Joel L« Shiner
A Thesis
submitted to the faculty of the
Department of Anthropology
In partial fulfillment of the requirements for the degree of
In the Graduate College, Tftiiverslty of Arizona
Direotor of G^sia
' /y
Joel L*^Shiner
This thesis has been submitted in partial fulfillment of
requirements for an advanced degree at the University of
Arizona and is deposited in the Library to be made available
to borrowers under rules of the Library,
Brief quotations
from this thesis are allowable without special permission,
provided that accurate acknowledgment of source is made.
Requests for permission for extended quotation from or
reproduction of this manuscript in whole or in part may be
granted by the head of the major department or the dean of
the Graduate College when in their judgment the proposed use
of the material is in the interests of scholarship.
In jail—
other instances, however^__pe^ini^4rS^rr^nTls^be^btained from
The Umatilla
The Sanpoil
Ethnographic Research ------- i -
- --
35-UM-5 (Hat Creek) - ^
Architecture - - - - - - - - - - - Material Culture ---------Economy - - - - - - - - - - - - - 35-UM-3
35-UM-7 (Cold Springs) --------
Material Culture ---------Architecture - - - - - - - - - - - Economy - - - - - - - - - - - - - Further Observations --------
35-UM-17 (Techumtas Island) ------
Architecture - - - - - - - - - - - Material Culture ---------Eoonomy - - - - - - - - - - - - - -
Architecture - - - - - - - - - - - Material Culture ---------Economy - - - - - - - - - - - - - ii.
45-WW-6 (Wallula)
Architecture - - - - - - - - - - - - 7 4
Material Culture ------- - -— 74
Burial Customs
---------- 87
Historical Data ---------- 88
45-BN-S (Berrians Island) -------
Architecture - - - - - - - - - - - - 91
Material Culture ---------- 91
Burial Customs ---------100
Economy - - - - - - - - - - - - 101
Architecture - - - - - - - - - - Material Culture --------Burial Customs ---------Economy - - - - - - - 1 - - - - Historical Data
45-B1T-55 (Sheep Island.) --------
Culture Change in the McNary Region -
v. Comparative
plateau sites
- - -
The Dal-les Region - -- -- -- -- -
Hobo . Cave - - -
- - -------
John Day Reservoir
--- - -
The Wahluke Site
The Yakima Region ---------The Chief Joseph Reservoir
------ 132
The Upper Columbia Region ------
British Columbia
The Snake-Clearwater Region -----
Early Plateau Sites ---------
Summary - - - - - - - - - - - - - - - - -
Seasonal Migration
Material Culture
- - - - - - - - - - - - - - - - - 158
Workmen measuring artifact looation - - • Frontls.
Voloanio ash at Hat Creek (35.TJM«»5) - - • piece
Ash. stratum at site 35.TJM-3.......
Ash. stratum at Hat Creek (35-UM-5) - - -
Projectile points, Hat Creek (35.TJM.5) .
Flake scrapers, Hat Creek (35.TJM-5)•
Carved bone, Hat Creek (35.TJM.5) ••• Hammers and choppers, Hat Creek (35.UM.5)
Shell stratum in profile, Cold Springs
(35.UM.7) •
.» •
View of terrace, Cold Springs (35.TJM.7) -
Stone knives, Cold Springs (35.UM-7) -.
Projectile points, Cold Springs (35.UM.7)
House pit 4, Cold Springs (35.UU.7)
before excavation
Earth Oven, Cold Springs (35»ini»7) - - -
Artlfaots from Teohumtas Island (35.TJBI.17)
House pit 14, Teohumtas Island (35.UM-17)
Projectile points, Wallula (45-WW.6) - -.
a. Bone tools from Wallula (45.WW~6) - - - b. Antler wedges from Wallula (45.WW.6).
x. a. Disooid choppers, Wallula (45.W3fl»6) - - b. Cobble choppers and hammers, Wallula
(45-WW-6) ....
a. Stone hoes from Wallula (45.WW.6)•
b. Net weights from Wallula (45.WW.6) ...
a. Chipped stone from Berrian's Island...
b; Stone mauls from Berrian's Island
(45-BN-3) ..............
v; a
vi; a
Bone and Horn implements, Berrian's Island
Carved and polished stone, Berrian's
Island (45-BN-3)
a. Plank cyst burial, Berrian's Island
b. Flexed burial, Berrianfs Island (45-BTJ-3)
Ornaments from Berrian's Island (45-BN-3) - -
Projectile point sequences
a. Hobo Cave -------------- 123
b. McTTary Region ------------ 123
House pit
site 35-UM.7
House pit 4, site 35-UMV7
House pit
House pit 14, site 35-UM-17
site 35-UM-7
5. House pits 5 and 6, site 45«Btf«>6
- 102
House pit 7, 45-BN-6
Time distribution of chipped stone artifacts « « 117
Time distribution of ground stone and bone
Distribution of grooved net weights •»--»««. 117
Distribution of polished oelts
Distribution of stone mauls
12. Distribution of side notched points
Distribution of lozenge blades
14* Distribution of sandstone shaft smoothers • - • 117
Distribution of digging stick handles .
• - - • 117
Distribution of bird bone beads ........117
Landforms in the Paoific Northwest -----
Arohaeological sites in the McNary Region
Cold Springs (35-UM-7) -
Site 35-UM-17
Site 45-BN-53
Site 45-WW-6
Site 45-BN-6
In any extended program that involvea as much as four
years work in the field and laboratory, contributions of time,
labor and ideas will have been made by many persons#
It is
not possible to extend to each of these persons my gratitude
for a job well done, but for the record, the cooperation and
assistance given me at all times was all that I could ask for#
Through all of the period of field and laboratory work,
Dr. Frank H# H# Roberts jr# was the director of River Basin
Surveys, and for most of the period, my direct supervisor#
Dr# Roberts was a constant source of guidance for any problem
large or small#
I wish to acknowledge my gratitude for this
and for a most pleasant association#
My closest associate in the field work, and for part of
the period, ray immediate supervisor, was Douglas Osborne#
Dr. Osborne, who is now with the Tftiiversity of Washington,
was of invaluable assistance, and is a real student of the
Plateau area#
Without his understanding and frequent as<-
sistance work would not have gone on#
It was through the efforts of Dr# L# S# Cressman,
Chairman of the Department of Anthropology at the University
of Oregon, that River Basin Surveys was given office and
laboratory space on the campus of that University#
Cressman*s long experience in northwestern arohaeology was
of constant assistance to the River Basin Surveys program,
for he gave freely of his time and ideas.
Louis R. Caywood, National Park Service Archaeologist,
Region 4, acted in the capacity of liaison between the
National Park Service and River Basin Surveys.' His friendly
and oheerful cooperation in the face of our constant demands
on him for maps, transportation and information, contributed
greatly to the progress of the work.
Numbers of institutions cooperated closely with the
River Basin Survey Program.
Not only individuals of the
National Park Service, but the organization as a whole was
geared for close assistance to the salvage program.
The Corps of Engineers, United States Army, frequently
supplied transportation and maps, as well as geological and
geographical information.
The same assistance was received
from the Bureau of Reclamation.
The University of Oregon provided office and laboratory
space for River Basin Surveys, as well as library privileges.
The University of Washington furnished generous grants
for field research.
These permitted the extension of the
work beyond what would have been possible.
The State College
of Washington also granted funds for field research, and
supplied transportation for field parties.
Dean H. P. Hansen, Oregon State College at Corvallis,
made pollen analysis of soil samples.
Even though the
results were negative, his time and efforts are appreciated.
The volcanic ash at sites 35-UM-5 (Hat Creek) and
35-UM-7, was examined "by Dr. Harold Culver, Department of
Geology at the State College of Washington.
Dr. Culver's
special trip and his subsequent report were of great as­
Dr. C. D. Cambell, State College of Washington,
made microscopic and speotrographic analysis of the volcanic
To the many persons who examined, tested and gave expert
opinions on artifacts, my appreciation is hereby extented.
The list includes metallurgists, art and textile experts,
button collectors, historians and especially residents in
the vicinity of our excavations and surveys.
After the field excavations were finished and the
laboratory work was carried to completion^ came the writing
of the interpretation of the data.
My guidance through this
critical stage was in the hands of my thesis committee.
was composed of Dr. Emil W. Haury, head of the Department of
Anthropology, Dr. Edward Spicer, Professor of Anthropology,
and Dr. Edward B. Danson, Assistant Professor of Anthro­
pology; all of the University of Arizona.
The comments,
suggestions and questions contributed by these persons did
much to ease the burden of writing.
I am especially grate­
ful to Dr. Haury for his assistance in solving technioal
problems in illustration.
The burden of typing and proof reading fell upon my
wife, Maxine Shiner.
Without her endless hours of pains­
taking labor, I oan truly say that this thesis would never
have been finished.
Permission to use the data gathered by River Basin
Surveys investigations as material for this dissertation
was generously granted by Dr. Prank H. H. Roberts Jr.,
Director of River Basin Surveys.
Workmen measuring artifact location
Volcanic ash at site 35-UM-5, Hat Creek
Anthropological research in the Plateau Area of north­
western North America has failed to produce a dear picture
of Indian culture. Prom both the ethnological and archaeo­
logical viewpoint there has been insufficient research and
little synthesis.
While ethnographic investigation has
permitted certain generalities about Plateau culture,
archaeological research has not produced any sort of chro­
nology, not even a local sequenoe.
Since Wissler's classi­
fication in 1922,1 which set up a oulture area known as the
Plateau, very little ha3 been done toward filling in the
details that were not available then.
A preponderance of the ethnographic research has been
centered in specific geographical regions to the neglect of
Much of the effort has been expended on detailed
problems of almost purely aoademio interest while basic
problems of time, spaoe and process have largely been
ignored. Archaeological research in the Plateau has been
limited to a few major exoavatlons, and the time factor has
not been considered. They have turned up collections of
artifacts but no reconstruction of the aboriginal oulture.
This complaint has been made many times and in many places,
1. Wissler, 1922
"but relatively speaking, the Plateau remains one of the least
knowa areas in North. America.
Considering the faot that
thousands of aboriginal habitation sites exist in the area
and thousands of relatively unacculturated Indians still
survive, it is no exaggeration
to state that the Plateau
is muoh in need of anthropological research.
An opportunity to reopen one phase of the anthropologi­
cal study came with the availability of a significant body of
archaeological data.
These data came from the River Basin
Surveys program of salvage archaeology, which began in the
Pacific Northwest in 1947, and came to an end in 1952.
intensive program of survey and excavation led to the develop­
ment of a local sequence in one region and additional infor­
mation from several other regions within the Plateau.
data should permit generalization about the Plateau during
the prehistoric period since time and space dimensions on
parts of the material culture are beginning to be understood.
The phrases culture area and Plateau area have been used
again and again, but little has been done to clarify the
concepts behind them.
This study will follow the classi­
fication made by V/issler in The American Indian and aocept
the concepts agreed upon by Wissler and Kroeber.
The idea
and reasoning behind such a classification is relatively
It stems from what might be called an Amerioan
revolt against the older methods of collecting.
In the old
days a museum would exhibit collections of bows, canoes or
arrowheads with the emphasis on type of artifact and a ten­
dency to neglect its associations#
The teaohings of Franz
Boaz regarding the importance of context were heeded by his
Ethnological collections began to be made ac­
cording to areas, and associated artifacts were exhibited
Among those most responsible for the Culture Area
concept was Clark Wlssler,2 and it was his classification
that set up a Plateau Area in addition to a Northwest Coast
Area and a Plains Area*
the onset we oalled attention to the need for
classification in dealing with ethnological and archaeo­
logical material. No one has ever gone far withour
feeling the necessity for this, - - - In setting up
the areas for culture we grouped tribes or comnunities
according to more or less common traits* This is
classification by similarities* It so happened that
tribes having many traits in common tended to cluster,
.. - the habitat of a cluster could be defined In
geographical terms* Culture Area is a name for such
a cluster of oonmunitles*n3
Wlssler stressed environment in delineating culture areas
and stated,
"Environment does not produce a culture but
stabilizes it* Once adapted to an environment, a
culture tends to hold fast, to spread in area of
adaption rather than move to a new a r e a * w 4
In setting up culture areas, he could have divided North
Amerioa into hundreds of areas, or only three or four; but
he compromised on a reasonable and workable number*
Wlssler, 1922
Wlssler, 1938, p* 297
Wlssler, 1922, p. 372
perusal of the literature of our subject shows
It to be oustomary to divide the two continents into
fifteen regions or areas. If desired, most of these
can be subdivided, but it will best serve our purpose
to deal with a smaller number. Each area designated
has natural features peculiar to it and the tribes
living in one of these areas have many oultural traits
in common."5
in making these classifications, he agreed with Kroeber that
a Culture Area should involve
"Reasonably uniform culture and some degree of
environmental uniformity.w6
With so little to go on, it is remarkable that Wissler
could designate the Plateau as a separate area.
At the time
the classification was made, few data were available.
Kroeber did not wholly agree with Wissler as to the character
of Plateau Culture, and designated the area as one of mixed
He was impressed by an apparent strong in­
fluence from the Northwest Coast on one hand, and from the
Plains on the other. Since that time, however, people have
generally considered that in spite of outside influences
Plateau culture maintained a ^personality" of its own.
Among the reoent tendencies in American research is the
inquiry into time depths in culture areas. Since data are
available on time and space distribution of material culture
in the area it follows that an examination of prehistoric
oulture in the Plateau would be a reasonable study#
Wissler, 1938, p. 220
6. Kroeber, 1939, p. 3
7. Kroeber, 1939
the problem can be posed, MWas the Plateau in prehistoric
times a Culture Area, or was it peripheral to another area
or areas"? The approach to this problem can be stated
briefly. It will be to develop the local sequenoe in the
region in which River Basin Surveys concentrated its work
and expand the picture to the Plateau in general. It in­
volves detailed reporting of the results of River Basin
Arohaeology and comparisons with available materials from
other sources*
From the local sequence and comparisons
important conclusions can be drawn*
The procedure will be to present a physiographic and
ethnographic baokground for a general picture of the Plateau
before going into details. A brief history of significant
contributions to Plateau ethnography and archaeology will
be followed by a summary of the part River Basin Surveys
played in its program of research. The discussion must
then become speoifio before summing up and conclusions
are reached. The thesis will be that the Plateau in pre­
historic times was more Justifiably a separate culture area
than it was in the early historical period with whioh
Wissler was dealing.
Two of the McNary sites, the burial site 45-BN-3 and
the pit house village 45-BN-53, were excavated and reported
on by Douglas Osborne. The interpretation of these two
sites was used by Osborne for his dootoral dissertation
whioh also included a thorough study of the historical
In view of Osborne's concern with the historio
period in the McNary Reservoir, the present study will
touch lightly on that period. The material culture of the
historic period will be discussed only in the sense that it
is a continuation of what went before*
In Chapter I the physical description of the Plateau
is given without setting exact limits to the area*
limits could not be set because the physical area described
was that which corresponded to the cultural Plateau, and
the boundaries of Plateau culture were not sharp lines*
The treatment of the Plateau Area in the present discussion
is forced into a concentration upon that portion which lies
within the United States*
Investigations, both archaeo­
logical and ethnological, have been made in Canada, but
nearly all of them were either made a long time ago or so
reoently that they have not yet been published*
Thus, while
it must be recognized that the Plateau Area extends far
north of the Canadian-TJhited States border, little can be
done at the present time to desoribe and analyze the culture
of that portion of the Plateau*
The Plateau Is a large basin with mountain ranges on
all sides (see Map I)*
It includes the eastern slopes of
the Cascades from the Fraser River south to Croolced River*
Eastward, it extends to the edge of the Plains Area whioh
would be the Flathead River in the north and the continental
divide in the south; The exact boundaries on the north have
never been set*
On the south, the Plateau merges into the
northern Great Basin*
Most of the Plateau is characterized by horizontal
sheets of basaltic lava piled one on top of the other, the
result of Miocene eruptions; Pleistocene and reoent eruptions
have been numerous, but have not materially ohanged the
Erosion and deposition have gone on, outting
and filling, carving deep canyons and oreating flat alluvial
Still the thick layers of lava dominate the
The Columbia River, with its major tributary the Snake,
is the most Important feature in altering the landsoape*
Throughout the Plateau these two rivers have carved deep
oanyons along the main courses-and side canyons where each
tributary joins the main stream*
The Columbia Is an exotlo
'L^/v 'AA ^\^
> •"•
-^ /
/•. 4. /
AA VrAv/
V )sr {% T^')jgk» W^"Ui''^
> \*W\
^>iV fcv
I /^
^ ^ \5> ^ : ?)) >
\/ A^A*S A^A
/ V"
1 W ;> h>•••^AA
1 . \
'~- ^ A- A A''^x -<5ViK\***-*
•/ A"
i* V"^A/
PfN£J^Of*eil.Lf r • ^ *
^-'V ^V*••« *
y\ . <-:-/•
^ yii>" .-^
* ,<^aV
1 'L\ ^
yfr \ ^V £L :/ \^4^%
Ax ^
, > „v
I A^«vf7 J
<N%^\<s.% A Wv?fc'A-^L
A- ^ a
^V> ,
/V /yv<^
v>yV yV^^
(5 ^V-
/fx. ^
ypv. *
/fV^ A A~^ /^V V
•fiS . _v ,-^x
-iL- ,
Av^- ^., A/V- .
. „..,
'V^'vAv _ ^ ^ yf\A
>^-)j^:x <*• V>^
\ ^S. sO K X
' /,V ^
\ ^,x
•^1^. -^Vv. •^V ^'*S ,«.' ^-;
^i(v «<N
A \ rfll.
.^s/' ^
' ^ \ a ^^
^1- <>- ^
» . /\ A*. /V
A *X
^N- ';:iA-
,,v ,4
1*1 A P 1
stream In that It rises in a climate zone different from the
Plateau, and carries a tremendous volume of water through a
semi-desert Whose rainfall could never support more than
Intermittent streams*
The Columbia and its tributaries made
it possible for man to utilize an area whioh otherwise might
have been comparable to the Great Basin in its aridity and
The oourse of the Columbia River through the Cascades
is in the form of a deep gorge with high mountains on both
At The Dalles, as one leaves the coastal strip and
enters the Plateau, the oountry changes suddenly, and east
of there the trees disappear. The lush vegetation of the
gorge gives way to semi«desert, and almost true desert
As one proceeds upstream, he oan see little
change in several hundred miles*
The river is entrenched
in a basalt«*oliff lined canyon one hundred to one thousand
feet below the rolling and broken uplands*
Here and there,
some miles away from the river, the hills rise to a height
capable of intercepting moisture from the Pacific Ocean
winds, and thus supporting some vegetation*
Otherwise, the
canyon and its flanks are barren except for those trees and
gardens recently planted by white men*
Life Is possible
in this interior basin only where water la available, and
that is limited to those major streams that rise in the
mountains that surround the basin*
Not until the river traveler reached the Okanogan
highlands on his way upstream, would he notice much change*
Here the river leaves the basalt cliffs and winds through
rounded hills which feature bunch grass and pine in a typical
park landscape*
Farther to the north the vegetation increases
gradually to beoome more forest-like.
