X X X X 
Echo Peak X X X X X X X X X X X X 
South Peak X X X X X X X X X X X X 
Cone Peak X X X X X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X X X X X 
Jumpoff Joe X X X 
Squaw Peak X X X X X X X 
Twin Buttes X X X X X X X X X X X X 
Carpenter Mt. X X X X X X X 
• I 
Tidbits Mt, X X X X X X X X X X X X X 
Lookout Mt. X X X X X X X X 
Frissell Pt. X X X X X X X X X X X 
Castle Rock X X X X X X X X X X X X 
O'Leary Mt. X X X X X X X X 
Horsepas ture Mt. X X X X X X X X X X X X X X X X X X X 
English Mt. X X X X X X X X X X 
Lowder Mt. X X X X X X X X X X X X X X 
) Tipsoo Butte X X X X X X X X X X 
I Olallie Mt. X X X X X X X X X X X X X X X 
., Rebel Rock X X X X X X X X X X X X X X X XX X X X 
Indian Ridge X X X X X X X X X 
Sardine Butte X X X X X X X 
Huckleberry Mt. X· X X X X X X X 
Bohemia Mt.(reg.) X X X X X X X X X X X X B B X X X X B 
Hershberger Mt. X X X X X X X X X X X 
Grizzly Peak X X X X X X X X X X X X X X 
Maxwell Butte X X X X X X 
Little Nash Cr. X 
Hogg Rock X 
Hoodoo Butte X X X 
Sand Mt. 
\ Three Sisters I I X I I X X I X X I I X I I X X X 
Crater Lake w EE W X E X X W E W W E X X W X E E 
144 
,) 
....- 
\ 
r' 
' 
Monument Peak 
Bachelor Mt. 
Three Pyramids 
Crescent Mt. 
Echo Peak X X X X X X X X X X X 
South Peak X X X X X X X X X X X 
' ' Cone Peak X X X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X X X 
' Browder Ridge X X X X X X X X X X X X X X 
Jumpoff Joe X X X X X X X X 
Squaw Peak X X X X 
Twin Buttes X X X X X X X X X X X 
Carpenter Mt. X X X X X X X 
\ Tidbits Mt. X X X X X X X X X 
Lookout Mt. X X X X X X X X 
Frissell Pt. X X X X X X X X X 
Castle Rock X X XX XX X X X X X 
0 'Leary Mt. X X X X X X X X X X 
Horsepasture Mt. X X X X X X X X X X X X XX X 
English Mt. X X X X X X 
Lowder Mt. X X X X X X X X X X X X X X X 
Tipsoo Butte X X X X X X X X X X 
Olallie Mt. X X X X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X 
Sardine Butte X X X X X X X 
Huckleberry Mt. X X X 
Bohemia Mt.(reg,j X B X X X B B X X X B X X X X X X 
Hershberger Mt. X X X X X X X X X X X X X X 
Grizzly Peak X X X X X X X X 
Maxwell Butte X X X X X X 
Little Nash Cr . X X X X X 
Hogg Rock X X X 
Hoodoo Butte X X X X X X X 
Sand Mt. X 
Three Sisters I X I X X X I I X X X X X X X X 
Crater Lake E wwwww w w E W W X X X X X E X X X 
145 
Monument Peak 
Bachelor Mt. X X X X X X X X X 
Three Pyramids X X X X X X X X X X 
Crescent Mt. X X X X X X 
Echo Peak X X X X X X X X X 
South Peak X X X X X X X 
Cone Peak X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X 
Jumpoff Joe X X 
Squaw Peak X X X X 
Twin Buttes X X X X X 
. Carpenter Mt. X X X X X X X . Tidbits Mt. X X X X X X 
Lookout Mt. X X X X X X 
Frissell Pt. X X X X X X X 
Castle Rock X X X X X x: X X X 
0 'Leary Mt. X X X X X X X X X 
_Horsepas ture Mt. X X X X X X X X X X 
English Mt. X X X X X X X X 
Lowder Mt. X X X X X X X X X X 
Tipsoo Butte ·x X X X X X 
Olallie Mt. X X X X X X X X 
Rebel Rock X X X X X X X X X X X X XX 
Indian Ridge X X X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X X X 
Bohemia Mt.(reg.) X X X B B X B X X X B X X 
Hershberger Mt. X X X X X X 
Grizzly Peak X X X X 
,r Maxwell Butte X X X 
Little Nash Cr. X 
Hogg Rock X 
Hoodoo Butte X X 
Sand Mt. 
Three Sisters X I X X I I X X I X X X 
Crater Lake W W E E X X X W X X E W w X w 
4 
t 146 
I 
l 
Monument Peak 
Bachelor Mt. 
Three Pyramids X X X X X X X X X X 
Crescent Mt. X X X X X X X X X X X X 
Echo Peak X X X X X X X X X X 
South Peak X X X X X X X X X X 
Cone Peak X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X 
Jumpoff Joe X X X 
J\ 
Squaw Peak X X X X X X 
Twin Buttes X X X X X X X X X X 
Carpenter Mt. X X X X X X X X 
Tidbits Mt. X X X X X X X X X X X 
Lookout Mt. X X X X X X X X 
Frissell Pt. X X X X X X X X X X X X 
Castle Rock X X X X X X X X X 
0 'Leary Mt. X XX X X X X X X 
Horsepasture Mt. X X X X X X X X X X X 
English Mt_. X X X X X X X X 
Lowder Mt. X X X X X X X X X X X X X 
Tipsoo Butte . X X X X X X X 
Olallie Mt. X X X X X X X X X X X 
Rebel Rock X X X X x· XX XX XX X X X 
Indian Ridge X X X X X X 
Sardine Butte X X X X X X X 
Huckleberry Mt. X X X X X X X 
Bohemia Mt.(reg.) X X X X B X B B X X X X X X X X X 
Hershberger Mt. XX X X X X X X X X X X X X 
Grizzly Peak X X X X 
Maxwe 11 Butte. X X X X X X 
Little Nash Cr. X 
Hogg Rock 
Hoodoo Butte X X X X 
Sand Mt. X 
Three Sisters X X X X X I X X X X X I X X X X 
Crater Lake WE W X W W X W WWW X W W X X E W X 
..J 
l 
147 
.-"-l'  
::, 
'"O 
•r-l 
4-< 
::, 
H 
' 
Monument Peak X A X A X X X X X X X X 
Bachelor Mt. X X X X X X X X X X X X 
Three Pyramids X X X X X X X X X X X X X X 
Crescent Mt. X X X X X X X X X X 
Echo Peak X X X X X X X X X X X X X 
South Peak X X X . X X X X X X X X 
Cone Peak X X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X X X 
Jumpof_f Joe X X X X X 
Squaw Peak X X X X X 
Twin Buttes X X X X X X X X X X X X X 
Carpenter Mt. X X X X X X X X 
Tidbits Mt. X X X X X X X X X X X X 
Lookout Mt. X X X X X 
Frissell Pt. X X X X X X X X X X 
Castle Rock X X X X X X 
0 'Leary Mt. X X X X X X X X X X X X 
Horsepas ture Mt. X X X X X X X X X X X 
English Mt. X X X X X X X X 
Lowder Mt. X X X X X X X X X X X X 
Tipsoo Butte X X X X X X X X X 
Olallie Mt. X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X 
,.. Indian Ridge X X X X X X X X X X X X X X X 
Sardine Butte X X X X X X X 
Huckleberry Mt. X X 
Bohemia Mt.(reg.) X X X X X X X X X X X X X X X X X B 
Hershberger Mt. X X X 
Grizzly Peak X X X X X X X 
Maxwell Butte X 
Little Nash Cr. 
Hogg Rock 
Hoodoo Butte 
Sand Mt. 
Three Sisters X I X X I X I XX 
Crater Lake WW X W X E W X w w w WWW XE 
148 
Monument Peak X X X X X X X X 
Bachelor Mt. X X X X X X X X X X X X 
Three Pyramids X X X X X X X X X X 
Crescent Mt. X X X X X X X 
Echo Peak X X X X X X X X X X X X X X X X 
South Peak X X X X X X X X X X 
Cone Peak X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X :X X X X X X X 
Browder Ridge X X X X X X X X X X X 
Jumpoff Joe X X X X X X X X 
Squaw Peak X X X . XX X X 
Twin Buttes X X X X X X X X X 
Carpenter Mt. X X X X X X X X 
Tidbits Mt. X X X X X X X X X 
Lookout Mt. X X X X X X X X X X 
Frissell Pt. X X X X X X X X X 
Castle Rock X X X X X X X 
0 'Leary Mt. X X X X X X X 
Horsepas ture Mt. X X X X X X > X X X X 
English Mt. X X X X X X 
Lowder Mt. X X X X X X X X X X 
Tipsoo Butte X X X X X X 
Olallie Mt. X X X X X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X X X 
Bohemia Mt.(reg.) X X X X X X X X X X X X X 
Hershberger Mt. X X X X X X X X X 
Grizzly Peak X X X X X X X X X 
Maxwe 11 Butte X X X X X X X X X X X X 
Little Nash Cr. X X X X X X X 
Hogg Rock X 
Hoodoo Butte X X X X 
Sand Mt. X 
Three Sisters X XX I X xxi X X X X X X X X X X 
Crater Lake w w X E X X X X E' X X W X X E X X X X 
149 
en 
::, 
H 
0 
..-I .,..; 
4-1 .,.., en 
\ -~ E (lj 
) .,.., (lj ::, ..-I 
(lj ~..... ..-I oJ) 
.,.., ..oo EE<ll 1 ::, 
..-l(lj ,.OQJ ;l S } ::,::,..c:: 0 
\ 
en O•M Op ~--~ ?.•.~ enQJ •, enQJ µ..C:: O.. ) en 
~ 
.,C../,l  \ en'H..-l(ljen..-l ~~-~ (ljµQJ ::, ::, J •M•MQµ::, C/len-,..;p(..)..-10..C/lPoDPU<ll•MC/l (/) 
..-I H ) enP'H<ll~C/l::,::,..-10::,..-1~pQJQJ<ll0~..-1::, ~ 
.,.., 0 C/l p..-1•,..; pH::,µ QJ O C/l~ QJ ol)QJµ.,..; ol)!-;•M O..C:: C/l C/l 
..0 ..-I (/) ::, QJ <ll H .,.., <ll ~ C/l u 4-1 H O ;3: P 0.. <ll H .,..; u O..'H µ ::, P 
(lj 4-1 ::, H..C:: ol)ol)O..•,..; ::,.,..;.,..; QJ HOO QJ H (lj..-l•M C/l•M P µ Q) 
µ .,.., p QJ u H QJ Ho o....o H µ~ o....c::..-1 Ho..> o E•,.., en <ll <ll~ 
(..) > .,..; u µ QJ µ <ll u QJ H QJ <ll P ..C:: C/l H P <ll 
QJ H en o .,.., .,.., PE C/l ..-1 <ll en ..-1 <ll EE EE EE EE <ll u P > 
0.. <ll H !-;en..C::•,..;QJ ::,::,::,::,::,::,::,::,QH•M•MQJ 
(/) 0.. ::, o.. u~ C/lt/lt/lent/len~~~~~~~~oDuSO..P 
C/lPE:C/l::,::,::,::,::,::,..-1..-1..-1..-1..-1..-1..-1..-1(1l 
en C/l C/l C/l ::, <ll QJ ::, C/l p p p p p o o o o o o o o u en C/l en C/l 
::,::,::,::,,.o..-1enp~~~~~~'H'H'H'H'H'H'H'H~::,::,::,::, 
..0 ..0 ..0 ..0 H QJ ::, µ 0.. 0.. 0.. 0.. 0..•,..; .,.., .,.., .,..; .,.., .,..; .,..; .,.., ~ µ µ µ µ 
::,::,::,::,oE !-;>,::,::,::,::,::,!-;!-;HHHHHHQJOOOO 
~~~~~< ~u~~~~~~~~~~~~~~~~~~ 
Monument Peak X X X X X X X A X X A 
Bachelor Mt. X X X X X X X X X X 
Three -Pyramids X X X X X X X X X X 
Crescent Mt. X X X X X X 
Echo Peak X X X X X X X X X X 
South Peak X X X X X X X X 
Cone Peak X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X 
Jumpoff Joe X X X X X X 
Squaw Peak X X X X 
Twin Buttes X X X X X X X X X 
Carpenter Mt, X X X X X X X X 
Tidbits Mt. X X X X X X X X 
Lookout Mt. X X X X X X X X 
Frissell Pt. X X X X X X X X X X 
Castle Rock X X X X X X X X 
0 'Leary Mt. X X X X X X X X X 
Horsepasture Mt. X X X X X X X X X X X 
English Mt. X X X X X 
Lowder Mt. X X X X X X X 
Tipsoo Butte XX X X X X X 
Olallie Mt. X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X X X X X X 
Bohemia Mt.(reg.) X X X X X X B X X X X X X X X 
Hershberger Mt. X X X X X X X X X 
Grizzly Peak X X X X X X 
Maxwell Butte X X X X X X 
Little Nash Cr, X X X X X X 
Hogg Rock X 
Hoodoo Butte X X X X 
Sand Mt. X 
Three Sisters X X XX I X I I X X X 
Crater Lake X W X X X X X X W X WW E E 
Monument Peak X X X X X X X X A X X A 
Bachelor Mt. X X X X X X X X X 
Three Pyramids X X X X X X 
Crescent Mt. X X X X X X X X X 
Echo Peak X X X X X X X X X X X X X X 
South Peak X X X X X 
Cone Peak X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X 
Browder Ridge X X X X X 
Jumpoff Joe X X X X 
Squaw Peak X X 
Twin Buttes X .x X X X X X X 
Carpenter Mt. X X X X 
Tidbits Mt. X X X X X 
Lookout Mt, x · X X X X X X X X 
Frissell Pt. X X X X X X X X X 
Castle Rock X X X X X X X 
, \ 0 'Leary Mt. X X X X X X X X 
.Horsepas ture Mt. X X X X X X X X X X X X X 
English Mt, X X X X X X X 
Lowder Mt, X X X X X X X X X 
Tipsoo Butte X X X X X X X X X 
Olallie Mt. X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X 
Indian Ridge X X X X X X 
Sardine Butte X X X X X X X X 
Huckleberry Mt. X X X X X X 
Bohemia Mt.(reg.) X X X X X X X X X X X 
Hershberger Mt. X X X X X X X X X X 
Grizzly Peak X X X X X 
Maxwell Butte X X X X X X X 
Little Nash Cr. X X X X X X 
Hogg Rock X X 
Hoodoo Butte X X X X 
Sand Mt. X X X 
Three Sisters I X X X X X X X X X X X 
Crater Lake w w E X E X X X E X X W X E 
151 
Monument Peak X X X X X X X X X X X X AX X 
Bachelor Mt. X X X X X X X X 
Three Pyramids X X X X X X X X 
Crescent Mt. X X X X 
Echo Peak X X X X X X X X X X X 
South Peak X X X X X X X 
Cone Peak XX X X X X X X X X X X X 
Iron Mt. X X X X X X X J XX X X XX X X X 
Browder Ridge X X X X X X X X X X X X X 
Jumpoff Joe X X X X X X 
Squaw eak X X X X X X 
Twin Buttes X X X X X X X X X X 
Carpenter Mt. X X X X X X X X X X 
Tidbits Mt. X X X X X X X X 
Lookout Mt. X X X X X X X 
Frissell Pt. X X X X X X X X X X X X 
Castle Rock X X X X X X X X 
0 'Leary Mt. X X X X X X 
Horsepasture Mt. X X X X X X X X X X X X X 
English Mt. X X X X X X X X X X 
Lowder Mt. X X X X X X X X 
Tipsoo Butte X X X X X X X 
Olallie Mt. X X X X X X X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X X 
Indian Ridge )< X X X X X X 
Sardine Butte X X X X X 
Huckleberry Mt. x·x 
Bohemia Mt.(reg.) X B X X X X X B X B X X X X X X X 
Hershberger Mt. X X X X X X X X X X 
Grizzly Peak X X X X X X X X 
Maxwell Butte X- -X X X X X 
Little Nash Cr. X 
Hogg Rock 
Hoodoo Butte X X X X 
Sand Mt. X 
Three Sisters I I X X X X I X I X X X 
Crater Lake W W X E X X E X X E X W W W X X W E X 
I 0 
\ 
152 
1 \ ) 
Monument Peak 
B'achelor Mt. X X X X X X X X X X X XX 
Three Pyramids X X X X X X X X X 
Crescent Mt. X X X X X X X X X X X 
Echo Peak X X X X X X X X 
South Peak X X X X X X X X 
Cone Peak X X X X X ' X X X X X 
Iron Mt. X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X 
Jumpoff Joe X X X X 
· Squaw Peak X X X X r 
Twin Buttes X X X X X X X X ·X 
Carpenter Mt. X X X X X X X X 
Tidbits Mt. X X X X X X X X X X X X X X X 
Lookout. Mt. X X X X X 
Frissell Pt. X X X X X X X X 
Castle Rock X X X X X X X X X X X 
O'Leary Mt. X X X , X ·x X 
Horsepasture Mt. X X X X X X X X X X X X 
English Mt. X X X X X X X X 
Lowder Mt. X X X X X X X X X 
Tipsoo Butte X X X 
Olallie Mt. X X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X X 
Bohemia Mt.(reg.) X B X X X B X B X B B X X 
Hershberger Mt. X X X X X X 
Grizzly Peak X X X X X X X X X X X X 
Maxwell Butte X X X X X X X X 
Little Nash Cr. X X X X 
Hogg R.ock 
Hoodoo Butte X X X 
Sand Mt. X X 
Three Sisters X X X X X I X X X I I X X I I X X X 
Crater Lake X X X X E E E E X W X W X W W X W E X 
153 
Monument Peak X X X X A X X X X X A 
Bachelor Mt. X X X X X X X X X 
Three Pyramids X X X X X X X X X 
Crescent Mt. X X X X X 
Echo Peak X X X X X X 
South Peak X X X X X 
Cone Peak X X X X X 
Iron Mt. X X X X X X X X 
Browder Ridge X X XX' X X X 
Jumpoff Joe X X X 
Squaw Peak X X X 
Twin Buttes X X X X X 
- Carpenter Mt. X X X X X 
Tidbits Mt. X X X X X X X X X X 
Lookout Mt. X X X ·x 
Frissell Pt. X X X X X X 
Castle Rock X X X X X X X X X X 
0 'Leary Mt. X X X X X 
Horsepas ture Mt. X X X X X X X X X 
English Mt. X X X X X X 
Lowder Mt. X X X X X X X X X X X 
Tipsoo Butte X X X X 
Olall ie Mt. X X X X X X X 
Rebel Rock X X X X X X X X X X X 
Indian Ridge X X X X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X X X X X 
Bohemia Mt. (reg. ) X B X X X B X X B B X X 
Hershberger Mt. X X X X X X X X 
Grizzly Peak X X X X X X X X X X 
Maxwell Butte X X X X X X X X X 
Little Nash Cr. X X X 
Hogg Rock 
Hoodoo Butte X X X X 
Sand Mt. X 
Three Sisters XXXXI XX X X X X X X X X X X I X 
Crater Lake W W E W E X X W XX X W X E X X X 
154 
\ 
,. 
Monument Peak 
Bachelor Mt. X X X X X X X X X X X 
Three Pyramids X X X X X X X X X 
Crescent Mt. X X X X X X 
Echo Peak X X X X X X X X X X X X X X X 
South Peak X X X X X X X X X X X X X 
Cone Peak X X X X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X X 
Jumpoff Joe X X X X 
Squaw Peak X X X X 
Twin Buttes X X X X X X X X 
Carpenter Mt. X X X X X 
Tidbits Mt . X X X X 
Lookout Mt X X X X X 
Frissell Pt. X X X X X 
Castle Rock X X X X X X 
0 'Leary Mt. X X X X X X X X X 
Horsepasture Mt. X X X X X X X ·x X ·x X X X X 
English Mt. X X X X X X X 
Lowder Mt. X X X X X X X X X X X X 
Tipsoo Butte X X X X X 
Olallie Mt. X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X X X X X 
Indian Ridge X X X X X 
Sardine Butte X X X 
Huckleberry Mt. X X 
Bohemia Mt.(reg.) X X X X X X X X X X B X X B B X X 
Hershberger Mt. X X X X X X X X X X X X 
Grizzly Peak X 
Maxwell Butte X X 
Little Nash Cr. X X 
Hogg Rock 
Hoodoo Butte X 
Sand Mt. X X 
Three Sisters X X X X X X X X I X 
Crater Lake W XX X w W X WE X X E X w 
CJ) 155 
. :, 
ii 
p 
m 
m j ~ .,s t \ l \ ..;  m , 1 L ~ x-+ \ .,..;.,..;_ J 
CJ) \ 1-<m ·,..; -,..; ---r--"f ) .o ~) ~ _} ) ",..;...-1U) 
~ Ul ) 0 H .,..; .,..; " CJ) U):J "00~ .,..; .,..; ...-1 0 ...-1 p m:i · Ulm:i.µ -Y w .Y. <llm...-1~8 
\ .j.J,,"I CJ) U)~...-1...-1 0 ...-1 CJ) CJ) :i ...-1 u m CJ) m L,,1/ .1 w P o .,..; µ 
w l mUl:J:i·,..;~wU) 0 0 :, H O :, ...-1 :, H °J Ul H '1l •.-,...-1 0 
'1l 1-<'1!1-<Ul"O•,..;~"O u1-<.µwum:,.µw~~:,:,u~...-1~ 
>.m 0...-1 o O..P :>"O·,..; .,..; o Ul u-,..;...-1 ,_. m~ ,-..,..; .µ o..•,..; Ul >.·,..; 
W"O"O ol),-,1 mm H H :> o..S:JOPMH~~WUlml-<1-<80..p 
...-1-,..;o:i~..c:1-<mmm :, WW H .,..; :, W <.J Ul ~ .,..; .µ:, W µ H W 
OM o,,..;.µMO..<.J"O I-< P"O O..<.J W Ul Ul...-1 W ~.µ O..S OW I-< 
o-,..;m"OP 0 :, H W :, m ~ Ul 
u1-<...-1 :iSmmPP PPPPPPPS0...o...-1o0Ul Ul 
...-1 m :i .,..; .,..; o o o o o o o o o .,..; mm m .,..; 
CJ) CJ) w •. ..., CJ) <.) CJ) CJ) ~ ~ S S S S S S S Ul Ul Ul Ul Ul ...-1 u u · u 1-< 
>. >, "O w :, CJ) p · p ... ... w w w w w w w :i :i :i :i :i m .,..; .,..; .,..; >. 
..C:..C: H H o..m•,..;•,..;.µ.µ .µ.µ.µ.µ.µ.µ.µ...-1...-1...-1...-1...-1-1,Jppp..C: 
(.)(.)(1!:J0.0...-1...-IU)U) Ul Ul Ul Ul Ul Ul Ul:,:,:,:,:, .,..; 0 0 0 .µ 
m<11p-1,J<.J1-<...-1...-1pp P P P P P P P S S S S S oO H H H P 
.µ.µom>.wooww wwwwwww~~~~~~www>, 
Cl)Cl)~Cl),_;jp,UUP.,P., p.,p.,p.,p.,p.,p.,p.,~~~~~O>>>C/) 
Monument Peak X X A X X X X X X X X X X A 
Bachelor Mt. X X X X X X X X X 
Three Pyramids X X X X X X X X X X X X 
Crescent Mt. X X X X X X 
Echo Peak X X X X X X X X X X X X X 
South Peak X X X X X X X 
Cone Peak X X X X X X X X X X X X X X 
Iron Mt, X X X X X X X X X X XQ X X X X 
Browder Ridge X X X X X X X X X X X X 
Jumpoff Joe X X X XX X 
Squaw Peak X X X 
Twin Buttes X X X X X X X X .x 
Carpenter Mt. X X X X X X X X X 
Tidbits Mt. X X X X X X 
Lookout Mt. X X X X X X 
Frissell Pt. X X X X X X X X 
Castle Rock X X X X X X X X X 
0 'Leary Mt. X X X X X X X X X X X X 
Horsepas ture Mt. X X X X X X X X X X X X X X 
English Mt. X X X X X X 
Lowder Mt. X X X X X X X X X X X 
Tipsoo Butte X X X X X 
Olallie Mt. X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X X 
Indian Ridge X X X X X X 
Sardine Butte X X X X X X 
Huckleberry Mt. X X X 
Bohemia Mt.(reg.) X X X X X X X X X X X B X B 
Hershberger Mt. X X X X X X X X X 
Grizzly Peak X X X .x 
Maxwell Butte X X X X- X X 
Little Nash Cr. X X X X X 
Hogg Rock X 
Hoodoo Butte X X X X X 
Sand Mt. X X 
Three Sisters X IX XX X XX X X I X I X IX X 
Crater Lake E X E X W XX E X X X X X W X W 
I 
Monument Peak X X X X AX X X X 
Bachelor Mt. X X X X X X X X X X 
Three Pyramids X X X X X X X X X X X X 
Crescent Mt. X X X X X X X 
Echo Peak X X X X X X X X X X X 
South Peak X X X X X X 
Cone Peak X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X 
Jumpoff Joe X X X X X 
\ Squaw Peak X X X 
Twin Buttes X X X X X 
r _,,. 
Carpenter Mt. X X X X X 
Tidbits Mt. X X X X X X X X X 
-Lookout Mt. X X X X X X 
) 
Frissell Pt. X X X X X X X 
Castle Rock X X X X X 
0 'Leary Mt. X X X X X X X X X 
Horsepasture Mt. X X X X X X X X X X X 
English Mt. X X X X X X X 
, Lowder Mt. X X X X X X X 
Tipsoo Butte X X X X X X X X 
Olallie Mt. X X X X X X X X X 
Rebel Rock X X X X X X X X X X X X X 
Indian Ridge X X X X X X X X 
Sardine Butte X X X 
Huckleberry Mt. X X X X X X 
Bohemia Mt.(reg.) X X X X X X B B X X X X X B X X X 
Hershberger Mt. X X X X X X X X X X X 
Grizzly Peak X X X X X X 
Maxwell Butte X X X X X X X 
Little Nash Cr. X X X X 
Hogg Rock 
Hoodoo Butte X 
Sand Mt. 