Much of the land away from the Columbia River and in
the basin is arid, windswept and supports little more than
cactus and sagebrush*
Major tributaries include the
Deschuttes, Just east of The Dalles, the Yakima nearly
opposite the mouth of the Snake and a number of small streams
In the northeast*
A division of the Plateau could be made
on the basis of aridity*
The southern and western portion
of the area, except for the mountains, is hot In sunnier,
dry treeless and basalt covered*
The northeastern part,
on the other hand, is forested and dotted with lakes*
has considered designating the latter region a
separate culture area*
This point will be discussed later*
In the hills and mountains that are on all sides of the
Plateau the size and numbers of trees are proportional to
the elevation*
The lower eastern slopes of the Cascades are
more or less in a rain shadow but benefit somewhat from the
run off*
On the other hand, the western slopes of the
eastern ranges receive sufficient direct rainfall to support
heavy forests*
The most common trees are Jackplne, Pitch—
pine, Douglas Fir and Maple In the higher elevations*
M&louf, personal oonrounication, 1953
Cedar and Oak dominate much of the forest edges whloh quiokly
give way to the ubiquitous sagebrush*
Mean Fahrenheit temperatures for different parts of
the Plateau as well as annual precipitation are taken from
Ray's charts*9
July Mean
in degrees
Kalispel .......
Qaatllla ------
Jan* Mean
in degrees in inches
68 •»....
66 .
71 .- - - 75 .....
-. 18
22 ••.
. 12
25 • -. 9
33 ... 7.5
Absolute maximums are not listed but temperatures exceeding
110 degrees Fahrenheit are not uncommon and temperatures as
low as minus 20 degrees are not unusual in the winter*
9* Bmj, 1936, p* 105
Map II gives the location and territory of the major
Plateau Tribes as of about 1825 to 1850, based on work done
by Berreman10, Ray**, and Osborne*2• The boundaries may not
be exact for many reasons; hunting territories are seldom
precise especially if hunting is not the major economic
There is also considerable evidence of movement
in the period between 1800 and 1850*13
Osborne's plan was
to place ethnic boundaries, wherever possible, on the nearest
topographio feature, stream or ridge to the land indicated
by native informants.
Three linguistic stocks are represented in the Plateau*
The Sahaptins are in the oentral portion along the Columbia
River and lower Snake River. The Interior Salish are in the
northern half of the area, all the way to the Canadian
The Northern Shoshoneans occupy the Snake River
south of Lewiston, Idaho, as well as the mountains that
fringe the northern Great Basin.
In order to gain an understanding of aboriginal life
Berreman, 1937
Ray, 1936
Osborne, 1952
Teit, 1928
k l / Tc k i t a t
in the Plateau it would be worth while to review one or
two typical Plateau cultures#
Ethnographic material is
available on several Sahaptin and Salish tribes but almost
nothing is written on the Plateau Shoshoneans. The Umatilla
are more or less typical of the Sahaptins and the Sanpoil
are representative of the Interior Salish#
Since the
Utaatilla occupied the lower portion of the McNary Reservoir
in 1800 a description of their culture is particularly
The Umatilla*
The Utaatilla occupied both sides of the Columbia River
from just east of Arlington, Oregon, to just west of the
mouth of the Walla Walla River. Prom what has been learned
in limited contacts it is believed that their nearest kin,
linguistically and culturally, were the Tenino who lived
on their western boundary.
Their political ties were few
but very friendly relations and a war alliance existed
between them and the Nez Perce' to the east*
Their "natural"
enemies were the Paiute in the desert to the south.
Tribal organization and centralization of political
power were undeveloped before the influx of Plains influence
which apparently began between 1750 and 1800. Local
# Most of this material is taken from H. D. Osborne,
"Excavations Near Utaatilla, Oregon; The Archeology of the
Columbia Interraontane Province",Unpub. Ph. P Dissertation.
autonomy was more or less typical of the whole Plateau
before the era of the horse and gun*
Local chiefs formerly
inherited their position but later achieved it by their
deeds* Concepts of ceremony probably changed at about this
The Umatilla shaman received his power from a guardian
spirit and both sexes could practice the art. Laymen also
sought guardian spirits, usually just prior to puberty*
Puberty rites were restricted to the first menarche iso­
lation of girls*
They believed that illness could be
caused by intrusions, both spirit and material or by loss
of the soul*
The corpse was prepared for burial by washing
and dressing but was never buried in the village*
War was important, at least after the horse and gun
became common*
With it was a complex of raids, coups,
scalps and slaves*
The bow and arrow whioh was used in
warfare was the principal hunting weapon*
Deer, elk,
rabbits, antelope and probably bison were hunted with both
the drive and stalking techniques used*
The Columbia was one of the main food sources*
salmon were said to be the most important fish and they were
taken by spear, nets, traps and weirs*
also numbers of shellfish*
Prom the river came
Surely as important as fishing
in the economy was the gathering of Camas and Kouse
(tuberous roots), berries, pine nuts as well as seeds, bark
and sap*
Cooking was done by roasting in an earth oven or
by boiling.
The Umatilla are said to have had dugout canoes and
rafts. Their houses were normally multifamily dwellings
60 feet long and 16 feet wide.
These were constructed of
poles and mats over a shallow excavation.
Conical semi-
subterranean lodges were also used, as were sweat lodges
and drying raoks for fish.
Other traits listed include armor, drums, flutes and
whistles, pipes, outdoor sports and a first salmon ceremony.
The Sanpoii-*
This group of people shared a language and culture but
were not a political unit.
The 1200 to 1300 Sanpoii lived
in autonomous villages along the Columbia in northeastern
Washington. Their territory of some 16,000 square miles
included about 85 miles of the Columbia as well as the
Sanpoii River drainage.
Much of their river frontage which
included all of their villages is now under the pool of
Grand Coulee Dam.
Theirs was a classless democratic society with no
slaves. All married men could vote and claim citizenship.
Anyone was eligible for chieftanship, a position of advisory
powers only. Since the village was the political unit there
was no real tribal organization.
There was no unity in war
» Material taken from Ray, "The Sanpoii and Nespelemj
Sallshan Peoples of Washington",U of Wash ,Pub in Anthro, 1932
for the Sanpoil did not make war#
Pacifism was stressed to
the point that even enemy raids were not retaliated.
The permanent habitations were along the Columbia River
whioh in this area is still a desert*
Just north of the
river there were forested hills, but they were used only for
hunting. It was the river that provided food, firewood
(driftwood), water and transportation#
Shellfish were also
used, as well as plants that grew along the banks of the
Pishing began in May and lasted until about the end
of November.
The most important species was the Chinook
Salmon but others were readily taken and dried.
methods of fishing were employed with the large traps pro­
bably the most important. Seines, dip nets and spears saw
considerable use particularly in conjunction with canoe
fishing. Large quantities of fish were dried on racks and
stored for use during the winter.
In the winter, hunting was the only practical economic
pursuit, but it too was carried on throughout the rest of
the year.
Hunting parties sought to drive the game which
was killed with the bow and arrow. Deer were hunted for
the most part but antelope, elk, bear and rabbits were
important food animals.
The spring and summer months were important ones for
gathering, and large quantities of many kinds of plants
were collected.
Camas was particularly favored with
bitterroot, service berry, chokecherry and sunflower seeds
filling in. Prickly pears and pine nuts were gathered in
The implements used were digging sticks, baskets
and carrying bags#
During the winter when fishing and
gathering wpre impractical much of their time was spent in
weaving baskets and manufacturing tools#
Most of the houses were of two types, the summer mat
house which was shallow and open to ventilation and the
winter lodge which was semi-subterranean and more solidly
built. Sumner clothing was rather scanty. The men wore
a breechclout and little else, not even sandals. The women
usually wore a woven poncho or often only a breecholout.
Winter apparel included the addition of fur robes, moccasins,
fur leggins and blankets. Snowshoes were known and used.
Both men and women painted their faces and braided their
hair on occasion.
A number of taboos were observed by the parents before
a child was born. Soon after birth the child was placed on
a cradleboard and seldom removed until he began to walk.
The usual isolation ceremony was observed by girls at
puberty while the boys performed exeroises such as swimming
and running without any real ceremony. After death the body
was immediately removed from the house and was buried as
soon as possible. A talus slope was usually chosen for
Guardian spirits were sought by all young boys and
some of the girls. They were sent out at night at an early
age in order to meet their spirits
Both men and women could
become shaman. They were well paid for curing but had to
perform certain public duties such as conducting funerals#
The Sanpoil had pipes, dice, the sweat lodge and wooden
mortars. The usual household might contain ten to twenty
Ethnographic Research
Two studies by Verne F. Ray3-4 give the most compre­
hensive treatment of tribal distribution in the Plateau.
These are "Native Villages and Groupings of the Columbia
Basin", and "Cultural Relations in the Plateau of North­
western America". Ray is quick to point out that the word
tribe does not apply to the Plateau until sometime after
1750 or 1800 because of the widespread local autonomy.
The boundaries shown on Map II thus are more dialectical
than political. Although there are hints of movement and
perhaps migration in the Plateau just prior to 1800 they
have not been worked
The groupings shown in Map II are concerned with two
languages, Sahaptin and Interior Salish.
The following are
Sahaptin groups: Tenino, Klickitat, Utaatilla, Cayuse,
Nez Peroe', Yakima, Wallula, Wanapam, Kittitas and Palus.
14. Ray, 1936 and 1939
15. Teit, 1928, p. 98
The Interior Salish include the Wenatchi, Columbia, Chelan,
Methow, Okanogan, Nespelem and Sanpoil, Spokane, Coeur
B'Alene, Colville, Lakes and Kalispel.
At the present time there are not sufficient data with
which to show significant differences between one group and
the next or between Sahaptin and Salish. A few generalities
can be made, however.
The Sahaptins apparently received
influence from the Plains Area earlier and were perhaps more
receptive to it. The ttaatilla were close allies to the Nez
Perce' who in turn were in touch with such Plains tribes as
the Blackfeet.
Through such contacts the Sahaptins learned
of horses, guns, tribal organization, war honors and other
typical traits of the Plains Area.
The Salish, without
suoh direct contact, remained what might be termed conserva­
They were slower and more selective in their
borrowing. There are probably many other differences*
is reason to believe that the Salish were better boatmen,
there being more water in the northeastern Plateau. Suoh
things as bear worship, exposure of the dead, grooved mauls
and "potato masher* pestles seem to be oonfined to the
Salish tribes, as head deformation, funerary houses, stone
fishing weights and stone fetishes seem to be more southerly
centered. All of these traits oould be studied fruitfully.
The distribution charts, Figures 7 to 16 cover only a few
sites for only a few site reports have been published in
the Plateau.
Scientific research, in the Plateau began with the Lewis
and Clark expedition of 1800 to 1806. Their Journals**"® are
full of ethnographic observations and faotual reporting
which showed these explorers to be far ahead of their time#
Lewis and Clark saw only the southern part of the Plateau
but recorded a most valuable account of what they witnessed.
Robert Stuart kept a diary of his travels but was not
given to recording many details. He may have been a keen
observer but left little in writing to preserve what he saw
in 1812. A full account on almost all parts of the Plateau
as of 1832 can be found in Parker*s Travelogue3'"''. Despite
the fact that he saw the Indians through the eyes of a
missionary intent on conversion, he gave excellent des­
criptions of clothing, houses, ornaments, economy and social
The journals of Alexander Ross cover a period between
1818 and 1828. Certain parts of his material are valuable
but they cover only a limited area.
There are many other traveler's accounts which can
yield specific information but none to compare with Lewis
and Clark or Parker.
It was not until the turn of the oentury that work In
the Plateau was begun by trained anthropologists. The main
contributions in archaeology were by Harlan I. Smith in
16. Lewis and Clark, 1905
17. Parker, 1844
Southern British Columbia^8 and the Yakima Valley^*®, Herbert
Krieger in the middle Columbia River20, Strong, Sohenck and
Steward at The Dalles2-1, and Collier, Hudson and Ford on the
upper Columbia River22•
Cressman did some work in the
Plateau23, but the majority of his research has been in the
Northern Great Basin.
The work of each of these men will
be used for comparative material.
Ethnological research in the Plateau has been oarried
on to a slightly greater extent#
The earliest investi­
gations were ethnographic and dealt with a single group.
Later work was of a comparative nature and in some cases
made contributions to anthropological theory as well as fact.
The first major ethnographic report published on the
Plateau was by James Teit and consisted of a study of the
Interior Salish24. It was followed shortly by Spinden'3
excellent monograph on the Nez Perce125. Leslie Spier and
Edward Sapir studied the Wishram2
near The Dalles about
the same time that Verne Ray'27 was working among the Sanpoil
and Nespelem in Northeastern Washington.
Linguistic studies
Smith, 1899
Smith, 1910
Krieger, 1927 and 1928
Strong, Sohenck and Steward, 1930
Collier, Hudson and Ford, 1942
Cressman, 1950
Teit, 1900
Spinden, 1908
Spier and. Sapir, 1930
Ray, 1932
have been made by Melville Jacobs for the Sahaptins* , Saplr
for the Wishram?9, and Gladys Relohard for the Interior
Minor contributions have been made by a number of
authors. There is no intention of slighting the works of
the many contributors to Plateau ethnography and archaeology,
but special mention can only be made of those whose research
has had fairly broad coverage of the area, or is pertinent
to this study.
Joel Berreraan published a study on the
distribution of tribes in Oregon
, and Walter Cline
published a study of the Southern Okanogan
The most active ethnologist in the Plateau has been
His work and major publications have been listed
earlier but it is worth mentioning that he has contributed
much to the ethnographic understanding of the Plateau Area.
His "Cultural Relations in the Plateau" is the only work of
its kind on the area.
Jacobs, 1931
Sapir, 1909
Relohard, ^published
Berreman, 1937
Cline, 1938
About the time that it was dear that World War II
would oorae to a suooessful conclusion, plans were made by
the federal government for a nation-wide program of flood
control and irrigation#
The realization that these plans
would be carried out caused some concern among historians,
archaeologists and paleontologists, in that the work would
inundate vast areas in which little or no scientific investi­
gation had been undertaken#
In 1945, one of the first steps was taken to remedy
the situation*
Members of the Committee on Basic Needs in
Amerioan Archaeology of the National Research Council, and
the Smithsonian Institution met and discussed plans for
initiating a program of salvage. First, a committee for the
Reoovery of Archaeological Remains was formed to study the
Later, agreements were reached between the Smith­
sonian Institution, the National Park Service, the Corps
of Engineers and the Bureau of Reclamation for a salvage
The National Park Service is the agenoy officially
responsible for the preservation of historical and archaeo­
logical sites; it therefore became the responsible agency
in the salvage program#
In October, 1945, a working agree­
ment was reaohed "between the Park Servioe and the Smithsonian
Institution, whereby the latter would undertake scientific
responsibility for work done. The Park Servioe would, under
the agreement, advise the Smithsonian Institution of Federal
projects whioh might involve salvage, and in turn advise the
agenoy responsible for the construction that archaeological
and or paleontologioal sites were threatened if such were
the case*
River Basin Surveys v/as actually organized in the fall
of 1945, with Dr. Frank H. H. Roberts Jr. as Director. It
was set up as a unit of the Bureau of American Ethnology.
In 1946, the Missouri River Basin established the first
field office at Lincoln, Nebraska.
In 1947, field head­
quarters were made at Eugene, Oregon, for the Pacific Coast
Region, and Dr. Philip Drucker was appointed Director.
Initial archaeological surveys were begun in the
Pacific Coast Region during the summer of 1947.
One of the
first surveys accomplished was that of the McNary Reservoir
on the Columbia River (see Map II).
Clarence Smith and
Franklin Fenenga did the field work and reported 120 archaeo­
logical sites within the pool area.
They recommended that
22 or more of these should be given priority for excavation.
Numbers of Reservoirs were surveyed in subsequent years, and.
reports on the results were mimeographed for distribution to
cooperating agencies and institutions.
During the summer of 1948 excavations were initiated
at three sites*
At site 45-BN-3 (Berrian,s Island), a
burial ground was almost oompletely excavated, at site
45-BN-53, a pit house village was partially exoavated and at
site 35-UM-8, an early occupation site was tested.
In 1949,
site 35-UM-8 received additional testing, and at site 35-UBS-7
(Cold Springs), a pit house and midden area was partially
In 1950, two pit house villages were partially
These were 45»BN»6 and 35-UM-17 (Techumtas
In 1951, an early site 35~UM*>5 (Hat Creek), and
a late site 45-WW»6 (Wallula) were each partially excavated*
In 1952, further testing was done at sites 35-TJM«*7 and
Sites 35-UM-3, 35-TJM-10 and 45*>BN~55 were scenes
of minor excavations*
After eaoh suntner in the field, the exoavated materials
and data were taken to the laboratory in Eugene for study*
As soon as possible, reports were written on the results of
eaoh excavation*
These reports were little more than a
factual account of the material recovered with an attempt
to relate it to other cultures In time and spaoe*
of their limited distribution illustrations were kept at a
Techniques In the field were kept as flexible as
possible and only a few were standarlzed*
was the grid system*
One of these
A site was staked with seotions of
reinforcing Iron with wooden caps which oould be lettered*
A single row of stakes ten feet apart were set across a
site in a north-south direction and numbered consecutively
from south to north. Facing north for directions right and
left, rows of stakes parallel to the first row were given
the additional designation R. or L. The first stake tc
right of stake 1 would be 1-R-l. The seoond stake to the
left of stake 3 would be 3-L-S and so forth. Since engi­
neering surveying equipment wa*s provided, and it was cali­
brated in feet and tenths of feet, these units were used
in the field.
Artifact locations were recorded in three dimensions,
with horizontal measurements taken from the nearest stake.
Vertical measurements taken from both the surface of the
ground and by alidade. Animal bones and other specimens
were normally recorded by arbitrary levels within a grid
Sites were mapped and contoured with excavations,
houses and other features shown. It was not deemed neces­
sary to relate each map to sea level but only to a single
benchmark arbitrarily chosen within the site.
Each house
was exoavated as a unit, and was controlled by an interior
grid system of five foot squares.
Normally, a house was
trenohed from north to south to determine what condition
the floor or floors were in.
Further excavation could be
made by parallel trenches or by quartering. A few houses
were exoavated by horizontal plane sections one foot apart
so as to distinguish the floor from the fill#
It was not
feasible to follow floors for they were far too thin and
It proved to be easier to find them in profile
by the ohanges in soil color.
It was seldom possible to do more than sample a site.
Funds were limited, and there were dozens of sites to ex­
cavate before the water began to rise behind the MoNary Dam#
Thus, four or five houses were usually excavated in a village
of several hundred dwellings, and three or four tests were
made in the midden. The only site that was extensively
excavated was 45-BN-3, a burial site with a remarkable
array of material culture*
M A P 2.
Ash stratum at site 35-UM-3
The dark soil is the midden beneath the ash#
Ash stratum at Hat Creek (35-UM-5)
During the extensive surveys and excavations in the
IvIcNary Reservoir, River Basin Surveys found no evidence of
the well known "Early Man" tools suoh as have "been related
to Folsom, Yuma and other cultures.
No fossils or remains
of extinct animals were recognized in association with
cultural material.