Three Sisters X X X X X I X I X X X X I X I X X 
Crater Lake X X X X WW X W W X X \'.l E X X W W X 
'-: 
157 
\ ~ J 00 
} .j.J co ::, 
CO 00 .-I 
co bJJ O \ .-I 
H •.-l H j >. 
i!!1L1 1~ 13~ ~'~·! 1 l ~ 
0;:J.µ •.-l•.-lcll •.-lO OW 1-< ~ >,•.-lOO O .-1 ) , ,.; \ 
.-1..ccooo.-100.j.J H..C:•.-l4-<00 (l.l•.-l(l.1(1.l•M ) H bl) 00 ) 
.-11-<•r-lcllPOO cll.µH•.-l;:J S•r-lppoo 4-< J oo . ~ 
oocoUOO•.-l(l.1(1.1 .-lpW'OCO Q,-1.j.J(l.lp oo::, =(1.1 
•.-l ;:JpbJJ..C:bJJ ;:J(U.-!pp 0.-1•.-l(l.l(l.lOOOOOO•.-l.-l 'OCO 
.-100.-l(l.l;:JUpOOUp::,::, .-ICO..C:H'O::>::>::>OO.-l~(U(l.l 
CO O O ..C: .,...., .µ 0 ::, QJ (1.1 0 .µ 00 00 ..C ..C: ~ bl) CO P W P P >. •.-l U U 
WO..:>nlp•.-lUP0..1-<UO;:J•.-l pWU•.-lW..C:POO•.-l 
1-<1-<p.µ000 coooo..001-<p.-1oooooooocobJJco.-1.µo..n,co.-1 
0 CO 'M::, U 00 bl) E:::, ;:J::,::,::, U •.-l 'M ;:J 00 0 6 UCO 
..CU nl~QJOOnlcllnlnlnl0..0..0..0.. 0...-l'OCO'O~ 
~n,n,n,p.µ1-<~.-1.-1.-1..c:uo..o..o..o..o.-10ni..c:wopp 
CO H H H H co •.-l O 00 ::, ::, ::, .µ •.-l CO CO CO CO bl) CO 4-< H 00 .-I bl) 0 0 
WOQJWQJ~l-< O..PPPO..C0..0..0..0..CO 1-<H 
nl,.C: U U UH .µ..C: 0 CO co CO 00.-1 0 0 0 0'0 H H H H H H WW 
i:Jo..~~~wuni..c:o..o..o..>.ni.-1.-1.-1.-1~waJwwwaJMbJJ 
PEPPP.-!W1-<.µSSB1-< 0..0..0..0...-1.µ.µ.µ.µ.µ.µ~~ 
•.-l :,:., 0 0 0 CO .-I CO •.-l CO CO CO ..C: CO CO CO CO O 00 00 00 00 00 00 H H 
~~~~~>~~0UUUU XXXX~<< -<<<<~~ 
Monument Peak X X X X A x\ X X 
Bachelor Mt. XX X X X X X 
Three Pyramids X X X X 
Crescent Mt. X X X X 
Echo Peak X X X X X X X 
S9uth Peak X X X X 
Cone Peak X X X X X X X X X 
Iron Mountain X X X X X X X X X X X X 
Browder •Ridge X X X X 
Jumpoff Joe XX X X 
Squaw Peak X X 
Twin Buttes X X X X X X X 
Carpenter Mt. X X ){ X 
Tidbits Mt. X X X X X X X 
Lookout Mt. X X X 
Frissell Pt. X X X X X X X 
Castle Rock X X X X X X 
0 'Leary Mt. X X X X X X X 
Horsepasture Mt. X X X X X X 
English Mt. X X X 
Lowder Mt. X X X X X X X 
Tipsoo Butte X X X 
Olallie Mt. X X X X X 
Rebel Rock X X X X X X X X X X X X X X 
Indian Ridge X X 
Sardine Butte X 
Huckleberry Mt. X X 
Bohemia Mt,(reg.) B X B X X B B X X X X X X 
Hershberger Mt. X X X X X X X X X X 
Grizzly Peak X X X X X X X 
Maxwell Butte X X X X X X X X 
Little Nash Cr. X X 
Hogg Rock 
Hoodoo Butte X X X X 
Sand Mt. X X 
Three Sisters X X X X X X X X X XIXXIIX X 
Crater Lake E W X E X X E W X X X X X X X 
158 
m 
; \ ~ \ ,g 
µ 0 M 00 
m <.) M f3 •M ;:J 
t·8 \ s ) a~§~~ 
m•M ::, . m"OQJM \ S 
M~OO M m 00 M~~ ) M 
MP~ m QJ ~ \ oµm m 
oM \ ~ u M J s·MPPmo. 
oou•M o. mm 1-<m·M f:l::,1-<mmµ -'5' 
::, ,0 QJ µoo Oµ•M ;:J•MmU•MmOO 
µOOQJOO<.) ~0 MP~ µM<.);:J<.)µ:::, mMm 
•M ;:J "O "M O I-< f3 mQJ ~ ~ .~ j m O "M QJ "M P "O \ m •M O "M 
0000 OOH mQJ vv~ p~l-<M>QJ~ ) ~M~M 
0 0 00 p <.) M <.) 00 "M •M 00 00 m QJ µ 0 "O M M O "M 0 
0. •M"M QJ •M I-< m Q,O u ;:J•M ' MM m S"O•M"M m•M 0~ 00~ 
SMl-<"OS cdl-<1-< <.)<.)Mm Mf3;:J;:Jl-<1-<µ>,~ •MH"M 
ooum QJS po ~::,s~ SMµ~u~~MQJ"O 
u~mpf3M mmO oouM::JE:mQJ •Ml-<µP>I-< 
m::,moo~~Mm;:Jm~M ~~mmoo1-<mmO~O 
p p p <.) •M Ei "M I-< I-< ;:J "M ~ I-< X M m H µ •M •M QJ µ 0. M M "O <.) 
ooo M~Mmmm~µMQJ>,QJmpooooµoo 
I-< I-< I-< cd cd QJ cd p P <.) <.) P ~ M <.) m "M "M "M mm cd cd cd 
QJQJQJN~~~ppOQJcommO.M~oossoocouuuuu 
MMM>.O.µO.QJQJp,O~~~O~J-<>,QJQJmp~~~~~ 
~~~Pm mµµQ)"OM"O"O~~µi-,µµµ~ppppp 
I-< I-< I-< 0 p p p p "O ::, QJ m m I-< u m ~ I-< I-< QJ ::J I-< I-< I-< I-< I-< 
~~~U0 <<<<~~~Z~<~U<<~~<<<<< 
Monument Peak X X X X X X X X X X X X 
Bachelor Mt. X X X X X X X X X X X X X 
Three Pyramids X X X X X X X X X X X X 
Crescent Mt. X X X X X X X X X X X 
Echo Peak X X X X X X X 
South Peak X X X X X X X X 
Cone Peak X X X X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X X X X 
Browder Ridge X X X X X X X X X X X X X 
Jumpoff Joe X X X X X X X 
Squaw Peak X X X 
Twin Buttes X X X X X X X X 
Carpenter Mt. X X X X X X X X X 
Tidbits Mt. X X X X X X X 
Lookout Mt X X X X X X X 
Frissell Pt. X X X X X X X X 
Castle· Rock X X X X X X X X X 
0 '.Leary Mt. X X X X X X X 
Horsepas ture Mt •. X X X X X X X X X X X X 
English Mt. X X X X X X X X 
1..._ Lowder Mt. X X X X X X 
Tipsoo Butte X X X X X 
-Olallie Mt. X X X X X X 
Rebel Rock X X X X X X X X X X X X X X 
Indian Ridge X X X X X X X 
Sardine Butte X X· X X X 
I 
Huckleberry Mt. X X X X X X 
Bohemia Mt.(reg.) X X X X B X · X X X X X X 
Hershbe,rger Mt. X X X X X X X XX 
Grizzly Peak X X X X X X 
Maxwell Butte X X X X X X 
Little Nash Cr -. X X X 
Hogg Rock X 
Hoodoo Butte X X 
Sand Mt. x 
Three Sisters I X X X X X X I I I 
Crater Lake X E W X X W XX W E W X X E E X X X W 
J 159 
Monument Peak X X X X X X A X X X 
Bachelor Mt. X X X X X X 
Three Pyr,amids X X X X 
Crescent Peak X X X X X 
Echo Peak X X X X X X X 
South Peak X X X X X 
Cone Peak X X X X X X X X X X 
Iron Mt. X X X X X X X X X X X 
Browder Ridge X X X XX X X X 
Jumpoff Joe X X X X X 
Squaw Peak X X 
Twin Buttes XX X X X 
Carpenter Mt. X X X X X X 
Tidbits Mt. X X X X X 
Lookout Mt, X X 
Frissell Pt. X X X X X X X X X X 
Castle Rock X X X X X 
0 'Leary Mt. X X X X X X X 
Horsepasture Mt. X X X X X X X X X X 
English Mt X X X X X 
Lowder Mt. X X X X X X X X X 
Tipsoo Butte X X X X X X 
Olallie Mt. X X X X X X X X 
Rebel Rock X X X X X X X X X X X 
Indian Ridge X X X X X X 
Sardine Butte X X X X 
Huckleberry Mt. X X X X 
Bohemia Mt.(reg.) X X X X B , X X X X X X X 
Hershberger Mt. X X X X X X 
Grizzly Peak X X X X 
Maxwell Butte X X X X X 
Little Nash Cr. X X 
Hogg Rock 
Hoodoo Butte X X X 
Sand Mt. 
Three Sisters X X I I X X I X X I 
Crater Lake X X X W E X X W X X 
APPENDIX B 
Introduction 
This ap pendix contains summaries of the taxonomy and ecology of 
85 disjunct or endemic Western Cascade species and dot distribution 
maps of 79 of these species compiled from field, herbarium, and liter-
ature searches. 
Monographic references have been studied wherever possible. In 
many cases where no such works are available and where the species seem 
to comprise well-differentiated taxa, the taxonomic treatment of stand-
ard manuals (Hitchcock and others, 1959, 1961, 1964; Cronquist, 1955; 
Peck, 1961; Munz and Keck, 1959; Jepson, 1925; Abrams, 1940, 1944, 1951; 
Abrams and Ferris, 1960) is followed. In other instances . where taxa 
seem to be poorly understood, additional taxonomic studies were under-
taken on species or larger groups in order to determine and help in the 
under standing of present distribution patterns. 
Families are ordered according to the standard Englerian system 
to correspond with most manuals and herbaria. The order of genera and 
species follows Peck (1961), with the exception of the genus Polygonum, 
where species have been shuffled to achieve a more coherent discussion. 
References to descriptions of taxa cited in this appendix will not be 
listed in the bibliography. Synonymies are not complete; only names 
which. are occasionally still used for these taxa occur in the lists of 
selected synonyms. 
160 
161 
Citations of herbarium specimens follow the code of Lanjouw and 
Stafleu (1964) except for specimens from the Herbarium of the Univer-
sity of Oregon, which are not followed by an herbarium code. 
Dot distribution maps contain information from several West 
Coast herbaria, including those of the University of Oregon (ORE); 
Oregon State University (OSC); Peck Herbarium, . Willamette University 
(WILLU ) ; Dudley Herbarium, Stanford University (DS); University of 
California, Berkeley (UC); California Academy of Sciences (CAS); Uni-
versity of Washington, Seattle (UW); Washington State Univer'si ty (WSC); 
and Crater Lake National Park. Collections by the author from new 
localities are noted with circles rather than dots. Occasional speci-
mens cited in published literature or previous dot maps are also in-
cluded. An index map for the peaks of the Cascade Range included in 
the inset is found immediately following this introduction. 
. ' 
162 
MONUMENT PK, 
+ MT. JEFFERSON 
GRIZZLY PK. 
BACHELOR MT. + 
+ 
THREE PYRAMIDS 
+ 
I THREE - FINGERED JACK 
CRESCENT PK. + MAXWELL BUTTE 
+ LITTLE NASH CRATER 
~NO;TH PK. + 
CO~E PK,+ + .:ECHO PK. HOGG ROCK 
IRON, MT.+ SOUTH PK. + 
..-,- -.... + + HOODOO BUTTE 
JUMPOFF JOE + BROWDER RIDGE SAND MT. · 
SQUAW PK. 
TWIN BUTTES + + 
MT. WASHINGTON 
CARPENTER MT. 
TIDBITS MT. f) 
+ STEPTOE$ 
+ 
LOOKOUT MT.+ + FRISSELL PT. 
CASTLI:: ROCK 
+ 
O'LEARY MT.+ +HORSE\PA STURE MT. 
Tt:IREE SISTERS 
. 
LOWDER MT. +LAMB BUTTE 
TIPSOO IBU .TT E+ + + ·+ ENGLIS!-f MT. YANKEE MT. + 
INDIAN RIDGE OLLALIE MT. 
+ 
----' REBEL PK . 
SARDINE BUTTE + + 
+ REBEL ROCK 
Index Map to Western and High Cascade Range Mountains. 
163 
Polystichum andersonii Hopkins, Am. Fern Journ. 3:116, pl . 9. 1913. 
Selected synonyms: 
Polystichum braunii subsp. andersonii (Hopk.) Calder and 
Taylor, Can. Journ. Bot. 43:1388. 1965. 
This rarely collected species is readily di s tinguished from 
other ferns of the Pacific Northwest by its centrally peltate sori and 
large, doubly-pinnate fronds, the teeth of which are long-awned. It is 
closely related to f. alaskenss Maxon , a boreal spec ies; and to P. 
californicum (Ea t .) Underw. and f. dudleyi Maxon, plan ts of the Coast 
Ranges of central California. Originally described from Vancouver 
Island, the center of distribution off. andersonii seems to be in the 
northern Cascades of Washington (Thompson, 1931; Abrams, 1923). It has 
also been collected from Glacier Na t ional Park, Montana, and from three 
sites in the Oregon Cascades, two of which lie within the study area. 
A spec ies of rather obvious boreal derivation, P. andersonii is 
often found in wet soil under alder thickets. I n t he Western Cascades, 
it is also assoc iated with alder (Alnus sinua t a), but wi t h a number of 
other boreal species as well, notably Chamaecyparis nootkatensis and 
Mertens i a bella. The two Western Cascade localities are at the margins 
of springs which are wet throughout the summer. Other Wet Meadow 
species pres ent include Senecio triangularis, Ribes bracteosum, 
Oplopanax horridum, Mitella breweri, Hydrophyllum fendleri albifrons, 
Montia sibirica, Valeriana sitchensis, and Veratrum viride. 
+ • \ ----\ 7 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I + 
+ • \ \ 
+ + + \ \ 
+ 
+ \ 
+ \ 
\ • 
+ 
+ + JJ + 
I 
I 
A I 
+ I 
++ •. . I 
+ I 
+ + ++ I 
+ I 
+ I I 
+ ++ I 
I 
110 mi. 
GEOGRAPHIC DISTRIBUTION OF 
POLY STI CHUM ANDERSONII 
I-' 
---- a, ---- .c:--
165 
Cheilanthes siliguosa Maxon, Am. Fern Journ. 8:116 . 1918, 
Selected synonyms: 
Onychium densum Brack. in Wilkes, U. S. Expl. Exped, 16:120. 
1854. 
Pellaea densa Hook., Sp. Fil. 2:150, pl. 125. B. 1858. 
Cheilanthes siliguosa is immediately recognizable by its small 
ovate fronds, marginal sori, the inrolled edges of the narrowly lan-
ceolate leathery pinnules, and the lustrous brown stipes.. Its. center 
of distribution is in the Klamath Region of southwestern Oregon and 
-,. 
northwestern California, where it is typically restricted to serpentine 
rocks and soils. It also is found in serpentine habitats in the 
Wenatchee Mountains of central Washington (Kruckeberg, 1964) and in 
I 
Quebec but occurs in scattered non-se~pentine sites throughout much 
of the Western United States. Evidently this species has responded 
adaptively to different soil types. 
_£. siliguosa is common in the Western Cascades, where it grows 
in light brown rocky loams on steep open ~lopes. It is typically found 
under the lower side of a boulder that is ; creeping down slope. Whether 
the boulder provides a favorable habitat for the gametophyte or whether 
the boulder is simply arrested in its downhill slide by the dense root 
and rhizome system of the fern remains open to question, Other ferns, 
especially Cheilanthes gracillima and Cryptogramma acrostichoides have 
been noted under boulders on well stabilized slopes, suggesting that 
the former hypothesis may be correct . _£. siliguosa is also found rooted 
in crevices of scoriaceous basalt or andesite with such members of the 
Outcrop Ridge association as Delphinium menziesii pyramidale, Castilleja 
+ \ 7 
\ 
C!) • + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
0 \ 
\ 
+ + \ + \ 
+ 
+ \ 
+ 
-+ 
+ • + 
• I A / 
A / • 
A. I • • 
I 
I 
+ I 
I 
I 
I 
I • I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
CHEILANTHES SILIQUOSA I~ 
f-----"'------ °°''  
167 
hispida, Sedum oregonense, Sedum stenopetalum, Sedum divergens, Haplo-
pappus hallii, Calochortus lobbii, Collinsia parviflora, Lomatium 
martindalei, Erigeron foliosus confinus, and Penstemon procerus brachy-
anthus. 
Selaginella scopulorum Maxon, Am. Fern Journ. 11:36. 1921. 
The taxonomy of low matted selaginellas is in need of much basic 
morphological and experimental work. Populations are extremely variable, 
and the methods of sexual reproduction in this group promote random 
fixation of adaptively neutral genes in small populations. In addition, 
there is strong selection for ecotypic differentiation in the extreme 
environments which these plants occupy. The morphological type here 
recognized as.§_. scopulorum is distinguished from the other widespread 
Oregon matted Selaginella, .§_. wallacei Hieron., with which it often 
occurs, by its thicker, more compactly leafy stems; dense, club-like 
strobili; and larger, obliquely imbricate leaves, which bend up from 
the side of the prostrate stem closest to the ground. Typical S. 
wallacei gives a much more wiry appearance, with smaller, more closely 
appressed leaves. Ciliation of the vegetative leaves is variable in 
both types and does not aid in their segregation. Although it is pos-
sible that material here referred to S. scopulorum should be considered 
an extreme form of.§_. wallacei, Western Cascade material compares favor-
ably with specimens from other localities annotated or cited by Maxon 
(Wheeler, 1937). It is relatively common in the Western Cascades on 
the summits or high outcrops - of many of the peaks. It has been collected 
____ T ____ 
+ A \ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
e \ 
A 
+ \ 
+ + \ + • \ 
+ 
. + \ 
+ \ 
\ 
+ + 
+ ;p 
+ + 
+G 
I 
A I 
+ A I 
+E> .. I 
+ I 
CD+ + 
+ I 
+ I 
I 
+ I 
+ .,e I 
I I 
11omi. I 
GEOGRAPHIC DISTRIBUTION OF I 
SELAGINELLA SCOPULORUM I~ 
---- 0------ 0) 
169 
from scattered localities in much of northwestern North America, espe-
cially the Rocky Mountains, but in Oregon, specimens have heretofore 
been taken only from the Wallowa Mountains and the Siskiyou region. 
All of the matted selaginellas are pioneer species with mosses 
and lichens on. nearly bare rock. They are most abundant where outcrops 
are close to the surrounding ground level or are otherwise disposed to 
afford some protection from drying winds. Few other plants besides 
moss species are associated with Selaginella in the Western Cascades. 
Pinus ponderosa Dougl. in Lawson, Agric. Man. 354. 1836. 
In the Pacific Northwest Pinus ponderosa is mostly restricted 
to a narrow, horseshoe-shaped band around the edge of the intermountain 
province (Detling, 1968). It is also common in the Siskiyou region of 
southwestern Oregon, through the Sierra Nevada, and in isolated small 
stands in the Willamette Valley and Puget Trough. It is the most char-
acteristic tree of the Arid Transition Zone, and typically forms open, 
park-like forests (Abrams, 1923). Except where it overlaps with P. 
jeffreyi Murr., ponderosa pine is easily recognized by its reddish 
platy bark, three-needled fascicles, and sharply pungent cone scales. 
Detling (1953) has noted many typical associates of ponderosa 
pine in "xeric islands". in w.estern Oregon, and finds it difficult to 
explain the absence of the tree from th' ese areas. There is• no reason 
to assume that the Arid Transition flora would have migrated as a unit, 
but there are now two localities for native pondero$a pine known from 
the "xeric is~ands" in the , Western Cascades. One is Iron Mountain, 
170 
where a single stunted, but old specimen was found growing in fine 
gravel scree on a southeast-facing slope. The other has been reported 
by J. F. Franklin (personal communication, 1967) from Frissell Point, 
where three specimens were found on rocky, south-facing slopes. These 
occurrences are noteworthy, but no map of the total distribution of 
P. ponderosa is given here. 
Chamaecyparis nootkatensis (Lamb.) Spach, Hist. Veg. 11:333. 1842. 
Selected synonyms: 
Cupressus nootkatensis Lamb., Pinus 2:18. 1824. 
Alaska cypress is normally an easily recognizable tree. The 
small peltate cones; scarcely flattened, slightly rough branchlets; and 
lack of glands set it off from other members of the Cupressaceae. From 
a di stance it can often be confused with Thuja plicata D. Don, which has 
similar drooping branches and shredding bark. However, in most speci-
mens of C. nootkatensis the branch tips curve up again, while those of 
Thuja do not. Alaska cypress is a self-layering species which repro-
duces vegetatively more often than sexually, occasionally establishing 
large clones of trunks from a single specimen. All seed collected from 
this species has failed to germinate even after long periods of strati-
fi cati on. In windswept rocky habitats branches may trail horizontally 
over the rocks for a distance of 20 m, producing numerous small vertical 
side branches. In southeastern Alaska the species is found from tide-
water to timberline, usually with Tsuga mertensiana, and has been known 
to reach a diameter of 2 m and a height of 40 m, although su.ch stature 
cC • A \ ----- 7 \ 
+ · + • \ I \ 
3mi. \ I 
. 1--'-----i 
\ 
\ I 
\ 
\ I 
0 \ 
A 
+ \ 
.p + \ + 
~Bo \ + + \ 
t9 -0 
+ 
+ • 
+ 'P 
•+Q 
I 
A I 
+ A I 
+ () A I 
0 I 
++ e I 
+ I 
I 
I 
++ I 
I I 
11omi. I 
GEOGRAPHIC DISTRIBUTION OF I 
CHAMAECYPARIS NOOTKATENSIS I~ 
---- -...J ----- ~ 
172 
is unusual (Heusser, 1960). It ranges from the Kenai Peninsula south 
along the coast of British Columbia and through the Cascades to the 
Klamath Mountains of Siskiyou County, California. It is highly disjunct 
in the southern part of its range, typically occupying exposed, wind-
swept ridges at high altitudes from which other conifers are excluded. 
Occasional clones are found beneath th~ Mesic Forest canopy. Popula-
tions are scattered but not uncommon in the Western Cascades. Although 
most trees are rather scrubby, one stand on Twin Buttes contains trees 
over 1.5 min diameter and over 30 min height. The tops were broken 
out of these trees, which were the largest of any species in the imme-
diate area. A single stand of large trees has been found in eastern 
Oregon near Fields Peak, Grant County (Cronquist 7646: DS). 
Allium crenulatum Wieg., Bull. Torr. Bot. Club 26:135, pl. 355. 1899. 
Selected synonyms: 
Allium cascadense Peck, Proc. Biol. Soc. Wash. 49:109. 1936. 
Allium watsonii Howell, Fl. N. W. Am. 1:642. 1902. 
Low onions growing on dry barren summits in western Oregon pre-
sent difficult morphological and taxonomic problems (Ownbey, personal 
communication, 1968). Typical!• crenulatum from the Olympic Mountains 
and!• siskiyouense Ownbey constitute the ends of a north-south cline. 
Intermediate specimens have been segregated as A. watsonii and!·~-
cadense. All of the numerous isolated populations of this complex show 
some distinc tive morphological differences, produced either by local 
selective factors or by genetic drift . Ownbey, who has examined the 
author's materi11 ' from the Western Cascades and has verified their 
• .& \ ---- 7 \ 
+ 
c!> \ I 
\ 
3mi. \ I 
1-------1 \ 
\ I 
---- \ 
\ I 
+ \ 
.& 
+ \ 
+ + \ + 
+ 
+ 
+ + 
+ ;p 
~ + 
+~ I 
I 
A I 
+ .& I 
++ A I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
ALLIUM CRENULATUM '~....,  ---- ------- w 
174 
identification, concludes that the pr oposed segregates cannot stand as 
distinct species, or even as varieties, and has broadened the concept 
of A. crenulatum to accommodate them. ~- siskiyouense at present 
remains a distinct taxon. A. crenulatum is characterized in general 
by lack of bulbcoat reticulations; two long, flattened leaves wh~ch 
exceed the flowe r ing stem in leng th and are normally somewhat falcate; 
and a flattened, somewhat winged upper scape. Clos~ly related species 
include~- tolmiei, a species restr icted to the eastern slopes of the 
Cascades and eastward. 
In the Wes t ern Cascades, A. c r enulatum is a member of the Fine 
Gravel Scree association, occupying sites which have abundant moisture 
during the first part of the growing season but desiccate by midsummer. 
These are normally open, south-facing slopes which are subject to con-
siderable frost heaving and have little vegetative cover. It is commonly 
associated with Lotus nevadensis douglasii, Aster gormanii, Trifolium 
productum, Crepis occidentalis, Ivesia gordonii, Gilia aggregata, Sedum 
oregonense, and Linanthus harknessii. 
Populus tremuloides Michx., Fl. Bor. Am. 2:243. 1803. 
Quaking a s pen is a characteristic Canadian Zone tree throughout 
North America. Although taxonomy in Populus is generally difficult due 
to extensive hybridization, this species is easily recognized by smooth 
white or yellow bark; oblong-conical capsules; and broadly ovate, finely 
serrate, small leaves with strongly f lattened petioles. It is found 
along streams and in other moist habitats and is closely related to, 
but di s tinct f r om, f. tremula L. of Eurasia. This species. is known from 
175 
only one population west of the Cascade crest. Thirteen large trees, 
averaging over 100 years of age, are found in the wet clay loam of 
Quaking Aspen Swamp, in the French Pete drainage. No map of the total 
distribu~ion off. tremuloides is given here, but this single locality 
within the study area is noteworthy. The trees are all of about the 
same age, although the heartwood of many is rotten, and exact ages 
cannot be determined. They are not reproducing by seeds, but numerous 
stunted root s prouts are found around the bases of the trees. The 
trees are located on somewhat higher ground at the west end of the 
swamp and are surrounded by such Mesic Meadow species as Mertensia 
pani culata , Senecio triangularis, Rubus parviflorus, and Rudbeckia 
occ identalis. Within a few meters to the east are Wet Meadow species, 
inc luding Veratrum viride, Ligus ticum grayi, Mer tens ia bella, Habenaria 
dilatata, Mitella breweri, and Kalmia polifolia. 
Populus trichocarpa T. and G. ex Hook., Icon . Pl . 9: pl. 878. 1852. 
Selected synonyms; 
P. balsamifera L. var. cali fornica S. Wats., Am. Journ. Sci. 
115:135. 1878. 
f. trichocarpa, or black cottonwood, is the western form of P. 
balsamifera L. , a characteristic tree of open environments along streams 
in northeastern North Ameri ca. Both species have rough yellow-green or 
gray bark, but f. trichocarpa is differentiated from all other closely 
related poplars by its densely hairy capsules. Widespread throughout 
western North America, this species is normally confined to lake margins, 
stream banks, and other sites with abundant moisture. It is especially 
176 
abundant along the rivers of west~rq Oregon up to elevations of about 
450 m. Specimens are not uncommon along the South Fork of the McKenzie 
River up to 900 m, and one small population has been located in evi-
dently dry, blocky talus of a small, east-facing cirque on Olallie 
Moun t ain ridge at more than 1500 m. This group of relatively large 
trees (about 20 m) was growing with Sambucus caLlicarpa, Abies procera, 
Epilobium angustifolium, Actaea rubra, and Anaphalis 1mprgaritacea. 
Al t hough this constitutes a notable entry of a mesic lowland species 
into t he high Western Cascades, no map of the total range of P. tri-
chocarpa is presented here. 
Quercus garryana Dougl. ex Hook., Fl. Bor. Am. 2:159. · 1838. 
The only white oak of the northern half of Oregon is normally 
restricted to dry habitats in the valleys west of the Cascade Range 
from Vancouver Island to California • .9.. garryana also is found in the 
Columbia Gorge and north and south from its eastern end to Yakima, 
Washington, and Warm Springs, Oregon, in the Columbia Basin grassland 
and juniper scrub (Ornduff and French, 1958). Isolated individuals or 
small stands of oaks have been discovered in clearings in the Mesic 
Conifer F;rest in the Western Ca~cades at elevations of about 700 m; 
the summit of Castle Rock (1150 m) supports a grove of moderate-sized 
trees (6-10 m); and a population of stunted individuals was discovered 
near Musick Mine, Bohemia Mountain, southern Lane County, at an eleva-
tion of over 1500 m. Although no map of the distribution of .9.. garryana 
is presented in this work, these disjunct occurrences of a common low-
land species well into the mountains ar~ noteworthy. 
177 
Polygonum spergulariaeforme Meisn. ex Small, Bull. Torr. Bot, Club 
19:366, 1892. 
Polygonum douglasii Greene, Bull. Cal. Acad, Sci, 1:125. 1885, 
Polygonum minimum Wats., Bot, King Exp. 315. 1871, 
Polygonum cascadense Baker, Madrono 10:62. 1949. 
The genus Polygonum is of special interest in this study, It 
is the most highly diversified genus occurring in the Western Cascades 
of Oregon: t he nine species present include members of three subgenera 
which are found in markedly different habitats, Both widespread and 
narrowly restricted species occur together in the Western Cascades, 
Comparative examination of their biologies is proving of help in inter-
preting disjunctions and endemism in the area. 
Further comments will center on members of the subgenus Duravia: 
small-flowered annuals which are sometimes spring ephemerals, but more 
often persist into the driest parts of the summer. Anthers often 
dehisce before the flowers open, indicating modal autogamy. Most 
species are represented by typical material, but many intermediates 
are known. This indicates that allogamy and interspecific hybridiza-
tion occur at frequent intervals. Breeding barriers are primarily 
autogamy and phenological and ecological differentiation, 
The taxonomic problems in Duravia attain considerable propor-
tions when considered over the total range of the species involved. 
Intergrades in as many as five separate directions from a single 
"specie s " may occur, clinally connecting species which are perfectly 
distinct where they coexist in other regions. 
178 
Oddly, there are no taxonomic problems in Duravia as it is found 
in the Western Cascades. The various species are here morphologically 
and often ecologically quite distinct, and few intermediate forms have 
been observed. The most striking feature of the group in this region 
is the ability of such closely related species to coexist in small 
areas. Occasionally four distinct species may be found within one 
square meter. 
Duravia is divisible into relatively coherent groups having 
app r oximately the status of sections. Figure 11 presents the author ' s 
conception of the composition and inter-relationships of the groups 
and the species comprising them. Heavier connecting lines indi cate 
mor e complete morphological in t ergradation . These groups will be dis-
cussed separately in the following pages. 
R_. spergulariaeforme, whi ch forms a major plexus of intergra-
dation within Duravia, is normally found on low rocky or gravelly hill-
sides at low elevations in the valleys west of the Cascades wi t h other 
members of the Lowland Xeric Meadow association. It t ypically blooms 
after many of its close relatives have set seed and withered, hence the 
common names "fall knotweed" and "farewell to summer." Flowers are 
large for t his group, often 7 mm in diameter, and are frequently visited 
by small flies and beetles. R_. spergulariaeforme is the only species 
of Du r avia known by the author to have animal pollinator s . This species 
and R_. majus (Meisn.) Piper are the only two species in this section 
with stamens equalling the perianth tube or exserted and are tradition-
ally differentiated only by the abrupt downward flexion of the pedicels 
shor t ly af t er anthesis, a character shared by R_. majus, P. douglasii, 
CASCADENSE MINIMUM 
/ ~ 
PARRY I NUTTALLII AUSTI NIAE 
~ 
CALI FORNI CU M ~ 
~ ? '----- . 
HETEROSEPALUM ------ SPE RGULAR I AEFORME MA JU S 
"'?" ------? ,~ / 
WATSON II SAWAT CHENSE ---DOUGLASI I 
KELLOG GI I 
I 
CON FE RT I FLOR UM 
I 
POLYGALOI DES 
' ' 
Figure 11. Relatiom,hips of Species of Polygonum , Subgenus Duravia ( line width indicates 
t--' 
- degree of relatedness ). ........ 
\.0 
180 
and P. austiniae Greene. The most recent student of this group has 
identified specimens with a single reflexed pedicel (and otherwise 
indistinguishable from f. . s pergulariaeforme) as P. majus (Coolidge, 
1964). The latter species is normally confined to areas east of the 
Cascades, but the two co-occur in the Columbia Gorge and the Klamath 
Mountains of Oregon and California. While concurring with Coolidge 
that these species should be considered dist i nct taxa, at least at 
the subspecific level, I disagree with his manner of segregating them. 
In this case, phenology and geography must be considered at least as 
important as the variable morphological trait of pedicel reflexion. 
P . spergulariaeforme also merges freely with f.. douglasii, the 
most widespread species of Duravi a and the only member found in the 
central and eastern United States. The two species are ecologically 
r-
and phenologically separated where they occur in close proximity in the 
Western Cascades. One variable series of intermediates between P. 
douglasii and f.. spergulariaeforme, which ranges through the inter-
mountain regions of the Western United States, has been segregated as 
P. sawatchense Small. Coolidge has proposed a further segregate, P. 
triandrous Cool., for the slightly less widespread three-anthered phase 
of P. sawatchense. The latter species typically has eight anthers, 
although t he present author has discovered a number of specimens having 
five anthers. Anther number, while important elsewhere, is obviously 
of dubious taxonomic value in this portion of Duravia. 
Leaf, flower, and achene size separate f. . douglasii from P. 
majus. These are variable characters, and the whole range of inter-
mediates is known. Geographical distinctions between these two species 
181 
are not clear, except that P. douglasii is common west of the Cascade 
crest, while f. majus is rare in this region. P. austinae has broader 
leaves than P. douglasii, crowded at the base of the stem, with an 
almost leafless terminal spike of flowers. It is confined to eastern 
Oregon in its pure form, but is partially represented in the Western 
Cascades by material transitional to P. douglasii from Bohemia Mountain 
(Hickman 0-120). 
In yet another direction, R_. spergulariaeforme tends toward 
forms with smaller flowers and leafier stems, the leaves being broader 
as well as more numerous. These have been segregated as P. nuttallii 
Small, a poorly-collected or rare species that is found through the 
Cascade region of Washington and Oregon. Although it has not been 
found in the Western Cascades, its leaf morphology and long internodes 
indicate that it is a transitional form between P. spergulariaeforme 
and two species which are common there, P. cascadense and P. minimum. 
Both are separable from f. nuttallii on the basis of leaf form, P • . 
minimum having broadly elliptic to almost orbicular leaves, and P. 
cascadense having more narrowly oblanceolate, obtuse, inrolled leaves 
and more flowers per axil than P. nuttallii. P. minimum is widespread, 
occurring in montane habitats throughout the Western United States. It 
is almost ubiquitous in high elevation moist to dry meadows in the Cas-
cade Mountains. f. c~scadense is a highly restricted species (Baker, 
(1949b) presently known only from the region of Crater Lake north to the 
Echo Mountain ridge in Linn County, and is confined, except for one 
locality from the southwestern boundary of Crater Lake National Park 
(Anderson and Simpson 116: OSC), to the central Western Cascades of 
+ • \ ---- 7 \ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ 
+ • \ \ 
+ + \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
++ I 
..•  1•• I 
++ A I • • 
+ I • 
++ + 
+ I 
+ I 
I 
+ 0 
0 ++ I I 
I I 
11omi. I 
GEOGRAPHIC DISTRIBUTION OF I 
POLYGONUM SPERGULARIAEFORME I~ 
---- 00 ---- N . 
+ • \ ---- 7 \ 
+ + \ I 
\ 
3mi. \ I 
\ 
- \ I \ 
\ I 
+ \ 
A 
+ \ 
+ + \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
++ I 
I 
A I 
+ A. I 
++ A. I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
++  I 
I I 
11omi. I 
GEOGRAPHIC DISTRIBUTION OF I 
POLYGONUM MAJUS 
---- '~co ---- lu  
e •• \ ---- 7 \ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
0 \ 
+ \ 
+ \ + 
+ 
+ 
+ + 
(!) 
+ + 
0 
(!)9 • •A  
I 
-oe A I 
.p + I + 
++ I 
_p · ; 
I 
+ I 
Gd" I 
I I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
POLYGONUM DOUGLASII 
e I~ 0:, ----- ------ ~ 
+ A \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
--- \ 
\ I 
+ \ 
+ A \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + . ... • 
-+ 
+ + 
++ I 
I 
I 
+ A•  I 
++ I 
+ I 
+ 
++ + I 
+ I 
I 
+ I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I . 
POLYGONUM AUSTINIAE I~ 
---- co ---- V, 
+ • \ ------\ 7 
+ + \ I 
\ 
3 mi. \ I 
\ 
\ I 
\ 
+ \ 
I 
\ 
+ \ 
+ + + \ \ 
+ 
+ \ 
+ \ 
\ 
+ + 
+ JJ 
+ + 
++ 
I 
I 
•A  I + I ++ I 
+ I 
+ 
++ + I 
+ I 
+ I I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
POLYGONUM NUTTALLII 
---- en ------- '~°'  
0 \ ----- 7 
\ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
-0 \ 
+ \ 
0 + C \ 
o&>t o + 
e +o 
+ 0 
O+ 
6 
+ 
+ 
~ I 
+ I 
+O t> 
0 c!> I 
.p I 
I 
Cf) • I 
(!) Oot- I 
I I 
110 rni. I 
GEOGRAPHIC . DISTRI BUTION OF I 
POLYGONUM MINIMUM I~ 
o:> 
---- ----- -.J 
+ A \ ------ 7 
\ 
+ + \ I 
\ 
3mi. .\ , I 
\ I 
\ 
.. \ I + \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ + 
+ 
+ + 
++ 
.... .. 
I 
I I 
I 
+ I . I 
~  I 
+ I 
o<!> t> 
+ I 
0 + I 
I 
+ I 
ot5' I 
I I 
' ' 11orni. I 
GEO.GRAPHIG DISTRIBUTION OF I 
PO_LYGONUM CASCAOENSE 
---- '~CX)  ------ CX) 
189 
Oregon. 
In the Western Cascades f. douglasii, f. minimum, and P. cas-
cadense exhibit only subtle ecological differentiation and are fre-
quently found growing together in open rocky spots on southwest-facing 
meadow slopes between 1400 and 1800 m. Typical coexisting members of 
the Xeri c Meadow association include Gilia aggregata, Collomia linearis, 
Orthocarpus imbricatus, Gayophytum diffusum parviflotum, Navarretia 
divaricata, Linum perenne lewisii ; Lupinus arbustus neolaxiflorus, 
Eriogonum nudum, Microsteris gracilis, Cerastium arvense, and Artemisia 
ludoviciana latiloba. 
Polygonum kelloggii Greene, Fl. Fran. 134. 1891. 
The Polygonum kelloggii complex may be an offshoot from the P. 
spergulariaeforme complex, but it obviously forms a relatively distinct 
and compact grouping within the subgenus Duravia and perhaps deserves 
separate sectional status. f. he terosepalum Peck and Ownbey forms a 
morphological link between P. watsoni Small and the group including f. 
californicum Meisn. and P. parryi Greene, which lack the distinctive 
joint at the juncture of stem and leaf which is common to all other 
species of Duravia. 
The f. kelloggii complex is adapted to vernal pool environments 
in arid regions. Typically of very small size, individuals mature seed 
rapidly during the early spring season of abundant moisture and quickly 
desi ccate and disappear as the volume of snowmelt begins to decrease. 
Several factors, especially localization of vernal pool habitats and 
190 
autogamy, tend to isolate small populations of these knotweeds even 
when they occur in the same region. Thus, ecotypic variation and 
random fixation are likely within the various species. This isolation 
is not absolute, however, as is evidenced by occasional truly inter-
mediate forms. 
The range off. kelloggii includes much of the Western United 
States. It has not been reported previously from west of the Cascade 
crest except for a single specimen from southernmost Oregon and several 
sheets from Hand Lake, near McKenzie Pass. This species has long been 
confused with P. watsonii, a species with eight anthers and deeply 
lacerate ochreae, which approaches f. heterosepalum and P. californicum. 
Of ten collections formerly labeled P. watsonii in the Herbarium of the 
University of Oregon, one is indeed watsonii, s even are f. kelloggii, 
and f. polygaloides Meisn. and P. confertiflorum Nutt. account for the 
remaining two. P. watsonii is evidently quite distinct, and would not 
have been so r eadily confused with other taxa if authors of manuals had 
more carefully segregated and accurately described morphological types. 
This confusion is perhaps due to the fact that good mate-i:-ial of P. 
wa tsoni i has seldom been collected. 
All these species are habi tally quite different • .. Growth habit, 
leaf and brac t shape and color, anther number, and ochrea differences 
are all us eful in segregating them . After studying several thousand 
sheets of Polygonum, the present author has yet to experience grave 
diffi culty in assigning material of t his complex to one of the four 
described species. This statement is not intended to obscure the fact 
t hat many of the characteristics are gradational and that the individual 
+ \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ 
+ + 
G +++ + 
+ 
+ +o 
+ + 
0 
+ 
+ + 
++ 
+ 
++ I 
.p + I + 
++ I 
+ I 
I 
+ 
~a I I 
I I 
' 110 mi. I 
GEO.GRAPHIC DISTRI BUTION OF I 
, POLYGONUM KELLOGGII '~\0  ----- ------ ~
+ A \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
..-- \ 
\ I 
+ \ 
+ • \ 
+ + \ + \ • 
+ 
+ \ 
+ \ 
+ + 
+ ~D • 
+ + I 
•••• •Q•t • 
++ 
/ .. 
I 
• I + A  • 
++ • I I 
+ I 
++ + 
+ I 
+ I 
I 
+ I 
++ I 
I I 
. 
' ' 110 rni . 
I 
GEOGRAPHIC DISTRIBUTION OF I 
PO-LYGONUM CONFERTIFLORUM I~ ---- \D ---- N 
193 
taxa are highly variable, but rather to indicate that the existing 
taxonomy, as described by Wheeler (1938), is entirely workable. 
In the Western Cascades ,_!:. kelloggii is found on open gravelly 
banks and slopes, and in well-established rocky trails where snowmelt 
runs in the early spring. Although_!:. cascadense, _!:. minimum, and_!:. 
douglasii are frequently found in closely adjacent areas, the flowering 
time of _!: . kelloggii scarcely overlaps that of the former group, mem-
bers of whi ch continue to bloom until well into August, and occasion-
ally un t il Oc tober~ Other members of the Ro cky Melt Seep association 
include Lewisia triphylla, Mimulus breweri, Linanthus harknessii, 
Allium amplectens, Gayophytum humile, and Romanzoffia .sitchensis. 
Po lygonum newberryi Small, Bull . Torr. Bot. Club 21:170. 1894. 
A broad-leaved species of subgenus Aconogon, _!:. newberryi is 
characterized by its alpine habitat; yellow-green axillary flowers; and 
pe tiolate, densely soft-pubescent leaves. Its c losest relative is P. 
davisi ae Brewer, a plant of similar habitat in southern Oregon and 
California, which has mainly sessile glabrous leaves. Al though _!:~ 
newberryi is a variable taxon, it seems to remain distinct from P. 
davisiae even where the two co-occur in southern Oregon and northern 
California. With the exception of spec imens from the Olympics; the 
Warner Mountains, Modoc County, California; and central Idaho, this 
species is limited to the Cascade Range from Washington to California. 
It has previously been reported only from high elevations around the 
major peaks, where it is often abundant in areas of pumice or fine 
scoriaceous gravel. It has also been found to occur in several 
+ \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
Ci) \ 
+ \ + 
+ 
+ 
+ 
+ + 
+ 
+ 
++ I 
I 
I 
+ I 
++ I 
+ I 
e + ++ I 
+ I 
I 
+ ' I 
+ ++ I 
I i 
11omi. I 
GEOGRAPHIC . DISTRI BUTION OF I 
- POLYGONU M NEWBERRY! 
----- I..O ------- '~.... 
195 
ecologically similar spots at somewhat lower elevations in the Western 
Cascades on or near the gravelly surmnits of some of the higher peaks, 
often in snowbed areas or spots that receive runoff. It is often found 
with Luetkea pectinata, Ribes binominatum, Penstemon cardwellii, Lotus 
nevadensis douglasii, Crepis occidentalis, Eriogonum umbellatum, 
Dicentra uniflora and other species of varied floristic affinities. 
Eriogonum nuduffi Dougl. ex Benth., Trans. Linn. Soc. 17:413. 1837. 
Eriogonum compositum Dougl. ex Benth., Lindl. Bot. Reg. 21: pl. 1774. 
1835. 
Eriogonum umbellatum Torr., Ann. Lye. N. Y. 2:241. 1828. 
Eriogonum is a large, fascinating, and difficult genus which has 
undergone great and evidently very recent differentiation in the Western 
United States. Species are extremely variable and tend to be poorly 
delimited from one another. The genus as a whole inhabits dry environ-
ments, and two of the three perennial species listed here have been 
called disjunct xeric indicators by Detling (1953). Since they are 
more continuous in their distribution ranges than many other species 
encountered and are well known in western Oregon, no maps of their 
total distributions are given here, but they are briefly described and 
discussed below, 
E. nudum lacks pseudostipes (stem-like extensions of the bases 
of the perianth) and pubescence around the inflorescence, and has large, 
umbellately compound, naked flowering stems which arise from a basal 
rosette of prostrate leaves. The flowers are white, and the leaves are 
green above and densely woolly beneath. This species is highly variable 
196 
and is sometimes lumped with the Californian E. latifolium, which then 
contains about eight subspecific taxa which are differentiated with 
difficulty. Throughout much of its range,_§.. nudum is essentially a 
coastal species, but the type locality is the Willamette Valley. It 
is found from southern Washington to southern California and in Oregon 
is common in the Cascades, as well as in other dry spots west of the 
Cascade crest. It also occurs in the Sierra Nevada of California. 
Both Eriogonum compositum and E. umbellatum have pseudostipes, 
reflexed or spreading involucral lobes, and naked flower stems topped 
by occasionally compound umbels of involucres. _§.. compositum has large 
(4-10 cm) cordate leaves and creamy white flowers in large heads, while 
E. umbellatum has smaller cuneate leaves and sulphur yellow flowers. 
Both species are variable, but_§.. umbellatum is especially so, compris-
ing at least 50 proposed taxa which intergrade in many different direc-
tions. Hitchcock (in Hitchcock and others, 1964) recognizes only seven 
poorly differentiated geographical races. Western Cascade material 
seems to be intermediate between var. hausknechtii (Dammer) Jones and 
var. umbellatum. The latter variety tends toward_§.. marifolium T. and 
G., which differs mainly in its erect, shorter involucral lobes. E. 
umbellatum is found in varied habitats throughout almost all of the 
Western United States. While it is more common east of the Cascade 
crest in Ore~on, it is abundant in both the High and Western Cascade 
Ranges. E. compositum is more limited in distribution. It occurs 
widely east of the Cascades through the Pacific Northwest, but is also 
found in dry areas throughout western Oregon. 
19 7 
Occasionally all three of these species are found coexisting in 
the Western Cascades on dry, gravelly, south or west-facing slopes. In 
general, however, E. nudum and!• compositum are found in slightly 
damper sites with finer, more highly organic soil, and E. umbellatum is 
restricted to crevices of small outcrops or deflation armor flat • 
Lewi sia triphylla (Wats.) Rob. in Gray, Syn. Fl. 11 :269. 1897. 
Selected synonyms : 
Claytonia triphylla Wats., Proc . Am. Acad. 10:345. 1875. 
·A diminutive Lewisia that occupies areas near snowbanks or in 
runoff rivulets, this s pecies is the only member of the genus with a 
globose corm--a feature characteristic of t he genus Claytonia. It 
differs from Claytonia, however, in regularly lacking basal leaves, and 
in its circumcissile rather than laterally dehiscent capsule. 
L. triphylla is widespread in arid regions east of the Cascade 
Range but has only rarely been collected from these mountains, and never 
previously from the Western Cascades. It is relatively common i n high 
elevation Rocky Melt Seeps, together with Mimulus breweri, Romanzoffia 
sitchensis, Dodecatheon jeffreyi, Polygonum kelloggii, Allium amplectens, 
Linanthu s harknessii, Gayophy tum humile, and Saxifraga occidentalis 
rufidula. 
Spraguea umbellata Torr. in Smith, Contr. Knowl. 6:4. 1853. 
Thi s typically high elevation species is immediately recognized 
by its basal ros ettes of red or green fleshy leaves and i. t s often 
+ • \ ------\ 7 
+ 
0 \ I 
\ 
3 mi. \ I 
\ 
\ I 
-~ \ 
\ I 
+ 
+ • • \ \ 
+ + + \ \ 
+ 
+ \ 
+ \ 
\ 
+ + 
C!) Jl I + •• + 
• • • I 
• 
++ 
I 
I • 
•.• .  1•  + I +E> I I •• • • + I  • 
++ -+O 0 I • 
.p I 
+ I 
'9+ I I 
I I 
' ' 110 n,i. I 
GEOGRAPHIG DISTRIBUTION OF I 
' LEWISIA TRIPHYLLA I~ 
---- I.O ------- CX> 
+ • \ ---- 7 \ 
+ + \ I 
\ 
3mi . . \ I 
t---1 \ 
\ I 
,,,- \ 
• \ 
I 
+ \ 
' ~ 
+ \ 
+ \ + • \ + 
+ \ 
+ \ 
\ • 
+ + 
+ •
+ + ~~. . •  • • • e 
++ ... •• • I I I ,.. 
+ ~ I 
++ •• I • 
+ • I ++ + () I ' •• 
.p I 
• •I • •
 • 
I 
++ I 
I I 
I 
' ' 110 mi. 
GSOGRAPHIC DISTRIBUTION OF I 
_ SP~AGUEA UMBELLATA '~..i:, ---- ---- ..i:,  
200 
prostrate globular heads of membranous white or pink flowers. It is 
variable, particularly with regard to robustness and longevity. Wide-
spread in subalpine or higher habitats east of the Cascades, it has 
seldom been reported from west of the Cascade crest. Like Polygonum 
newberryi, it is found in open gravelly soils or pumice sand both in 
the high mountains, where it reaches 2600 m, and in the Western Cas-
cades. Like Nothocalais alpestris, it has been found in the lower 
mountains only in the vicinity of 0lallie Mountain, where lava flows 
from the High Cascades have created a bridging plateau between t he two 
ranges at almost 1500 m elevation . 
S. umbellata normally occupies gravel scree areas which are in 
a state of flux due to freeze- t haw cycles. Few other plant s can estab-
lish in these areas, and Spraguea is not closely associa ted with any 
other Western Cascade species . 
Arenaria rubella (Wahl.) Smith, Eng. Fl. 4:267 0 1828. 
Selected synonyms: 
Arenari a propinqua Rich., in Frankl., 1st Journ . Bot.a App . 
738. 1823. 
Arenaria verna of American authors, not A. verna L. 
Arenaria verna L. var. pubes c ens Fern. Rhodora 21:210 1919. 
The section Alsine, the perennial matted forms of Arenaria in 
whi ch the capsuie dehisces by t hree valves, poses difficult taxonomic 
problems; but these seem to be less acute than those encountered in the 
section Eremogone. Hitchcock (in Hitchcock and others, 1964) recognizes 
only four species in section Alsine from the Pac ific Northwest and 
201 
defines these very broadly. A. rubella is the only species in this 
group forming small compact mats with a single well-defined taproot. 
More robust open forms from lower or more sheltered habitats may 
approach~- nuttallii Pax or~- obtusiloba (Rydb.) Fern. in habit. 
This is especially true of material from Saddle Mountain, Clatsop 
County, Oregon, and from the Olympic Peninsula. 
~- rubella is a cliff-dwelling plant, with its taproot firmly 
anchored in crevices of vertical volcanic faces, normally exposed to 
prevailing wind~ ~ According to Maguire (1951), this species is circum-
polar and circumboreal, occurring in North Ameri ca from Greenland, the 
Arctic Archipelago, and Alaska south along the Atlantic Coast, in the 
Rockies and Cascades to the Gaspe Peninsula, northern Arizona, and 
northern California. It is extremely polymorphi c when considered 
throughout its range. Compact forms, such as t hose found in the West-
ern Cascades, are normally restricted to Arcti c~Alpine elevations at 
this latitude, but the species is occasionally found lower than 300 m 
above sea level. It has not often been collected in the United States, 
and except for one locality on the Three Sisters, the spec imens col-
lected in this study constitute the only Oregon Cascade material known 
to the author. 
Other members of the Vertical Outcrop association, such as 
Penstemon rupicola, Saxifraga bronchialis vespertina, Erigeron cas-
cadensis, Douglasia laevigata, Castilleja rupicola, Polypodium hesperium, 
Polemonium pul cherrimum, and Heu chera micrantha, are associated with 
Arenaria rubella in the We s tern Cascades. 
+ \ -- - - 7 
\ 
+ + \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
\ 
+ \ 
+ \ + 
+ 
+ 
+ + 
-+ 
+ + 
0 
++ 
0 
+ 
e + h. I 
+ I 
+ 
+ + -+ I 
+ I 
I 
+ I 
+ + 0 I 
I I 
110 mi. I 
GEOGRAPHIC . DISTRIBUTION OF I 
AR ENAR IA RU BELLA IN 
0 
,___  _..,__ -- - - - - N 
203 
Arenaria capillaris Poir. subsp. americana Maguire, Bull. Torr. Bot. 
Club 74 : 41. 194 7. 
Se l ected synonyms: 
Arenaria nardifolia of American authors, not A. nardifolia 
Ledeb. 
Arenaria formosa of American authors, not A. formosa Fisch. 
Arenaria capillaris Poir. subsp. formosa ( Fisch.) Mag., 
Madrono 6 : 24. 1941. 
Arenaria pumicola Cov. and Leib. , Proc. Biol. Soc. Wash. 11:169. 1897. 
Relationships among members of t h e American Capillaris-
group are complex and obscure. With i n it little clear-cut spe-
cific differentiation ha s taken place, or at l east most species 
are so variable within thems e lves as to transgress all specific 
lines, removing discontinuity , and to leave only emphatic notes 
in a basally continuous series . . •• The interpretation of 
populations, specific and subspecific, depends largely on accu -
rate evaluation of the interplay of sepal, leaf, and inflorescence 
form . Gland size and disposition , seed, and rootstock offer sup-
porting characterization . Of no little aid is the final summation 
of intangibles that comes of familiarity with the plants, and that 
gives populations an "appearance" or "look"- - indefinite properties 
that defy descriptive analysis. (Maguire, 1947, p. 38-40) 
Thi s s tatement by the most recent student of Arenaria briefly summar-
izes the pres ent state of taxonomy in the genus. Even those characters 
mentioned by Maguire , however , do not always stand up on close scrutiny. 
Large - scale geographical considerations may help, but presumably diag-
nostic characters may occur far outside the natural range of the 
" species" they characterize . 
A. cap i llaris ameri cana has been obs erv ed ( Hickman, in prepara-
tion ) to intergrade with all other species of section Eremogone that 
are found in Oregon. Occas ional s pecimens show obvious diagnostic 
characters of th r ee or perhaps more of the s e species (~. kingii, 
204 
!_. aculeata, and!_. congesta, in addition to the two under discussion 
here). !_. capillaris americana is found in pure form from southern 
British Columbia and Alberta south in the Cascades to the Three Sisters 
and in the Rocky Mountains south to northern Idaho and western Montana. 
Intergrades with!_. pumicola, !_. aculeata, and!_. kingii extend the 
range of certain characteristics diagonostic of!_. capillaris americana 
south to Josephine County, Oregon, and White Pine County, Nevada. 
Since A. capillaris has the northernmost range of any American species 
in this circumboreal section, it is generally considered the most prim-
itive American taxon, from which other, more southerly species have been 
derived. 
A. pumicola is characterized by straight, rigidly ascending, 
deciduous basal leaves; relatively wide, fleshy cauline leaves which 
are abruptly reduced at the inflorescence; and a naked, stout, erect 
caudex which branches just below the soil surface. Numerous intergrades 
with A. capillaris americana are found, and intermediates between this 
species and A. aculeata and!_. congesta have been collected on rare 
occasions. A. pumicola is abundant at Crater Lake, the type locality, 
and is also found south and west to the Siskiyou Mountains of Curry 
County, and north to the north slope of Mr. Jefferson, Marion County. 
Intergrades with!_. capillaris americana, especially alpine forms, 
carry some diagnostic characters of this species as far north as Mt. 