The earliest cultural evidence comes from two sites,
35-UM-5 (Hat Creek) and 35-UM-3. The former, which is
situated on the east bank of Hat Creek near the Columbia,
was extensively excavated.
Site 35-UM-3 was only tested
in several areas. In order to determine the relative age of
the culture represented at the Hat Creek site, the geologi­
cal stratigraphy must be considered. At both 35-UM-5 and
35-UM-3 all of the cultural materials were found beneath a
thick (one to two feet) mantle of pure volcanic ash.
continuous layer of ash was unbroken except for small rodent
holes and forms an effeotive isolating mantle for the
cultural material beneath it.
Disappointingly little is known of the origin of the
volcanic ash.
Efforts to tie it definitely to a specific
volcano have not succeeded mainly because there are too
many of them#
Williams notes,
"Too much space would be occupied by listing the
signs of postglaoial volcanic activity elsewhere along
the crest of the Cascades, for there are youthful flows
and cinder cones by the score*"33
For an idea of what takes place in volcanic eruptions
of this sort oonsider the example of Mt* Katmai in Alaska
whioh erupted violently in 1912.34 This was one of the most
dramatic explosions known to history and occured in three
stages, one on the sixth of June and two on the seventh*
The result of the explosions was something like that which
took place when Mt* Mazama (Crater Lake) blew apart*
Several oubic miles of rock were pulverized to finely
divided dust and were blown high into the air*
At the town
of Kodiak, 100 miles from the mountain, darkness lasted
60 hours*
Fifteen miles from the volcano the ash was five
feet deep, 118 miles away it was three and one half inches
deep, and some of it drifted 900 miles to the east*
On the island of Kodiak, 50 to 100 miles east of Katmai,
brush and trees were burled, but a high percentage came up
through the ash and survived*
Marine life apparently
suffered more than did the land dwellers, for fish and
shellfish died in great numbers*
A few birds and animals
died on Kodiak. Islanders themselves took refuge in houses,
but there were many oases of sore throats and eyes*
Williams, 1948, p. 51
Martin, 1913
Two or
three men died from the effects*
For several years there
was a shortage of game fish, but things eventually returned
to normal.
The story of the eruption of Mt. Katmai has interesting
parallels with what must have taken place in the McNary
Region. At least two villages or camping spots were buried
under from one to one and one-half feet of ash*
The pattern
of the ash deposit itself is remarkably similar to that
at various places east of Mt. Katmai.
Martin illustrated
a chart indicating the depth and composition of the Katmai
ash at several stations east of the mountain.
The column
at Middle Bay, Alaska, 101 miles east of Mt* Katmai was
ll£ inches high, and the oross section is diagrammed below.
The McNary ash was 17^ inches high with the pattern depicted
Mt. Katmai
fine white
fine brown
medium brown
coarse gray
35. Martin, 1913
The MoNary pattern suggests that either it was closer
to its eruption than 101 miles, or that the eruption that
deposited the ash was of greater intensity or duration#
The present writer is inclined to believe that there were
two major explosions in the same volcano, and that they were
a matter of a few days or perhaps a week apart*
For example,
there is a repetitive pattern in the MoNary ash:
Furthermore, between the lower white layer
and the brown layer, there is a slight admixture of sand
indicating perhaps some wind deposition between ash falls.
Other than a physical description, little can be said
of the MoNary pumicite. Its existence is ignored by the
geological literature, and inquiry around departments of
Geology at northwestern universities yielded no information*
In 1951, Dr. Harold Culver, of the Department of Geology at
the State College of Washington, examined the deposits at
Hat Creek and other exposures in the vicinity. He concurred
in the opinion that all of the exposures were homologous,
and that the source lay in the Cascade Mountains about 100
miles to the west. The volcanic souroe could not have been
much closer than 100 miles for the ash itself is composed
entirely of minute particles. Were the souroe closer the
ash would have contained larger particles including pumice.
The fact that the ash was air-borne over a considerable
distance wa3 verified by Dr. Culver.
Dr. Culver found no basis on which the precise time
of the a3h fall could be determined*
He postulated a post­
glacial age and late rather than early post-glacial.
nation of the ash itself was made by Dr. C. D. Campbell,
also of the State College of Washington*
He desoribed the
material as very fine volcanic dust with an index refraction
of 1.53 indicating andesitic rather than rhyolitic affinities.
Part of the material proved to be 90 percent glass with no
incipient crystallization. Another part of the column
showed 50 percent crystalline material, in part plagioclase
Estimates, all admittedly guesses, would place the time
of the ash deposition at between 2000 and 5000 years ago.
Without going into detail, it is the writers opinion that
it would be difficult indeed from a cultural perspective,
to account for as much as 5000 years since the eruption.
On the other hand, the changes in material culture and depth
of deposit seem to require no small amount of time. It is
dangerous perhaps to speculate with so little evidence but
the cultural data would tend to support a figure closer to
2000 rather than 5000#
Recognition of the layer of voloanic dust as a single
event permits the relative placement of several of the
MoNary sites in time. Sites 35-TM-3 and 35-111-5 (Hat Creek)
are stratigraphically earlier than the deposition; sites
35-UM-7 (Cold Springs), 35-UM-8 and 45-BN-6 are later sinoe
the ash stratum underlies the occupational remains.
35-UM-5 (Hat Creek)
On the south hank of the Columbia about six miles east
of McNary Dam, Hat Creek empties Into the river• The creek
is named for Hat Rook, a tall erosional remnant of columnar
basalt which looks like an old beaver hat#
This landmark
was seen by Lewis and Clark and noted on their map by the
name Hat Rook.
The site is located on both sides of Hat Creek, a little
distance from the river proper.
The principal exposure, on
the east bank of the creek, was 50 yards from the river; the
one on the west bank was 100 yards from it. Both exposures
were fairly high on the steep hillsides, and ware 40 or 50
feet above the normal river level*
Extensive testing of the
west bank uncovered cultural material beneath the pumioite,
but the pumioite stratum itself had been almost decimated
by rodent burrowing. Since nearly ideal conditions were
found on the opposite bank, the material on the west bank
may be ignored.
The exposure of the voloanio ash on the east bank was
a mere six or eight feet of badly slumped material#
excavation, after the establishment of a grid pattern for
horizontal control, was in the form of a broadface ten feet
wide. It was soon extended to 30 feet after oultural ma*
terials began to appear under the volcanic ash (see Plate la).
The ash stratum sloped slightly toward the river and
slightly more toward the oreek, thus dipping toward the
northwest at about three degrees.
Its mean thickness in
the excavated area was 17 inches, but a few hundred yards
up the oreek in a small depression, it had accumulated to a
thickness of seven feet*
The later deposit was due to wind
Beneath the ash was a stratum of sandy
loess soil thinly scattered with broken rock, charcoal, ani­
mal bones and an occasional hearth. This midden, averaging
three feet in depth, rested on a hardpan of undetermined
thickness. Some charcoal flecks and bone fragments were
embedded in the hardpan, but below the first tenth of a
foot, it was sterile,
Reoent erosion has eliminated part of the site, the
extent of which is not known#
As excavation proceeded and
the broadface advanced some 20 feet, the thickness of the
overburden increased to serious proportions, five and more
feet of sterile coarse sand.
The eastern extremity of the
site was not reached.
Primary consideration was given to the maintenance of
stratigraphic controls.
Every effort was made to be abso­
lutely certain that the excavated material was unquestionably
older than the ash stratum. This involved a careful in­
spection of the ash stratum to detect any penetration before
the midden beneath it was excavated. By using a step
pattern of excavation, one foot wide along the length of
eaoh ten foot square, the overburden was removed. The ash
along a ten foot horizontal atrip was examined and then re­
moved to expose the midden*
This procedure was repeated as
eaoh section of the broadfaoe advanced*
No penetrations
other than small rodent burrows were found.
The course sand overburden was tested repeatedly by
soreening and found to be absolutely sterile of cultural
The oourse sand rested conformably upon the ash,
so that there was no possibility of the cultural materials
having fallen down through the rodent burrows*
If the sheer volume of burned bones, flakes and fire
oracked rock was not proof enough, several artifacts,
hammers and choppers, were found to be considerably larger
than the rodent holes. Furthermore, hearths composed of
rocks and mussel shells were found in situ.
It was realized that the Hat Creek site was important
to the regional prehistory, especially since it was the
first clear example of pre»ash occupation*
Therefore, it
was deoided that as muoh as possible would be saved for
laboratory analysis*
After the slumped face was cleared,
every bit of the midden material was screened through a
one fourth inch mesh hardware cloth*
of bone and broken rock was saved*
Every flake, fragment
Arbitrary one foot
levels were maintained for bones and flakes, using the ash
stratum as the datum plane. Every artifact recovered during
the excavation was recorded In three dimensions to the
nearest one tenth of a foot.
Closer Inspection in the
laboratory turned up several small artifact fragments among
the flakes#
These might well have been overlooked had the
flakes not been saved*
Analysis of the three levels of flakes showed that
there was no significant variation in materials from one
level to the next*
A level between three and four feet
below the ash included some of the sterile hardpan, and for
that reason is not included here#
In numbers the flakes
were highest in the middle level and lowest in the upper
The same was true of the artifacts#
Perhaps the
intensity of occupation was dwindling before the volcanic
ash fell*
If there was any change in the material culture during
the occupation at Hat Creek, it did not show up in the
relative position of the artifacts. The similar proportions
of stone materials also bears this out. Therefore all of
the cultural material is treated as though it were the
result of a single occupation.
The following table shows the distribution of flakes
by materials, and is based on 3,000 flakes«
Table I
- - - --
Quartzite - - - -
Red ocher - - - -
-- - -
There was no evidence of habitations
at the Hat Creek site. If the small fireplaces found there
had been associated with dwellings, the latter left no
Sinoe the depth of the midden suggests an occu­
pation of some duration, some sort of shelters must have
been built. It is probable that such shelters were made
of brush or mats, and that they were light enough to stand
without large posts*
Material Culture; The Hat Creek site did not yield a
large sample of material culture, but nevertheless the
recovered materials permit certain conclusions*
Besides a
certain amount of non-oultural data, some 97 artifacts were
recovered from beneath the volcanic ash*
Most frequent among the artifacts recovered were pro­
jectile points and fragments thereof.
Twelve of these were
either complete or lacked only the tip of the point, so that
their shape could be determined with some aocuracy (see
Plate II b). Ten of the 12 were of a simple wleafw shape,
with no notohes, barbs or shoulders.
They were oval in
'v TWti
*J ^A
s. <MI
Projectile points, Hat Creek (35-TJM-5)
4 4 m•
4 # »
Flake scrapers, Hat Creek (35-TJM-5)
outline with convex bases. Two traits of manufacture were
Several of the projectile points were markedly
plano-convex in cross section, and several had deliberately
serrated edges. Some specimens had one or both traits, some
had neither. Those specimens which could be measured varied
from 39 to 51 mm. in length and from 17 to 24 mm. in width.
Another variety of projectile point was represented by
two specimens.
The only noticable difference between this
and the preceding type lay in the base, which is concave
instead of oonvex. Complete specimens of the second type
were not recovered,but there appeared to be a close simi­
larity in size and general appearance.
The 33 fragmental projectile points bore out the con­
clusions made earlier, that no notching or shouldering was
apparent and that piano-convexity and serration were conmon.
Workmanship throughout was considered excellent; the people
who made these artifacts knew what they were doing, and ex­
celled in flaking several kinds of material. Twenty of the
points and fragments were of basalt, and the rest were of
local quartz family rooks, jasper, agate or chert.
whioh became popular at a later time, was absent.
One fragment, tentatively classified as part of a
projectile point, merits some further discussion. It was
a midsection, of basalt, but wider and thicker than the
other fragments. When complete, it would have been larger
than the other projectile points, possibly much larger. The
flaking was not as well done as the rest of the points, and
there is reason to believe that it may have been a knife*
Large basalt knives were typical of the early occupation of
the Cold Springs site, stratigraphlcally later than the Hat
Creek culture (see Plate V b)*
They were associated there
with projectile points identical with those from beneath
the ash at Hat Creek*
Since every flake recovered from the Hat Creek site
was brought back to the laboratory, a large number of used
flakes were counted*
The majority of these were primary
flakes of jasper, basalt and chert that had been used for
cutting or scraping*
Usually, one edge showed a row of
tiny flake scars, the result of having been used*
scrapers that had been prepared before use, on the other
hand, showed larger and longer flakes as well as some in­
tentional shaping. Altogether, only six of the 27 scrapers
appeared to have been prepared for use. Size and shape
seemed to make little difference to the inhabitants of the
site, and the materials used were those available in the
river gravel*
Tools used for hammering and chopping were made from
river worn cobbles, for the most part (see Plate III b).
The hammers (5 specimens) were unaltered cobbles that showed
the scars and abrasions of use but no intentional shaping*
The chopping tools can be divided into three types*
One of
these (13 specimens) was the ovoid river oobble from which
Carved bone, Hat Creek (35-I3M-5)
Hammers and choppers, Hat Creek (35-TJM-5)
five to 15 large flakea had been struck#
It produced a
ragged but effective cutting edge. Another type (5 specimens)
was a heavy spall struck from a river worn cobble and
crudely chipped to a cutting edge. A third type (4 specimens)
was made from an exfoliated basalt slab which had a wedge
shaped cross seotion.
Characteristically, all of the heavy tools left something
to be desired in both the selection of stones and in the
chipping. The cobbles were seldom symmetrical and would
seem to fit the hand poorly. The chipping was orude for the
flakes removed were fewer and larger than would be character­
istic of later workmanship. This does not mean that each
hammer and chopper in the McNary region can be assigned
readily to a pre or post ash date by inspection, but, as
a group, those tools from beneath the volcanic ash are
distinguished by their orudeness«
A total of seven bone artifacts came from beneath the
volcanic ash.
Two of these were slender splinters, probably
from a deer long bone, that showed some wear and polish at
the point and along the shaft (see Plate III a). Little
or no effort had been expended in improving their appearance
or utility.
Two beads of bone had been fashioned from bird
tibia (see Plate III a). One was just under 2 cm. long
and 7 mm. in diameter.
The other was 3.5 cm. long and
4 mm. in diameter. Both had been decorated by incising
a few transverse lines on the surface. Two small sections
broken out of the shaft of a deer long bone had been inoised
with deep parallel lines. The complete artifact type is not
known but must have been ornamental.
A final piece of bone
is not actually an artifact but a residual fragment. A
section of long bone had been deeply grooved and snapped
off. What was recovered was the waste 6f the making of some
unknown artifact.
The simplicity of the bone tools and ornaments seems
to agree with that of the other artifacts at the Hat Creek
site. Only the projectile points are really objects of
craftmanship. It appears that, except for projectile points,
almost no effort at all was made to improve or finish an
Economy: The animal bones from the Hat Creek site
have not been thoroughly analyzed. A preliminary analysis
showed that numerically speaking rabbits were most frequently
killed. Deer were the second most frequent, followed by
Bird bones were present in the midden, but their
fragmentary condition did not permit an immediate identification as to species. Almost all of the animal bones
were broken and a great many of them were partially burned.
The proportion of fish bones found at the Hat Creek site
was smaller than at any other site excavated in the McNary
As will be shown in the description of material
culture, specialized fishing was apparently not developed
at the Hat Creek site. The relative scarcity of fish bones
bear out this observation.
Ethnographic reports tell of fish being taken with
olubs, with lances and with bow ana arrow. It is quite
possible that techniques of this sort were employed during
the occupation at Hat Creek since no net weigjhts were
About two miles west and downstream from Hat Creek on
the same (Oregon) side of the river is site 35-TJM-3.
is situated on a hillside that slopes toward the river and
is 30 or 40 feet above the normal river level.
The 3ite
itself is in and around a large 3and blowout, some 150
feet long and 75 feet wide (see Plate I a). In the center
of the blowout there was an erosional remnant of midden
capped with a thick stratum of volcanic ash which in turn
was capped with a stratum of sandy loess.
The volcanic ash is part of the same ashfall that is
present at Hat Creek and Cold Springs. At 35-UM-3 it was
undisturbed except for occasional rodent burrowing.
sand over the ash was sterile and contained no habitational
Beneath the ash there were some signs of occupation,
but it could hardly be called a midden. The animal bones
and flakes were thinly scattered through some two and one
half feet of sandy deposit.
On both sides of the erosional
remnant the wind had blown out about six or eight feet of
sand and ash. The surface of the blowout wa3 paved with
flakes, broken rook, mussel shells and occasionally an
artifact. It cannot be proven, but it is suspected that
the cultural debris carae from beneath the volcanio ash.
Test excavations beneath the ash recovered a few flakes,
a spall flake scraper or knife and a basal section of a
projectile point. The latter, leaf shaped with a convex
base, would fit easily into the collection from Hat Greek.
The test exoavation Was not large, and a very small sample
was recovered, but it demonstrated that the situation at
Hat Creek was not unique. On the surfaoe of the blowout a
small fragment of bone was found. It was incised in exactly
the same way as were the two fragments from the Hat Creek
site. The few chopping tools from the surfaoe were also
similar but contemporaneity cannot be proven.
55-UU-7 (Cold Springs)
The Cold Springs site (35-UM-7) lies about three miles
east and upstream from Hat Creek.
Situated on the south
bank of the river, the site can be detected by its numbers
of pit house depressions, mussel shell fragments and broken
rook on the surface.
The site ocoupies a terrace which was not inundated
except by the highest floods, such as occurred in 1948 and
1903 (see Plate IV b). Its composition is for the most
part fine silt and wind blown sand, although no small part
of the volume is made up of the midden debris of man, shells
and broken rooks*
This terrace Is roughly 150 feet wide and
1500 feet long. The midden material, where tested, averaged
some five to five and one half feet in depth.
Beneath the midden, but not altogether in stratigraphio
oonformity with it, lay a stratum of volcanic ash«
This was
the same stratum which was exposed at Hat Greek, and which
covered the midden there. Although there had been some
slight erosion of the ash at Cold Springs, and it had been
severely burrowed through by rodents, it formed an excellent
point of orientation in relating the Gold Springs artifaots
to those from Hat Creek.
Another stratum, somewhat weakly developed in certain
parts of the site, was composed of shells of the fresh water
mussel, Margaritlfera margaritlfera falcata. (Gould). This
stratum was more or less continuous over most of the site
in the form of contiguous lenses and lay between two and
three feet below the surface (see Plate IV a). Its thick­
ness varied from about one inch to nearly one foot, and it
was made up of tightly paoked shells, some burned and some
Praotically nothing else was to be found in the
shell layer.
Since the Cold Springs site was excavated prior to the
excavation of the Hat Creek site, the Cold Springs materials
were reexamined. In 1952, additional testing was done at
Cold Springs, and a larger sample of artifaots was recovered.
Shell stratum in profile, Cold Springs (35-UM-7)
View of terrace, Cold Springs (35-UM-7)
Altogether, some 2500 man hours were spent in excavating
the site, but only a small oolleotion of artifacts was
recovered. Less than 200 speoimens were catalogued#
Material Cultureg
In the earliest trash that accumu­
lated above the voloanio ash, the stone and bone artifacts
were practically identical to those reoovered at the Hat
Creek site*
These artifacts included oobble hammerstones,
cobble choppers, flake scrapers and leaf shaped projectile
The early trash in this case is considered to be
approximately the first 12 inches of midden which accumu­
lated above the ash.