Rainier. The variety californica Maguire occurs from Sierra to Mono 
Counties, California, centering around the Lake Tahoe region. It is 
morphologically intermediate between typical!_. pumicola and!_. kingii 
of the Great Basin. A. ursina Robins. of the mountains of southern 
205 
California may in turn be derived from!!_. pumicola californica. 
Morphological evidence points to a complex pattern of reticulate 
evolution in this group of species, where breeding barriers are evi-
dently poorly developed. A small number of genes may be responsible 
for many of the critical diagnostic characters, making the evolutionary 
and geographical significance of their occurrence in isolated popula-
tions at least disputable. Much further work needs to be done on the 
biology of these species. 
It is certainly significant that the study area includes the 
latitudinal limits of the pure forms of both!!_. capillaris americana 
and A. pumicola and is the only area in which such forms co-occur. A 
number of intergrades have been collected here. 
The narrowly linear leaves and perennial, more or less cespitose 
habit of all members of the section Eremogone are characteristic of 
plants growing in dry habitats. These species are found either high in 
the mountains or in dry plain or desert environments. In the Western 
Cascades,!!_. capillaris americana and A. pumicola are often rooted in 
crevices of volcanic rock in exposed but seldom south-facing situations. 
Although they have not been found together, they are both commonly as-
sociated with Penstemon procerus brachyanthus, Sedum stenopetalum, 
Sedum oregonense, Arctostaphylos nevadensis, Eriogonum umbellatum, 
Silene douglasii, Juniperus communis saxatilis, and other members of 
the Outcrop Ridge, Gravel Scree, or Small Boulder Creep Slope associ-
ations. 
+ \ 7 
\ 
+ \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
0 \ 
+ \ •• 
+ \ + • \ 
+ 
+ \ 
oO 
+ + • 
0 
+ + ore 
B o 
++ 
ii 
+ 
++ I • 
+ I 0 
A -0 I 
+ I 
I 
I 
I 
I 
' ' 11omi 
GEOGRAPHIG OISTRIBUTICN OF 
Aren':lria c3.:9illaris americ ':l.na • o ; A. pumicola ■ ; Intergrades : 
A. capillaris - A. pumicola •A; A. capillaris - A. a culeat a .---------------
A. capilla ris A. kineii gl3.brescensO. 
207 
Silene campanulata Wats. subsp. glandulosa Hitch. and Mag . , Univ. 
Wash. Puhl. Biol. 13:1-71. 1947. 
According to Hitchcock and Maguire (1947), Silene campanulata 
is not closely related to any other western Ameri can s ilenes . It is a 
member of the subgenus Melandyrum, which is primarily characterized by 
unilocular capsules. The calyx is broadly campanulate, and the pe t als 
are finely divided into linear lobes, making the flowers quite distinc-
tive. Three subspec ies have been described. S . campanulata subsp. 
greenei (Wats.) Hitch . and Mag. is known from northern California and 
southwestern Oregon. It is considerably less glandular than subsp. 
glandulosa. A few intergrades are known between these two subspecies. 
Subspecies typica Hitch. and Mag. is known only from its type locality 
on Red Mountain, Mendocino County, California . I ts leaves are linear, 
rather than broadly lanceolate, and the stem pubescence is shorter than 
that of subsp. glandulosa. 
~- campanulata glandulosa ranges from the Klamath Mountain and 
north Coast region of California through the Siskiyous and southern 
Cascades of Oregon. Specimens have been collected from dry rocky sites 
at the head of the Willamette Valley, from the vicinity of Mt. Hood, 
and at scattered locations through the sou t hern part of the Western 
Cascades, where it is commonly rooted in crevices of andesite or basalt 
on steep south-facing slopes, As a member of the Outcrop Ridge assoc i-
ation, it often occurs wi t h Haplopappus hallii, Comandra umbellata, 
Lomatium martindalei, Erigeron foliosus confinis, and Arenaria capillaris 
americana. 
+ • \ ---- 7 \ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ 
·~ \ + \ 
+ + \ + \ 
+ 
+ \ 
+ \ 
\ .. 
+ + 
+ 
(!) + ~ 
++ 
• 
A 0 
+ 
+e • • A I • 
+ I 
+ 
ec!> + I 
·• + I I 
"1' I 
%0 I 
I I 
' ' 110 mi. I 
GC:OGRAPHIG DISTRIBUTION OF I 
SILENE _CAMPANULATA GLANDULOSA L.) 
---- ------- 0 00 
209 
Silene douglasii Hook., Fl. Bor. Am. 1:88. 1830. 
Both this species and Silene campanulata glandulosa are inordi-
nately well defined for members of their genus and family (Caryophyl-
laceae). Silene douglasii, a member of the subgenus Melandryum, has 
perhaps been derived from stock of_§_. parryi (Wats.) Hitch. and Mag., 
since where the ranges of the two overlap, sterile hybrids are some-
times formed (Hitchcock, in Hitchcock and others, 1964). Specific dis-
tinctions are not clear in this genus as a whole, however, and rel.at.ion-
ships are notably difficult to determine (Hitchcock and Maguire, 1947). 
S. douglasii is widespread in Oregon, occurring both east and 
west of the Cascades, and extending to British Columbia, Montana, Utah, 
and central California. Throughout its range it is found in dry habi-
tats. In the Western Cascades, where it is abundant, it grows in 
crevices of rock on well-drained slopes with either a southern or west-
ern aspect. It is typically less exposed to desiccating influences 
than_§_. campanulata. Other members of the Outcrop Ridge association 
with which_§_. douglasii occurs include Penstemon procerus brachyanthus, 
Haplopappus hallii, Arenaria capillaris americana, Sedum stenopetalum, 
and Polygonum douglasii. 
Arabis holboellii Hornem. var. retrofracta (Grah.) Rydb., Contr. 
U.S. Nat. Herb. 3:484. 1896. 
Selected synonyms: 
Arabis secunda Howell, Erythea 3:33. 1895. 
Arabis holboellii var. secunda Jeps., Man. Fl. Pl. Calif. 
430. 1925. 
\ 
.\ 7 
+ \ I 
\ 
\ I 
\ 
\ I 
\ 
0 \ 
I 
\ 
+ \ 
+ \ + \ 
+ 
+ \ 
\ 
+ 
• ~µh• .•  
C 
0 + 
II• • • :• • • • • 
• • 0 
I • • 
I •• • I 
I 
I 
I 
I 
I I 
I 
' ' 110 mi. 
GEOGRAPHIC DISTRIBUTION OF I 
SI LENE DOUGLAS! I I .N... . 
0 
• A \ ------ 7 \ 
+ 
ct \ I \ 
3mi. \ I 
\ 
\ • I \ 
\ I 
~ 
A \ 
+ \ 
+ \ + \ •• • 
0 + • 
+ \ 
e \ 
\ • • • • 
+ 
-+ 0 • 
+ • ~-~ . 
•• • ••  
••• • 
+c, • • 
• 
• 
a • • 
+ A • • 
Ii> A. I • • I 
.p e I 
• • • 
+ 
lo I \ 
e I • I \ 
I 
-+ •• I 
e+ I • 
I I 
• 
' 110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
ARABI$ HOLBOELLII RETROFRACTA L,) 
------ >--' ---- >--' 
212 
Arabis is one of the most difficult genera of the Cruciferae, 
and certainly the most highly diversified genus of this family in west-
ern North America. All species are variable, and complete morphological 
delimitation is sometimes impossible. A. holboellii is one of the most 
variable species of Arabis, as a complete synonymy would testify. Rel-
atively conservative workers (Rollins, 1941) have recognized only five 
subtaxa, all of which occur in the Pacific Northwest, Variety retro-
fracta is recognized by its auriculate, revolute, cauline leaves; 
sharply reflexed fruiting pedicles; and narrow siliques. It ranges 
throughout most of the Western United States and has been reported from 
as far east as Quebec. It typically inhabits sagebrush or ponderosa 
pine communities. 
Only rarely collected from west of the Cascade crest , this 
species is a dominant member of the Outcrop Ridge and Small Boulder 
Creep Slope associations of the hot, summer-dry south or west-facing 
slopes of the Western Cascades. It is commonly rooted in crevices of 
andesite or in dry gravelly loam with Delphinium menziesii pyramidale, 
Castilleja hispida, Penstemon procerus brachyanthus, Calochortus 
lobbii, Haplopappus hallii, Silene douglasii, Arctostaphylos nevadensis, 
and Juniperus communis saxatilis. 
Arabis platysperma Gray var. howellii (Wats.) Jeps., Man. Fl. Pl. 
Calif. 432. 1925. 
A wide-podded (4-6 mm) species,~- platysperma is distinguished 
from its closest relative, A. suffrutescens Wats., by its erect rather 
than reflexed siliques and its more scapose stem , The common Oregon 
+ • \ ----
+ •· 
7 
\ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
e • ., 
.p + \ + 
e 0 
+ 
+ + • 
+ • 
+ e • 
++ .. • • + -: 
++ • • A I 
+ • I + + ++ • I 
+ I 
I •
+ ••  I 
++ -. I 
I I 
' ' 110 mi. I 
GEOGRAPHIG DISTRIBUTION OF I 
- ARABIS PLATYSPERMA IN....  
---- ---- w 
214 
phase, varo howellii, is glabrous and glaucous. It merges with the 
stellate-pubescent var. platysperma south and east from Mt. Shasta. 
The species as a whole ranges from Mt. Hood south to eastern Nevada 
and southern California . The two varieties are mapped together in 
this work . 
!!_ . platysperma is abundant and has often been collected from the 
Three Sis t ers, Lane and Des chutes Counties, but heretofore has not been 
reported from the Western Cascades . Two localities have been dis covered 
there, plus several sites in the lower peaks of the High Cascades of 
Lane and Linn Counties. In all sites, plants occur in moist rocky soil 
near the summi t s, partly shaded by boulders or other vegetation. A 
perennial,!!• platysperma sheds its broadly winged seeds before summer 
moisture stresses be come great . It grows in association with Phyllodoce 
empetriformis, Erigeron a cris debilis, Luetkea pectinata, Polygonum new-
berryi, Cas t illeja parviflor a oreopola, Senecio triangularis, and 
Eriogonum pyrolaefolium. 
Cardamine integrifolia Greene var . sinua t a (Greene) Hi tch., Vas co 
Pl. Pac. No Wo 2:469. 1964. 
Selec ted synonyms: 
Dentaria integrifolia Nutt . , in To and G., Flo No Am. 1:88. 
1838. 
Dentaria californi ca Nutt o, in T. and G. , Fl. N. Am. 1:88. 
1838 . 
Cardamine integrifolia Greene, Bull . Calif . Acad . Sc i . 2:389. 
1887. 
Cardamine californi ca Greene, Fl. Frano 266. 18910 
Dentaria sinua t a Greene, Pitt . 3:123. 18960 
215 
Cardamine californica var. sinua ta Schulz, Eng. Bot. Jahrb. 
32:387. 1903. 
Dentaria integrifolia var. ca lifornica Jeps . , Man. Fl. Pl. 
Calif. 426. 1925. 
Dentaria californica var. integrifolia Detl., Am. Journ. Bot. 
25:576. 1936. 
This is a complex species, comprising several varieties in Cali-
fornia, but i ts nomenclature is far more confusing than its taxonomy. 
Hitchcock (in Hitchcock and others, 1964) notes that Dentaria, a segre-
gate of those species having larger flowers, tuberous rhizomes, and 
fewer cauline leaves, merges completely with Cardamine s.s. through C. 
constancei and _g. rupicola, both of which coul d be placed in either 
genus (Detling, 1936). Following Hitchcock , I recognize only t he older 
genus in t his work. There has also been confusion in the li terature 
(Jep s on, 1925; Detling, 1936, 1937) as to which of several proposed 
specific epithets is valid for the present taxon. Again, I concur with 
Hitchcock • 
.£. integrifolia differs from .£. pulcherrima (Robins.) Greene in 
its larger subglobose tubers, the leathery texture of the foliage, and 
more numerous flowers. Var . sinuata is distinguished from other vari-
eties by its simple rhizomal leaves, which have sinuate margins; and by 
its mostly compound cauline leaves. 
This is normally a coastal vari ety 9 entering the Cascades of 
s outhern Oregon, and occurr i ng s poradi cally northward through the lower 
mountains to Tidbits Moun tain, Linn County . In the Western Cascades it 
always occurs in dense blocky talus on north-facing slopes at the base 
of c liffs with the tubers deep among the boulders. Other members of 
+ • \ ---- 7 \ 
+ + · \ I 
\ 
3mi. I -\ 
1-------i \ 
\ I 
,..,... \ 
\ I 
+ 
+ ·• \ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
0 + + 
++ 
• 
+ ..A  I 
++ .. I 
+ I 
+ 
++ + I 
+ I 
-.- I I 
• ++ I I I 
' ' 11omi. 
I 
G!::OGRAPHIG DISTRIBUTION OF I 
CARDAMI_NE I NTEGRI FOLi A SINUATA IN 
---- ---- f--1 °' 
217 
the Blocky Talus association include Sambucus racemosa pubens arborescens, 
Acer circinatum, Aquilegia formosa, Thalictrum occidentale, Erigeron 
cascadensis, and Campanula rotundifolia. 
Sedum stenopetalum Pursh, Fl. Am. Sept. 324. 1814. 
Selected synonyms: 
Sedum douglasii Hook., Fl. Bor. Am. 1:228. 1834. 
This species has commonly passed under the name_§_. douglasii. 
Hitchcock (in Hitchcock and others, 1964) reports the discovery that 
Pursh's holotype of_§_. stenopetalum actually comprises two distinct 
species, and that the description most closely approximates not the one 
which has been since referred to as S. stenopetalum, but to_§_. douglasii 
of Hooker. This discovery neces si tates shifting several widely accepted 
names. S. stenopetalum, as presently constituted, is characterized by 
alternate dorsally-keeled leaves, the bases of which are broadened and 
membranous. The scarious leaf bases and midr i bs are persistent, giving 
the stems a ragged look late in the season. Single yellow flowers · are 
borne by many stems and are visited by bees but appear to set no seed. 
Reproduction is accomplished by bulblets, whi ch form in the axils of 
the upper cauline leaves. These bulblets drop off when the stems desic-
cate in August and overwinter on the soil surface. The many non-flower-
ing stems are biennial, appearing in the spring as fragmented pieces of 
the previous season's matted stems, and blooming during their second 
growing season. They are recognizable by being considerably larger 
than the axillary bulblets. 
+ ... \ ---- 7 
\ 
+ + \ . I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ ... \ 
+ \ 
+o + \ + 
+ 
+ 
+ + 
••o .,,. + 
+ + 
+C!) • 
+ ...•  
~ • I 
+ I 
+O + 
'bt- I 
+ I 
I 
-t I 
~ I 
I I 
11omi. I 
GEOGRAPHIC . DISTRIBUTION OF I 
- SEOUM STENOPETALUM IN I-' 
----- ---- co 
219 
~ . stenopetalum is common in dry areas of eastern Oregon and 
Washington east to Montana and in the Siskiyou region. It has occa-
sionally been reported from "xeric islands" in the Willamette Valley, 
and from the Western Cascades. It was frequently encountered in the 
latter area on rocky, dry, south-facing slopes . Other members of the 
Boulder Creep association include Penstemon procerus brachyanthus, 
Delphinium menziesii pyramidale, Castilleja hispida, Cerastium arvense, 
Poten ti lla glandulosa typic.a, Sedum divergens, Eriogonum umbellatum, 
Lomatium martendalei, and Arenaria capillaris americana. 
Sedum divergens Wats., Proc. Am. Acad. 17:372. 1882. 
A bright reddish stonecrop with subglobose opposite leaves and 
strongly divergent carpels,~. divergens is easily recognized in the 
field or herbarium. The perennial stems are often matted and root at 
the nodes. Flowers are bright yellow and are borne profusely. Viable 
seed is set, but most reproduction is evidently vegetative by r ooting 
of broken pieces of stem. This species is characteristic of the Wash-
ington Cascades and the Olympic Mountains and has been reported in the 
literature from only as far south as Mt . Hood (Hitchcock, in Hitchcock 
and others, 1964). However, old undated herbar i um specimens are avail-
able from a s far south as the Oregon Caves region, Josephine County 
Mr s , Clarice Nye). This Sedum has been found to be almost ubiquitous 
on gravelly, open, south-facing slopes in the northern half of the cen-
tral Western Cascades with a much more sporadic distribution to the 
south. A member of the Boulder Creep Slope association, it often occurs 
with Penstemon procerus brachyanthus, Delphinium menziesii pyramidale, 
• ... \ ---- 7 \ 
+ 
t> \ I \ 
3mi. \ I 
\ 
\ I 
--- \ 
\ I 
0 ... \ 
+ \ 
+ \ + \ 
+ 
+ \ 
\ 
+ ... 
+ ;\  
+ 
++ I 
.... .. 
I 
I 
+ 
++ ... I I 
+ I 
+ 
O+ + I 
+ I 
I 
E> I 
++ I 
I I 
' ' 110 mi. I 
GEOGRAPHIC' _DISTRIBUTION OF I 
~ 
SEOUM DIVERGE NS IN 
-N 
---- ------- 0 
221 
Castilleja hispida, Cerastium arvense, Potentilla glandulosa typica, 
Sedum stenopetalum, Eriogonum umbellatum, Lomatium martindalei, and 
Arenaria capillaris americana. 
Ribes binominatum Heller, Cat. N. Am. Pl. ed. 2. 5. 1900. 
Selected synonyms: 
Grossularia binominata Cov. and Britt. N. Am. Fl. 22:218. 
1908. 
The three species which form the compact section Watsonianae 
should perhaps be considered only subspecifically distinct. R. binom-
inatum is the most widespread and abundant of these three species, and 
it is proposed here that it has spread from its center of distribution 
in the Siskiyou Mountains of southwestern Oregon and the Coast Ranges 
and Klamath Mountains of northwestern California north and south through 
the mountains, differentiating slightly with di stance from its ancestral 
region. The extreme morpho-geographical forms have been described as 
separate species. 
R. tularensis (Cov.) Fedde is presently found in the Sierra 
Nevada of Tulare County, California. It is differentiated from R. 
binominatum in having slightly shorter filaments, slightly stronger 
internodal spines, and more glandular hairs on the leaves and ovaries. 
It has the trailing habit, tripartite nodal s pines , silky leaves, and 
large berries armored with yellow-green spines of~- binominatum. The 
latter species was formerly thought to extend no farther north than 
Douglas County. R. watsonianum Koehne occurs from the eastern slopes 
of Mt. Hood north through the Cascade Mountains of Washington to Chelan 
+ \ 
\ 7 
+, \ I 
\ 
3mi. \ I 
t------1 \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + 
+ 
+ 
o+ 
+ + 
0 .. I 
A. I 
I 
I 
I 
.p I 
I 
+ I 
I 
11omi. 
GEOGRAPHIC' .DISTRIBUTION OF 
- RISES BINOMINATUM 
r-------------- ----
223 
County. It has not been abundantly collected but is differentiated 
from R. binominatum by erect habit, slightly shorter filaments, gland-
ular hairs on the leaves (but not twigs), and berries with somewhat 
more slender glandular spines. 
R. binominatum has been found in this study to be widespread and 
common in the Western Cascades. Specimens are partially intermediate 
to both!· watsonianum and!· tularensis with regard to abundance of 
glandular hairs on stems, leaves, and ovaries; and the weak, ephemeral, 
internodal bristles. It is not unlikely that populations of R. binom-
inatum and!• watsonianum are quite continuous, and that morphological 
intergradation continues in more northerly populations than those dis-
covered in this study. Further field work is needed to solve this 
problem. 
In the Western Cascades, R. binominatum is found trailing over 
fallen logs or partly shaded by overgrowing Mesic Meadow species such 
as Pteridium aguilinum, Rudbe ckia occidentalis, Lupinus arbustus neo-
laxiflorus, Luina stricta, Aquilegia formosa, and Erigeron aliceae. 
It may become so extensively buried in taller vegetation that specimens 
are difficult to locate late in the season, even in spots where it is 
known to be common . This species is occasionally an alternate host to 
the white pine blister rust. 
Ribes erythrocarpum Cov. and Leib., Proc. Biol. Soc. Wash. 10:132. 
1896 . 
This distinctive spineles s currant, with trailing glaucous stems 
and erec t branches, sa lmon- colored flowers, and bright red 
+ \ 
oe 7 \ 
+ \ I 
\ 
3mi. \ I 
1------i \ 
\ I 
\ 
\ I 
+ \ 
O+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ + 
++ 
+ 
++ I 
+ I 
+ + I 
.p I 
I 
+ I 
0 ++ I 
I 
11omi. 
GEOGRAPHIC' DISTRIBUTION OF 
- RISES ERYTHROCARPUM ~------ -- -- -- -
225 
glandular-pubescent berries was formerly thought to be closely restricted 
to the Crater Lake region of the southern Cascades. A single locality 
was known from Gearhart Butte, Lake County. It is now known from six 
sites in the central Cascades, five of which are new with this study. 
The sixth site is represented by a specimen pr eviously identified as 
R. cereum from near Mt. Jefferson . Large populations of this species 
were found only on Bachelor Mountain and Olallie Mountain. Other local-
ities represent isolated occurrenc es of non-flowering material. 
The large populations grow in light moist soil under Abies 
procera or Abies amabilis. Pockets of snow remain in both sites until 
July. Associated species include Xerophyllum tenax, Ribes viscosissimum, 
Smilacina sessilifolia, Vaccinium membranaceum, Tsuga mertensiana, Abies 
amabilis, Valeriana sitchensis, Ac tea rubra, Achlys triphylla, Ribes 
lacustre, Asarum caudatum, and Taxus brevifolia . 
Luetkea pectinata (Pursh) Kuntze, Rev. Gen. 1:217. 1891. 
A monotypic genus of northwestern North America, Luetkea is 
closely related to Spiraea and Petrophytum from which it differs in its 
bi- or triternate leaves and its rhizomatous habit. Primarily a species 
of the Cascade Range, L. pectinata is also found on Vancouver Island, in 
the Olympi c s, through the Canadian Rockies, and in the Bitterroot Range 
of Idaho and Montana. In the Cascades it is almost completely restricted 
to the highest peaks, where it is common around snowbeds, and in and 
around rivulets of snowmelt. Uncommon in the Western Cascades, Luetkea 
has been collected twice at the edges of summit snowbanks. It is also 
+ A \ ---- 7 
\ 
0 \ 
• I \ 3mi. \ I 
\ I 
,,..- \ 
\ 
\ I 
+ 
e I 
\ 
\ 
+ + \ + \ 
+ ++e + \ 
+ + 
+ \ 
\ • 
+ + 
-+ 
+ + /
• / .~  
++ 
• I 
~ I • 
+ AO I 
++ A. I • 
+ I 
+ + ++ I 
+o I 
I 
+ I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC' -DISTRIBUTION OF I 
LUETKEA PECTINATA IN 
---- N ---- °' 
227 
reported here for the first time from some of the lower peaks of the 
High Cascades. Other Snowbed species with which 1_. pectinata occurs 
include Orogenia fusiformis, Trillium ovatum, Erythronium grandiflorum 
pallidum, Castilleja parviflora oreopola, Arabis platysperma howellii, 
and Senecio triangularis. 
Ivesia gordonii (Hook.) Torr. and Gray in Newberry, Pac. R. R, Rep. 
63:72. 1857. 
Selected synonyms: 
Horkelia gordonii Hook,, Journ. Bot, and Kew Misc. 5:341, 
pl. 12. 1853. 
Ivesia alpicola Rydb, ex Howell, Fl. N. W. Am, 182. 1898, 
This remarkable species, while widespread in western North Amer-
ica, has never previously been collected in the western half of Oregon, 
It occurs in abundance on the ridge connecting Iron Mountain and Cone 
Peak in Linn County, from which the nearest localities for the species 
are Mt. Adams, 230 air km to the north, and Drake's Peak, 300 air km to 
the southeast. Throughout its range!, gordonii is found at very high 
elevations, seldom below 2400 m and typically from 3000 to 4200 m, The 
disjunct Western Cascade populations occur between 1300 and 1700 m. 
The Washington Cascade populations are fully as disjunct . as those in 
the Western Cascades, but they are found only above 3000 m. 
Ivesia is separated from the closely related genera Horkelia and 
Potentilla by its stamen number (5 in!• gordonii), shallow hypanthium, 
few achenes, undilated filaments, low receptacle, yellow petals (often), 
and lack of distinctive odor (Keck, 1938), Although Ivesia as consti-
tuted by Keck contains 22 western American species, only two of these 
+ .6. \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
1------1 \ 
\ I 
\ 
\ I 
+ 
A \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ \ • 
\ 
+ + J] • 
-+ 
+ + 
++ 
~ 
+ .6. 
++ .6. I • 
+ I 
+ + ++ I • 
+ I • 0 
I 
-+ I 
++ I 
I I 
11omi. I 
GEOGRAPHIC' DISTRIBUTION OF I 
IVESIA GORDONII INN  
---- ---- co 
229 
are found in Oregon,.!.• baileyi Wats. being restricted to the Basin and 
Range mountains in the southeastern corner of the state • .!.• gordonii is 
distinguished from other congeners by its campanulate hypanthium, white-
hirsute receptacle, and by details of achene morphology. This species 
has many more pairs of leaflets than.!.· baileyi. 
In the Western Cascades (and on Steens Mountain, Harney County) 
this spec ies is a member of the Gravel Scree association, and is also 
present in crevices of dikes or other outcropping rocks, fully exposed 
to sun and wind. It is associated with Lotus nevadensis douglasii, 
Chrysothamnus nauseosus albicaulis, Cilia aggregata, Allium crenulatum, 
Trifolium productum, Aster gormanii, Crepis occidentalis, Sedum 
oregonense, and Sanicula graveolens. 
Horkelia fusca Lindl., Bot. Reg. 23: pl. 1997. 1837. 
This species is complexly polymorphic. Keck (1938) was forced 
to admit six subspecies with few clear geographic distinctions. Four 
of these geographical subspecies occur in the ponderosa forests and high 
lava plains of Deschutes County east of the study area. The most wide-
spread is subsp. parviflora (Nutt.) Keck. For this and various morpho-
logical reasons, Keck believes that subsp. parviflora is the ancestral 
type from which the other subspecies are derived. Although Western Cas-
cade material is of subsp. parviflora grading into subsp. pseudocapitata 
(Rydb.) Keck, material of this species will be treated as a unit in both 
this discussion and in the following map. 
Horkelia is closely allied to Ivesia and Potentilla. It is 
+ \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ 
A \ 
+ \ 
+ + \ + 
0 + 
+ 
+ 
+ + 
+ 
+ + 
++ 
6. 
+ .. • 
++ A 
+ 
+ • 
I 
I 
++ + I 
+ I 
I 
+ I 
+ ++ I 
I I 
11omi. I 
GEOGRAPHIC' _DISTRIBUTION OF I 
HORKELIA FUSCA IN 
w 
---- ---- 0 
231 
distinguished from both by having 10 stamens and a deep hypanthium • 
.!:!• fusca is set apart from other horkelias by its 15-25 carpels, erect 
pedicels, 5-10 pairs of leaflets, entire stipules, and glabrous fila-
ments and inner hypanthium. One collection of the species has been 
taken from west of the Cascade crest , high on the flank of Mt. Hood 
(Gorman, in 1891 ). In the Western Cascades, there is a large popula-
tion of robust individuals at Lost Prairie, growing in an open gravelly 
flat at an elevation of 1000 m. A frost pocket, the area is surrounded 
by a Picea engelmannii-Pinus monticola dominated forest. Associated 
species in the meadow are Solidago canadensis salebrosa, Potentilla 
drummondii, Erigeron aliceae, Viola bakeri, Anemone oregana, Sambucus 
racemosa pubens arborescens, Polygonum douglasii, and Prunella vulgari s 
lanceolata. 
Lupinus arbustus Dougl. ex Lindl . subsp. neolaxiflorus Dunn, Leafl. 
West. Bot, 7:254. 1955. 
Selected synonyms: 
Lupinus laxiflorus Dougl. ex Lindl., Bot. Reg. 14: pl. 1140. 
1828. 
Lupinus silvicola Heller, Muhl. 6 : 81. 1910. 
Lupinus laxiflorus var. silvicola C. P. Smith ex Jeps., Man. 
Fl. Pl. Ca l if. 527. 1925. 
Lupinus arbustus subsp . silvicola Dunn, Leafl. West. Bot. 
7:255. 1955. 
Lupinus arbustus subsp . arbustus var. montanus (Howell) Dunn, 
Leafl. West. Bot. 7:254. 1955 . 
This taxon has a long and confusing nomenclatural history, due in 
part to a presumed later mislabelling of the holotype and in part to the 
232 
difficult nature of taxonomy in the genus Lupinus. These problems are 
reviewed by Phillips (1955), Dunn (1955, 1957), and Hitchcock (in Hitch-
cock and others, 1961). While Hi tchcock concedes that Dunn's arguments 
for reducing the complex taxon L. laxiflorus to a subtaxon of L. arbustus 
are probably correct, he continues to use the traditional nomenc lature, 
Dunn's arguments are convincing to the present author, and for the most 
part, his nomenc lature is followed here , 
Dunn (1957) admits that intermediates are known among almost all 
combinations of the six sub t axa he proposes for the 1· arbustus group. 
In addition he cites hybridization between members of this species and 
L. caudatus, l• sulphureus, 1· burkei, and l • rubricaulis, as well as 
close morphological affinities with L. lepidus and l • argenteus . Al-
though all these taxa are highly variable, l • arbustus subsp, neolaxi-
florus is, in the main, a distinctive perennial lupine with an abaxially 
pubescent banner petal, a slightly spurred calyx, and a small patch of 
hairs near the upper tip of the wing petals . Material from the Western 
Cascades is morphologically subsp. neolaxiflorus according to Dunn's 
key (1957), but occurs in what he considers the range of subsp. silvicola 
Dunn and subsp . arbustus var , montanus Dunn, I see no morphological or 
geographical justification for segregating these three subtaxa, and the 
taxonomy, but not the nomenclature, of Hitchcock (in Hitchcock and 
others, 1961) is followed in t his regard . The following map treats L. 
arbustus subsp . arbustus, subsp. silvicola, and subsp . neolaxiflorus as 
a single taxon under the last name , Subsp . calcaratus (Kell.) Dunn and 
subsp. pseudoparviflorus (Rydb.) Dunn are here recognized as separate 
taxa, not inc luded on the map . 
+ \ 7 
\ 
+ \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
-0 \ 
+ \ 
+ \ + 
+ 
+ 
+ + 
+ + 
+ 
I 
I 
I 
,O I • 
I 
I 
I 
I I 
I 
' ' 110 mi. 
GCOGRAPHIC- DISTRIBUTION OF I 
LUelNU~ ARBUSTUS NEOLAXIFLORUS IN w 
w 
234 
Detling (1953) referred t o this plant as a "xeric indicator." 
Its major distribution is in t he arid lands east of the Cascades in 
Oregon and Washington, extending south through the Sierra Nevada and 
east to Montana and Utah. I t also occurs in dry places in the Willamette 
Valley and has been previously collected several times from the Wes t ern 
Cascades. Here it is found in rela t ively dry meadows, roo ted in light 
but rocky soils whi ch are continually churned by frost action, rodents, 
and mass wasting . This taxon i s abundant and widely distributed, oc~ur-
! 
ring with Gilia aggregata, Linum perenne lewisii, Collomia linearis, 
Erigeron aliceae, Orthocarpus imbricatus, Sedum stenopetalum, Polygonum 
douglasii, Eriogonum compositum, Eriogonum nudum, Artemisia ludovi c iana 
latil oba, and other members of the Xeric Meadow assoc iation. Moisture 
tensions in this plant were the lowest of any species measured in the 
Western Cascades during t he summer of 1967. 
Trifolium produc tum Greene, Erythea 2: 181. 1894. 
Selected synonyms: 
Trifolium kingii var. productum Jeps., Fl. Calif. 2:304. 1936. 
1_. productum is an easily recognized perennial clover, character-
ized by its large (11-18 rnm), reflexed, purpli sh flowers; narrow, 
spinulose-serrate leaflets; and distinc tive rachis, which is produced 
well beyond the head of the flowers, where it often branches and is 
occasionally foliose . 
Two specimens of T. productum have previous ly been collected 
from the central Cascades: l· ~- Leiberg, in 1934, Horsepasture Mt . , 
Lane County; and A. R. Sweetser, i n 1927, "Jack Creek, Metolius Region, 
+ \ 7 
\ 
+ + \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
+ 
A \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ + 
+ + 
++ 
+ 
+O I 
+ I 
+ + I 
+ I 
I 
I 
++ I 
I 
110 rni. 
GEOGRAPHIC' DISTRI BUTION OF 
.._ TR~FOUUM PRODUCTUM 
236 
Crook County." It is the last specimen on which Martin (1943) bases 
the northwestern limit of his range description, but Jack Creek is in 
the southwestern part of Jefferson County, not Crook County as stated 
on Sweetser's label. In addition to the population on Horsepasture Mt., 
three other large populations have been collec ted in the Western Cas-
cades by the author: Bohemia Mountain, Iron Mountain, and Cone Peak. 
In all of the We stern Cascade localities , this clover is found rooted 
in crevices of dark volcanic outcrops or in the scoria gravel derived 
from them. A plant of hot, dry, south-facing slopes,!· productum is 
typically associated with Crepis occidentalis, Aster gormanii, Ivesia 
gordonii, Gilia aggregata, Allium crenulatum, Lotus nevadensis douglasii, 
and Castilleja rupicola. 
Trifolium howellii Wats., Proc. Am. Acad. 23:262. 1888. 
A distinctively robust clover of damp, often marshy places, 
T. howellii can be separated easily from other c lovers by lack of an 
involucre, glabrous calyx, globose heads, large flowers (11-15 mm), 
trifoliate leaves, and large leaflets (2.5-5 cm in breadth). Its 
relationships to other members of this large genus are not clear. 
The range of T. howellii is rather restricted. It occurs west 
of the Cascade crest from Lane County south to the Si skiyou Mountains 
of northern California. It reaches its northernmost limit in the cen-
tral Western Cascades, near the summit of O'Leary Mountain, where it is 
occasionally encountered in moist shady spots on the north and west-
facing slopes under Abies amabilis and Pseudotsuga menziesii. The 
+ \ ----
\ 7 
+ + \ J 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
++ I 
I 
I 
+ I 
e+ I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
- TRLFOLIUM HOWELLII IN 
---- w ---- -..J 
238 
author has also collected this species on Bohemia Mountain, Lane County, 
where it is relatively abundant in open springy areas along with Caltha 
bifolia, Mimulus moschatus, Veratrum viride, and Boykinia major. 
Lotus nevadensis (Wats.) Green var. dougla s ii (Greene) Ottley, 
Britt. 5:81. 1944. 
Selected synonyms: 
Lotus douglasii Greene, Pitt. 2:149. 1890. 
Hosackia decumbens Benth., Bot. Reg. 15: under pl. 1257. 1829. 
The most recent monographer of Western American loti recognizes 
three varieties of L. nevadens is (Ottley, 1923, 1944). The variety 
congestus has shorter branches and internodes, and is found along the 
northern California coast. Variety nevadensis, which is smaller in all 
respects, is found throughout southern California and adjacent Nevada. 
Variety douglasii, the most northern subspecific taxon, grades in t o both 
other varieties in northern California. It is found primarily through 
the Sierra-Cascade chain, with some localities in the Willamette Valley, 
near Puget Sound, and in eastern Oregon and Washington. 
It normally grows in dry rocky places, but its prostrate habit 
and indehiscent beaked legumes make it well adapted to lif~ as a weed. 
In Oregon it is typically found in soil disturbed by human activity. 
It is especially abundant along trails and has been called "mule clover" 
by observant outdoorsmen. Occasionally it occurs along trails in shady 
woods, where it becomes erect and larger in all respects but does no t 
flower. It is not consistently associated with any other species, al-
though it is commonly a member of the Gravel Scree as s ociation with 
+ \ 7 
\ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I e \ 
+ \ 
+ + \ -+ \ 
+ 
+ \ 
+ 
-+ 
+ + 
+ 
~ I 
+ I 
-+ + 
-+ + I 
+ I 
I 
+ I 
I 
I 
' ' 110 mi. 
GEOGRAPHIC, DISTRIBUTION OF 
LOTUS NEVADENSIS DOUGLASII 
,--_ __,__ - - - - - - - -
240 
Chrysothamnus nauseosus, Gilia aggregata, Crepis occidentalis, Ivesia 
gordonii, and Eriogonum compositum. 
Lathyrus lanszwertii Kell o var. aridus (Piper) Jeps., Fl . Calif. 
2:389. 1936 . 
Lathyrus is one of the more diffi cult genera of the Leguminosae, 
exhibiting a great deal of what Hitchcock (1952) terms "for tu itous 
variation." Leaf shape, flowe r color , pubes c ence, and other poten-
tially diagnostic characters vary widely, to the taxonomist's distress. 
L. lanszwertii aridus is, however , an easily recognized taxon, charac-
terized by six narrowly linear leaflets appressed in pairs, a non-func-
tional or completely absent tendril, and small white flowers, often with 
lavender or ochroleucous high ligh ts. The variety lanszwer t ii has more 
leaflets; a functional tendril; and larger, more highly colored flowers. 
It is widespread in e~stern Or egon and extends t o Washington and Idaho, 
Utah, and cen tra l California. Variety aridus is more limi t ed in dis-
tribution but is more abundant where it does occur. It is found in a 
narrow belt on t he east side of the Cascades from Mt . Adams to Mto 
Shasta, approximately occupying t he ponderosa pine bel t. In California 
it occurs in the Klamath Mountains and the northern Sierra Nevada. A 
single locality has been recorded from Owyhee County, Idaho, where this 
variety is typically replaced by variety lanszwertii. 
New locali t ies in the Wes tern Cascades include dry gravel scree 
slopes at high elevations and dry partly-shaded forest floors on the 
lower eastern slope of the range . One other collection from the junc-
tion of the Santiam highways, Linn County (Peck 25911: OSC) has been 
+ \ 7 
\ 
(!) \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
+ \ 
\ 
0 \ + \ 
+ \ 
+ + 
+ 
+ + 
++ 
I 
+ I 
++ I 
+ I 
+ + I 
+ I 
I 
+ I 
++ I 
I 
' ' 110 mi. 
GEOGRAPHIC' .DISTRIBUTION OF 
LATHYRUS LANSZWERTII ARIDUS 
,____,_...__ -- -- -- -
242 
discovered from west of the Cascade crest in northern Oregon. 
Linum perenne L. subspo lewisii (Pursh) Hulttn, Fl. Alas. 7:1122. 
1947. 
Selected synonyms : 
Linum lewisii Pursh, Fl . Am. Sept. 210. 1814. 
Linum perenne var. lewisii Eat. and Wright, N. Am. Bot. 302. 
1840. 
Although l o lewisii of western North America has traditionally 
been considered specifica lly distinct from 1• perenne of Europe and 
Asia, there is only one significant difference between them: Eurasian 
material is heterostylous, while American populations are homostylous. 
Thus, it seems most accurate to consider the morpho-geographical forms 
as taxonomic subspec ies. 
1• perenne is the only perennial Ameri can flax with blue flowers. 
A widespread, variable, montane species, it ranges from Alaska south to 
Cerro Potosi in the Sierra Madre Orient al of Mexico (Beaman and Andresen , 
1966), where it occurs above 3500 m. Prior to this study, the species 
had been collected twice from the central Cascades west of the crest 
(Gilkey, in 1936: OSC; 1· .§_ . Detling 7021). It is also common in the 
dry southwest-facing meadow slopes of Iron Mountain, Echo Mountain, and 
Browder Ridge, where it is associated with Artemisia ludoviciana 
latiloba, Gayophytum diffusum parvi florum, Gilia aggregata, Eriogonum 
compositum, Eriogonum nudum, Polygonum douglasii, Or thocarpus imbricatus, 
Lup inus arbus tus neolaxiflorus, Cerastium arvense, and other members of 
the Xeric Meadow association. 
+ \ ---- 7 
\ • 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ + 
++ 
+ 
+ I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
4+ I 
I I 
11omi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
- LINLJM PERENNE LEWISII IN 
---- .z:-.. ---- w 
244 
Gayophytum diffusum T. and G. subsp. parviflorum Lewis and Szweyk., 
Britt. 16:386. 1964. 
Selected synonyms: 
Gayophytum lasiospermum Greene, Pitt. 2:164. 1891. 
Gayophytum ramosissimum T. and G. var. strictipes Hook., 
Lond. Journ. Bot. 6:224. 1847. 
Gayophytum intermedium Rydb., Bull. Torr. Bot. Club 31:569. 
1904. 
Gayophytum nuttallii T. and G. var. abramsii Munz, Am. Journ. 
Bot. 19:774. 1932. 
The Gayophytum nuttallii complex as discussed by Hitchcock (in 
Hitchcock and others, 1961) has long been puzzling to taxonomists. The 
large number of synonyms for the present taxon is an indication of the 
variety of opinions concerning its identity and status (Munz, 1932). A 
significant breakthrough in the understanding of the genus came with 
the application by Lewis and Szweykowski (1964) of cyto-evolutionary 
techniques. These workers verified the existence of both diploid and 
tetraploid taxa within the group and thereby discovere? that many of 
the traditional key characters for these species are of no taxonomic 
value. Four of the diploid species (Q. decipiens, Q. ramosissimum, 
Q_. oligospermum, and Q. eriospermum) are quite closely related, the two 
former being widespread through the western United States and the latter 
pair being restricted to different portions of California. They have 
retained the ability to interbreed freely, as evidenced by the exist-
ence of Q. heterozygum, a postulated diploid hybrid between Q. oligo-
spermum and Q. eriospermum. This proposed hybrid is found through the 
Cascade-Sierra chain and along the California coast. The variable and 
widespread tetraploid, Q. diffusum, seems to have been derived by 
+ • \ ------ 7 \ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + 
+ 
+ 
+ + 
+ 
+ + 
+Q 
+ 
++ I 
+ I 
+ + ++ I 
9 I 
I 
+ I 
o+ I 
I I 
11omi. I 
GEOGRAPHIC' .DISTRI BUTION OF I 
GAYOPHYT-UM DIFFUSUM PAf1VI FLORUM IN 
---- ~ ---- V, 
246 
contributions of genes from all of these species. The small-flowered 
subspecies parviflorum is the most widespread taxon in the genus and is 
so variable that it is easily confused with any of the parental diploids 
if chromosome counts are not made. An indication of its morphological 
variability is that in their careful key to the 10 taxa of the genus, 
Lewis and Szweykowski arrive at G. diffusum parviflorum through six 
different dichotomies. Western Cascade material is all readily identi-
fiable as G. diffusum parviflorum of the "lasiospermum" type. 
It is my opinion that because the geographical range of G. diffusum 
diffusum is completely contained within that of Q. diffusum parviflorum, 
the two entities should be considered as varieties and not subspecies, 
but no new combinations are proposed here. The two taxa are commonly 
found growing together, but subsp. diffusum, with large flowers, is mod-
ally outcrossing, while subsp. parviflorum is almost completely self-
fertilizing. Intermediates between these two forms are common in some 
localities, indicating that some outcrossing must normally occur in 
populations of Q. diffusum parviflorum. 
All members of the genus except Q. humile occur in dry habitats. 
In the Western Cascades, Q. diffusum parviflorum occurs in ilie .loose 
soil of dry meadows, flowering late in July or August. It is easily 
missed in the spring. Common Xeric Meadow associates include Polygonum 
douglasii, Polygonum minimum, Galium bifolium, Cirsium centaurea, 
Orthocarpus imbricatus, Collomia linearis, and Microsteris gracilis. 
Gayophytum humile Juss., Ann. Sci. Nat. I. 25:18, t.4. 1832. 
Gayophytum humile occurs in widely-separated areas in North and 
0 \ ----- 7 
\ 
+ \ I 
\ 
3mi. · \ I 
t--'------1 \ 
\ I 
\ 
\ I 
t> \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ \ 
\ . 
+ 
e 0 + If 
0 
oe 
+ 
-tP I 
+ I 
-P t) 
0 6 I 
Q I 
I 
+ 
ao I I 
I I 
110 mi. I ' ' 
GEOGRAPHIC' DISTRI BUTION OF I 
- GAYOPHYTUM HUM I LE IN..,...  
---- ---- -..J 
248 
South America. Numerous other distantly-related species exhibit sim-
ilar amphi-tropical distributions (Raven, 1963) . A number of popula-
tions of this species exist in the Andes of Chile and Argentina at t he 
approximate latitude of Santiago. They are morphologically identical 
to North American material (Lewis and Szweykowski, 1964) . 
Because of its dense branching pattern, low habit, and numerous 
seeds aligned obliquely in the capsule, Qo humile is the most distinc-
tive member of this confusing genus . It is, however, very closely 
related to other members of the group and seems to have produc ed the 
tetraploid G. racemosum by amphidiploidy with Qo decipiens (Lewis and 
and Szweykowski, 1964). In North America this species occurs in the 
Klamath Mountain-Sierra arc of California and southern Oregon and again 
in the Columbia Intermontane Provinc e of Washington, northeastern Oregon, 
and adjacent Idaho. A few specimens have also been reported from the 
Rocky Mountains of Idaho, Montana, and Wyoming. The species has evi -
dently never before been collected from west of the Cascade crest north 
of Jackson County . 
In the Western Cascades G. humile occurs in t he spring in run-
ning snowmelt or in the dampest and shadiest portions of Mesic Meadow 
slopes. I t is ephemeral, disappearing when wa ter becomes limiting o 
Other common species in the Rocky Melt Seep association are Mimulus 
breweri, Lewisia triphylla, Polygonum kelloggii, Allium crenulatum, and 
Romanzoffia sitchensis. 
Orogenia fusiformis Wats., Bot . King Exp. 120 0 1871. 
Orogenia is a small and poorly collected genus of western North 
\ - . ___ \ __ _ 7 
\ 
+ \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
\ 
\ 
+ + p 
+ + 
++ 
I 
+ I 
+ E) I 
+ I 
,0 e 
++ I 
.p I 
I 
I 
e+ I 
I 
110 mi. 
GEOGRAPHIC' .DISTRIBUTIOI-.J OF 
, OROGENIA FUSI FOR MIS 
;------.C..-- - - - -
250 
America. Evidently closely related to Lomatium, the genus is charac-
terized by subterete glabrous fruits with corky lateral wings and obso-
lete stylopodium and carpophore. The two component species, O. 
fusiformis and Q. linearifolia Wats., differ in overall size and shape 
of the root, Q. fusiformis having a spindle-shaped taproot and larger, 
more fully developed leaves, while O. linearifolia is characterized by 
a globose root and smaller leaves. The latter species occurs in the 
Rocky Mountains from southeastern Washington and Ravalli County, Mon-
tana, south to Utah and western Colorado. O. fusiformis has a disjunc t 
distribution similar to that of Mertensia bella. It occurs in north-
central Idaho and in the Western Cascades from Marion County south to 
northern California, where it enters the Siskiyou Mountains. Popula-
tions are also known from the northern Sierra Nevada in the Lake Tahoe 
region. The present study has approximately doubled the number of 
collecting sites known for this spec ies. 
Both species of Orogenia are snowbed plants, blooming with the 
earliest vernal assemblages and growing almost without fail at the edge 
of slowly melting snowbanks. Some specimens have been observed pushing 
flowering stalks up through compacted snow. Such early blooming times 
are undoubtedly factors contributing to the paucity of collections. O. 
fusiformis is commonly found with Di centra uniflora, Trillium ovatum, 
Erythronium grandiflorurn pallidum, Senecio triangularis, Hydrophyllum 
occidentale, and other Snowbed association species. 
Pterospora andromedea Nutt., Gen. Pl. 1:269, 1818. 
Pterospora is a monotypic genus of saprophytic heaths. Although 
251 
the type was collected near Niagara Falls, it is typically western, 
occurring normally in dry open forests, usually under Pinus ponderosa 
in the Pacific Northwest (Copeland, 1941). The stem is viscid and 
woody, growing to a height of nearly two meters. It does not, there ~ 
fore, make a manageable herbarium subject and has been poorly collected. 
No map of the distribution of Pterospora is presented in this work. 
The occurrence of the genus Pterospora west of the Cascade crest 
has been documented previousl y by at least one worker (Hopson, 1946). 
During the course of the present study, this species was found growing 
under a Mesic Conifer Forest canopy dominated by Pseudotsuga menziesii 
on Castle Rock and Tidbits Mountain. 
Rhododendron albiflorum Hook., Fl. Bor. Am. 2:43. 1834. 
This interesting species reaches its southernmost limit of known 
range within the study area. It is distinct from all other rhododen-
drons, having 10 stamens, deciduous leaves, and relatively small (2 cm) 
white corollas. !• albiflorum ranges from the Selkirk Mountains of 
British Columbia south in the Rocky Mountains to Baker County, and to 
Indian Ridge in the central Western Cascades. It grows in damp sites, 
either along streams or lakes or on shaded north-facing slopes where 
snowmelt runs late in the season. It is typically a high montane spe-
cies, occurring as low as 1200 to 1500 m only in the Western Cascades, 
where it has now been collected from four localities. 
A member of the most mesic subdivision of the Mesic Conifer 
Forest,!• albiflorum is commonly associated with Menziesia ferruginea, 
+ \ 
0 \ 
.+O + 
----77 
\ 
\ 
\ I 
\ 
\ I 
\ 
+ \ I 
\ ,--..___  _J__ _- -.--_.-1 ~ 
+ \ 
+ + + \ 
\ I + 
+ \ 
\ I 
\ 
+ 
+ + fl 
+ I 
I 
++ 
I 
I 
+ I 
++ I I 
+ I 
++ + I 
+ I 
I 
++ I I 
I I 
' \ I 
GEOGRAPHIC' DISTRI BUTION OF 110 mi. I 
RHODODENDRON ALBIFLORUM IN 
r-------- - - - - V, ------- N 
253 
Vacc inium membranaceum, Senecio triangularis, Ligusticum grayi, and 
Valeriana sitchensis. 
Menziesia ferruginea Smith, Pl. le. Ined, pl. 56, 1791. 
Selec ted synonyms: 
Menziesia glabella Gray, Syn. Fl. 21 :39. 1878. 
Menziesia ferruginea var. glabella Peck, Man. High. Pl. Oreg. 
542. 1941. 
Menziesia ferruginea subsp. glabella Calder and Taylor, Can. 
Journ. Bot. 43:1398. 1965. 
Recent work has shown that the two extreme forms of western 
North American Menziesia must be considered conspecific, and further 
that there are no justifiable grounds for drawing coherent subspecific 
lines (Hi ckman and Johnson, 1968). M. ferruginea is most closely re-
lated to the Appalachian .~. globularis Salis., although there are also 
several species of this genus native to Japan. !:!· ferruginea occurs 
from the Kenai Peninsula south along the coas t to Humboldt County, Cali -
fornia , and through the Cascades and Rocky Mountains to Tidbits Mountain 
and northwestern Wyoming, respectively. 
!:!• ferruginea is always found in moist habitats similar to those 
occupied by Rhododendron albiflorum, with whi ch it is often associated 
in the Western Cascades. Other members of the Mesi c Conifer Forest 
frequently found in the vicinity inc lude Vacc inium membranaceum, Senec io 
triangularis, Ligusticum grayi, and Valeriana s itchens is, !:!• ferruginea 
is evidently quite localized and highly disjunc t 1n the central Western 
Cascades , Populations occurring here constitute the southernmost known 
occur r enc es of the montane form. 
.. 
• \ ----\ 7 
·+ 
+ \ I 
\ 
3mi. \ • I \ 
-- \ I 
\ 
\ I 
+ 
.6. \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
QO + 
+ 
++ 
..... .. + ++  I 
+ I 
+ 
++ + I 
+ I 
I 
+ I 
++ I 
I 
' ' 
' 110 mi. 
GEOGRAPHIC -DISTRIBUTION OF 
MENZIESIA FERRUGINEA N 
---- ---- V, ~ 
255 
Douglasia laevigata Gray, Proc. Am. Acad. 16:105. 1880. 
Selected synonyms: 
Douglasia laevigata var. ciliolata Const., Am. Mid. Nat. 
19:254. 1938. 
Douglasia laevigata subsp. ciliolata Cald. and Tayl., Can. 
Journ. Bot. 43:1398. 1965. 
Constance (1938) notes that generic distinctions in this portion. 
of the Primulaceae (Androsaceae-Primulinae) are weak. Douglasi~'s 
closest relatives are probably Androsace and Primula, all three of 
which have obvious arctic affinities. There are six species of Douglasia. 
Two are arctic American, three are from the higher western Cordillera, 
and the remaining, which is sometimes segregated into the genus Gregoria, 
is from the Alps-Pyrenees chain. Constance believes that Douglasia is 
probably a pre-glacial genus whose distribution can now be considered 
relictual. D. nivalis, a form with stellate pubescence and an umbellate 
inflorescence, occurs from the northern Rocky Mountains to the Wenatchee 
Mountains of Washington and is probably the cloiest relative of D. 
laevigata. Constance divides the present species into two varieties, 
the typical form occurring in the Columbia Gorge at elevations as low 
as 30 m. The more widespread variety ciliolata, differentiated by the 
number and size of the simple cilia on the leaf margins and the more 
compact umbellate inflorescence, is found in the mountains of Vancouver 
Island to the Olympic Peninsula, where it is abundant, and also on 
Saddle Mountain, Clatsop County. In the Cascades its previously known 
range was from Snohomish County, Washington, south to Mt. Hood. The 
Western Cascade populations are intermediate between the two named var-
ieties, and their separation is questioned by the present author. 
+ • \ ---- 7 \ 
+ + \ I 
\ 
3mi. \ I 
\ 
- \ I \ 
e \ 
I 
A \ 
+ \ 
+ \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
++ 
A 
+ A 
++ A I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
+ ++ I 
I I 
' ' 110 mi. I 
GEOGRAPHIC' _DISTRIBUTION OF I 
- DOUGLAS IA LAEVIGATA IN 
---- 1./1 ----· 0\ 
257 
Accordingly, they are treated as a single taxon here. 
In the Western Cascades, as on Saddle Mountain, Douglasia grows 
on vertical faces of basalt or andesite, rooted in crevices. An inter-
esting aspect of its local habitat is that it grows only on faces that 
are well exposed to the prevailing winds. I t seems likely that moisture 
in the forms of mist and fog, common in all the areas where the species 
occurs, is a critical factor in its distribution. Other species of the 
Vertical Outcrop association include Saxifraga bronchialis vespertina, 
Penstemon rupicola, Castilleja rupicola, and Heuchera micrantha. 
Convolvul,us nyctagineus Greene, Pitt. 3:327. 1898. 
Selected synonyms: 
Convolvulus atriplicifolius House, Muhl. 4:54. 1908. (not 
C. atriplicifolius Pair.) 
A typically Oregonian species, C. nyctag i neus is distinguished 
from other native morning glories by its broad floral bracts, which con-
ceal the calyx; its deltoid-hastate leaves; and i ts trailing to erect 
(but not twining) habit. Although it has been collected from the coast 
mountains of Curry County and Del Norte County, California, it is pri-
marily found in dry places in the Willamette Valley and in the Western 
Cascades. It has also been collected occasionally from the eastern end 
of the Columbia Gorge and north to the vi c inity of Mt. Adams on the 
eastern slope of the Cascades. Three new localities are reported here, 
two of which (O'Leary and Horsepasture Mountains) include the highest 
elevations yet reported for the species (1650 m). 