This mu3t be approximate for there
was no even distribution of trash over all of the site at
all times*
The closest similarities between the artifacts of Hat
Creek and Cold Springs are t"o be found in projeotile points
and scrapers*
Both were made of like materials, jasper and
basalt, and both were flaked in the same manner*
Five leaf
shaped points were recovered in the first foot of the postash deposit (see Plate V b, bottom row). They varied from
41 to 48 mm. in leftgth, which compares favorably with the
39 to 51 mm. of the projectile points from Hat Creek. The
same tendencies toward piano oonvexity and serrated edges
were present at Cold Springs, as had been notioed at Hat
All in all there is so close a similarity between
the points of the two sites that a continuity must be
1 2 3 4 5
Stone lenives, Cold Springs (35-UM-7)
4* » A
• M « A
Projectile points, Cold Springs (35-UM-7)
Some differences must be noted.
At Hat Greek two of the
twelve classifiable points had concave bases, a variety that
was not found in the early midden at Cold Springs. Further­
more, at Cold Springs, associated with the five projeotile
points just desoribed, was one side notched point with a
concave base. It will be recalled that none of the points
from Hat Creek were notched or barbed. This single specimen
might be termed the earliest (in our collection)of a type
that became more numerous, and soon replaoed the leaf shaped
Not much oan be said of the hammer-stones and choppers.
These crude tools apparently were used and disoarded at will.
The most frequently used material was a tough fine grained
basalt. Hammers were either tonaltered cobblestones from the
river gravel or choppers whioh had been worn and dulled.
The tools classified as choppers were also river oobbles,
but had been roughly flaked to a Jagged cutting edge.
is little chance that either the hammerstones or the ohoppers
had been hafted, their shapes would have made it most dif­
ficult, and there was no sign of grooving.
In comparing
these tools with those from Hat Creek, a few general obser­
vations may be made. The Hat Creek specimens were of in­
ferior manufacture and showed less use than those from
Cold Springs. In most oases, the oobbles selected for use
at Cold Springs were more symmetrical and would have been
easier to handle*.
A new type of nohoppingm tool appears at Cold Springs,
and stratigraphically shows up in the early part of the
occupation. It is made on flat oval or round river cobbles.
The diameter is usually between 60 and 100 mm. and the
thickness between 10 and 15 mm. Basalt was usually the
material used, but it is not df the tough fine grained
variety that was preferred for the oobble choppers. This
tool which may well have been used for scraping and fish
scaling, as well as chopping was made in several forms.
The edge of the flat stone was flaked from both sides to
produce a tool with a single bit, or one with a double bit,
or occasionally one with a cutting edge all the way around.
The early or deepest portion of the midden showed only the
type with the single bit, and the others appear to develop
somewhat later. Altogether ten of these tools were re­
covered at the Cold Springs site, and the minority, about
four, were assignable to the early portion of the midden.
Three kinds of net weights were found at the site, and
each type appeared early in the occupation. The type of
which the most specimens were recovered was the notohed net
weight (see Plate XI b). It was made from the same kind
of f^.at river worn rooks that were used for the chopping
tools described above. The materials selected were the
same as for the ohopping tools, but there was a preference
for more oval stones in preparing net weights, and more
circular stones for the choppers. The notched net weight
was prepared as the name suggests by notohlng both ends of
the flat oval stone, so that it could be secured to the net*
In most cases there Is evidence that the sharp edges of the
notch were abraded in some way, so that they would not cut
the binding cords*
Perhaps five notched net weights could
be assigned to the early part of the occupation*
At least
two specimens apparently were discarded or lost shortly
after the occupation began at the site*
Another type of net weight was of the same general
proportions but about twice as large*
It was bored through
with a hole 15 mm. In diameter placed near one end, but
otherwise was unaltered*
Several artifacts of this type
have been picked up on the surface along the banks of the
Columbia, but none besides this one seem to have been found
in direct association with other archaeological materials*
The third type of net weight was an ovate river worn
stone with dimensions 16 cm* by 13 cm* by 7 cm*
shows a similar specimen*
Plate XI b
The material is quartzite, and
it has a shallow groove circling the smaller diameter.
with the other types of net weights, the only alteration
of the stone was in creating a means of fastening it secure­
ly to the net*
One girdled and one pierced net weight were
found, both of them associated with the early part of the
Stone knives used during the early occupation of the
Cold Springs site were of two general types (see Plate V a)»
One type was a basalt blade, long and slender, with overall
The ohipping was rough and appeared to be of the
percussion type. Each of the three complete specimens was
pointed at each end although the bases were blunter and
somewhat thicker in cross-section.
One of the three had a
deliberately made shoulder on one side only. This specimen
bears a striking resemblance to the Sandia point of New
Mexico, however, it is longer, more slender and probably
considerably later in time. The lengths of the complete
artifacts varied from 12,7 cm. to 14.5 cm.
There was a
slight curvature of the longitudinal axis prominent in two
of the knives, but nob in the specimen which was shouldered.
In addition to these, there were two fragmentary knives
(mid-sections) that seemed to conform in size and shape,
and were also made of basalt.
The second knife type, represented by two specimens,
is considerably shorter and wider than the first.
One of
these was 7.7 cm. long and 3.8 cm. wide, and the other was
8.3 cm. long and 4.9 cm. wide. Both were of basalt, were
well made and had convex bases.
Besides these two knife types there were found two
large rough blades made on primary flakes. Only one edge
of the flake was deliberately shaped, so these specimens
were not classified as a diagnostic type of knife.
Stratigraphically later than most of the material
described above were fifteen side notched projectile points
(see Plate V b, top two rows). This point type apparently
oame into use after the site had been occupied for a while,
existed side by side with the leaf shaped point, and then
replaced it entirely#
Besides the five leaf-shaped points
described for the early part of the midden, three more of
them were associated with the side notched points. There
is considerable variation in the size of the side notched
projectile points. The range is from 3.2 to 6.6 cm. in
length, but with no significant size variation according
to depth below the surface.
Even the smallest of these
points are larger and heavier than the small types found
in late sites.
Thirteen of the fifteen side notched points
have concave bases, and the rest of the bases are either
straight or slightly convex.
All kinds of materials were
used for the side notched points, with basalt most frequently
Obsidian, jasper and chalcedony were about
equally distributed.
There were two pieces of carved stone associated with
the upper part of the midden.
One of these was a small
oval piece of steatite that had been drilled for suspension
but had not- been finished.
steatite tubular pipe.
The second was a carefully made
It was bell shaped with a short
stem which had a small flanged mouthpiece.
A nearly identi­
cal specimen found at site 45-BN-3 is shown in Plate XIII b.
Bone tools were relatively scarce.
Since large numbers
of animal bones were recovered at all levels, preservation
is probably not a factor. Those tools that were recovered
were for the most part crude and showed that little at­
tention had been paid to their preparation.
They included
a tip of a broken flaking tool and the lip of an antler tine
that may have been used for the same purpose.
Two slender
tips of splinter awls had been ground to fairly sharp points
but reoeived little treatment along the shaft.
One bone
bead, apparently from the leg bone of a large bird, was
nearly identical in size and shape to one recovered beneath
the volcanic ash at the Hat Creek site.
and 0.6 cm. in diameter#
It was 1.9 cm. long
Another bird leg bone had been
girdled for breaking but had split during the operation*
The only specimen of marine shell found was a bead of
Ollvella biplioata with the tip ground off so that it could
be strung. It was probably assooiated with the later part
of the occupation of the site, and represents the earliest
ocourance of sea shell in the McNary region that has been
One of the most unusual artifacts found at the Cold
Springs site was a bone projectile point.
It had been
carved in such a way that it appeared to have been chipped
from stone.
Its size and shape were identioal to several
of the side notched projeotile points, and had it not been
for its light weight, it could easily have been mistaken
for a stone projectile point.
I 8
No traces of dwellings had been found
at the early sites 35-UM-5 (Hat Creek) or 35-TJM-3.
It was
at the Cold Springs site that the first evidence or archi­
tecture was found#
Houses in the McNary Reservoir region
are poorly preserved, and not much information can be re­
covered from their excavation*
The more recent houses
showed a surface depression that was saucer shaped*
older houses often had no surface indications at all#
of the dwellings in the region had traces of floors which
could be picked up only near the center of the houses.
floor level, when present, was found as a charcoal and
ash stain which did not extent to the edges of the house#
House pit 3, at the Cold Springs site, showed a de­
pression on the surface of the ground and was oarefully
Although there was little direct evidence of it,
there were apparently two or more occupations of the house
During the excavation, no trace of a floor was found
until a depth of three feet was reached.
At first it was
believed that the deep floor belonged to the surface de­
However, the contour of the depression plotted
before exoavation did not match the oontour of the deep
The floor center proved to be more than seven feet
southwest of the oenter of the depression.
artifacts found just above the deep floor were of a type
associated with the earliest occupation of the site; they
included two of the large basalt knives.
It follows
therefore, that the depression that plainly showed on the
surface belonged to a house which had been occupied at
a later time.
Another reason for this conclusion was that
the thick layer of disoarded mussel shells which blanketed
this portion of the site was not present in the fill of the
house pit.
This showed that the house had been oocupied
during or after the period in which the shells accumulated#
Since an average of about three feet of midden had accumu­
lated before the deposition of the shells, the shell layer
was stratigraphically later than the early artifact complex
found both in the midden and in the earliest house.
brief, the house was occupied at a time when the earliest
material culture at the site was being used.
thick layer of shells accumulated.
Later, the
The final occupation
of the house came after the shells had accumulated, and
the occupants apparently cleaned the trash out of the house
pit before living in it.
Time did not permit the full excavation of the deep
floor but a good profile was obtained.
It showed the house
to be bowl shaped, sloping up toward the surface with no
vertioal walls.
No features were observed.
This negative
evidence which was to be repeated in nearly every house
excavated in the region was actually of some significance.
It limited the type of superstructure to something that may
be visualized.
Obviously, there was no heavy structure of
dark miodem
SITE 35-UM-l
I N T E R V A L : .5 FT.
wood, for there were no post holes.
A roof entrance would
hardly have been possible without interior posts.
A flat
roof flush with the ground w'ould not have left more than two
or three feet of spaoe under it, even in the center of the
What is indicated here is light framework, perhaps
of small poles, covered with mats.
Mats are mentioned in
all of the early travelers reports, and traces of them have
been found in most of the McHary sites, including Cold Springs.
In most oases there is a slight rise or mound around the
perimeter of the house pits.
While it might be no more than
back dirt from the house, there is reason to suspect that
this earth may have been banked against the lower edges of
the matting.
If this were the case it would eliminate the
necessity of sinking posts into the ground for stability.
There is no direct evidence of a doorway in any of the
However, in three houses that were carefully con­
toured, one side, the one facing the river, was slightly
lower than the rest of the perimeter.
This may reflect
only the slope of the terrace toward the river, but it is
the only evidence indicative of an entranceway.
Reoccupation of old houses seems to have been a pattern.
It would be far easier to clean out an old house depression
than it would to dig a new one.
This could account for the
lack of material culture found in the house fill.
parently, every so often as the house began to fill with
trash, it was cleaned out to the depth desired.
Traces of another house that probably should be assigned
to the early period were found during the excavation of the
pit 3.
It was approximately 50 feet southwest of house
Badly disturbed by later occupation, the floor that
remained consisted of only a thin charcoal stain about three
feet in diameter.
No surface depression was evident and
practically no information could be gained from working
about the remains of the floor.
House pit 12, located at the eastern extremity of the
site, had been occupied at least two times.
The earlier
floor was in the form of a deep bowl (see Figure 3), and
the later floor was more saucer shaped.
Unfortunately, the
only artifacts associated with the deeper floor were ubiqui­
tous hammers and choppers, so that the house could not be
assigned to any period. The shell layer, which was so
prominent at the western end of the terrace, did not extend
that far east so it could not be involved in the stra­
tigraphy. As near^as could be determined, the earliest
indistinct floor of house pit 12 was of about the same size
and shape as the early floor in house pit 3.
The later
occupation of the pit left almost no traces; only a slight
discoloration near the center of the depression.
House pit 4 (Plate VI and Figure 2) presented another
picture of multiple occupation.
The earlier floor of the
house was saucer shaped as far as it could be traced, and
had a small fireplace in the center.
There were two strata
House pit 4, Cold Springs (35-TJM-7) before excavation
Earth oven, Cold Springs (35-IM-7)
2-5 • 111
of mussel shells just outside of what seemed to be the edge
of the dwelling, and they appear to represent two periods
of occupation of the house.
A later house utilized at least
part of the depression of the first house, but was centered
approximately seven feet to the east.
The later house was
larger (43 feet in diameter), and was stratigraphically
later than the two thin layers of mussel shells.
If it was
correct to correlate the shell layers with the occupation
of the earlier floor level, that occupation may in turn be
correlated with the deposition of the thick shell layer
which covers the central portion of the site.
The shell
layers near the house were proven to be parts of the major
shell deposit.
A.fter the later house had been abandoned, the depression
of House pit 4, was used for the construction of a large
earth oven.
Large lens-shaped concentrations of fire craoked rock
were found at three different places on the site.
In each
case the concentration was just beneath the surface and for
this reason they are believed to belong to the latter part
of the occupation of the site.
The first of the features which have been called earth
ovens was approximately four feet in diameter and about five
inches in depth.
The second (Plate VI b) was about six feet
in diameter and the third was eleven and a half feet long
and five feet wide.
Small quantities of charcoal were
scattered through the first two features and a large con­
centration of charcoal was present in the third feature.
While the first two earth ovens were unassociated with other
features at the site, the large one was found in the center
of house pit 4.
There is no question as to its relationship
to the house as is shown in the discussion of that house#
It obviously was constructed in the depression after the
house had been abandoned.
To understand the use of an earth oven it would be
best to quote the experience of Collier and his associates
in the Grand Coulee Reservoir area.
"We observed the process of cooking cama3 in
an earth oven at the house of Rosie Seymour, 84 year
old Okanogan-Lakes woman living at Kelly Hill in the
hills north of Kettle Palls. A pit four feet square
and ten inches deep was dug. The pit was covered with
timbers and rooks were piled on top of the timbers.
The timbers were fired, and when they had burned down
and the rocks had fallen into the pit, the latter
" were leveled and covered with green tule. The six
sacks of camas were placed on the tule and covered
with tule and damp grass, then with a layer of earth,
and finally with a carefully laid layer of sod. Over
the resulting mound was piled wood, which was in turn
covered with green willow branches and leaves to pre­
vent rapid oombustion. The top fire was kept going
nearly forty-eight hours and then the camas bulbs
were removed. The remaining pile of burned and
cracked stones resembled precisely the burned rook
areas described above."36
Here and there about the Cold Springs site were evi­
dences of a late occupation.
The materials associated with
it were indicative of a period post European contact, large
Collier, Hudson and Ford, 1942, p. 38
glass beads and rolled copper tubing.
The type of bead found
here is not the earliest kind that was traded into the area,
and is believed to be of the historic period. The material
culture of the Cold Springs site, excluding this late occu­
pation, is typologically earlier than that of two other sites,
35-UM-17 (Techumtas Island) and 45-BN-52.
These last two
in turn, judging from trade goods, are earlier than the
last occupation of the Cold Springs site.
What is indicated then is an abandonment of the Cold
Springs site and a later superficial reoccupation.
houses, Nos. 5, 7 and 14 were apparently cleaned out and
Besides the glass and copper several new arti­
fact types came into the picture.
Projectile point3 associ­
ated with this occupation were small and thin, averaging
between 20 and 25 mm. in length.
Three speoimens were side
notched, two were barbed and the other was leaf shaped.
Choppers were carefully made with symmetrical double bits,
and a chipped stone fetish was associated with these
This last artifact requires some discussion
(see Plate VII a, bottom row, center).
Admittedly, there is some question as to the use of
the artifact, but there is no reasonable explanation of
it as a tool form.
It was made by splitting a small discoid
pebble of quartzite that was from 3 to 7 cm. in diameter.
Two opposite edges of the spall were then chipped away until
the outline is wider at one end than at the other.
Two arcs
of the original circle are left, one larger than the other.
In the edge and at the center of the larger arc a small
notch was cut.
Usually, when preservation permits, red
ochre is found rubbed into the notch.
An Indian at Celilo Falls, Oregon, identified one of
these objects as a "good luck stone" formerly carried by
young men.
He stated that it was in effigy of a man (actu­
ally it looked more like a headless torso), and that the
broader arc represented the shoulders and the notch repre­
sented the neck.
Unfortunately, the large collection of animal
bones recovered at 35-UM-7 were not examined thoroughly by
an expert.
A preliminary examination, however, led to the
identification of a number of animals that were probably
used for food.
of all.
The bones of fish were the most numerous
Salmon, trout and sucker were identified.
the mammal bones, the most frequent were those of deer,
followed by the brush rabbit and the jackrabbit.
Bones of
elk, mountain sheep, porcupine and several kinds of birds
were also present.
Due to incomplete analysis, it is not
possible to state whether or not there were changes in the
type of game sought through the duration of the occupation.
It was noted, however, that fish bones were numerous at all
depths in the midden.
The river mussel was used for food at all times.
During One short span, apparently toward the end of the
major occupation, there was heavy reliance upon river
mussels for food.
At this time, the thick layers of shells
were discarded over a large portion of the site.
this was caused by a failure of the annual salmon run or
not it is impossible to determine.
Specialized tools that could have been employed in
gathering vegetable foods were not recovered.
Several flat
stones showed evidence of having been used for grinding
some substance.
slabs of basalt.
These were not mortars, but slightly used
Specialized hunting and fishing artifacts
were numerous, as can be seen in the section on material
Both the long slender basalt knives and the dis­
coid choppers could have been used in preparing meat and fish
for drying.
In spite of the lack of data on gathering, it
must be assumed that a balance existed between hunting,
gathering and fishing.
The latter was most likely the most
important to the economy.
Further Observations;
Rodent activity at the Cold
Springs site had been extensive.
It showed clearly in all
profiles, for many of the burrows contained the dead white
volcanic ash from the substratum which contrasted sharply
with the grey sandy midden.
One thing did a great deal to
prevent wholesale mixing of the midden} that was the thick
dense layer of mussel shells whioh interposed itself ap­
proximately halfway between the volcanic ash and the surfaoe
of the site.
Thus burrowing above the shell layer rarely
penetrated "below it#
The burrows below the shell layer
seemed to be confined to the period prior to the shell
In spite of the labyrinth of holes, most fire­
places, house floors and other features were still discernable.
Fish bones were often still articulated, and
fragments of fiber matting were in situ#
A peculiar feature that did not lend itself to cate­
gorical treatment was found at a depth of between two and
one half and three feet.
It consisted of a carefully piled
mass of seventy nine flat river worn cobbles.
The stones
were from four to seven inches in diameter and from one and
one half to three inches thick.
burning, flaking or hammering.
They showed no evidence of
Centered beneath the cairn
was a single scapula of an elk and nothing else.
There is,
of course, room for much speculation on the interpretation
of this feature.
Ho reference to such practice could be
found in the ethnographic literature.