+ .A \ ----
\ 7 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ 
A \ 
+ \ 
+ + \ + 
+ 
+ 
+ 
+ + 
+ + • + • 
++ 
• ... .. e 
ee ... I 
+ I 
+ + ++ I 
+ I 
+ I 
I 
++ I 
I I 
' ' 11omi. I 
GEOGRAPHIC, DISTRIBUTION OF I 
C.ONVQLVULUS NYCTAGINEUS L_, 
---- ---- Ul co . 
259 
In the Willamette Valley,~- nyctagineus is a plant of gravelly 
roadsides and dry fields and pastures. In the Western Cascades it is 
restricted to dry south-facing slopes of open gravelly loam. Here it 
occurs with Erigeron foliosus confinis, Phacelia linearis, Polygonum 
spergulariaeforme, Githopsis speculari~ides, Plectritis congesta, 
Gayophytum diffusum parviflorum, Microsteris gracilis, and other mem-
hers of the Lowland Xeric Meadow association. 
Polemonium pulcherrimum Hook., Curtis' Bot. Mag. 57: pl. 2979. 1830. 
Selected synonyms: 
Polemonium viscosum Nutt. var. pilosum Greenm., Bot. Gaz. 
25:263. 1898. 
Polemonium pilosum Jones, Univ. Wash. Puhl. Biol, 5:125. 1936. 
Polemonium shastense Eastw., Bull. Torr. Bot. Club 32:205. 1905. 
Wherry (1967) recognized "six major taxa" of temperate tufted 
polemoniums. Of these, four (perhaps five) should be treated as members 
of a single polytypic species. Besides typical f. pulcherrimum, they 
include f. parvifolium Nutt., f. delicatum Rydb., f. californicum Eastw., 
and possibly f. nevadense, all of which intergrade completely and to-
gether range through the mountainous regions of western North America. 
Although the most recent monograph of the genus retains these taxa at 
the subspecific level (Davidson, 1950), Cronquist (in Hitchcock and 
others, 1959) lumps all of these close relatives under the name P. 
pulcherrimum, recognizing three "varieties," two of which occur along 
the Cascade-Sierra axis. Variety ca lycinum (Eastw.) Brand is the more 
robust larger-flowered ecotype which is found at lower elevations, 
260 
especially on the high peaks, while var. pulcherrimum comprises the 
reduced alpine and subalpine forms. All intermediate stages are known 
between these two varieties, but in some localities (such as the Three 
Sisters region) they are ecologically and morphologically quite distinct. 
Davidson (1950) proposes that both the P. viscosum complex (in 
which he includes_!:. elegans Greene) and the~- pulcherrimum group are 
descended from the northern P. boreale. He does not, however, note the 
connection between these two major c lusters of specie s through f. elegans 
and the alpine forms of_!:. pulcherrimum, which approach one another so 
closely that most of the high alpine Oregon material of the former has 
been identified as the latter. Specimens from the central Cascades of 
Oregon are clearly intermediate between the more robust forms of P. 
pulcherrimum from Crater Lake Na t ional Park and southward and_!:. elegans 
from Mt. Rainier and northward. Although this c lose relationship needs 
to be emphasized, the two species do seem to be separable, mainly by 
subtle differences in growth habit. Due to the uniformity of_!:. elegans 
as it occurs in the Washington Cascades, it seems likely to the present 
author that on morphological grounds all Oregon material to date belong s 
in the more variable taxon _!:. pul cherrimum. It is, however, perhaps 
instructive to note the similar ities between the ranges of_!:. elegans 
(if the Western Cascade forms could be assigned to this species) and 
other boreal species such as Castilleja rupi cola . Bo t h species grow in 
quite similar habitats and locali t ies in the Washing t on Cascades (espe-
cially on Mt. Rainier) and reappear together in the Western Cascades of 
Oregon. 
A single population of P. pul cherrimum is known from the Wes t ern 
+ • \ ----~---- 7 
+ + \ I 
\ 
3mi. \ I 
\ 
-~- \ I 
\ 
\ I 
+ 
+ A 
\ 
\ 
+ + \ 
++++  
+ 
C, + 
+ + + 
+ + • 
+ 
+ e + 
++ • 
+ 
++ I 
+ I • 
+ + I 
+ I 
+ ' I 
I 
++ I 
• I 
11omi. 
GEOGRAPHIC' .DISTRIBUTION OF 
Pf' LEMONIUM PULCHERRIMUM PULCHERRIMUM 
r-----'------
262 
Cascades. This species is well-established in vertical crevices on the 
southwest-facing basalt cliffs of Iron Mountain. Morphologically iden-
tical forms are the highest flowering plants occurring on Middle Sister, 
Lane County, at elevations of over 3000 m. Lower elevation forms nor-
mally identified as f. californicum, but equivalent to f . pulcherrimum 
var. calycinum, have also been collected from the lower slopes of the 
High Cascade peaks throughout Oregon but have never been found in the 
Western Cascades. 
On Iron Mountain f . pulcherrimum is a member of the Vertical 
Outcrop assoc iation, and occurs with Saxifraga bronchialis vespertina, 
Penstemon rupicola, Selaginella wallacei, Saxifraga cespitosa, Heuchera 
micrantha, and Polypodium hesperium . 
Cilia aggregata (Pursh) Spreng., Syst . 1:626. 1825. 
Selected synonyms: 
Ipomopsis aggregata (Pursh) Grant, El Aliso 3:360 . 1956 • 
. , 
Although Grant (1956) presents arguments for the separation of 
Ipomopsis from Cilia for the sa~e of internal consistency within the 
Polemoniaceae, his arguments have not been gen~rally accepted by other 
taxonomists (Cronquist , in Hitchcock and o t hers, 1959) . Cronquist's 
treatment is followed here . Cilia aggregata is the showiest of the 
gilias, immediately re.cognizable by i ts large ( 2-4 cm), scarlet, long-
tubular corollas. It is, however , a variable taxon, divided at times 
into as many as seven barely delimited species (Grant, 1956; Wherry, 
1946, 1961a). In the Pacific Northwest, Q. aggregata is quite uniform, 
dry regions of the southwestern United States being the center of 
+ • \ ---- 7 \ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
• \ I + 
A \ 
o+ \ 
.p \ •• • + + 
+ 
+ • 
• 
+ 
oe 
+ e • 
0 + + • 
++ 
• 
A • 
+ ... . •• • • -dt • I •• •  • •'  •+ I •  
$0 + 0 I • • 
C9 I •e • 
I • • 
+ I • • • 0 
+ o+ I • 
I • • • I 
• • 110 mi. I 
GEOGRAPHIC' DISTRIBUTION OF • I 
GILIA AGGREGATA • • IN ---- O" --• -- w 
264 
variability for the species. All ecotypes studied have been shown to 
be self-incompatible (Grant, 1956) and are visited by hummingbirds (in 
the Western Cascades Selasphorus rufus) throughout the flowering season • 
.Q_. aggregata is widespread throughout the montane and intermon-
tane regions of western North America, but has heretofore been consid-
ered to be strictly limited to areas east of the Cascade crest in the 
Pacific Northwest (Cronquist, in Hitchcock and others, 1959). It has 
been found in the course of the present study that this species occurs 
in dry south or west-facing meadows or on open gravelly slopes in the 
Western Cascades where it is frequently a dominant. Other members of 
the Xeric Meadow association with which it is often found are Collomia 
linearis, Gayophytum diffusum parviflorum, Orthocarpus imbricatus, 
Polygonum minimum, Polygonum cascadense, Polygonum douglasii, Lupinus 
arbustus neolaxiflorus, and Linum perenne lewisii. 
Linanthastrum nuttallii (Gray) Ewan, Journ. Wash. Acad. Sci. 32:139. 
1942. 
Selected synonyms: 
Gilia nuttallii Gray, Proc. Am. Acad. 8:267. 1870. 
Linanthus nuttallii Greene ex Milliken, U. Cal. Puhl. Bot. 
2:54. 1904. 
This recently constituted monotypic gilioid genus is morpholog-
ically intermediate between the older genera Leptodactylon and Linanthus 
in habit, ecology, and leaf and calyx morphology. Linanthastrum may, 
in fact, be the modern representative of the stock from which Linanthus, 
an annual genus, has evolved (Ewan, 1942; Wherry, 1961b). A 
+ A \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
1--------1 \ 
\ I 
\ 
\ I 
+ \ 
+ A \ 
+ + \ + \ 
+ 
+ \ 
+ \ 
\ 
+ + 
+ 
+ jJ + 
I O • I 
++ • 
.. • + •  • 
++ • I - • 0 + I + ++ + I 
+ I 
I 
+ I 
+ +e 
•• 
I 
I 
110 mi. 
GEOGRAPHIC' DISTRIBUTION OF • 
L-1 NAN..THASTRUM NUTT ALLI I __• _ .__ • 
--• 
N 
-- V°', 
266 
suffrutescent, s ometimes matted perennial, 1· nuttallii is further 
characterized by its large white and yellow flowers, non-pungent 
falsely-whorled leaves, and lack of intercostal calyx membranes. 
Although widespread in high montane regions throughout much of 
western North America, 1· nuttallii has been collected from only three 
localities in the Cascades of Oregon. One of these is in the study 
area, on t he high south-facing cliffs of Rebel Rock. The other two 
localities are Fairview and Bohemia Mountains. At all three locali-
ties the species is rooted in crevices of outcrops of volcanic rock or 
in pockets of fine scree on steep south-facing slopes. A member of the 
Gravel Scree or Outcrop Ridge associations, it occurs with Lotus 
nevadensis douglasii, Gilia aggregata, Arctostaphylos nevadensis, 
Haplopappus hallii, Sedum stenopetalum, Lomatium martindalei, and 
Cheilanthes siliquosa. 
Linanthus harknessii (Curran) Greene, Pitt . 2:255. 1892. 
The annual species of Linanthus are, according t o Wherry (1961b), 
descended from perennial species of Linanthastrum and Leptodactylon. 
Wherry uses calyx features, which he considers to be of utmost import-
ance in this family, to support his arguments. Linanthus is most 
highly diversified in California, where over 30 species are found. Al-
though most of these species seem to be distinct , there is a good deal 
of confusion in the literature regarding their ranges (Peck, 1961; Munz 
and Keck, 1959; Hitchcock and others, 1959; Mas on, 1938). It is clear 
that a thorough monographic study of the genus is needed. 
+ A \ ---- ---- 7 
\ 0 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
++ 
• 
A 
+ .. 
i:X) A I • 
+ I 
+O + 
++ I 
+ I 
• 
I 
+ I 
-o+ I 
I I 
11omi. I 
GEOGRAPHIC' UISTRIBUTION OF I 
- LINANTHUS HARKNESSII IN 
.(..j..'.\,  
268 
L. harknessii as here treated is differentiated from its c losest 
relative,.!:_. septentrionalis Mason, by its habit of growth, smaller 
glabrous corolla and single seed per locule. Both species are found 
extensively throughout eastern Oregon, often growing in the same stand • 
.!:_. septentrionali.s becomes more common farther east and is the dominant 
species in the Rocky Mountains. L. harknessii occasionally extends 
westward across the Cascade crest into the Western Cascades. An annual, 
it grows in moist , open, gravelly sites on ridgetops or in trails. It 
rapidly desiccates as snowmelt diminishes. Other members of the Rocky 
Melt Seep or Grave l Scree associations with whi ch it frequently occurs 
include Allium amplectens, Lewisia triphylla, Mimulus breweri, Polygonum 
kelloggii, Polygonum cascadense, Navarretia divaricata , Gayophytum 
humile, and Saxifraga integri.foli.a. 
Navarretia divaricata (Torr.) Greene, Pi tt. 1:136. 1887. 
Selected synonyms: 
Gi lia divaricata Torr. ex Gray, Proc. Am. Acad. 8:270. 1870. 
Crampton (1954) recognizes two evolutionary lines within the 
gilioid genus Navarretia. Section Fragi.les , adapted to vernal pool 
environments, has uninervate coro llas, entire or shallowly lobed stigmas, 
indehi scent capsules, and shallow stamen insertion. This section finds 
its highest development in the southwestern United States . Section 
Eunavarretia, which includes!!• divari cata and 20 other species, com-
prises those forms with variable ecologies, trinervate corollas , deeply 
cleft stigmas, and low stamen insertion. N. divari cata is distinguished 
from other species of its sec t ion by its red or br own pigmented stems, 
+ \ 7 
\ 
+ \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
t) \ 
A 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ + 
-+ 
+ + 
+ • 
I • 
I ••• 
I 
.() I 
I 
-+ I 
ti' I • 
I • 
116 mi. 
GEOGRAPHIC' .DISTRIBUTION OF 
-NAVARRETIA DIVARICATA 
270 
which branch divaricately from just below t he terminal head of flowers. 
It is most closely related to Q. peninsularis Greene of southern and 
Baja California, Q. prolifera Greene of the Sierra Nevada foothills, 
and Q. breweri (Gray) Greene, a yellow-flowered form of the intermon-
tane regions from Washington to Arizona, 
Throughout its range, from southern Washington east to central 
Idaho and south to southern California, Q. divaricata occurs in dry 
open spots, typi ca lly rooted in fine mineral soil or gravel. It has 
not been previously reported f r om the Cascade Range north of the 44th 
parallel. It commonly grows in trails in the Western Cascades, and its 
slightly mucila&inous seeds may indicate animal dispersal. N. divaricata 
is a member of the Xeric Meadow assoc iation and commonly grows with 
Gayophytum diffusum parviflorum, Gayophytum humile, Orthocarpus imbri-
catus, Polygonum douglasii, Eriogonum nudum, Polygonum cascadense, 
Cerastium arvense, and Artemisia ludoviciana latiloba. 
Collomia linearis Nutt., Gen . Pl . 1:126 . 1818 . 
This easily recognizable Collomia is the most widespread member 
of the genus, occurring throughout the dry open lands of the Western 
United States and Canada, It has also been reported from Wisconsin, 
Ontario, and Quebec. C. linearis has evidently been introduced into 
some areas (Hitchcock, in Hitchcock and others, 1959). It is possible 
that recent introduction has established it in the Western Cascades, 
where the species is relatively common in some of the more heavily 
grazed, dry, south or west-facing slopes between 1350 and 1650 m 
+ A \ 7 
\ 
+ + \ 
• I 
\ 
3ml. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
A 
+ \ 
+ \ + 
+ 
+ 
+ 
+ + 
+ 
+ 0 + 
• 
++ I 
.. I A+ .  I • I ++ .. I •• 
+ I • 
++ + 
+ I 
+ I • • • 
I 
+ I 
+ + I 
I 
116 mi. 
GEOGRAPHIC' _DISTRIBUTION OF 
0 
- COLLOMIA LINEARIS • N 
,__ _ ......, _____....___ - - - - I-' 
272 
elevation • .£. linearis has previously been reported from only three 
localities in Oregon north of the 44th parallel and west of the Cascade 
crest. Two of these collections are from dry areas in the Willamette 
Valley, the third from the Western Cascades. 
Other members of the Xeric Meadow assoc iation found with C. 
linearis are Gilia aggregata, Gayophytum diffusum parviflorum, Luina 
s tri c ta, Orthocarpus imbricatus, Polygonum douglasii, Microsteris 
gracilis, and Collinsia parviflora. 
Hydrophyllum fendleri (Gray) Heller var. albifrons (Heller) Macbride, 
Cont. Gray Herb. n. s. 49:23. 1917. 
The genus Hydrophyllum is composed of two distinct assemblages of 
species; one occurs in the eastern United States and the other west of 
the Rocky Mountains. The two groups are partially linked through.!:!• 
tenuipes and.!:!· virginianum, which are geographi cally distinct but mor-
phologically very similar (Constance, 1942). H. fendleri is intermediate 
in leaf and inflorescence morphology between.!:!· tenuipes and H. occi-
dentale. Its range can be divided into three distinct parts. Variety 
fendleri occurs from the Medicine Bow Mountains of Wyoming south through 
Colorado and Utah to central New Mexico. Variety albifrons is found 
from the Cascade Mountains of British Columbia south through the Cas-
cades and Klamath Mountains. In the Snake River region of Oregon, 
Washington, and Idaho, a disjunct set of populations containing members 
of both varieties occurs (Constance, 1942). There is some disagreement 
in the literature concerning the existence of intermediate forms. If 
it is eventually shown that the two taxa represent biologically discrete 
+ A \ ----- 7 
\ 
-0 + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
-0 
+ A 
\ 
\ 
+ -+O \ + 
+ 
+ 
+ 
+ + 
+ 
+ + 
. 
++ 
..A  I + I 
-tO • A. I 
~ I 
++ +o I 
9 I 
I ~ .p 
I 
++ I 
I I 
' 11omi. I 
GEOGRAPHIC' DISTRIBUTION OF I 
HYDROPH.YLLUM FENDLER! ALBIFRONS IN 
....i 
---- ------ v,) 
274 
species, this most unusual distribution pattern will prove easier to 
explain. 
In the Western Cascades Hydrophyllum fendleri albifrons is typ-
ically a plant of moist shady meadows or streamsides, growing only in 
areas of rich soil that does not dry completely during the summer. It 
is commonly associated with Mertensia bella, Mitella breweri, Ligusticum 
grayi, Veratrum viride, Valeriana sitchensis, Hydrophyllum tenuipes, 
Senecio triangularis, Rubus spectabilis, and other Wet Meadow species . 
It also occurs commonly beneath dens~ thi ckets of Alnus sinuata or Acer 
circinatum. 
Phacelia linearis (Pursh) Holz., Contr. U.S. Nat. Herb. 3:242. 1895. 
Although_!'.,. linearis has generally been considered a close rela-
tive of the P. franklinii complex (Howell, 1945), more recent experi-
mental work has indicated that it has probably constituted a monotypic 
evolutionary line within this complex genus for some time (Gillett, 
1962). Although s pecies of Phacelia do not seem to have developed gen-
etic isolating barriers to any considerable extent, P. linearis does 
not set seed when crossed with any· other species yet tried . Plants are 
self-compatible, but protandry results in modal outcrossing for this 
species (Gillett, 1962). 
_!'.,. linear is i :s a widespread and common species. Al though sev-
eral localities are known from the Willamette Valley and the area imme-
diately around Puget Sound, it is primarily confined to dry open areas 
east of the Cascades, extending from central British Columbia to 
+ \ ----
\ 7 
+ + \ 
\ I 
3ml. \ I 
\ 
\ I 
\ 
+ \ I 
\ 
+ \ 
+ + + \ 
+ \ 
+ \ 
+ 
+ + ... 
+ + 0 
+ 
++ 
A 
. ++ A. I 
+ I 
++ + 
+ 
I 
+ I 
+ I 
8++ I I 
I I 
' ' I 
GEOGRAPHIC' DISTRIBUTION 110 mi. OF I 
PHAGE LIA LINEAR IS IN 
---- ---- -...J V, 
276 
southern Utah and as far east as eastern Wyoming. It might be expected 
that_!:. linearis would be relatively common in the dry south-facing 
slopes of the Western Cascades, but it is presently known from only 
three localities in this region and is evidently confined to the driest 
windswept exposures at the tops of south-facing cliffs or steep rocky 
slopes. The author has found it to be common in only one site--near 
the summit of Castle Rock. 
A member of the Gravel Scree or Outcrop Ridge associations, this 
species has been found with Artemisia tridentata, Crepis acuminata, 
Plectritis congesta, Polygonum spergulariaeforme, Sanicula graveolens, 
Sedum stenopetalum, and Comandra umbellata. 
Cryptantha affinis (Gray) Greene, Pitt. 1:119. 1887. 
Cryptantha affinis, an annual borage, is distinguished by its 
small flowers, bristly herbage, and four obliquely compressed nutlets 
(Johnston, 1925). Although abundant east of the Cascade crest, it had 
previously been collected only twice west of this line in the Pacific 
Northwest. It occurs as far east as Albany County, Wyoming, and as far 
south as southern California. 
C. affinis is found occasionally as a constituent of the Xeric 
Meadow association with Polygonum cascadense, Polygonum minimum, Nav-
arretia divaricata, and Orthocarpus imbricatus. It -rarely occurs in 
moister habitats with Luina stricta and Galium bifolium. Unlike many 
borages C. affinis has smooth and shiny fruits which are retained for 
long periods in the spiny dry calyx, which may act as a bur. 
+ A \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
t-------t \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ e \ + \ 
+ 
+ \ 
+ 
+ + A 
+ + + 
++ 
+ A • 
++ A I • 
+ I 
+ + ++ I 
+ I 
I 
+ I 
+ ++ I 
I I 
' ' 110 mi. I 
GEOGRAPHIC' _DISTRIBUTION OF I 
- CR'lPTANTHA AFFINIS IN 
---- -...J ---- -...J 
278 
Plagiobothrys scouleri (H. and A.) Johnst., Contr. Gray Herb. n. s. 
68:75. 1923. 
Selected synonyms : 
Allocarya scouleri Greene, Pitt. 1:18. 1887. 
Allocarya media Piper, Contr. U. S. Nat, Herb, 22:107. 1920. 
Plagiobothrys medius Johnst., Contr. Arn, Arb, 3:58, 1932. 
Allocarya granulata Piper, Contr. U.S. Nat. Herb. 22:109. 
1920. 
Plagiobothrys granulatus Johnst., Contr. Arn. Arb. 3:57. 1932. 
Allocarya fragilis Brand, Fedde Rep. Sp. Nov. 18:312. 1922. 
This species, widespread as a whole, is polytypic and has a con-
fusing nomenclatural history (Piper, 1920; Cronquist, in Hitchcock and 
others, 1959). There are two modes of variation. The first and most 
common, var. penicellatus (Greene) Cronq., consists of vernal pool 
forms which are prostrate in habit and show great diversity of nutlet 
sculpturing. These forms are widespread throughout western North Amer-
ica east of the Cascades (Ornduff and French, 1958). Variety scouleri 
is more constant in morphology and is limited to areas west of the Cad-
cade crest except for one locality on the eastern end of the Columbia 
Gorge. It ranges from Humboldt County, California, to Vancouver Island 
and is the form of interest here. It is erect in habit, larger-flowered 
than var. penicellatus, and is not found in vernal pool habitats but in 
dry rocky areas, especially in the Willamette Valley and occasionally 
along the Pacific Coast. Only variety scouleri is included in the fol-
lowing map. 
One population of P. scouleri is known from the Cascade Mountains 
of Oregon. It occurs in dry meadow and scree areas on the south-facing 
+ \ ------ 7 
\ 
t + \ I 
\ 
3mi. \ I 
1----i \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ - + \ 
+ + \ 
+ +++ + 
+ \ 
+ +o 
+ + A 
+ 
+ + 
++ 
I 
+ I 
++ I 
+ I 
++ + 
+ I 
+ I 
I 
+ I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
PLAGIOBOTHRYS SCOULERI SCOULERI IN 
'-J 
---- ---- \0 
280 
slope of Browder Ridge, where it i s associated with such members of the 
Xeric Meadow and Gravel Scree associations as Sanicula graveolens, Eri-
ogonum umbellatum, Sedum oregonense, and Juniperus communis saxatilis. 
Mertensia bella Piper, Proc. Biol. Soc. Wash. 31:76. 1918. 
This beautiful and distinctive member of an extraordinarily 
difficult genus comprises the monotypic section Neuranthia (Williams, 
1937). It is distinguished by its cormlike globose root and campan-
ulate corolla, which flares from the base without any constriction 
marking the division between tube and limb that is characteristic of 
all other members of the genus Mertensia. It seems to have no close 
relatives (Piper, 1918). 
The distribution of M. bella is remarkably disjunct. It has 
been collected from several localities near the Oregon-California 
border in Josephine and Siskiyou Counties; from the Western Cascades 
(type locality: Horsepasture Mountain); and from one locality each in 
Idaho and Clearwater Counties, Idaho. It has been poorly collected and 
was thought until the present to be rare throughout its range. However, 
15 additional collecting localities for the species have been found in 
the Western Cascades. It appears that this spec ies, while local in its 
occurrence, is a common and well-established member of the Cascade flora. 
Mertensia sites can often be located by attention to other spe-
cies which are more easily noti c ed and occur faithfully with this borage. 
They include members of the Wet Me a dow association, particularly Veratrum 
viride, Senecio triangularis, Valeriana sitchensis, Ribes bracteosum, 
+ A \ ----- 7 
\ 
+ \ I 
\ 
3mi. \ I 
t---1 \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ 
+ + \ 
0 + 
+ \ 
+ \ 
\ 
o+ 
+ c!> 
b 
+ 
++ 
A 
+ 
~ I 
+o i> 
I 
++ I 
~ I 
.p I 
I 
tP I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
MERTENS IA BELLA IN 
CP 
----- ---- ~ 
282 
Rubus spectabilis 9 Hydrophyllum fendleri albifrons, and Mitella breweri. 
Monardella odoratissima Ben th., Lab. Gen, and Sp. 332. 1834. 
The genus Monardella continues to be problematic. The member 
taxa compose a highly complex network of var iable and interfertile pop-
ulations, Little has been done in the way of cytogenetic or breeding 
analysis. It seems highly unlikely to the present author that the num-
erous morphologically defined entities, wh ich show poor geographical. 
differentiation, can represent biologica lly discrete species or sub-
species. Epling's 1925 monograph is the most recent thorough study of 
the genus. He considers M. odoratissima one of the points of central 
tendency around which sections of the genus are established. This spe-
cies, though relatively distinctiv e for the genus, seems nevertheles s 
to intergrade completely with the~- villosa complex. M. odoratissima 
subsp. odoratissima is normally found along the western edge of the 
Rocky Mountains from northeastern Washington and northern Idaho south to 
central and northeastern Oregon and cen tral Idaho. Ep l ing al s o inc l uded 
isolated specimens from Nevada, New Mexico, and the Columbia Gorge in 
this subspecies, which Western Cascade material most closely approxi-
mates. Interv ening between the Rocky Mountain and Cascade populations 
of M. odorati ssima odoratissima in the dry regions of the Columbia 
Plateau and the High Lava Plains of Des chutes County , Oregon, is the 
much more densely pubescent subspecies di scolor. Subspecies g l auca , 
also closely related to subspecies odoratissima 9 occurs in the Si skiyou 
region and the Wallowa Mountains . It is extremely difficult to imagine 
any of these forms as discrete evolutionary units. 
,-
+ \ ----
\ 7 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
+ \ 
I 
\ 
+ \ 
+ + + \ 
+ 
+ 
+ 
+ + 
+ + + 
++ 
+ h. 
++ A I 
+ I 
+ + ++ I 
+ I 
+ I 
I 
+ ++ I 
I I 
' 110 mi. I 
GEOGRAPHIC -OISTRIBUTION OF I 
MON ARDELLA ODORATI SSI MA IN 
---- ---- 00 w 
284 
Perhaps all three "subspecies" have contributed to the Western 
Cascade gene pool, Leaf pubescence characteristics are in large part 
diagnostic of the described subspecific taxa. Western Cascade specimens 
have glabrous leaves for the most part, with southern specimens showing 
a distinct tendency toward hirsuteness along t he veins, especially on 
the undersurface of the leaves. This tendency may represent an influx 
of "discolor-like" genes into glauca-odoratissima populations, The 
latter taxa are distinguished by Epling on the basis of bract shape and 
pubescence. Subspecies odoratissima typically has round pubescent 
bracts, and subspecies glauca has elliptical puberulent to glabrous 
bracts. These populations thus seem to represent a combination of genes 
such as are presently typical of t he Columbia Gorge and Columbia Plateau, 
the Siskiyou Mountains, the Great Basin, and the Wallowa Mountains. 
The genus Monardella is characteristic of dry rocky habitats 
throughout its range, In the Western Cascades it grows in crevices in 
volcanic rock on south-facing slopes or along barren ridgetops. Typ-
ical Xeric Meadow associates are Castilleja pruinosa, Penstemon procerus 
brachyanthus, Sedum stenopetalum, Comandra umbellata, Lomatium martin-
dalei, Gilia aggregata, and Sanicula graveolens. 
Penstemon deustus Dougl. ex Lindl., Bot . Reg, 16: pl. 1318. 1830, 
Penstemon deustus is the only common member of its immediate 
group, which includes three narrowly endemic species of subsections 
Deusti and Arenarii of Section Graciles. P. tracyi, a montane endemic 
of Trinity County, California, is the other member of the subsection 
285 
Deusti and is the single closest relative off. deustus (Keck, 1940). 
P. deustus is easily distinguished by its sharply toothed leaves and 
small, purple-veined, creamy-white corollas. It is common in arid 
country from central Washington and western Montana and Wyoming south 
to central California and Nevada . Douglas' type locality, "scorched 
rocky plains in the interior of northwest America ••• ," is the basis 
for the common name "scorched penstemon." In the Western Cascades it 
grows in andesitic crevic es on south-fac ing outcrops that become very 
hot and dry in midsummer when these plants bloom. A member of the 
Outcrop Ridge association, it occurs with Chrysothamnus nauseosus 
albicaulis, Eriophyllum lanatum, Gilia aggregata, Arenaria capillaris 
americana, Lotus nevadensis douglasii, Silene douglasii, and Haplopappus 
hall ii. 
P. deustus has been div{ded into four intergrading varieties 
which are treated together here. All Western Cascade material is of 
the widespread variety deustus, which occurs commonly east of the Cas-
cades and in two sites in the Willamette Valley. Variety variabilis 
(Suks.) Cronq., which has frequently whorled narrower leaves which are 
not as sharply toothed as those of var. deustus, is found in and east 
of the Columbia Gorge in Washington and Oregon. Variety heterander (T. 
and G.) Cronq., with a completely or mostly glabrous corolla, occurs in 
south-central Oregon and adjacent California and Nevada. Variety sudans 
[subsp. sudans (Jones) Penn. and Ke ck ] is limited to volcanic rocks and 
soils in southwestern Oregon and south to Lassen County, Cai'ifornia. It 
is characterized by 1prominent _glandulos ity of corollas and herbage. Al-
though Cascade material is c learly representative of variety deustus, 
+ \ ------
\ 7 
+ 
+ \ I 
\ 
3ml. \ I 
1---i \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
-+ 
+ + 
++ 
0 
0 
++ I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
%+ I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
, PENSTEMON DEUSTUS IN 
---- ---- co a, 
287 
the geographic distribution of thi s variety cannot be considered out-
side the context of the other varieties, and all four have been mapped 
a s one entity in this work. 
Mimulus pulsiferae Gray, Proc . Am. Acad. 11: 98 . 18 76. 
This small annual monkeyflower is characterized by the following 
attributes: corolla yellow, only s lightly bilabiate, 8-16 mm in length; 
leaves tapered to a sessile base. More common in the northern half of 
California, especially in the foothills of the Sierra Nevada, it has 
been collected only rarely north of Crater Lake, although it ranges 
north almost to Mt. Adams. The closest relative of~- pulsiferae is 
~- washingtonensis Gand . , from whi ch it differs in its smaller les s 
bilabiate flowers and glabrous style. The ranges of the two spec ies 
overlap along the eastern base of the Cascades in Klickitat County, 
Washington, from which point~- washingtonensi s sp r eads east through 
southern Washington, northe r n Oregon, and west- central Idaho (Grant, 
1924; Cronquist, in Hitchcock and other s, 1959). 
Throughout its range -M. pulsiferae occurs in moist open areas . · ' 
In the Western Cascades it is a rather ephemeral member of the spring 
flora, occupying open gravel flats and steep rocky slopes while snow-
melt runs, desiccating after midsummer. In some areas it occurs abun-
dantly in dense patches but has been fo und in only four distinct local-
ities in the Oregon Cascades north of Crater Lake to date. ~- pulsiferae 
occurs in the Rocky Melt Seep or Gravel Scree associations where it 
grows with Gayophytum humile, Mimulus breweri, Polygonum kelloggii, 
+ \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
1--------i \ 
\ I 
\ 
\ I 
+ 
A \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + l 
-+ 
+ + I 
++ 
l 
I 
-t>+ I 
+ I 
++ -+ 
+ I 
+ I 
I 
+ 
% I + I 
I I 
I 
' ' 11omi. 
GEOGRAPHIC' .DISTRIBUTION OF I 
- MIMULUS PULSIFERAE IN 
co 
--·-- ------- 00 
289 
Linanthus harknessii, Eriogonum nudum, Artemisia tr identata, Navarretia 
divaricata, and Polygonum cascadense . 
Mimulus breweri (Greene) Coville, Contr. U. s; Nat. Herb. 4:171. 1893. 
M. breweri is a glandular anthocyaniferous annual with very 
small purple and red flowers. It is evidently closely allied both to 
the larger yellow-flowered!:!• suksdorfii Gray and to !:!• rubellus Gray, 
a species of the Rocky Mountain reg ion which may have either red or 
yellow corollas. Under dry or otherwise adverse conditions M. breweri 
may attain a height of less than 2 cm with one pair of leaves beyond 
the cotyledons and a single flower. Many rapidly-maturing minute seeds 
are produced in each capsule. The se plants are markedly ephemeral under 
most conditions but will attain a height of 15 cm and produce hundreds 
of thousands of seeds if sufficient moisture remains available for con-
tinued growth. Other members of the Rocky Melt Seep association with 
which!:!• breweri is frequently found include Lewisia triphylla, Linanthus 
harknessii, Polygonum kelloggii, Gayophytum humile, Galium bifolium, 
Allium amplectens, Polygonum minimum, and Collinsia parviflora. 
Castilleja pruinosa Fern., Erythea 6:50. 1898 . 
There is no satisfactory treatment of Ca s tilleja in the litera-
ture. It is one of the most difficult genera in the Pacific Northwest. 
Although Castilleja is a relatively well-defined genus, the possibili-
ties for hybridization among the numerous species seems limitless. The 
morphological evidence points to reticulate evolution in at least 
B \ ----\ 7 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
0 \ 
+ \ 
t) + \ 
+ + \ 
O(\,O 0 + \ 
+ +o 
+ 0 
+ 
(t) 
+ + 
(!)8 
+ 
-&P I 
+ I 
+O t) 
(!)(!) I 
~ I 
I 
I 
~ I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
MIMULUS BREWER I IN 
---- ---- '0°  
291 
portions of the genus . Our commonest species is.£. miniata Dougl., 
which evidently intergrades to some extent with most other species of 
Castilleja occurring within its range. Two students of the genus, 
Noel Holmgren and Rimo Bacigalupi, have examined my material and have 
helped greatly with its interpretation. 
According t o Ownbey (in Hitchcock and others, 1959), .£. pruinosa 
is the only member of the Sec tion Pruinosae (characterized by branched 
hairs on the leaves and stem) occurring in the Pacific Northwest. It 
is evidently closely related to t he shorter galead more tomentose C. 
mollis Penn . of the coastal dunes of San Luis Obi s po County, California, 
although on the California coast the two species are separated by 
nearly 640 km. 
Typical C. pruinosa has numerous branched hairs, giving t he 
lower stems and leaves a singular whitish cast even from a distance . 
The leaves are long and narrowly linear and the inflorescence rela-
tively long and open. Some specimens from t he Western Cas ades (e .g. 
Hickman 8-20, 124-9, 272-1) are perfectly good.£. pruinosa, but other 
material with some branched hairs is quite problemati cal. Bac igalupi 
hypothesize s that these specimens contain genes of.£. pinetorum Fern . 
and C. affinis H. and A. Holmgren cites the congested inflorescence 
and broad leaves of these spec imens as support for his opinion that 
they represent intergrades with£. miniata and a r e actually closer to 
that species than to.£. pruinosa . Recent work by Heckard (1968), which 
centers on the role of polyploidy in the evolution of this genus, has 
implicated.£. pruinosa in the complex centered around C. peckiana Penn . 
The latter species approaches C. miniata on one hand and C. hispida 
292 
Benth. subsp. acuta Penn. on the other. Ownbey (in Hitchcock and 
others, 1959) found a discontinuity in this clinal progression and re-
duced£. peckiana to synonymy with£. minnata. Heckard has proposed 
a more complex relationship, however, in which C. chromosa and C. 
pruinosa also contribute genie material to the complex. Two speci-
mens were collected ( Bacigalupi and Heckard 7900, 7905: UC) from 
· south-central Wasco County which, though otherwise like typical£. 
peckiana, have a number of branched trichomes. Heckard also notes 
that these records help substantiate the presenc e of£. pruinosa (at 
least in its chief diagnostic characteristic) north of the Siskiyou-
southern Cascades region. A single recor d had previously been reported 
from Hunt's Cove near Mt. Jefferson (Leach 4601), which was doubted as 
to locality by Ownbey (in Hitchcock and others, 1959). Other specimens 
from Grizzly Peak (Hickman 438-1, 438-2), only one kilometer from the 
Hunt's Cove locality, also help confirm accurate labelling of the Leach 
collection. These plants show some charac teri sti cs of£. miniata and 
£. hispida. Thus presently available evidenc e indi cates to this writer 
that all of the Western Cascade specimens should be cons idered part of 
the£. peckiana complex, comprising a polymorphi c group of intergrades 
among£. miniata, £. hispida, and£. pruinos a. 
In the Western Cascades these castillejas are found rooted in 
crevices of andesite or basalt on precipitous , dry, sou th-facing cliffs 
and slopes. Plants are never encountered in large populations, and the 
collecting localities are few. As members of the Outcrop Ridge associ-
ation, they usually occur with Erigeron folio sus confinis, Silene 
campanulata glandulosa, Cheilanthes siliguosa, Arabis platysperma 
+ \ ------ 7 
8 \ 
+ \ I 
\ 
3mi. \ . I 
\ 
\ I 
\ 
\ I 
+ \ 
+. \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ + 
++ 
I 
+ I 
++ I 
+ I 
+ 
<!>+ + I 
Q. I 
I 
+ I 
+ o+ I 
I I 
' ' 110 mi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
' CASTILLEJA PRUINOSA IN 
---- ---- \0 w 
294 
howelli.i, Eriogonum umbella tum, Alli um amp lee tens, and Juniperus 
communis saxatilis. 
Castilleja rupicola Piper, Erythea 6:45. 1898. 
Selected synonyms: 
Castilleja andrewsii Henderson, Madrono 3:31. 1935. 
Several members of the Section Parviflorae occur in the Cascades. 
They inc lude.£. hispida, .£. rupi co la, .£. parviflora oreopola, .£. suks-
dorfii, _£. glandulifera, and.£. applegatei. All of these but the last 
occur at the latitude of the Three Sisters, and numerous intergrades 
are known, not only within this group, but with more distantly related 
species as well • .£. hispida is common in rocky dry sites throughout 
the Cascades as far south as Douglas County. The other species are 
more restricted in both range and habitat • .£. rupicola occurs rooted 
in crevices of north or west-facing outcrops in the northern Cascades 
from southern British Columbia to Mt. Rainier , and again in the Western 
Cascades of Oregon. In the fresh state its leaves are markedly pur-
plish; this color is often retained on drying. This characteristic is 
shared with only one other taxon, _£. parviflora var. oreopola (Greenm.) 
Ownbey, a close relative. This variety of.£. parviflora is found only 
in the High Cascades at this latitude, being common on the Three Sisters 
and occasionally collected from lower more westerly peaks in the High 
Cascades (Detling 3325, 5393; Hi ckman 439-5, 499-2). It also occurs in 
the high Olympics, and is distinguished from.£. rupicola by fewer leaf 
lobes and purplish rather than scarlet bracts and corollas. The habi-
tat it occupies is also quite different; it is _t ypi cally found in damp 
295 
scoria gravel or pumice flats near snowbeds. 
Ownbey (in Hitchcock and others, 1959) notes two other species 
as being close relatives of£· rupicola. C. suksdorfii occurs in high, 
wet, mountain meadows from Mt. Adams to Crater Lake in the Cascade 
Range. Like£. parviflora oreopola, it has fewer leaf lobes than C. 
rupicola, is much more robust, and seldom branches from the base, as is 
characteristic of both£. parviflora oreopola and£. rupicola . Two 
specimens off. suksdorfii are known from near Fish Lake, Linn County 
(_E!. R· Sheldon 12570, 12555), and one from a perennially wet meadow 
seep at the top of Tombstone Meadow on Cone Peak in the interior of the 
Western Cascades (Hickman 190-7). £· rupicola also occurs on this peak 
at higher elevations. C. covilleana is strikingly similar to£. rupi-
cola except for its shorter and less exserted galea. It occurs at high 
altitudes in the mountains of central Idaho and southwestern Montana. 
Most collections of C. rupicola from the Western Cascades are 
essentially identical to Piper 2071 or Gorman 2891 from Mt. Rainier and 
Mt. Baker, respectively. A few specimens tend morphologically toward 
£. hispida as shown by a widening of the undivided base of the leaves 
(Hickman 493-2, Tidbits Mountain). The type specimen of Henderson's 
£. andrewsii (Andrews 233) from Hors epas t ure Mountain consists of two 
stems, one of which is£. rupicola so completely introgressed with C. 
hispida as to be almost identifiable as that species. Leaves of this 
stem are wide, the lobes are broad and rounded, as in£. hispida . This 
portion of the type is entirely lacking in purple leaf pigments. The 
other smaller stem is more typical off. rupicola but still shows sev-
eral hispida characters. Locality information is too vague to permit 
+ \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
\ 
+ It. \ 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
C!) + 
+ 
++ 
+ 
-f I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
++ I 
I 
11omi. 
GEOGRAPHIC' .DISTRI BUTION OF 
,CASTILLEJA RUPICOLA N 
---- \.0 ---- Cl' 
297 
accurate recollection of this aberrant population. Further material 
from Horsepasture Mountain (Hickman 305-5) is nearly identical to 
mate r ial of C. rupicola from Mt . Rainier, the type locality. 
Another population from Tidbits Moun ta in (Hi ckman 556-1) has a 
matted webby pilosity about the inflorescence that is characteristic 
of C. arachnoidea Greenm., which is common 50 km to the east on the 
Three Sisters . A single collection of C. arachnoidea has been taken 
from Olallie Meadows at the western edge of the High Cascade Range. 
This species, of the Section Pilosae, is strikingly different from.£ . 
rupicola in its floral morphology, but some crossing between these 
taxa may nevertheless have occurred. 
The conclusion that must be drawn from presently available in-
formation is that C. rupicola, a species of basi cally northern ancestry, 
is beginning to lose its iden t ity in the southernmost parts of its 
range in the Western Cascades of Oregon, where it freely interbreeds 
with.£. hispida and occasionally with more di stantly related species. 
Nevertheless, most material from this area compares favorably with 
typi cal.£. rupicola. 
Vertical Outcrop associates include Saxifraga bronchialis 
vespertina, Penstemon rupicola, Selaginella scopulorum,. Selaginella 
wallacei, Gentiana calycosa n. subsp., Saxifraga occidentalis rufidula, 
Valeriana sitchensis, Campanula rotundifolia, and Douglasia laevigata. 
Galium bifolium Wats., Bot. King Exp. 134. 1871. 
This species is one of the most distinctive of all the bedstraws. 
+ \ ----- 7 
\ 
+ \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
-0 \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
+ c!> + 
+G 
+ 
~ I 
() I 
++ +i t> I 
--'? I 
+ I I 
+ oo I 
I I 
11omi. I 
GEOGRAPHIC' .DISTRIBUTION OF I 
GALIUM BIFOLIUM IN 
---- ----- 'c°o  
299 
It is an annual, having only one or two pairs of leaves at each node 
(if two, the pairs are of unequal size), glabrous stems, and strongly 
divaricate pedicels with solitary nodding fruits. Its affinities to 
other species of Galium are not clear. 
G. bifolium occurs throughout the mountainous regions of the 
Western United States but has only rarely been collected from the High 
Casca des or further west . In the course of this study, however, it was 
found to be relat i vely abundant on all of the Western Cascade peaks 
that have been mapped (Peck and others, 1964) as Plio-Pleistocene High 
Cascade vol canic rocks. It i s generally a spring ephemeral but may 
persist for some time shaded by taller species in mesic middle or high 
altitude meadows where the moist soil i s loose and continually mixed by 
mass wasting and the action of rodents. Other members of the Mesic 
Meadow as·oc iation are Rubus parviflorus, Pteridium aquilinum, Rudbeckia 
occidentalis, Aquilegia formosa, Erigeron aliceae, Gayophytum humile, 
Polygonum minimum, Luina stricta, and Ribes binominatum. 
Lonicera conjugialis Kell., Proc. Cal. Acad. Sci. 2:67. 1863. 
In hi s detailed synopsis of the genus Lonicera , Rehder (1903) 
recognized 157 species, most of which are restricted to subtropical 
regions of the world. The taxonomy and nomenc lature of this group are 
diffi cult, but the western North American species do not present grave 
problems. L_. conjugialis i s the only species from this region having 
dark reddish flowers. The largely united ovaries are also distinctive . 
This species ranges from Mt. Adams south through the High Cascades and 
+ \ ---- -- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ \ 
\ 
+ + 
+ 
+ + );; 
/ 0 0 
++ 
+ 
++ A I 
+ I 
++ + 
+ I 
+ I 
I 
-+ 
q. + I + I 
I I 
110 mi. I 
GEOGRAPHIC' .DISTRI BUTION OF I 
LO N I CE RA GONJUGI ALIS l w 
---- 0 ---- 0 
301 
Great Basin to the Siskiyou Mountains of Oregon and California and 
southward through the Sierra Nevada to central California . It had not 
previously been collected from west of the Cascade crest except in the 
Siskiyou region. 
In the Western Cascades l· con jugialis occurs uncommonly on south 
or we s t-facing slopes above 1500 min open grass-Carex meadows or in 
partial shade along forest-meadow ecotones. It has been found roo ted 
among rocks or in dry meadow soils. Associated species include Bromus 
carinatus, Eri ogonum umbellatum, Vera trum insoli.tum, Luina stricta, 
Abies amabilis, Sambucus callicarpa, and Calochortus lobbii, 
Loni cera utahensis Wats . , Bot. King Exp. 133 . 1871 . 
This species shows a strangely disjunct di s tribution. Most 
collections have been taken from moist mountain slopes in the Olympi cs 
north and east through Washington, British Columbia, and Idaho, and 
south in the Rocky Mountains reportedly to Utah . Most Oregon ma t erial 
has been collected in the Wallowa Mountains. Anothe r disjunct series of 
populations extends from the central Cascades of Oregon south in the 
moun ta ins t o northern California . Only three locali t ies in this area 
were previously known; this study adds four more Cascade populations. 
L. u t ahensis is di stinguished by its yellow axillar y flowers and 
small bractlets, whi ch do not enclose the ovaries, I t has been found 
only in the saturated soil of bogs and on north-fac ing seepage slopes 
where snow remains until midsummer or later, Typically a high elevation 
species, l · utahensis occ urs as l ow as 1300 min t he Weste r n Cas cades . 
+ \ ---- 7 
\ 
+ + \ I 
\ 
3ml. \ I 
t----------1 \ 
\ I 
\ 
\ I 
+ \ 
O+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ 
++ 
+ 
++ I 
+ I 
+o + ++ I 
+ I 
I 
+ I 
+ ++ I 
I I 
' ' 11omi. I 
GEOGRAPHIC' DISTRIBUTION OF I 
- LONJCERA UTAH ENS IS lw 
0 
__.; ---- ---- N 
303 
On north-facing cinder slopes it is associated with Anemone occidentalis, 
Pinus contorta murrayana, Polygonum newberryi, Vaccinium membranaceum, 
and Penstemon euglaucous. In boggy areas associated species include 
Boykinia major, Kalmia polifolia, Aster alpigenus, Hypericum anagalloides, 
Mertensia bella, Habenaria dilatata, Drosera longifolia, Valeriana 
sitchensis, and Viola palustris 
Chrysothamnus nauseosus (Pall.) Britt. var.albicaulis (Nutt.) Rydb. 
Mem. N. Y. Bot. Gard. 1:385. 1900. 
Selected synonyms: 
Chrysothamnus speciosus Nutt., Trans, Am, Phil. Soc. II 7:323. 
1840. 
Chrysothamnus nauseo sus var. speciosus Hall, U, Cal. Puhl. 
Bot. 7:169. 1919. 
This species is composed of a complex of poorly defined varieties 
of which the present one is the most widespread and common, It is dis-
tinguished from other taxa of Chrysothamnus by the shape and pubescence 
of the involucral bracts, by the presence of a felt-like tomentum on the 
twigs, and by pubescent achenes, It occurs throughout the Western United 
States and southern Canada in sub-desert or high cold desert regions. 
Hall and Clements, in their 1923 monograph, recognize 20 subspecies of 
Chrysothamnus nauseosus with 83 listed synonyms. The relationships be-
tween these subtaxa are highly complex, and it is unlikely that they all 
represent natural evolutionary group s . 
Throughout much of its range,.£. nauseosus albicaulis occurs with 
sagebrush, Artemisia tridentata. Stems have been shown to die back under 
great and prolonged moisture stress by McKell (1956) and in the present 
+ \ - - -- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
\ 
\ I 
+ 0 \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
-+ 
+ + 
++ 
+ 
++ I 
+ I 
++ -+ 
+ I 
+ . I 
I 
+ 
o+ I I 
I I 
I 
' ' 110 mi. 
GEOGRAPHIC- DI STRIBUTION OF I 
CHRYSOTHAM NUS NAUSEOSUS ALBICAULIS lw 
--- - 0 --- - ..,.. 
305 
work, but resprouting from the crown has occurred with application of 
water in every instance. Deeply penetrating root systems insure that 
such drought stress will be infrequent. A large population has been 
collected by the author on gravelly terraces of the South Umpqua River 
near Canyonville, although this species is typically part of the Juniper-
Sage association of the Great Basin. In the Western Cascades, Chrys-
othamnus is restricted to a few sites on exposed south-facing cliffs, 
where it roots in crevices of andesite or in pockets of fine gravelly 
talus. One specimen has been found growing in a gravel highway embank-
ment just below precipitous rocky slopes. Although areas in which this 
species grows are harsh enough that it has no close associates, Eriogonum 
compositum, Eriogonum umbellatum, Lotus nevadensis douglasii, Penstemon 
deustus, and other Gravel Scree species frequently grow in the general 
vicinity. 
Haplopappus hallii Gray, Proc. Am. Acad. 8 :389 . 1872. 
Hall~s 1928 monograph of the genus Haplopappus has been called 
"one of the landmarks of American taxonomy" (Cronquist, 1955, p . 211) . 
Hall calls attention to the fact that Haplopappus is an ancient and com-
plexly diversified genus with several presumably derivative genera of 
the Astereae such as Chrysothamnus, Grindelia, and Chrysopsis, and sev-
eral closely related groups with which hybridization may have occurred 
in the past. The latter include such presently problematical genera as 
Solidago and Aster. 
Hall recognized 16 sections of Haplopappus. One of the small sec-
tions is Hesperodoria, which contains H. halli and an eradiate species 
\ ----
\ 7 
(!) + \ 
\ I 
3 mi. \ I 
\ 
\ I 
\ 
(!) \ I 
\ 
+ \ 
+ + + \ 
+ \ 
+ \ 
+ \ ':,-
\ 
+ 
-+ 
+ 
I f 
+Cl) 
I 
I 
+ I 
I 
(!)(!) 
I 
-+o + + I 
++ I 
+ I 
+ I 
+ qo I I 
I I 
I 
GEOGRAPHIC' DISTRI BUTION OF 110 mi. I 
- HAPLOPAPPUS HALLI! lw 
----- ------ 0 °' 
307 
of the southern Colorado Plateau,.!!• scopulorum (Jones) Blake. The two 
species are differentiated from other species of Haplopappus by their 
perennial habit, narrowly campanulate or turbinate inflorescences, and 
fine white pappus. The last character allies this section with certain 
species of Solidago. Both species are quite distinctive and confined 
to rather narrow ranges. Hall was aware of only one locality outside 
of the eastern end of the Colu~bia Gorge for.!!• Halli (Applegate 2755: 
US; near Breitenbush Hot Springs), but the present work has shown the 
species to be quite common in the Western Cascades as far south as 
Hershberger Mountain, southern Douglas County. It occurs in crevices 
of volcanic outcrops or on gentle deflation armor slopes in exposed 
south-facing habitats. Flowers appear late in August, making this one 
of the last Western Cascade species to bloom. It often occurs with such 
members of the Outcrop Ridge association as Sedum divergens, Castilleja 
hispida, Erigeron foliosus confinis, Cheilanthes siliguosa, Silene 
douglasii, Eriogonum umbellatum, and Juniperu s communis saxatilis. 
Aster gormanii (Piper) Blake, Rhodora 30:228. 1928. 
This highly restricted and distinctive member of a most complex 
genus is characterized by a small number '(8-13) of white ray flowers, 
typically solitary heads, and sessile entire leaves reduced but little 
to the inflorescence. It differs from the closely allied~- paucicapi-
tatus Rob., which occurs in the Olympic Mountains and on the south end 
of Vancouver Island, in its smaller size and wider involucral bracts. 
A. gormanii has previously been reported from only two localities: the 
+ \ ----
\ 7 
+ + . \ 
\ I 
3 m1. \ I 
\ 
\ I 
\ 
+ \ I 
\ 
+ \ 
+ + + \ 
+ 
+ 
+ \ 
\ 
+ + 
1 · 
+ -+ 
Jj 
+ I 
+ + 
I 
I 
+ I 
+ + I I 
+ 
+ I 
++ + I 
+ I 
+ I 
++ I I 
I I 
I 
GEOGRAPHIC' -DI STRIBUTION OF 110 mi. ,. 
ASTER GORMAN II lw 
---- ---- 0 co 
--, 
309 
type locality on the northern slope of Mt. Jefferson, Marion County, 
and from cliff faces around Harvey Lake about six km to the north. A 
third well-established population was discovered in a gravel scree on 
the northeast side of Iron Mountain, almost 50 km southwest of the type 
locality. 
The Western Cascade population occurs with Ivesia gordonii, 
Trifolium productum, Cilia aggregata, Lotus nevadensis douglasii, Crepis 
occidentalis, Sanicula graveolens , and Linum perenne lewisii. 
Erigeron cascadensis Heller, Muhl. 1:6. 1900. 
Selected synonyms: 
Erigeron spatulifolius Howell, Fl. N. W. Am, 1:317. 1900 . 
Erigeron pachyrhizus Greene, Leafl. 2:216. 1912. 
Erigeron cascadensis is one of many species of the Section 
Euerigeron that evidently has been derived from E, peregrinus stock 
(Cronquist, 1947). Its closest relatives are E. cervinus, endemic to 
the Klamath Mountain region, and!• leibergii, a species of the eastern 
slope of the Cascades of Okanogan, Chelan, and Kittitas Counties, Wash-
ington. It is morphologically intermediate between these two species. 
!• cascadensis, although quite distinctive, has been poorly collected 
and evidently is very locally distributed in addition to having a limited 
range, It has been collected from fewer than half a dozen sites between 
Fairview Peak in the Calapooya Range and Crater Lake, and at Pansy Camp 
on the Marion-Clackamas County Line. One specimen (Lloyd, in 1893) is 
reportedly from the "45th parallel, Cascade Mountains," but since early 
+ \ ------ 7 
\ 
+ + \ I 
\ 
3mi. \ I 
1-------i \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ \ 
\ 
+ + 
+ p 
(!) + 
++ I 
I 
IA I 
+ I 
(!)+ I 
+ I 
+ + 0+ I 
+ I 
I 
+ I 
-o,._ +O I 
I I 
' ' 110 mi . .. I 
GEOGRAPHIC' DISTRI BUTION OF I 
- ERIGERON GASGADENSIS lw 
---- f-1 ---- 0 
311 
collectors evidently used this designation to include most of the 
central Cascades of Oregon, the exact locality is open to question, 
This species grows in rock rubble and in crevices of weathering 
volcanic outcrops at the summits of several Western Cascade peaks, It 
occurs both on moist north-facing slopes and on hot, exposed, south-
facing cliff faces. Its phenology is greatly affected by exposure as-
pect, but its lack of habitat specificity, except for substrate texture, 
makes its geographical restriction puzzling, Commonly found with 
Erigeron cascadensis in Fine Gravel Scree or Vertical Outcrop associa-
tions are Penstemon rupicola, Erigeron foliosus confinis, Selaginella 
wallacei, Silene douglasii, Silene campanulata glandulosa, Arenaria 
capillaris americana, and Linanthastrum nuttallii. 