There were one or two things evident in the settlement
pattern at the Gold Springs site.
First, the material which
seemed to be relatively early was concentrated at the western
and west central portion of the site.
The later materials
were found there as well as considerably farther to the east.
The exact limits of the earlier part of the occupation were
not determined.
The pit houses on the site, that is the
surface depressions, were divided into two groups (see Map
There were eleven houses at the western end, six
houses at the eastern end, and a space of five hundred feet
in betweeno
If there were houses in the space between the
two groups, no surface indications remained#
55-TJM-17 (Techumtas Island)
The Teohumtas Island site is located about one and one
half miles northeast and upstream from the Cold Springs site.
The island, which is also known as Hoodoo or Sweitzer*s
Island, is approximately nine miles upstream from the McNary
Prom east to west it is just under three miles in
length, and it varies between one forth and one half miles
in width.
During most of the year, the channel on the
southern side is dry, so that only during the late spring
and early summer does it really become an island#
On the ohannel side and near the downstream end of the
island there is a small exposure of basalt bedrock.
wise, the island appears to be built up from a large gravel
bar with a topsoll of river sand and eolean deposits#
Actually, the whole island is a flat terrace#
Recent culti­
vation has taken place near the central part of the island,
"but the rest is covered with sage, Artemesia trldentata.
grasses and weeds0
There has been no little wind erosion,
and duning is well developed in several places#
Along the
main river ohannel the bank slopes steeply to the water»s
The shingled beach is composed of river gravel, well
sorted by the current.
l umb/a
s c a l t <n f e f f t
SITE 35-UM-17
© f 7
The site itself is located about 500 yards from the
downstream end of the island, and is on the side facing the
main channel.
Nineteen well preserved house pits and several
discontinuous midden areas are scattered along the top of the
bank for a distance of 850 feet.
The house pits are divided
into two groups, with 14 pits situated at the downstream end
of the site and five pits at the upstream end*(see Map IV).
In the intervening space of 275 feet there was a small amount
of midden debris but no trace of architecture.
A close
examination of the area led to the conclusion that it was
all one site.
Some trash had been scattered about on the surface, but
there are indications that most of it probably went over the
bank toward the river.
This pattern was recognized from the
results of a great many small test pits which were excavated
in a grid pattern over the length and breadth of the site.
They revealed that the maximum depth of trash on the terraoe
was slightly more than one foot.
The deepest penetrations
into the terrace were the houses themselves, which ranged
from two and a half to three feet.
The shingle beach ex­
tended up to within a few feet, (three to six), of the level
of the terrace, and clearly showed that any over-the-bank
dump would be swept away by the spring floods.
House pit 14 wa3 oval shaped, 40 feet
long and 30 feet wide, according to surface measurements
(see Plate VII b).
Like the majority of houses in the
SCALE IN f e e t
McNary area, the long axis was parallel to the river as well
as to the prevailing wind.
The first floor encountered was found at a depth of
less than one foot below the surface (see Figure 4). It
was a clearly defined saucer shaped area of oharooal and
ashes, with a fire depression in the oenter.
however, it did not extend to the extremities of the house,
but was confined to an area approximately 15 feet in diameter#
As near as could be determined, the floor followed the sur­
face contours of the house pit#
Two and one half feet below the surface and roughly
at the center of the house was the remnant of an earlier
It was from six to seven feet in diameter and less
dearly defined.
As usual, it was recognized by a thin
dark stain of charcoal.
Below the earlier floor the soil
was sterile.
House pit 15 was nearly circular and about 14 feet in
The occupation level, or floor o£ the house, was
found below the surface, and was in nearly every respect
similar to house pit 14.
Below the floor of the house only
sterile soil was found, so that it is believed that there
was a single period of occupancy.
Tests in houses 1, 9 and 18 failed to discover a
living level, and excavation was discontinued.
These three
structures failed to produce any diagnostic artifacts.
an occasional flake or spall could be found, although the
Artifacts from Techumtas Island (35-UM-7). Top row
Projectile points; Middle row: Bone awls; Bottom
row; Scrapers except ?fl03, a chipped stone fetish.
House pit 14, Techumtas Island (35-UM-17)
depressions clearly were the remains of dwellings#
The houses of the Techumtaa Island site were slightly
smaller than those at Cold Springs; the latter averaged
30 feet by 28 feet, and the Teohumtas Island houses averaged
28 feet by 26 feet*
At both sites the longer axis of the
house normally was parallel to the river.
Furthermore, at
both sites there was a dual division of the village into
two parts with the smaller group at the upstream end of the
Naturally, it was disappointing not to recover in­
formation on the type or types of superstructures employed
at the site, but such data were not to be found.
It is
almost certain that the superstructures were of light
flimsy materials such as poles and mats.
The midden trash at the Teohumtas Island site was thin
and unevenly distributed.
It would appear that since the
occupation was a relatively short one, trash had not ac­
cumulated to any extent on the terrace.
The over-the-bank
dump, being subject to stream erosion, was not permitted
to accumulate.
However, the heavier artifacts were not
readily swept away, and some collections were made on the
surface of the bank immediately below the house pits.
These will be treated in the disoussion of material culture.
Material Culture:
It was hoped that significant dif­
ferences could be found between the artifact collections
associated with the two floor levels in house pit 14.
fortunately, they were not apparent.
The lower floor had
12 artifaots associated with it, and six of these were
unaltered spalls that had probably been used for cutting
and scraping.
Also found at this level were a small cobble
chopper, two crude flake scrapers and a notohed net sinker.
On the upper or most recent floor, 22 artifaots were
Six of them were unaltered spall flakes, and
eight others were crude flake scrapers.
Only two of the
scrapers had been prepared for use; the others were primary
Four projectile points came from the upper floor
(see Plate VII a).
Three of them were side notched with
concave bases and averaged 20 mm. in length.
The fourth,
simply triangular with a straight base, was 31 mm. long.
There were also three complete bone awls on or near the
floor; each awl a different typef (see Plate Vila).
made from a long bone of deer.
One was
It tapered slightly, and had
a conical point much like that of a sharpened pencil.
Another was made of a scapula spine, probably of deer, and
was similarly pointed.
The third, much smaller, was made
from a deer fibula, and tapered smoothly to a point.
first awl described above was finished along the shaft and
had a small hole about 3 mm.
the 3haft near the butt.
in diameter drilled through
The others had been only slightly
smoothed along the shaft.
The last item from house pit 14 was a typical chipped
stone fetish similar in size and shape to the one described
for the late Cold Springs occupation (see Plate VII a).
House pit 15 had six artifacts associated with the
single floor.
They included two rough cobble choppers and
a cobble hammer, all of basalt.
A tip of a bone tool and a
thumbnail scraper were also associated with the house.
small glass trade bead was recovered near the floor of house
pit 15.
It was of the tubular type which is believed to be
early in the region.
Although there is always the possi­
bility that it could be intrusive (via a gopher hole), it
suggests a late pre-contact date for the house.
The other
artifaots at the site, small points and fetishes, would tend
to confirm this dating.
A number of artifacts were found in the midden trash
adjacent to the houses.
One test section just south of
house pit 9 uncovered what may have been a cache of fishing
However, the 18 notched net weights and one
grooved weight were in no particular order, and were associ-*
ated with a quantity of discarded mussel shells and some
fire-cracked roclcs.
Also associated with the group were two
cobble hammers, a oobble chopper and a spade shaped stone
whiob may have been a boring tool.
is unique in the region.
This flat basalt tool
All of these artifacts were closely
associated on a single level less than one foot below the
Tests on the bank between the houses and the river pro­
duced very little.
Artifacts recovered included a small
triangular projectile point, a thumbnail scraper, a stone
fetish and a few hammers and choppers#
Although there was
some stratification in the trash, the cultural materials
recovered were far too few for any conclusions to be drawn#
Two years after the excavations were made on Teohumtas
Island there was new erosion of the bank immediately below
the lower group of houses.
The river swept away the sand,
but dropped the larger stones almost in place.
The artifacts
collected from the surface there almost certainly belonged
to the over-the-bank dump from the houses, and should be
considered part of the complex.
This group of artifacts
included nearly 100 notched net weights and six ovoid grooved
net weights.
There were about 40 discoid choppers and 60
spall flakes as well as 12 or more cobble hamners and chop»«
Three stone fetishes were also found on the eroded
All of the above were artifact types which matched those
found in the houses.
There were a few artifacts in the sur­
face collection whioh did not duplicate those from in or
near the house pits.
They included a miniture stone mallet
three and one half inches (9 cm.) in height, and part of
a full sized mallet.
The former, whioh may have been a toy,
is similar to, but smaller than those illustrated in Plate
XII b»
Two small projectile points found there were corner
notched with straight stems*
This is the first appearance of the stone mallet in
the local sequence.
Mallets, or mauls as they are sometimes
called, are among the most carefully made artifacts in the
Ground and polished from basalt, porphyry and
diorite, these tools are about the size and shape of a milk
bottle with the handle often suggesting a phallic symbol*
The earliest mention of mallets in the literature is from
the Lewis and Clark journalsWhen these explorers were
on their way to the coast in 1805, Clark visited a village
near the juncture of the Snake and Columbia Rivers.
reported seeing a stone mallet used in conjunction with an
antler wedge for splitting timber.
His description makes
it clear that the artifact was of the type being discussed
Pine ground mallets have been reported from The
Dalles, Oregon,®® from the upper Columbia,®® and from the
Yakima River*"0
The small, rather delicate, corner notched projectile
point becomes the typical one for the historical period.
How much earlier than that it oocurs is difficult to say,
but corner notched points appear in quantity in sites that
have no European trade goods.
Most of the tools that could be assigned to
a particular economic pursuit were obviously designed for
The large number of net weights in and around the
37• Thwaites, 1904
38. Strong et
39. Collier et al, 1942
40. Smith, 13T0
houses attested to a considerable reliance on fishing as
did the numbers of fish bones.
Animal bones included those
of deer, rabbit, various birds, antelope (Antllocapra
americana), bison but no horse. A Projectile points were
present although not really plentiful, but there was no
direct evidence of vegetable foods#
Shellfish were apparent­
ly consumed in quantity, although not to the extent that was
evident at Gold Springs (35-TJM-7).
This large village site was exoavated by Douglas Osborne
in 1948, and was reported in full in his doctoral disser42
It was not a large or a lengthy excavation, and
not many diagnostic artifacts were recovered.
Because of
these things only a brief summary of the results will be
presented here.
One of the largest villages anywhere in the middle
Columbia region, 45-BN-53 is situated on a long flat terraoe
about five and one half miles upstream (east) of the McNary
Dam^ (see Map II)•
The terrace is on a s&mi«detached island
on the north (Washington) side of the Columbia River and is
nearly opposite the Hat Creek site. The local physiography
is almost identical to that of Teohumtas Island, sandy loess
soil, steep shingle beach and some shifting sand dunes*
Osborne, 1953, p# 261
Osborne, 1951
SITE 45-BN- 53
The village itself is approximately 300
yards long, parallel to the river and 35 yards wide.
consists of 182 houses or house depressions.
Since only a
few were excavated, almost nothing can be said about the
relative ages of the houses or the duration of the occupation.
The evidence tends to indicate that it did not extend over
a long period of time but that evidence is not conclusive.
The house pits of various sizes are scattered along
the terrace in no particular pattern.
No streets or signi­
ficant groupings of houses could be deteoted, although some
information might be forthcoming if the ages of most of the
houses were known.
Six of the house pits at 45-BN-53 were excavated or
Houses number 11, 49, 70 and 168 were completely
excavated, while numbers 52 and 169 were only trenched.
The results of the excavations were nearly identioal to
those at Cold Springs and Techumtas Island.
levels were found, but no post holes or traces of superstructure could be detected.
East-west diameters were
generally greater than north-south diameters, making the
long axis in line with the prevailing wind.
Median diame­
ters were north-south, 16 feet, and east-west, 19 feet.
This is smaller than the average for the houses at Teohumtas
Island and Cold Springs.
Generally, there were several occupational levels in
each house and apparently the houses were not cleaned out
from one occupation to the next*
Fireplaces were usually
in the center of the saucer shaped floors.
There were in­
formal fireplaces, a few rocks and a concentration of oharcoal and ash#
Entrances were not detected, and the only
trace of building material was in the form of mat fragments
found on the floor.
Occasionally pairs of houses were side
by side so as to suggest contemporaneity, but the poor pre­
servation of floors and structural details prevents their
Material Cultures
Moat of the work at 45-BN-53 was
concerned with the excavation of house pits.
About 12 work­
ing days with a crew of twelve men was spent at the site,
so the list of materials recovered is small.
Of the 209
artifacts recovered, 129 were of the large coarse type such
as hammers, choppers or net weights.
The others were so
similar to the materials from Techumtas Island (35-TJM-17)
that there is little reason to go into detailed description*
They included projectile points, awls, scrapers and drills.
Although no trade goods of European origin were enoountered
in the houses, a few glass trade beads were found in the
midden tests.
This is, of oourse, no proof that the site
should be dated by these finds.
Later utilization of the
terrace could have been responsible for the occurrence of
the trade beads.
However, the material culture, especially
the small projectile points, strongly suggests that the site
was oocupied during the late prehistoric period.
Trade beads
might have found their way into the area as early as 1740,
or 1750*
Osborne^3 would date the site as early eighteenth
century at the latest.
Among the animal bones recovered at 45-BN-53
were those
of deer, elk, rabbit, various birds, idog, fish
and bison.
Horse and antelope bones were absent. Pish
bones were not as numerous as one might have expected, and
suckers were more common than salmon*
The horse is absent
in all pre-contact sites in this region; it was not found
at Techumtas Island (35-UM-17), a site believed to be con*
temporaneous with 45-BN-53.
The economy, therefore, was
apparently identical to that described for site 35-UM-17#
Hunting, fishing and gathering seem to have had nearly equal
45-WW-6 (Wallula)
The Wallula site (45~WW~6) is located at the confluence
of the Walla Walla and Columbia Rivers, in the southeast
corner of the state of
The village was built
on a slight elevation about 50 yards from the east bank of
the Columbia, and the same distance from the north bank of
the Walla Walla.
About a mile downstream from the Walla Walla River the
Osborne, 1953, p. 263
Ibid., p. 261
SITE 45-WW- 6
Columbia flows through a considerable gorge, the result of
cutting through the Horse Heaven Hills.
Local tradition has
it that the hills once dammed a deep lake within what is now
a large open valley*
Thousands of years ago, the water
burst the natural dam, out the gorge and drained the lake,
all according to local "authorities"#
How much of this is
based on scientific examination of the geological facts it
was not possible to ascertain*
At any rate, the broad flat
valley is there and supports the small cities of Pasco,
Kennewick and Hanford*
The site itself is approximately 200 feet long and
perhaps 60 to 80 feet wide.
The subsurface material is
coarse river gravel, with a two or three foot layer of
yellowish sand on top of the gravel*
The midden is about
three feet in depth near the center of the site, and it
rests directly on the sterile sand*
Toward the edges of
the site the ground slopes away gradually, and the midden
thins out proportionally.
Pothunting has created a great deal of damage at the
Wallula site.
At least one third of the surface has been
potted, and no fewer than four burials were robbed*
expression is appropriate in this case, for only the crania
and long bones were taken, along with most of the artifacts*
Soreening in the irregular disturbed areas recovered vast
quantities of beads and small human bones.
the excavations carried on during the two weeks that River
Basin Surveys spent on the site did not encounter a single
undisturbed burial.
In spite of the beliefs of the local residents, the
Wallula site was not primarily a burial ground, but a fish­
ing village of the late prehistoric and early historic
In some plaoes the soil was black, almost greasy,
with charcoal, and salmon vertebra were present in vast
Most of the artifacts were utilitarian, and the
500 square feet of surface excavated encountered only midden
Architecture; Site 45-WW-6 is subject to spring floods.
In May, 1950, the present author was forced to abandon work
on the site and move out by a normal spring flood. Surface
erosion was evident over all of the site, and it is believed
that traces of houses had been obliterated.
If the Wallula
site had been a summer fishing village, the surface mat
house (to be desoribed later) would have been used.
slight depression left by a ^urnirer mat house would be erased
quiokly by flood erosion. Large areas of charcoal were
encountered all through the midden, but none of them con­
formed to a house floor.
Therefore, no definite information
was obtained on architecture.
Material Culture:
Projectile points were numerous at
45-WW-6 in spite of other indications that it was primarily
a fishing village.
Out of 70 specimens recovered, 48 were
sufficiently intaot to permit classification (see Plate
VIII' a ))•
Mo3t of the points were small, between 18mm. and
27 mm. in length, and were barbed.
Nine specimens were of
a simple leaf shape with rounded bases.
They were similar
in outline to the projectile points from the Hat Creek site,
but were generally smaller.
They lacked the serrated edges
and were not plano-convex in cross-section.
Table II gives
the distribution according to types set up by Strong, Sohenck
and steward.
Table II
The small stemmed and barbed projectile points are
typical of the late prehistoric and early historic periods
all over the Plateau.
Those at the Wallula site were made
of petrified wood (17 specimens), basalt (12 specimens),
agate (9 specimens) and jasper(5 speoimens).
Only two
projectile points were made of obsidianThere is nearly always room for argument over whether
certain chipped tools are to be called projeotile points or
45. Strong et al, 1930
At the present time there is no way of being certain
about some specimens.
With the materials from the Wallula
site there seemed to be a preference for small barbed points.
Others were leaf shaped and small, and some were intermediate
in size.
There were still others, five in number, that were
considerably larger and heavier, and in this oase have been
called knives.
Ten more blades were of shapes that have
not been associated with projectile points, but have been
found (at 45-BN-3) with knife handles still intact.
All ten
of these knives are variations on a lozenge shape with
straight, concave or convex bases.
The materials included
basalt (7 specimens), petrified wood (6 specimens), agate
(1 specimen) and jasper (1 specimen).
Blades similar to
the lozenge shape knives were found in north central Oregon,
by L. S. Cressman.
He said in referring to hi3 specimens,
believe these latter to be end scrapers on
the basis of experience in classifying numerous simi­
lar specimens in early collections for The Dalles
region of the Columbia River. Microscopic examination
showed pitch on the sides of the contracted portion,
obviously evidence of hafting."46
Cressman's explanation may be correct, but the specimens
from 45-BN-3 were hafted with the pointed end exposed for
There is no reason why a tool of this sort might not
be used either way.
Plake scrapers found at the Wallula site were not
significantly different from those described earlier.
Cressman, 1950, p. 378
*• *
Projeotile points, Wallula (45-WW-6)
• ."
were of essentially the same materials as the projectile
points, and there was no attempt at shaping#
A few of the
flakes may have been deliberately removed, but the majority
are the result of having been used for soraping.
The 22
specimens recovered at the site are primary flakes, with an
average diameter of about 3 cm#
Spall scrapers, which were abundant at the Techumtas
Island site (35-UM-17), were represented at the Wallula
site by only four specimens.
These tools are almost always
made of quartzite, perhaps because the quartzite cobbles
in the Columbia River tend to have such well defined cleavage
The spalls are always thin, having one surface which
is weathered and one which is the cleavage plane.
usually show some crumbled battering.
The edges
It is likely that the
spall was a sort of general utility artifact, easy to make,
handy to use and expandable.