Erigeron compositus Pursh, Fl. Amer. Sept. 2:535. 1814. 
Selected synonyms: 
Erigeron compositus Pursh var. discoideus Gray, Am. Journ. 
Sci. II 33:237. 1862. 
Erigeron compositus Pursh var. glabratus Macoun, Cat. Can. 
Pl. 2:231. 1884. 
Erigeron composit11s Pursh var. submontanus Peck, Torreya 
28:56. 1928. 
This highly variable species has been the center of much taxo-
nomic and nomenclatural confusion (see Macbride and Payson, 1917; Payson, 
1926; Cronquist, 1947, 1955). Slight variants have often been recognized 
as distinct species, and the taxon as a whole shows a history of the 
progressive lumping of subspecific taxa after an initial spurt of spe-
cific descriptions. The three varieties listed above have been to the 
312 
present maintained as merely morphological groupings (Payson, 1926). 
They evidently constitute at best a series of ecotypes adapted · to pro-
gressively higher elevations. Variety glabratus is the most common and 
shows the greatest elevational range. This intermediate form is found 
from 1000 m to 3500 m. Since all three of these varie ties have been 
collected in the central Cascades of Oregon, and none has a discrete 
geographic range, they must here be lumped and discussed as a unit. 
Experimental work is needed in this group. 
Other closely related species include such members of the series 
Multifidi as!· flabellifolius Rydb., E. vagus Payson, and E. pinnati-
sectus (Gray) A. Nels. !· compositus is distinguished from these taxa 
by having a much heavier less branched caudex and more deeply and finely 
divided leaves. 
Erigeron compositus is widely distributed throughout western 
North America as far south as central California and Colorado and is 
primarily restricted to high elevations. Previous specimens from the 
Cascades have all come from the Arctic-Alpine zones of the highest peaks, 
with the single exception of the type collection of E. compositus var. 
submontanus (Peck 14804: WILLU), a robust form with finely divided 
leaves from near Detroit, Marion County, at an elevation of about 400 m 
(Peck, 1928). This is now considered to be the only collection of v~r-
iety compositus from Oregon, others being confined to the eastern Columbia 
Basin of Washington. 
A sizable population of!· compositus occurs in crevices and on 
coarse gravel scree on the south slope of Browder Ridge. These specimens 
are typical subalpine to alpine forms and are found here at an unusually 
+ A \ 7 
\ 
+ + \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + \ 
+ 
+ \ 
\ 
\ 0 
+ + ;1 
+ 
+ + 
++ 
/ 0 
I 
I 0 
"-+ I 
++ I 
+ I 
++ + + I 
0 
+ I 
I 
+ I 
++ I 
I 
' ' 110 mi. 
GEOGRAPHIC' DI STRIBUTION OF 
- ERIGERON GOMPOSITUS 
J-- -'------
314 
low elevation (1700m). They are associated with Eriogonum compositum, 
Eriogonum umbellatum, Allium amplectens, and Delphinium menziesii 
pyramidale. 
Erigeron foliosus Nutt. var. confinis Jeps., Man. Fl. Pl. Calif. 
1056. 1925. 
This taxon is readily recognized by its narrowly linear, entirely 
cauline leaves and strigose to scabrous pubescence. A more robust and 
less pubescent form, E. foliosus var. hartwegii Jeps., is also found 
throughout the range of var. confinis but has not been collected in the 
Western Cascades. These two varieties are common in,  northwestern Cali-. 
fornia and the southwestern corner of Oregon, east to Crater Lake, and 
north in the Cascades to Mt, Jefferson. Several other varieties are 
found in the southern half of California and Baja California. Another 
closely related species, into which 1:_. foliosus confinis passes in the 
southern part of its range, is 1:_. breweri, which occurs from Mt. Shasta 
through the Sierra Nevada to the San Bernardino Mountains (Munz and 
Keck, 1959). 
In the Cascades most collections of this species have come from 
the high peaks, with the exceptron of Castle Rock, where a specimen was 
collected early in the present century (!2:_. l· Sweetser, in 1903). E. 
foliosus confinis seems, however, to be more common in the Western Cas-
cades and is commonly firmly rooted in crevices of dark volcanic rock on 
dry, exposed, south-facing slopes from 1000 to 1600 m. A member of the 
Outcrop Ridge association, it grows commonly with Sedum stenopetalurn, 
Eriophyllum lanatum, Arctostaphylos nevadensis, Silene campanulata 
+ \ ----- ----
\ 7 
+ \ 
\ I 
3mi. \ I 
\ 
\ I 
\ 
0 \ I 
\ 
+ \ 
+ + + · \ 
+ \ 
+ \ 
+ 
+ 
-t 
+ 
+Q) 
d)<!> I 
.p (i) I + 
++ I 
+ I 
+ . • I 
tJ) I I 
I I 
I 
GEOGRAPHIC' .DI STRIBUTION 11omi. OF I 
ERIGERON FOLIOSUS CON FINIS l w 
---- 1--' ------ V, 
316 
glandulosa, Haplopappus hallii, and Penstemon procerus brachyanthus, 
Helianthus cusickii Gray, Proc. Am. Acad, 21:413, 1886, 
H. cusickii is distinguished from other Pacific Northwest sun-
flowers by its loose, narrow, acuminate involucral bracts; its narrowly 
lanceolate leaves; and its thickened taproot, the crown of which bears 
numerous simple stems. A species of the dry regions of the Columbia 
Basin and the Basin and Range Province,~. cusickii is known from only 
one locality west of the Cascade crest, where it may have been intro-
duced within the last century , The upper south-facing slopes of Bachelor 
Mountain support an extensive population of plants which are morpholog-
ically variable and which occur in diverse habitats on open south or 
west-facing slopes near the summit, Stems from a single crown were 
noted to bear leaves in either strictly opposite or completely alternate 
arrangement , Some specimens were encountered in wet meadow loams par-
tially shaded by other vegetation while others were rooted in dry, 
rocky, mineral soils on exposed ridges. Individuals were not consis-
tently associated with any other plant spec ies . 
Artemisia ludoviciana Nutt . var . latiloba Nutt . , Trans, Am. Phil, 
Soc , II 7:400, 184 1. 
Selected synonyms: 
Artemisia candicans Rydb., Bull . Torr, Bot, Club 24:296, 1897, 
Artemisia ludoviciana Nutt . s ubsp. candicans (Rydb,) Keck, 
Proc . Cal. Acad. Sci . IV 25 :44 7. 1946 . 
As a tetraploid member of the Artemisis vulgaris complex (Keck, 
+ \ ------ 7 
\ 
+ \ I 
\ 
3mi. \ I 
t-------1 \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ + 
++ 
+ 
++ I 
+ I 
+ + + + I 
+ I 
I 
+ I 
++ I 
I I 
' ' 110 mi. I 
GEOGRAPHIC- DISTRIBUTION OF I 
, HELi ANTHUS I -CUSICKII I..,.) 
------ ---- I-' -...J 
318 
1946), this species is closely related to all other members of the group 
of herbaceous perennial wormwoods. Members of the complex, which are 
differentiated mainly on the basis of leaf morphology, are sufficiently 
confusing that they have received a variety of treatments in the taxo-
nomic literature. In 1916 Rydberg recognized 54 distinct species from 
North America. Seven years later Hall and Clements (1923) proposed only 
the European!!_. vulgaris L. as a distinct species and considered the 
American material to constitute 15 morphological and geographical sub-
species. With the addition of cytological evidence, Keck (1946) found 
Hall and Clements' morphological work to be basically sound, but follow-
ing a more moderate species concept than either of the previous monog-
raphers, he recognized 9 spec ies with 11 subspecies, 7 of the latter 
included under A. ludoviciana. Using a more geographically rigorous 
definition of subspecies, Cronquist (1955) admits only a northern and a 
southern subspecies, keeping Keck's seven subtaxa of A. ludoviciana as 
varieties. Both authors admit that since no type of A. ludoviciana var. 
latiloba Nutt. has been located, it may well prove to be nomen nudum, 
but since I agree with Cronquist that the varietal level is the proper 
one for this taxon, and not here wishing to propose a new combination 
using Rydberg's later epithet candicans, I will accept Nuttall's name 
as correct. 
The geographical range of the complex as a whole includes the 
temperate and boreal portions of the nor t hern hemisphere with the West-
ern United States as its present center of diversity. Here, superim-
posed on a small number of ancient and stable diploid species, relatively 
recent evolution has produced a vast polyploid network of morphological 
319 
and ecological types, mostly poorly isolated from one another, f:... 
ludoviciana is the most widespread and polymorphic species of the com-
plex. The variety latiloba is found throughout the Columbia Plateau 
region of Washington and Oregon, in the northern portions of the Basin 
and Range province, and through the Rocky Mountains from northern 
Montana and Idaho to southwestern Wyoming. It has not previously been 
reported in the literature from west of the Cascade crest. On the lower 
western slope of the Cascades and in the Willamette Valley it is replaced 
by f:... douglasiana Bess., a hexaploid (n=27) whose origin seems to have 
been by amphidiploidy from f:... ludoviciana (n=18) and A. suksdorfii Piper, 
a diploid (n=9) of the immediate coast (Clausen, Keck, and Hiesey, 
1940). f:... douglasiana is morphologically, ecologically, and geograpb-
ically intermediate between its proposed parent species. Two varieties 
off:... ludoviciana (var. ludoviciana and var. latiloba) are known to occur 
presently with f:... suksdorfii in the Columbia Gorge, supplying a possible 
point of origin for the hybrid, Occurrence of A. ludoviciana in the 
Western Cascades, although presently in disjunct populations well re-
moved from the closest populations off:... suksdorfii, may offer an alter-
native explanation for a hybrid origin of A. douglasiana. Here -A. -lud-o-
viciana latiloba occurs in moderately dry meadows from 1300 to 1700 m, 
where it may constitute the dominant vegetation over small areas. It is 
also sometimes weedy, occurring at lower elevations along dry gravelly 
roadsides. Connnon associates include such Xeric Meadow species as Gilia 
aggregata, Collomia linearis, Gayophytum diffusum parviflorum, Orthocarpus 
imbricatus, Lupinus arbustus neolaxiflorus, and Linum perenne lewisii. 
+ \ ----
\ 7 
+ \ 
\ I 
3ml. \ I 
\ 
\ I 
\ 
t> \ I \ 
+ \ 
+ + \ 
+ \ 
+ \ 
+ + 
+ (!) 
+ 
o' 
++ 
I 
I 
I 
+ I 
+ I 
+ I 
+ + ++ I 
+ I 
+ I 
++ I + I 
I I 
110 mi. I GEOGRAP IC' DISTRIBUTION OF I 
ARTE M ISIA LUDOVICI ANA LATILOBA lw 
----- N --- -- -- 0 
321 
Artemisia tridentata Nutt., Trans. Am. Phil. Soc, II 7:398 . 1841. 
Recent work in the Sec tion Tridentatae (Ward, 1953; Holbo and 
Mozingo, 1965) has indicated that the taxonomic relationships in this 
group are exceedingly complex. Cytotaxonomi c and chromatographic studies 
have resulted in alternative taxonomic schemes , but conc lusions based on 
these types of evidence are often at odds with each other as well as 
with conc lus ions based on morphology. A conservative taxonomic treat-
ment is therefore used here. 
Artemisia tridentata subsp. tridentata is the common sagebrush 
through most of the Western United States. Diplo i d and tetraploid races 
occur throughout the total range of the subspecies, and while the two 
may be ecotypically differentiated within a given region, the pattern of 
differentiation may be quite different or even reversed in other regions. 
Ward (1953) notes that it is impossible to predict accurately the rela-
tive chromosome number of a given individual. The tota l range of morpho-
logical and ecological variation in t hi s subspecies is greater t han in 
any other member of the complex. 
A. tridentata is alkali-intolerant, but grows i n extremely diverse 
habitats in the arid portions of the West. A single loca lity for this 
species is known in the Wes tern Cas cades--on ridgetops and on west-facing 
rocky meadow slopes in the Rebel Ro ck Geo log ica l Area. The plants are 
robust, and many flower annually. The population is well-established 
and individual plants occupy a variety of micro-habi t ats and are associ-
ated with numerous species. Thi s indicates to t he author that Artemis ia 
may have been introduced to this spot within the last 70 years. 
+ \ ----
\ 7 
+ + \ 
\ I 
3mi. \ I 
\ 
\ I 
\ 
+ \ I 
\ 
+ \ 
+ + +· \ 
+ \ 
+ \ 
+ 
+ + 
+ + 
+ 
+ + 
+ 
++ I 
+ 
+ + 
I 
++ I 
+ I 
+ I 
I 
+ 4 + I 
I I 
0 01 
GEOGRAPHIC' .DI STRIBUTION 110 mi. OF I I 
, ARTE ISIA TRI DENTATA lw 
---- ---- N N 
323 
Luina stricta (Greene) Rob., Proc. Arn. Acad. 49:514. 1913, 
Selected synonyms: 
Rainiera stricta Greene, Pitt. 3:291. 1898 . 
L. stricta is easily distingui shed from other members of the 
Tribe Senecioneae by its moderately large eradiate heads borne in a 
narrowly racemose inflorescence and its large, narrowly oblanceolate 
leaves. It was originally thought to be endemic to Mt. Rainier (hence 
Greene's generic name) but has since been found to be common in the Cas-
cades as far south as Bohemia Mountain, Lane County, In some localities 
it forms a sub-dominant in the moist soils of Mesic Meadow associations, 
occurring with Orthocarpus imbri catus, Ribes binominatum, Erigeron 
aliceae, Polygonum douglasii, Lupinus arbustus neolaxiflorus, Lupinus 
latifolius, Mi crosteris gracilis, Ar temisia ludoviciana latiloba, Linum 
perenne lewisii Mertensia paniculata, and Polygonum phytolaccaefolium. 
A single collection (Q • .£. Ingram, in 1920: OSC) is labeled "6th 
Umatilla Collection, Umatilla National Forest . " Due to the lack of spe-
cific locality, and since Ingram collected largely in the Cascades rather 
than in the Umatilla region, the author believes that this collection is 
probably mislabeled. If not, this constitutes an important first col-
lection of the species from eastern Oregon . The locality has not been 
included in the following map of Luina stricta. 
Arnica parryi Gray, Arn. Nat. 8 :213. 18 74. 
A. parryi is eradiate throughout most of its range. In the 
Sierr a Nevada of California t here oc cur s a radiate form, var . sonnei 
+ \ ----- 7 
\ 
8 \ I 
\ 
3ml. \ I 
t-----------; \ 
\ I 
\ 
\ I 
+ \ 
()I- \ 
+ \ + \ 
+ 
+ \ 
+ + 
+ c) + 
+ + I 
I 
I 
+ I 
++ I 
+ I 
+ + ++ I 
+ I 
I 
+ I 
+ ++ I 
I I 
110 mi. I 
GEOG APHIC' DIST IBUTION OF I 
LUINA STRICT A lw 
---- N ----- .t,-
325 
(Greene) Maguire. Except for this variation the present species ts ·well 
marked by its open cymose inflorescence, nodding young heads, and narrow 
long-petioled leaves. Its relationships, however, are not clear. 
Maguire (1943) places Arnica parryi in the subgenus Chamissonis, arguing 
from the fine, abundantly rooting rhizomes; the tawny pappus color; and 
the width of the leaves. However, there are a number of characteristics 
which tend to ally f::.. parryi more closely to subgenus Austromontana, 
such as the typically eradiate heads, three pairs of cauline leaves, 
turbinate-campanulate involuc res, and long-petioled lower leaves. This 
taxon in intermediate between these proposed subgenera. Evidently, 
Arnica varies in too complex a fashion to lend itself well to such neb-
ulous subgeneric categories as Maguire proposes . Perhaps A· parryi, f::.. 
grayi Heller, f::_. parviflora Gray, and A• discoidea Benth. (if they are 
considered distinct) represent a series of intermediates derived both 
from A. mollis Hook. of subgenus Chamissonis and from various members of 
the subgenus Austromontana, including radiate A· cordifolia Hook. and 
eradiate A· spathulata Greene. f::_. parryi does overlap geographically 
with such eradiate forms as A. spathulata, A· parviflora, and A• discoidea 
(sensu Maguire) in the southern Cascades, Klamath, and California Coast 
Ranges, and with f::_. grayi in the Columbia Gorge. These species are still 
confused in the literature (Cronquist, 1955). Neither Maguire nor Cron-
quist admit the existence of A· parryi in the central or northern Cas-
cades of Oregon, but as far as the author can determine, the present 
material is identical to typical f::_. parryi. A more precise statement of 
relationships in this group must await further work. 
In the Western Cascades, f::_. parryi is found in meadow-forest 
+ \ ------ 7 
\ 
+ + \ I 
\ 
3m;. \ I 
1----------i \ 
\ I 
\ 
\ I 
+ \ 
b. 
+ \ 
+ \ + \ 
+ 
+ \ 0 
+ 
+ + 
+ 
+ + 
++ 
+ 
,ct I 
G I 0 
o + + 
+ I 
+ I 
I 
+ I 
+ d I 
I I 
110 mi. I ' 
GEOG APHIC· DISTRI BUTION OF I 
ARN I CA PARRY ! ____ L ____ lw N 
(J"\ 
327 
ecotones, rooted in relatively dry gravelly loam in areas of partial 
shade. It is often associated with Loni cera conjugialis, Xerophyllum 
tenax, Carex spp., Vaccinium membranaceum, Sambucus racemosa pubens 
arborescens, and Aster ledophyllus. Seeds set during the course of the 
present study appear shriveled and non-viable but germinate freely after 
an extended stratification period. 
Microseris nutans (Geyer) Schulz-Bip., Pollichia 22-24:309. 1866, 
Selected synonyms: 
Scorzonella nutans Geyer, Lond. Journ. Bot. 6:253. 1847. 
Ptilocalais nutans Greene, Bull. Cal. Acad. Sci. 2:54. 1886. 
M. nutans is a distinctive species which may someday be placed in 
a smaller genus. The re-establishment of the genus Nothocalais from 
Microseris (Chambers, 1955, 1957) perhaps presages such an event. Its 
pappus is unique, consisting of bidentate basal paleaceous scales sur-
mounted by a long, white, plumose awn. It is also unusual for this group 
in having entire leaves and a stem that branches several times near the 
base. Like several closely related species, it has a fusiform perennial 
taproot. 
Contrary to recent statements (Cronquist, 1955), the species seems 
to be fairly common west of the Cascade crest in Oregon, although it has 
heretofore been rarely collected. The species is abundant in eastern 
Oregon and extends to Montana, Colorado, and central California. In the 
Western Cascades it occurs in open rocky meadows that are moist in the 
spring but dry after midsummer. It occurs with such members of the Xeric 
Meadow or Small Boulder Creep Slope associations as Cilia aggregata, 
+ \ ---- 7 
\ 
+ + \ I 
\ 
3m,. \ I 
1---------i \ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
+ 
+ + 
+e 
+ 
ee> I 
I 
.p + + 
<!>+ I 
+ I 
e I 
oe I I 
I I 
. ' 11omi. I 
GEOG AP IC- _DISTRIBUTIO OF I 
LCROSERIS NUTANS lw 
N 
---- ---- 00 
329 
Lotus nevadensis douglasii, Chrysothamnus nauseosus, Comandra umbellata, 
Arctostaphylos nevadensis, Juniperus communis saxatilis, Cheilanthes 
siliquosa, and Bromus carinatus. 
Nothocalais alpestris (Gray) Chambers, Contr. Dud. Herb. 5:66. 1957. 
Selected synonyms: 
Troximon alpestre Gray, Proc. Am. Acad. 19:70. 1883. 
Agoseris alpestris Greene, Pitt. 2:177 . 1891. 
Microseris alpestris Q. Jone s , Vase. Plants Pac. N. W. 5:267. 
1955. 
This small genus of northwestern American composites is closely 
allied to Agoseris but lacks the distinctive beak to the achene. Pappus 
bristles are essentially capillary or sometimes flattened, with palea-
ceous bases. B_. alpestris is the only species of the genus occurring as 
far west as the Cascade Range, others being sagebrush scrub or prairie 
species. It is typically found at very high elevations, although it oc-
casionally descends to 1500 min the high mountains. Two stations have 
been discovered in the Western Cascades, where flows from the high peaks 
have resulted in a continuous high plateau between High and Western 
Cascades. 
A variable plant in the high mountains, B_. alpestris is uniformly 
robust as it occurs in the Mesic Meadows of the Western Cascades with 
Sambucus racemosa pubens arborescens, Ribes erythrocarpum, Valeriana 
sitchensis, Aster ledophyllus, Galium bifolium, Abies amabilis, and 
Lupinus latifolius. 
+ • A \ ---- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
\ 
\ I 
---- \ 
/ \ I 
+ \ 
A 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ \ 
\ • 
+ + 
-+ 't, 
+ + • /I• • - . 
++ ,.o 
•• le I 
+ •.•..  I • ++ I 
+ I 
+ + ++ / -
_p I .. 
+ • I I 
++ I 
I I 
110 mi. I 
GEOGRAPHIC DISTRIBUTION OF I 
' NOTHOCALAIS ALPESTRIS lw 
---- w ---- 0 
331 
Crepis acuminata Nutt. subsp. typica Babe. and Steb., Carn. Inst. 
Wash. Pub. 504:170. - 1938. 
E. B. Babcock's monumental work on the genus Crepis (1947) is one 
of the classic studies in modern taxonomy. The complexity of this group 
is well known, and will not be discussed in deta il here. Many species of 
Crepis include both a fertile outcrossing diploid and a series of poly-
ploid apomictic forms which have been derived from them. C. acuminata is 
one of these. The typical subspecies includes all of the diploids and 
16 apomictic races. As opposed to all other western species of Crepis, 
here the diploids constitute the most widespread and generally successful 
forms. They are extremely variable but can be differentiated from asexual 
forms by the smaller size of the individual bracts and the amount and reg-
ularity of pollen produced. Usually apomicts produce no pollen at all. 
The range of fertile diploids includes the northern half of the Western 
United States, but the species has never before been collected west of the 
Cascade crest in Oregon or Washington • .£. pleurocarpa Gray, a closely 
related species, is common in the Klamath Mountains and the southern Cas-
cades of Oregon. Only the diploid form of.£. acuminata is plotted on the 
following map. 
Throughout its range.£. acuminata grows in dry open habitats in 
foothills or mountains. In. the Western Cascades rare populations of 
fertile diploids are found in exposed, dry, south-facing environments at 
the tops of precipitous rocky slopes associated with Castilleja pruinosa, 
Phacelia linearis, Eriogonum umbellatum, Arabis platysperma howellii, 
Selaginella wallacei, Lotus nevadensi s douglasii, and Arctostaphylos 
nevadensis, members of the Outcrop Ridge association. 
+ \ 7 
\ 
+ + \ I 
\ 
\ I 
\ 
\ I 
\ 
\ I 
+ \ 
+ \ 
+ + \ + \ 
+ 
+ \ 
+ + 
0 
+ 
+ + 
0 
++ 
0 
0 
+ 
++ I 
+ I 
+ + 
++ I 
+ I 
I 
+ I 
I 
I 
' ' 110 mi. 
GEOGRAPHIC DISTRI BUTION OF 
CRfPIS_ ACUMINATA TYPICA 
333 
Crepis occidentalis Nutt, subsp. pumila (Rydb.) Babe, and Steb, 
(related to apomict rydbergii) x subsp. conjuncta Babe. and Steb. 
(related to apomict jepsonii)? Carn, Inst, Wash. Puhl. 504:128, 
130-131, 134, 136-137. 1938. 
The involved name listed above is the best determination possible 
with the finely discriminating keys of Babcock and Stebbins (1938). The 
name is necessary to communicate the fact that the Wester,n Cascade popu-
lations are composed of individuals intermediate between the two sub-
species pumila and conjuncta. Babcock and Stebbins realized that such 
intermediates occur, and erected the apomictic taxon rydbergii for the 
form of pumila most like subsp. conjuncta and apomict jepsonii for the 
converse. The Western Cascade individuals are far outside the described 
ranges of both of these apomicts; furthermore, they are morphologically 
intermediate between them and cannot be considered to belong in either 
described taxon. Perhaps more is known of the complex morphology and 
evolution of Crepis than can be satisfactorily expressed in taxonomic 
terms, All of the subspecies of.£, occidentalis intergrade with one an-
other and with other species as well, The various apomictic forms have 
been considered nodes on the clines connecting the larger taxa. That the 
subspecies themselves are probably not evolutionarily discrete entities 
is suggested by their complexly disrupted ranges. For example subsp. 
pumila is a rather widespread taxon, occurring in a ring around the cen-
tral valley of California, in the Siskiyou Mountains, in the Basin and 
Range of Oregon, Nevada, and Utah and at scattered localities in Idaho 
and Montana. A series of populations disjunct by over 480 km from the 
other pumila populations occurs in the Olympic Mountains. All of the 
three other subspecies are found in the intervening area. Subspecies 
334 
conjuncta is even more striking in this respect. Most individuals occur 
in the northern Sierra Nevada and in the Siskiyou Mountains of northern 
California and southern Oregon. However, one locality is known from the 
Wenatchee Mountains of central Washington, several from the southeastern 
corner of that state, and one from the Teton Range in Wyoming. Again, 
other subspecies now fill the intervening areas. 
Thus, with regard to the Western Cascade specimens, I feel that 
it is impossible to discriminate between the two subspecies in considering 
the parent taxon. Obviously both subspecies have contributed to the gene 
pool of the Western Cascade plants and apparently in about equal propor-
tions. Since the two are known to merge completely in the Siskiyou Moun-
tains, it is here proposed that this is the parental area for Cascade 
forms, and that a combined stock of the two subspecies has spread north-
ward through the Western Cascades. The material from Bohemia Mountain 
is most similar to subsp. pumila; that from Iron, Cone, and South Peaks, 
farther to the north, more closely approximates subsp. conjuncta. The 
differences between these specimens are, however, much less than is ad-
mitted for a single apomictic form. 
Like.£. acuminata, .£. occidentalis grows in dry sites normally on 
poorly developed soil throughout its range . In the Western Cascades 
specimens are found on steep slopes of fine scoria gravel, or more rarely, 
rooted among denser volcanic rocks on high exposed areas of deflation 
armor. Several of the following species of the Fine Gravel Scree associ-
ation may be found in the immediate vicinity: Ivesia gordonii, Trifolium 
productum, Lotus nevadensis douglasii, Gilia aggregata, Eriogonum umbel-
latum, Allium crenulatum, Arabis platysperma howellii, and Gilia capitata. 
+ \ --- 7 
\ 
+ + \ I 
\ 
3mi. \ I 
1----f \ 
\ I 
\ 
\ I 
+ \ 
+ \ . 
+ \ + \ 
+ 
+ \ 
+ 
+ + 
++ 
+ 
++ I 
+ I 
+ 
++ + I 
+ I 
I 
+ I 
++ I 
I I 
11 0 mi. I 
GEOGRAPHIC' DIST IBU ION OF I 
CREPIS OGGIDENTALIS PUMILA AND l w 
w 
GRE IS OGCIDE TAUS GONJUNGTA --.---
\Jl 
1/ 
Typed by 
Mary L. Armes 
Mu 1 ti 1 ithed 
by 
Margaret Pluid