The same was probably true of
the other chopping and hantnering tools.
Stone weights for seines and nets were abundant in the
midden, on the surface, and on the gravel beach of the
Columbia River (see Plate XI b)«
The description of the
two-notched stone weight given for those from the Cold Springs
site hold perfectly well for 31 specimens excavated at the
Wallula site*
A variation on the type makes its appearance
there with 18 specimens that have four notches instead of
two. These quadrilaterally notched weights are about 25$
longer and wider than the laterally notched variety, but
Bone tools from Wallula site (45-WW-6). The dark object,
3rd from right is a projectile point, the others are awls
or needles.
Antler wedges from Wallula site (45-WW-6)
otherwise do not differ significantly*
No evidence could be
found to show that either variety was stratigraphically old­
er than the other, for both were randomly distributed through
the midden*
There does seem to be a proportional inorease
in the four notched variety as one prooeeds upstream from
the McNary Dam.
Girdled net weights were made by peoking a shallow
groove around the smallest diameter of a large ovate river
cobble (see Plate XI b, center).
Granite prophery and
quartzite were preferred or at least were readily available
in the river gravel.
The three grooved weights from the
midden averaged 16 cm. in length and 13 cm. in width.
Coarse stone tools such as hammers and choppers were
particularly abundant all over the region, and certainly
no less so at Wallula (see Plate X b).
Of the 78 speci­
mens that had been used for hammering, 34 had been intended
for use as hammers only.
The remaining 44 hammers had
originally been chopping tools which, after becoming dull,
were used as hammers.
As has been explained earlier, these
basalt river cobbles were roughly flaked to a single edge
for chopping tools, or used without alteration for hammers.
Twenty six cobble choppers saw no subsequent use as hammers,
and were discarded with no further alteration.
All of the
cobble tools were apparently fashioned for use without hafting.
No grooves or other hafting devloes were employed.
The discoid chopping tools, flat river worn pebbles
Discoid choppers, Wallula (45-WW-6). Top rows 3/4 bit;
middle rows Double bit; Bottom rows Single bit
Cobble choppers and hammers, Wallula (45-WW-6^. Top
rows Choppers; Bottom rows Hammers
with chipped edges, were represented by 38 specimens (see
Plate X a)*
A trial division of these tools into four sub­
types seem to be justified, for there is no gradation or
overlapping in the division.
The simplest, and incidently
the smallest, sub-type is chipped on one edge, and the bit
takes up approximately one fourth of the circumference#
Another, slightly larger, has bits of the same size on
opposite sides*
A third, still larger, is chipped three
qruarters of the way around the circumference, and the last
is chipped all of the way around.
The gross size of each
tool is thus directly proportional to the amount of the bit
or cutting edge.
It is perhaps significant that discoid
choppers are often associated with notched net weights,
suggesting some use in the fishing complex.
Two problematical objects that were recovered in the
midden could have been used as adzes or hoes (see Plate XI
They were flaked from basalt spalls of exfoliation and
utilized the plane surface of exfoliation (or perhaps frost
wedging) as one side of the tool.
By rough peroussion
flaking from the flat side, a plano-convex tool with an
outline like that of an axe blade was formed.
The broad
end of the tool was retouched and quite sharp, but the sides
of the tool were blunted.
If the tool were hafted by lash­
ing a T-shaped handle to the flat side, the blunted edges
would lessen the chance of the lashings being out during
use, and there are slight polished areas on the surface
Stone hoes from Wallula (45-WW-6)
W W!
Net weights from Wallula (45-WW-6)
where the lashings might be expected to cross#
One of the
tools is 22.8 cm. long and 9 cm. wide, and the other is
21 cm# long and 8.6 cm. wide.
Such-a tool would be most
useful in hollowing out oanoes, digging houses and graves
and perhaps other things.
Nearly identical tools have been
found in the region during surfaoe surveys, and one was
found associated with the late occupation of the Cold Springs
Two smaller basalt artifacts were probably used as
One had been prepared by striking two long flakes
from a more or less cylindrical pebble, and the other was
a basalt fragment whose planes of frost wedging created a
natural wedge.
Both had been battered at the blunt end.
The chipped specimen is 5.1 om. long and 2.2 cm. wide while
the other is 11.8 cm. long and 5 cm. wide.
Occasionally some of the large river cobbles were used
as lap—stones or anvils.
The two stones used in this manner
at the Wallula pite were river worn cobbles of granite
porphyry, each about 24 om. in diameter and 10 cm. thick.
The battered surfaoe was in the center of one of the nearly
flat sides and perhaps 6 cm. in diameter.
Had the battering
soars not been in the form of sharp pits, these objects
might have been called mortars.
One of the anvils had been
flaked to a rough outting edge over a third of its diameter
in the manner of the cobble ohoppers.
The heavy two-handed
chopper may have been made before or after its use as an
anvil; it was not possible to determine which toolc precedence.
There were two types of chipped stone drill bits found
at the Wallula site, but.only one representative of eaoh
One of these is long (3.3 cm. without the tip) and
slender (9 mm* in diameter)*
almost circular.
In oross seotion it is oval,
The other type is shaped like an auto­
mobile key, having a broad rounded base with a sharp bit.
The basal diameter is 2.1 cm., but the tip of the point is
Both types were probably hafted, but it would be
entirely possible to use the
type unhafted.
No data
are available on whether a bow was used with a hafted drill
bit or the shaft was twirled between the hands.
Objects made of ground and polished stone are rarely
found in midden deposits in the McNary region.
Those re­
covered there are usually fragmentary. We were fortunate in
recovering six such artifacts at the Wallula site, even
though five of the six were broken and incomplete.
fragments of basalt pesttes that were found consisted of the
blunt grinding ends only.
They had been peoked to shape but
worked only to a fair polish.
but of poor workmanship.
A third specimen was complete
It has been pecked to shape but
the peoking soars had not been polished away.
The length
was 18.3 cm., the diameter was 6.4 om., and it tapered
Another pestle, not classed as ground stone, was
a long cylindrical river stone of basalt that had been used
with no alteration.
Two fragments of stone bowls were In the midden deposit#
Though not part of the same bowl, the two pieces are of
nearly the same size and shape, each being about one quarter
of the original speoimen#
One would have had an outside
diameter of from 14 to 16 cm#, and the other would have
been between 13 and 15 cm. in diameter#
Overall height
of each would have been approximately 7 om«
Both bowls had
been hollowed out of river cobbles, one of quartzitio sand­
stone and the other of granite#
The interiors are ex­
ceptionally well smoothed, and the bases were flattened for
One of the bowl fragments was in close associ­
ation with several cobble hammers and one of the pestle
Only one fragment of a stone pipe was found at Wallula,
and it was a part of the bowl of a two-pieoe type#
type oonsists of a parabolic bowl with a female flange
fitting for the insertion of the stem#
The stem is in line
with the pipe, so that it is merely a variation of the
regular tubular type#
The specimen reoovered in the midden
was of dark grey talc sohist, well polished, but too fragmenary for measurement#
Several pipes of this kind were found
on Sheep Island by Garth in 1950#
Altogether, 34 artifacts of bone and antler were re­
covered at 45-WW-6#
As with nearly every class of artifact
Garth, 1952
that is represented in the area, many of these tools are
generalized, and one can only speculate as to their use or
Nevertheless, they have been divided tentatively
into six categories:
wedges, bone awls, flaking tools, bone
projectile points, needles and dice.
The wedges, eight in number, are all of deer or elk
antler (see Plate IX b).
The antler, in most cases, was
split longitudinally, ground and polished to a rounded bit
at one end, with a blunt striking surface at the other.
Wedges of this sort, as previously mentioned on page 68,
were observed in use and described by Lewis and Clark in
October, 1805, a few miles from 45-WW-6.
They were used in
conjunction with stone mauls for splitting timber#
All but
one of the Wallula wedges are flat in cross-seotion and
have oval bits.
The exception is round in cross-section
with a pointed rather than a flat bit.
Antler, expeoially
the cancellous tissue, is extremely susceptible to decay,
and for this reason, many of the specimens are not complete.
The average wedge appears to have been between 8 cm. and
12 cm. long and between 2 cm. and 4.5 cm. at the maximum
A considerable vatiation oan be noted in the quality
of workmanship, in that some of the wedges are well made
and polished, while others are exceedingly rough.
Within the class of bone awls there is no end of
variation (see Plate IX a). Some are almost certainly awlsj
others may have had another use.
Some are carefully shaped
and polished, while others are only abraded at the tip.
There seems to be little point in going into the individual
details of all the speoimens since no two are really identi­
It should be pointed out that besides the variation in
shape and size a number of materials were used.
made from various bones of deer:
carpal or metatarsal.
Awls were
scapula, fibula and meta­
Deer ribs and one of the long bones
of a large bird were also used.
Thirteen specimens of
awls were recovered, and less than half of them could be
considered well made.
Flaking tools were presumably used in flaking projectile
points, knives, scrapers, and drills.
Two of them were
tines of deer antler showing wear at the tip; the third
flaker was fragmentary and of bone.
Only the tip remained
but it approximated the antler tines in shape and wear.
The three bone artifacts that had been classified as
bone projectile points were all approximately the same size:
55 mm. to 58 mm. long, and 8 mm. to 10 mm. wide.
One of the
group was but a slightly abraded bone sliver whioh was
keenly pointed at one end and beveled at the other.
other two were more oarefully fashioned, and were quite
symmetrical (see Plate IX a). In cross-section they were
rectangular with only a slight beveling of the edges.
points had been ground all over with a rough abrasive suoh
as scoria or sandstone but not polished.
It would be
possible to mistake them for flaking tools except that the
tips show no greater degree of abrasion or wear than do the
While they may be spear tines, there is no apparent
way for them to be hafted.
Bone points or the trident fish
spear usually show beveling near the base in order to facili­
tate hafting.
Sinoe these points were apparently constructed
to be hafted singly, they have been classified as projeotile
In the use of the term needle to describe the four long
slender artifacts of bone, it is not meant to imply that they
were necessarily employed in sewing.
The two specimens that
were complete had no eyes, nor was there any direct evidence
that eyed needles were used at the site.
These tools were
simply too long, slender, and fragile to have been used as
ordinary awls (see Plate IX a). It is suggested that they
may have been employed in constructing twilled or coiled
basketry, twilled mats and in making small perforations in
Sinoe the average width is but 5 mn. and the average
thickness only 3 mm#, rough usage would result in breakage.
The complete specimens probably averaged over 8 cm. in
One incomplete specimen, oarved and polished from deer
or elk antler, is obviously a gambling device. Its outline
is a symmetrical long slender ellipse, and it is also el­
liptical in cross-section.
One surface is smooth and plain
while the other is covered with an estimated 18 small
shallow pits that were drilled into the surface.
It is
estimated that the die had an original length of 8.5 cm. and
a width of 2 cm.
The only ornaments of non-European manufacture that
could be found were two small shell beads#
One of these
was a short section of Dentaliura from the Pacific Ocean;
the other was a small disc from river mussel shell and
drilled slightly off center.
Because of their small size,
both of these ornaments could have been lost during the
They were not associated with the late burials
in the site.
The European trade goods were restricted to a depth
of only one foot below the surface#
The other two and one
half to three feet apparently dated from pre-contact times.
Of the four glass beads found in the midden, three were of
the cylindrical type that is believed to be the earliest in
the region.
The other was shaped like a doughnut and was
identical to those beads found in the burials and those
recovered in the ruins of Ft. Walla WallaThere were
two bits of sheet copper, one an irregular flat fragment,
and the other a rolled tubular bead.
Another copper object
is believed to be a stud used for joining leather harness
of some sort.
The last item of European origin was a broken
fragment of a clay pipe stem.
The small assortment of trade goods is nevertheless
Garth, 1952
significant, since it is highly indicative of a time relationship with site 45-BN-3 on Berrian's Island#
aboriginal artifacts are also very similar.
Burial Customss
The burials at the Wallula site consti­
tute a minor problem.
Their presence, before being robbed,
is clear because of the numbers of small human bones and
ornaments left behind.
About all there is left to speculate
on is the large collection of glass and porcelain beads
that were salvaged.
The group of beads differs considerably
from those found in the graves at 45-BN-3.
The latter were
nearly all blue and white with the blue beads constituting
64# of the total.
Most of them were of cylindrical shape,
although there was considerable variation.
Site 45-BN-3
and its trade material is believed to date from the middle
and late 1700»s and possibly in the early 1800»s.
dating appears to be upheld by the Wallula site where the
earliest trade goods are identical to those from Berrian*s
The beads from the burials at Wallula are nearly all
of the doughnut shape*
About 50% are white and the rest
nearly equally divided between black, blue, green, yellow,
red and pink.
The best chanoe of a date comes from Ft.
Walla Walla which was established in 1818.
The midden at the Wallula site contained
Osborne, 1952
more animal and fish bones than any of the sites described
so far.
Most of the bones were salmon vertebra, but bones
of deer and elk were numerous#
Sinoe no complete analysis
of the animal bones has been made, nothing more can be said
of that phase of the economy.
The presence of pestles and
stone bowls suggests that vegetable foods such as seeds
and berries may have been prepared by grinding.
is known to have been important in historic times, and was
probably no less so at Wallula.
Historical Data; The Wallula site is the first of the
series under discussion that can be related to historical
Therefore an attempt will be made to date the site
by use of early journals.
In October, 1805, Lewis and Clark left the mouth of
the Snake River, and proceeded down the Columbia by boat.
They mentioned Indian camps on the islands and on the main­
land, but at the mouth of the Walla Walla River, they
noticed only a "small rivlet'1.50
Had lodges been there at
the time, Lewis and Clark could have hardly missed seeing
On April 29, 1806, Lewis and Clark landed at the con—
fluence of the Walla Walla and the Columbia, and proceeded
about a mile up the north bank of the former.
They wrote,
"There are twelve other lodges on this river
a little distance below our camp."51
Thwaites, 1904
Ibid*, p. 337
This may have been an occupation of the Wallula site for in
the first mile upstream from the Columbia River, only the
Wallula site is a suitable camping spot.
Prom these data
we can imply that the site was not oocupied in Ootober, 1805,
but that it may have been used the following April#
Ootober, 1805, when Lewis and Clark first reached the Co­
lumbia River, they remarked that the Indians wore quantities
of bright blue and white beads, copper and brass.
these data, we may speculate that the Wallula site had
ceased to be a permanant village by the time Lewis and Clark
came by.
If the Wallula site was a seasonal camp, the vast
quantity of fish bones and fishing equipment in the midden
would indicate that it was a fishing camp and should have
been occupied during the summer and fall.
Furthermore, the
shortage of European trade goods in the midden seems to
prove that it was not a permanant village in 1806 and there­
In 1812, Robert Stuart mentioned the Walla Walla River,
this stream and its neighborhood live the
nation from whom it derives its name."52
Stuart, however, did not specifically mention the mouth of
the Walla Walla.
Port Nez Perce'(later oalled Port Walla Walla) was
built in 1818 on the east bank of the Columbia River about
Rollins, 1935, p. 62
500 yards north of the Wallula site#
It was a trading post
surrounded by a stockade and protected by a company of armed
Some years after he was in command of Port Nez Perce',
Alexander Ross wrote an exxjellent account of life at the Post.
He tells that by 1818, the Indians had guns and horses, iron
knives and all sorts of trade goods.
Most of these had been
acquired since 1805, for Lewis had reported the Indians to
be terrified at the sound of a gunshot.
Ross mentioned that
the nearby territory belonged to the Cayuse and the Walla
Walla tribes, but he did not mention a permanent village
at the mouth of the Walla Walla River.
By 1818, the Indians were probably using part of the
Wallula site as a burial ground.
The beads found in the
graves are of bright colored procelain, and are identical
to beads found in the excavation of Port Walla Walla by the
National Park Service.53
Following the evidence cited
earlier from Lewis and Clark, it would seem that the bodies
were interred some 15 years after the site was no longer
being occupied#
45-BN-3 (Berrlan1s Island)
Site 45-BN«»3, on Berrian's Island, is about five miles
east of MoNary Dam and opposite a point equidistant between
35-T3M-3 and 35-TJM-5o
Berrian's Island is separated from
Caywood, 1951, personal communication
the north "bank of the river by a narrow channel, and the
site is located at the down-stream end of the island faoing
the main channel on the river.
The site was used for both
habitation and burial, with the latter being the most im­
This is judged on the basis of few signs of habi­
tations and many burials.
The owner of the property on whioh the
site was located reported that there had been three or four
house pits on a slight ridge just north of the burial area.
However, during the spring flood of 1948, these pits were
washed away or covered up.
Test pits in that vicinity failed
to uncover any sign of houses.
Occupational debris was found
on the western side of the burial area, but no houses were
Material Culture:
The archaeology of the site has been
reported in detail by Douglas Osborne, who was in oharge of
the excavation.
The present discussion, therefore, will be
limited to such a description of the material oulture as
reported by Osborne and by use of field notes.54
Some trenching was done in the thin midden area east
of the main burial area, but little was found beyond a few
glass and copper beads.
The bulk of the artifacts came from
the graves themselves, which were 52 in number.
A comparison of the artifacts from the graves at
45-BN-3 with those of other sites is difficult because of
54. Oaboren, 1951
»»•* '
Chipped stone from Berrianfs Island (45-BN-3). Top
row: Drills; 2nd rows Points; Remainders Knives.
b. Stone mauls from Berrian's Island (45-BN-3)
the difference in quality of the artifacts lost or disoarded,
and those used as burial furniture.
The fine carved stone
and bone as well as most of the ornaments usually found their
way into graves.
Crude stone tools such as hammers, choppers
and net weights were seldom found in the graves, but were
very common in the midden trash.
There was a little overlap
in ohipped stone artifacts, and occasionally broken fragments
of the better made artifacts were recovered in trash.
The projectile points from the graves were mainly of
two general types, side notched and barbed (see Plate XII a).
These points were very similar to those found in the late
occupation at Cold Springs, at Techumtas Island and the
Wallula site.
They were small, thin and well made.
type of blade, considerably larger, may have served as a
projectile point or as a knife.
This type graded from a
simple triangular outline to that of a lozenge with a flat
Whereas the small points, the notohed and barbed type,
varied around 2.5 cm. in length, the triangular blades varied
around 4.0 cm. in length.
Several large knives were found (see Plate XII a).
Five or six of them were between five and seven inohes long,
(12.5 to 17.5 cm.) and were in the form of a long slender
The material (tabular petrified wood) and the
workmanship were practically identical to that of the tri­
angular blades described above.
A few blades, about half
a dozen, were leaf shaped with flat or rounded bases.
of the larger blades were found with remnants of wooden
handles still in place.
Nearly all of the larger blades
were made of petrified wood, while jasper, obsidian and
some petrified wood were used in making the small projectile
Three types of drill bits were used*
chipped stone, was pointed at both ends.
A small one, of
Another type,
long and slender, had a slight expansion at the base,
third was of the
type whioh was described for the
Wallula site.
Chipped stone scrapers were not found in the graves.
Those from the midden were of the usual primary flake type,
showing more the results of having been used than having
been deliberately prepared for use.
Nearly all of the coarse stone tools came from the
midden area or were being washed out of the river bank on
the edge of the site.
The list included hammers, choppers,
and spall flakes as well as two-notched and four-notched
net weights.
The choppers were of the cobble as well as
the discoid types, and four out of every five net weights
were of the two-notoh type.
Net weights littered the
gravel along the water line the whole length of Berrian*s
Island as well as the island adjacent to it on which
45-BN-53 is situated.
Objects of ground and polished stone were frequently
placed in the graves.
The large stone mallets which have
Bone and horn implements, Berrian's Island (45-BN-3)
M f l l M
l l - l i l l
t 3 4 ft '
i »«.•»•••
I j
" :...10I
Carved and polished stone, Berrian's Island (45-BN-3)
been described earlier were among these (see Plate XII b).
Pour mallets were recovered, as were five long tapered
pestles of diorlte.
The pestles were from seven inches
(18 om.) to 15 inohes (38 cm.) long.
Since no stone mortars
were found, it was assumed that mortars may have been made
of wood.
The pestles from the graves were better made and
more finely polished than those found at the Wallula site.
Five stone pipes of soapstone (steatite) had been
placed with the burials (see Plate XIII b).
Three of the
five were of the tubular type, slightly restricted near the
mouthpiece and flaring into a flanged mouthpiece.
pipe had a bulbous bowl and a straight narrow stem without
the flange.
The last one, only 3.5 cm. long, had a fore­
shortened bowl and flange, and must have had a separate
stem whioh fitted into it.
All of the pipes were carefully
made and exceptionally well polished.
Three pipes were
decorated with small notohes around the mouthpiece; one had
hatched triangles incised on the bowl.
Chipped stone fetishes were not found in the graves,
but were found in quantities on the eroded section of the
midden and on the terrace just east of the burial area.
There is every reason to believe that they are of the same
age as the burials even though they were not found in the
Antler wedges were represented by about six specimens
(see Plate XIII a).
There were no significant differences
between these and the wedges found at the Wallula site.
The antler was split,ground to a chisel point at one end,
and left blunt at the other.
It is believed that these
tools were used with the ground stone mallets for splitting
Bone awls were numerous in the graves, and several were
of excellent workmanship (see Plate XIII a).
A number of
awls were of the splinter type, unfinished except at the tip
of the point, and made of split sections of long bone.
was made of a deer ulna with the proximal end used as a
At least three awls were finished and polished all
They were round in cross-section and tapered smoothly
to a point.
Classifying these three as awls is only a pre­
sumption, for they may have been hair ornaments.
One or two
bone tools had been used as flakers and were partially
smoothed along the shaft.
Other bones, principally the leg
bones of large birds, had been out off into cylindrical
sections for use as beads.
They were of many sizes but
averaged around 3.0 cm. in length and less than 1.0 cm.
in diameter.
Beads made from the hollow leg bones of birds
were found at Hat Creek and Cold Springs, and are considered
typloal ornaments in the region.
This brief description of the material culture covers
those types of artifacts that have been found and described
for the other McNary sites up to this point.
The series of
artifacts to be described next are typical of the burials
at 45-BN-3, but are types that were not found in the other
It should be olearly understood that these are not
neoessarily unique at 45-BN-3, but were recovered there only
because that site was a burial area.
The same artifacts
were probably part of the material culture assemblage at
the Wallula site and other contemporary sites, but were not
lost or discarded in the trash; at least they were not re­
covered in the excavations*
Several polished stone blades that were found in graves
are believed to be celts or adze blades (see Plate XIII b).
They are flat and thin with the edges beveled from one or
both sides. The material is a very dense nephrite which is
greenish black in color.
Exoellent workmanship is exhibited
in these tools for the surfaces and beveled edges are well
In size these tools vary from two inches (5 cm.)
in length and one and one half inohes (4 cm.) in width, to
five inches (14 cm*) in length and two inches (5 cm.) in
Polished stone celts like these, without a hole or
groove for hafting, have been found all along the Columbia
River by collectors,
A few of these were found during the
archaeological survey.
Other pieces of carved stone inoluded a steatite ball
about 4 cm, in diameter with a hole drilled through it
(see Plate XIII b, left center)*
It may have been an orna­
ment, but it would have been a very heavy one to wear.
disc beads were carved from steatite, and had a single hole
Plank oyst burial, Berrianfs Island (45-BN-3)
Flexed burial, Berrian!s Island (45-BN-3)
drilled through the center (see Plate XV, third from the
They were about 1 to 2 cm. in diameter.
A spoon
made of soapstone was made in effigy of a sea shell, Glycymerous sp., and the hinge served as the handle.
A pendant,
also of steatite, was made from the flanged mouthpiece of a
tubular pipe.
Apparently the pipe had been aocidently broken,
and the mouth piece had been reworked into an ornament.
A new kind of fetish was found with the Berrian's Island
It was carved and polished from a piece of tubular
slate (see Plate XIII b, second from right).
The designs
inoised on the fetishes are more easily pictured than de­
Some of them are illustrated below.
In nearly all cases red ochre had been rubbed into the
Strangely enough, some of the chipped stone
fetishes had red ochre rubbed into the notch, and at a site
(35-WS-5) near The Dalles, Oregon, a great many chipped stone
fetishes had been similarly treated.
be forthcoming.
Some explanation should
It can be shown that no chipped stone
fetishes have been found in graves, and no polished stone
fetishes have been found outside of them.
Perhaps the
latter were reserved for burial furniture#
There were two kinds of sandstone shaft smoothers found
in the graves (see Plate XIII b)«
One was carefully made
in block outline with a diagonal groove.
This type was from
15 to 22 cm. long and was made of tuffaoeous sandstone.
second type was made of fine sandstone, and was probably
used by employing matched pairs and sliding the shaft between
the two.
The groove was parallel to the long axis, and this
type was slightly smaller than the other.
Among the objeots made of bone and antler were several
artifacts that had not been found in other McNary sites.
All of these things had been reported by ethnologists as
being typical artifacts of the area, and most of them were
seen by early travelers in the Plateau.
The antler digging
stick handle is one of these (see Plate XIII a)«
It is a
curved tine with a hole bored through it at about the center.
The digging stick was passed through the hole, so that the
antler served a3 a sort of crutch handle.
There is no ex­
planation as to why an antler tine was used, but that was
the case.
Three such handles were recovered from the graves.
A bone fleshing tool, used for removing hair and flesh
from hides, was made from a long bone of deer or elk#
handle was smoothed and one end was toothed like a small
hand rake.
Whistles were made from the wing bones of the
Golden Eagle in the form of a straight tube with holes out
in the shaft (see Plate XIII a, center)*
Numbers of teeth
Ornaments from BerrianU Island (45-BN-3)
Left to righti Bone beads, seed beads, stone beada
and u n known shell, glyoymerls and oltvella shell beids, obot?
beads, glass beads,
and claws were used as ornaments.
Elk teeth were drilled
for suspension in the manner of those worn by members of a
large fraternal organization#
Both the teeth and the olaws
of bears and wolves were used for ornaments#
Most of the shell at 45-BN-3 was of marine origin, and
all of it was ornamental (see Plate XV, center). Whole
shells of the Olivella blplaoata were strung into necklaoea.
Half shells of Glycymerls sp. were pierced at the hinge and
worn as neoklaoes or braclets.
Dentallum sp. was strung
whole or out into sections, and some of the shells were
Disc beads were cut from various unidentified
marine shells.
Most of the shell pendants were made of
Haliotis (Abalone), cut into various shapes, and usually
pierced for suspension.
Apparently the only non-marine
shells used were those of the river mussel, Margaritafera,
which were occasionally made into pendants, but were gener­
ally too fragile for beads.
All in all, shells were among
the most numerous Of things recovered at the site*
In the case of many of the artifacts desoribed earlier
there may be some question as to use, but most of the orna­
ments were found In situ; that Is, in association and posi­
tions that made their Identification conclusive.
The burials
were carefully excavated and position of artifacts were
most carefully noted.
Vast quantities of trade goods of European origin were
found in and around the burials.
After good statistical
SITE 45-BN-6
samples of the glass heads were taken, no special effort was
made to recover them all; they existed by tens of thousands
(see Plate XV, right).
Since the glass beads were discussed
earlier, no further mention need be made.
Nearly all of the metal recovered was of rolled sheet
copper, and was in the form of tubular beads or flat pen­
dants (see Plate XV, seoond from right).
The copper was
not all of the same gage, but every piece examined by metal­
lurgists turned out to be of European manufacture.
The beads
were usually 4 to 5 mm. in diameter and from 1 to 7 cm. long.
All sizes and shapes of pendants were found with various
decorative holes and "knobs punched into the metal.
Iron was apparently well known, but most of it was badly
Identification was possible on a few knives,
arrow points and bracelets.
Many other bits of iron were
too rusted for classification.
Several metal buttons that
were found appear to have been from (or for use on) military
They were of brass, and were either flat or hemi­
Other metal objects included a brass thimble,
bits of silver and fragments of pewter.
Burial Customs:
Some of the graves were simple in­
terments, others were in a plank lined cyst.(see Plate XIV a).
The latter form is interesting. Apparently, the grave was
dug and the mat wrapped body deposited in it.
The grave was
then lined with upright oedar planks about four Inohes wide
and one or two inches thick.
The grave was then filled in
and the protruding planks were burned off level with the
In one case glass beads and food had been thrown in
the fire, presumably as an offering.
Generally the bodies
were flexed (see Plate XIV b) or semi-flexed, on the back or
side with the head oriented to the west (downstream).
Most of the information that was reoovered
from site 45-BDT-3 pertained to material oulture and burial
While it is possible that a group large enough to
be responsible for all of the burials actually lived on the
site, there is reason to believe that the site was not
primarily residential.
Directly aoross the Columbia River from Teohumtas
Island there is a long flat terrace.
Site 45-BN-6, which
is situated on the terrace, stretches along parallel to the
river for one half mile (see Map V).
The site is nine
miles east of the MoNary Dam and one mile east of the small
town of Mottinger, Washington.
Except for being on the
mainland instead of on an island, 45-BN-6 is very similar
to site 45-BN-53.
The house pits are scattered along the
terrace at random with no apparent pattern.
Basalt cliffs
rise abruptly behind the houses and the terrace falls off
into the river with a steep bank and little or no beach.
The village included 59 semi-subterranean
pit houses and one long mat house.
Besides testing for
midden trash, three of the house pits and the mat house were
Apparently the village trash was dumped into
the river for no midden worthy of the name could be found®
Almost all of the artifaots came from the houses as was the
case at 45-.BN-53.
The architecture at 45-BN-6 was little different from
that described for other sites in the region#
House pit 7
was more typical of the houses in the village.
House pits
5 and 6 were possibly occupied at the same time and may have
been a single house over two pits.
House 59 was a very large
mat lodge and the only one excavated by River Basin Surveys.
House pit 7 had been approximately 17 feet long and 15
feet wide (see Figure 6). The remains of several floors
were clearly disoernible near the center of the house but
faded out near the edges.
As near as can be estimated, the
subterranean portion of the house was approximately two and
one half feet deep*
Eaoh of the floors was saucer shaped
without abrupt walls at the sides of the house.
There were
three distinot occupational levels in the house, eaoh marked
by a black accumulation of oharooal and ashes less than one
inch in thickness.
The floors were from six inches to one
foot apart vertically, and each one more or less paralleled
the others.
The earliest floor had been built on a slight
accumulation of trash which included a few animal bones,
some oharooal, ashes and flakes of stone.
Typioal of houses in the MoNary region, house pit 7
• •
W W W \ \ \ \\\ \ \
\ X <
45-BN- 6
gave no indication of the kind of superstructure used.
were no post holes nor remnants of construction materials#
The side of the house nearest the river was slightly lower,
indicating a possible entrance there, but there was no
substantiating evidence.
Houses 5 and 6 were both considerably smaller than the
average for the village or the region (see Figure 5).
House pit 6 was 13 feet long and 12^ feet wide, and house
pit 5 was 16 feet by 12-| feet.
The pits were tangent, and in
neither pit was there evidence of backdirt from the con­
struction of the other.
This led to the belief that the
two were occupied contemporaneously.
The few artifacts
found in the two houses showed no significant differences,
but there is still no proof that they represented a single
No timbers or other evidence of superstructure
were found nor were post holes detected.
As in house pit 7,
the side of house pit 6 nearest the river was lower, sug­
gesting an entranceway there.
None of the houses, 5,6 or
7 were deep enough for a roof entrance to have been used.
House 59 was quite different from any other structure
excavated in the McNafcy region.
It was nearly 65 feet long
and 16^ feet wide, in the form of a reotangle with rounded
ends (see Plate XVI a)#
The floor was less than one foot
deep except where a small firepit had been dug into the
The dimensions of this house were almost identical
to those of the multi-family units reported by Ray,
and Clark56 and others for the historic period.
informant reported that the mat lodge was usually 60 feet
long and 16 feet wide, reotangular in floor plan with rounded
The superstructure was an inverted V section covered
with mats, although there was often a gap in the roof, the
length of the house, through which smoke might esoape.
Several families could live in each mat house, each with its
own fireplace#
Material Culture:
Although there were several occu­
pations of houses 5, 6 and 7, the artifacts recovered from
the various levels of occupation were too complacent to
show any change.
In house pits 5 and 6 the non-European
artifaots were mostly cobble choppers and hammers.
An awl
made of wood, a thumbnail scraper and two elk rib fleshing
tools were also found (see Plate XVII).
The fleshing tools
were simply sections cut off a rib and used without special
In house pit 7 about the same kinds of materials were
Besides cobble choppers and hammers, a crude bone
awl and a notched net weight were recovered.
the long mat house, was more productive.
House pit 59,
Twelve projectile
points from the floor inoluded three that were small and
triangular (see Plate XVII), four that were small and corner
Ray, 1939
Lewis and Clark, 1805
notched, two that were side notched and three broken unclassi­
fied points.
These small projectile points made of obsidian,
jasper and petrified wood are typically those of the late
prehistoric and early historic periods.
A few flake scrapers
and a hammerstone were also reoovered in house pit 59, as
was a basalt pebble that appeared to have been used as an
arrowshaft polisher*
The assemblage of European trade goods associated with
the houses was significantly different from that of either
45-BN-3 or 45-WW-6. Instead of glass beads and copper, the
trade goods included fragments of tin oups in house pits
5, 6 and 7j bits of glass, possibly window glass, in house
pit 59; and a fragment of canvas in house pit 7.
Other trade
items included a large caliber rifle or pistol ball, a frag­
ment of a square nail, a horse shoe and a machine made
mother of pearl buttons(see Plate XVII)»
The horse shoe
and button were in the fill of the houses and could repre­
sent a utilization of the site after the main occupation
had ended.
However the general character of the assemblage
of trade materials suggests that the houses were occupied
after white men were living in the vicinity.
There were
four glass trade beads in house pit 59, and all were of the
type found at Wallula which are believed to be from Ft.
Walla Walla.
Scattered about the terraoe were four or five piles of
medium to large sized rooks.
Excavation revealed,nothing
beneath them and no structure associated with them#
up rooks was one of a number of tasks given to boys who were
in training for the spirit quest.0
The custom was based
on the idea that the spirit quest was not only a very serious
affair, but one whioh required careful preparation*
small boys were given tasks whioh were difficult as well as
frightening in order to test their strength and their
They would be sent out alone on a dark night to
some remote spot to leave some object or pile up stones to
show that they had been there.
This may be an explanation
for the rook piles.
Burial Customs3
Several test pits were excavated in
the steep bank at the river's edge in hope of finding an
over-the-bank dump.
In one of the test pits, however, a
complete burial was encountered.
The remains were of a
male somewhat over 50 years of age who had been interred
without any burial furniture.
All metrical measurements
and morphological observations fell within the expected
ranges, and no anomalies or pathological conditions could
be seen.
The teeth were in extremely poor condition, whioh
would be expected for an individual of that age.
of the bank also washed out a stone bowl that was praotioally identical to those recovered at the Wallula site.
A complete analysis of the animal bones was
Spier, and Sapir, 1930
not available at the time of this writing#
Osborne, however,
reported the presence of horse and bison bones in consider­
able numbers and the absence of antelope bones.58
Bones of
fish and deer were recovered in quantity, as is usual in
McNary sites.
In'addition to the bones, the artifacts showed
that both hunting and fishing were important.
Net weights
and projectile points were found in the houses, and were
eroding from the river bank.
Data on foods obtained by
gathering are always difficult to recover, and other than
one stone bowl, none were found at site 45-BN-6.
The stone
bowl may have been used as a small mortar for grinding
seeds and berries.
Even though evidence of gathering was
lacking, it is known from historical data that it supplied
a large share of the food that was consumed.
Historical Data;
There is one historical reference
which probably applies to site 45-BN-6.
In 1812, Stuart
was on his way up the Columbia River, and stopped at an
Indian Village to trade horses.
His biographer placed this
village just east of the town of Mottinger, Washington.5^
The location fits site 45-BN-6 for there are no other sites
within three miles in either direction and 45-BN-6 is
definitely post contaot.
The horse bones found in the house
pits offer further evidence that site 45-BN-6 was the village
that Stuarii visited.
Osborne, 1953, p. 262
Rollins, 1935, p. 61
45-3N-55 (Sheep Island)
Sheep Island is situated approximately one half mile
downstream from Techumtas Island and approximately one
fourth mile upstream from the Cold Springs site.
The island
is a large gravel bar about 300 yards long and 200 yards
Its size, of course, fluctuates with the height of
the Columbia River.
The downstream end of the island is
considerably higher and has a sandy loam deposit stabilized
"by vegetation.
There have been two archaeological investigations
carried out on Sheep Island.
The first of these was an
excavation in 1949, sponsored by the Whitman College Museum
and Department of History, and the field work was done by
Thomas R. Garth.
The second excavation was by a River
Basin Surveys crew in the summer of 1950, and was under the
direotion of Douglas Osborne.6-'-
Garth»s work brought to
light two large cremation pits and several burials with a
series of stone and bone artifacts associated with each
of them.
Osborne found a series of burials which were also
stratigraphically older than the cremations.
The burials
unfortunately had no artifacts associated with them.
Since the excavation of the cremation pits was not
done by River Basin Surveys crews, the associated materials
were not available for comparison with materials from other
Garth, 1952, p. 343
Osborne, 1952
McNary sites.
The illustrations in Garth*3 publication do
not show anything, however, that differs significantly from
late prehistorio sites in the region#
The stone materials included small delicate projeotile
points seemingly identical to those from site 45-WW-6 and
lozenge shaped knives similar to those from sites 45-BN-3
and 45-WW-6.
A tubular stone pipe and several fragments of
pestles were also found.
Artifacts of bone included several
small incised fragments of ornaments and a number of points
of awls or needles.
Garth found ten burials that were stratigraphically
older than the cremations.
Grave goods were found with six
of the burials, but the majority of the artifacts accompa­
nied just one of them#
These artifacts also consisted of
materials that were practically identical to those of late
prehistorio sites in the region.
The small projectile
points and knives are of identical shapes and were associated
with tubular pipes, stone mallets, pestles and antler wedges.
None of the materials that Garth found would be out of place
at the Wallula site (45-WW-6) or at 45-BN-3.
The pro­
portions of projectile point types are more like those of
45-WW-6 than of 45-BN-3. Sinoe no European trade goods
were found on Sheep Island, it would be expected that the
Sheep Island material culture would more closely resemble
Garth, 1952, p 348
that of the Wallula site which was in part prehistoric.
The burials found "by Osborne are apparently of the same
group as those excavated by Garth.
Osborne, however, found
no artifacts associated with the graves, so that materials
for comparison are still not available.
His analysis of the
skeletal material has not been published as yet.
Since no European trade goods accompanied the graves
and cremations, it is reasonable to assume a pre-contact
date for the site.
1750 A.D.
This would undoubtedly be prior to
Since material culture is similar in every
respect to late prehistoric sites which contain a few glass
beads and bits of copper, the site is probably not a great
deal earlier than 1750.
It is impossible however, to fix a
more precise date.
Culture Change in the McNary Region
In view of the fact that the time factor is understood
only in a relative sense, the rate of culture change in the
McNary region cannot be estimated.
It can be demonstrated
rather clearly that certain artifact types gave way to other
types as time went by.
During this span there were also
modifications of economic pursuits and of the residence
How much time is involved is not known. All that
can be done at the present time is to review the type and
nature of the ohanges as they are reflected in the archaeo­
logical record.
It must be understood then that patterns of
change must "be based on a certain amount of speculation which
is due to normal incomplete recovery of the cultural picture
at any one stage of development#
To speak of a development
in regard to the McNary region, I believe is justified for
the results of the exoavations demonstrate just that.
total change between the earliest material and the latest
is considerable.
Each site or sites which follow in sequence
add a few traits toward the development of a peak which was
reached around the year 1800 A. D.
At this time the abori­
ginal population appeared to be in its best adjustment to
its environment.
This adjustment was quite different from
that of the earliest group in the local sequence*
The people who lived along the south bank of the
Columbia River at a time prior to the ash fall utilized
fish but had no specialized fishing equipment. They must
have had a tradition of hunting, for their most skillfully
made tools were projectile points.
Other than their orna­
ments, the remainder of their material culture was of the
expendable variety, simple tools that required little or no
preparation prior to use*
In this category were the crudely
made hammers, ohoppers and flake scrapers.
These sorts of
tools are seldom carried along on a movement or migration.
They oan be produoed quickly as the need arises and as
readily discarded.
The people who lived on Hat Greek hunted rabbits, deer
and birds, caught a few salmon and shellfish, and probably
gathered a certain amount of vegetable foods such as seeds,
roots and berries.
They seemed to have lived in flimsy
houses of a type that left no remains, and they made consider­
able use of red oohre paint.
This is all that has been
learned of the earliest group of people whose record has been
pioked up in the MoNary region.
Following the settling of the volcanic ash or pumicite,
a group of people began to live on a terrace about five
miles to the east of Hat Creek, a site now called Cold
It is not possible to say how much time elapsed
between the two occupations, but it appears to have been a
very short time.
This is based partially on the close
conformity of ash and midden at both sites as well as the
close continuity of material culture.
Every type of arti­
fact that was found at the Hat Creek site was also found
deep in the midden at Cold Springs.
Before the village at Cold Springs had been in ex­
istence for very long, two kinds of change had taken place.
Certain artifacts had been improved, especially in work­
manship, and certain new artifacts had been added to the
Hammers and ohoppers were made by removing
more flakes and using more careful flaking techniques.
Specialized fishing equipment was added to the inventory.
Notched, grooved and perforated net weights were adopted,
and the nets that were used are responsible for the great
increase in the number of fish bones.
The projectile points were carried over from the Hat
Creek period unchanged except that the concave "based type
was no longer made, and obsidian was now used#
Scrapers and
knives were unchanged, but the long basalt knives became more
Red oohre was still used, but it had been refined
in some way and was pressed into tablets#
Bone beads and
bone awls were retained as they were into later times, and
into the historical period#
Later on, the leaf-shaped projectile point was replaced
by a side notched point#
This is the first appearance of
notching of any type in the region#
Along with side notched
points came the use of discoid choppers#
The earliest of
these were not specialized, in that the size of the cutting
edge or bit had not yet become stabilized#
Discoid choppers
did not replace the oobble choppers, but were an addition
to the inventory#
About that time, the long basalt knives
began to drop from use.
A few of them were associated with
side notched points but the majority were lost or disoarded
when the midden was still shallow.
It is not dear when the first permanent houses were
The earliest house found at Cold Springs was pro­
bably assignable to a time toward the end of the earliest
occupation or about the beginning of what might be termed
the middle part of the occupation#
From that time on,
semi-subterranean pit houses were standard.
The middle period saw the introduction of the first
marine shell and the first carved stone, the latter repre­
sented by a small steatite pendant and a short steatite
tubular pipe#
The rest of the material culture shows no
Thus in transition from the early part of the
occupation of Cold Springs to the middle, leaf shaped pro­
jectile points are replaced by side notched points, discoid
choppers, new ornaments and carved stone are added*
By the time people were living on Techumtas Island
and across the Columbia at 45-BN—53, more changes had taken
There may have been a time interval between these
villages and the abandonment of Cold Springs, but the record
is not clear*
At least the people were no longer making the
large side notched points, and had replaced them with small
delicate corner notched or triangular projectile points*
Some side notched points were made but they were half the
size of those used at Cold Springs (see Plate XVI a for
relative sizes).
Small thumbnail scrapers made their ap­
pearance at this time, although the flake scrapers continued
in use*
Chipped stone fetishes were found at the Techumtas
Island site for the first time, and the stone mallet was
probably introduced into the area them*
Discoid choppers
were numerous and probably specialized in form, although
too few specimens were recovered for analysis.
Spall flake
sorapers or knives became more numerous than ever*
There seems to have been little change in the economy
except for a deoreased use of shellfish*
Pishing and hunting
were important, and it must be assumed that gathering of
vegetable products was always carried on.
The only change
in architecture detectable is a slight decrease in the size
of houses#
The pattern of dual division of the village seen
at Cold Springs was continued at Teohumtas Island but was
not apparent at 45-BN-53.
Sometime during the occupations
of Teohumtas Island and 45-BN-53 the first glass trade
beads may have filtered into the region, probably well ahead
of actual European penetration*
Farther upstream, the Wallula site seems to have been
contemporaneous with the two villages just described#
is reason to believe that the village at the mouth of the
Walla Walla River lasted longer, possibly until about 1800
The material culture there shows no significant change
from the assemblage given for sites 45-BTT.-53 and 35-UM-.17.
There were some artifacts found at the Wallula site that
were not found at the other two.
These differences, however
are slight and most of them are probably due to chance
The large stone adzes or hoes and the stone bowls
may be"of more recent vintage than the artifact assemblages
of 45-BN-53 and 35-TJM-17.
A stone adze was associated with
the very late occupation of Cold Springs terraoe and a stone
bowl was found at 45-BN-6, a site of the historic period.
Chipped stone fetishes and stone mallets were not found at
the Wallula site; they should have been for both artifacts
existed both earlier and later than the occupation of the
Wallula site.
At Wallula there is no apparent shift in economic
pursuits except for evidence of more fish being caught.
The large numbers of projectile points and the quantity of
animal bones attests that hunting was still important#
Shellfish v/ere still used but in small numbers as they were
at 45-B1T-53 and Techumtas Island®
Since no trace of arclii-
tecture was found at Wallula, nothing can be said of it.
The burial site, 45-BN-3 (Berrian's Island) dates from
approximately the same period as does the Wallula site.
The elaborate collection of burial furniture includes a
number of artifacts that had not been found in earlier
These weres
Antler digging stick handles
Bone fleshers
Carved fetishes
Long triangular knives
Sandstone shaft smoothers
Marine shells in great quantity
Teeth and claws as ornaments
Of the group the antler digging stick handles and the
oarved slate fetishes seem more likely to be new additions
but without earlier graves it is not possible to be sure.
One thing is certains
with the appearanoe of trade goods
in quantity, the old ornaments and tools were not replaced
but only supplemented.
Practically all of the artifacts
that were made earlier were retained, with the result that
iron and stone knives were found in the same grave, and both
glass and shell beads were strung on the same necklace.
Whereas the occupation of site 45-EN-3, together with
the graves and their furniture, seems to be a climax of
aboriginal culture, site 45-BTT-6 appeals to show a decline0
Site 45-BN-6 is certainly later than the occupation at
The types of European trade material and the
existence of horse bones indicate a late site.
There is
evidence that the village existed in 1812 and may have
lasted until the middle of the nineteenth century.
The material culture of 45-BH-6 was more utilitarian
than that of earlier sites, and most of the artifacts were
of chipped stone.
They included the usual array found at
the other sites except for the better made artifacts such
aa pestles, mallets, celts, pipes and the like.
The presence
of iron, galvanized iron, glass and leather suggests that
artifaots of European origin were beginning to replace some
of the aboriginal tools.
By this time, white civilization
was probably exercising its usual influence on native
culture, and the latter v/as rapidly deteriorating.
In order to show at a glance how certain artifacts
appeared and disappeared in the McNary region, charts were
These are somewhat simplified and convention­
alized, but reflect the stratigraphy enoountered by River
Basin Surveys excavations.
Figure 7 shows the time distri­
bution of chipped stone artifacts, and Figure 8 does the
same for ground 3tone and bone.
The position of each arti­
fact vertically, is an indication'of the relative time that
it appears in the McNary region.
In most oases, the rela­
tive order of appearance of artifacts is better known than
the order of their disappearance.
There are only two points in time that are definite:
the time of the ash fall and the calendar year 1805.
reference to the ash fall as a point in time does not mean
that we are able to give it a calendar date, but rather that
it is an event that happened at one point in time and can
be reoognized at several different localities.
The in­
definite point in time is the beginning of the late pre­
historic period.
Prom all evidence discussed so far, this
point would appear to be closer to 1805 than to the time
of the ash fall.
The artifact drawings are not to scale since they in­
cluded specimens from less t}aan one inch to specimens as
large as 15 inches in size.
Figures 9 to 17 give the areal distribution of certain
Only artifacts from documented excavations are
shown, since amateur collectors so often forget where they
find their relics.
Only artifacts from the late prehistoric
period are shown in these figures.
poorly documented in the Plateau.
The earlier periods are
E_*«M o a
.. £*/y ^ qa
- • EJV1?* OA
. £1***
* 4S~H,
n— —
. £*** O A
— •
^Th .'
The scarcity of published material on archaeological
excavations in the Plateau is unfortunate.
Almost all of
the existing reports have been concerned with sites of the
late prehistoric and early historic periods.
This may be
due to the earlier sites being less numerous and difficult
to locate.
However, one should not overlook the fact that
late sites usually contain more spectacular artifacts and
larger quantities of them.
There are several ways in whioh the material culture
of one area may be compared with that of another.
One of
the most widely used methods is the trait list which records
presence and absence of artifact types.
sons are valid within limits.
Trait list compari­
If the two artifact col­
lections are of the same general time horizon and consist
of large enough samples, reasonably good conclusions can
be reaohed.
There are, of course certain pitfalls.
sence and absence listing usually does not acoount for the
abundance of an artifact type in one area and its rarity
in the other.
This must be done by some additional
Statistical treatment of trait lists for the derivation
of coefficients of correspondence usually is unsatisfactory#
The results can be no better than the data used, and who can
be sure that equal weight should be given to each trait?
The sample from each area is seldom, if ever, completely
random, and rarely are collections large enough to be good
statistical samples. For these reasons it is believed that
a list of traits for each area accompanied by observations
is suitable for the purpose of comparison.
In considering
the other documented sites in the Plateau this practice
will be followed.
Certain artifact types are found all over the Plateau;
especially are they associated with the late prehistoric and
the early historic periods.
To save repetition, a list of
artifacts common to all sections of the Plateau is given
It is to be understood that these are present at
all of the sites to be discussed, and are not, therefore,
useful for comparison#
Cobble hammers
Cobble choppers
Polished pestles
Stone mallets
Small corner notched points
Flake scrapers
Tubular pipes
Antler wedges
Splinter awls
Two notched net weights
Chipped stone drills
Leaf shaped blades
Dentalium shell
The Dalles Region: Downstream from the McNary Reservoir,
there are several documented sites which are comparable.
At d?he Dalles and vicinity, on the western edge of the
Plateau Area, a considerable amount of work was carried out
by Strong, Schenck and Steward.63
Their methods consisted
of extensive test excavations, surface collecting and exami­
nation of private collections#
Although the sites tested were of different ages, some
with European trade goods and some without, no local sequence
was developed#
The time factor was not ignored, but the
'report was one which considered the area and its artifacts
rather than any attempt at chronology#
Nevertheless, a
horizon comparable to the late prehistoric period in the
MoNary region is recognizable«
Earlier horizons are not
isolated in The Dalles region*
Table III lists artifacts
common to both regions as well as those exclusively found
in each#
Although Table III seems to show a number of differ­
ences in material culture between The Dalles and the MoNary
regions, the similarities are more striking.
Soulpture in
stone and bone, especially anthropomorphic forms, was in
vogue at The Dalles but rare in the McNary.
However, all
of the fancy stone art illustrated by Strong is from private
Elaborate pieces of soulptured stone have yet
to be found in documented excavations, but are very common
in private collections.
Sufficient samples of fairly good
stone sculpture have been found to show that sculpture was
practiced in the region, but the really elaborate pieces
Strong ejb al, 1930
Both Regions
Grooved net weights ------ X
Shaft smoothers
Lap "stones
Stone howls
Discoid choppers
Side notched points
Lozenge blades
Bone needles
Chipped stone fetish
Olivella and haliotis shells
Ground slate fetish - - - - - - - - - - - X
Polished celts
Pour notched weights
Digging stick handles
Bird bone beads
Glycymeris shell
Baked? clay - -- -- -- -- -- -Deep bowls
Fancy carved bone
Sculptured stone
Bone harpoons
Harpoon sockets
Polished stone chisels
Bone labrets
Shouldered points
The Dalles
are subject to question.
Anthropomorphic bone carving,
on the other hand has been documented.
Small clay tablets
with designs pressed in with a sharp instrument are docu­
mented for The Dalles region, and are unique there.
polished stone chisels described for The Dalles region are
very similar in material and workmanship to the McNary celts.
Apparently, however, they were used in some other manner.
Strong did not list chipped stone fetishes for the region
but the present writer found them in two sites including
one site tested by Strong,
Hobo Cave:
This deep stratified cave is situated on
the south bank of the Columbia River, about 35 miles east
of The Dalles and about five or six miles east of the John
Day River.
It was excavated in 1950 by the University of
Oregon, and under the supervision of the present writer.
The cave deposit, with a maximum depth of nine feet, showed
almost continuous occupation,
A local sequence, at least
for projectile points, was developed, and is the only
sequence in the entire Plateau available for comparison with
that of the McNary region.
Briefly stated, there were three distinct levels, each
characterized by diagnostic projectile points.
The top
level (0 to ,88 m. below the surface) yielded 58 projectile
points, nearly all of them small and corner notched (average
64, Strong et_ al, 1930, p, 20s This was Strong's site
14, In 1952, Shiner recovered 65 chipped stone fetishes among
some 1300 artifacts from the site.
4 I
$ -4
• i f f « l|
Hobo Cave
>*•' "
McNary regi 011
length, 25 nim.)#
They were practically identical to points
from Wallula (45-WW-6) and contemporary sites in the McNary
The exception is that there were no side notched
points recovered in Hobo Cave.
The seoond level (.88 to 1.65 m. below the surface)
was characterized by larger points (average length 38 mm.).
Of 35 points recovered in this level, 31 were corner notched,
and most of thein had straight bases.
The size difference
between the points of the first and seoond levels is clearly
shown in Plate XVIII.
There is an interesting parallel with
the situation in the McNary, for in both places the small
points are preceded by a point type that was half again as
Although all of the McNary points (from Cold Springs)
were side notched and nearly all of the second level points
in Hobo Cave were corner notohed, the trend is the same.
Between 1.65 m. and 1.70 m. there was a stratum of
sterile sand in Hobo Cave.
It stood in sharp contrast to
the rest of the cave fills, which is composed of silt, char­
coal, ashes and occupational debris.
Beneath the sand was
a third oultural level which was characterized by projectile
points with shoulders, but no notches or barbs (see Plate
XVIII, third row from top).
reoovered in level III.
Only 12 projectile points were
Ten of these were shouldered points
(average length 41 mm.), one was a basal fragment of a leaf
shaped blade, and the other was a corner notched point which
was typical of the second level.
A comparison between these
projectile points and those of the McNary region finds the
general trend of simple unnotched and barbless points being
earlier in both places.
Below the cultural level which contained the shouldered
points were several other artifacts.
Among these were two
obsidian points of the corner notched variety, but which
differed from those of the second level.
With them were
nine disc shell bead3 and a flake scraper.
This, the lowest
level of the cave, is poorly defined because of the scarcity
of artifacts and no attempt can be made to relate it to
other cultures.
Besides the projectile points, numerous other artifacts
were recovered in Hobo Cave.
They did not, however, differ
significantly from one level to the next#
Level I, which
had the small projectile points, is compared to the late
prehistoric material culture of the McNary region in
Table IV.
Since Hobo Cave was situated high on a hillside, it can
not be considered the same sort of occupation that would be
found on a river terrace.
The cave may have been used by
hunting parties, but was not a normal residence.
many of the utilitarian artifacts, especially the heavy ones,
would not be expeoted there.
This can explain the absence
of net weights and the scarcity of hammers and choppers.
The more carefully made artifacts such as tubular pipes,
mallets, celts and digging stick handles normally are found
Lozenge shaped knives
End scrapers
Discoid choppers
Bone needles
Bone flakers
Bird bone "beads
Disc shell beads
Side notched points Notched net weights
Grooved net weights
Chipped stone fetish
Ground slate fetish
Polished celts
Tubular pipes
Shaft smoothers
Lap stones
Fish spear tines - - Bone disc beads
Baked? clay
Both Regions
Hobo Cave
only in graves.
The absence of side notched points, however,
must be due to cultural preference.
One of the traits,
baked clay, cannot be explained fully.
Clay sherds, which
resembled pottery, were recovered in the second and third
The material was not true pottery, but no one has
been able to identify it.
Hobo Gave thus gives a sequence which follows the trends
shown by the projectile points in the McNary region.
latively simple points are replaced by notched points, and
those are followed by a sharp reduction in size.
At the
present time it is not possible to demonstrate contempor­
aneity for the changes in both regions, but European trade
goods are associated only with the small projectile points
in each region.
John Day Reservoir; The Columbia River Valley between
Hobo Cave and the McNary Dam is within the limits of the
proposed John Day Reservoir.
As a potential reservoir, it
was surveyed by River Basin Surveys in 1950.
Not very much
was recovered in the way of material culture during the
survey since the area is subject to constant search by
relic hunters.
However, 88 archaeologioal sites were re­
corded and the surface collections from those sites com­
pared closely with materials from the McNary region.
collections included many projectile points, net weights,
scrapers, hammers and choppers.
The Wahluke Site:
Upstream from the McNary region
aloiig the Columbia River, there are several archaeological
sites that have "been scientifically excavated and reported.
One of these is a site know as Wahluke, which was excavated
by Herbert Krieger65 in 1926.
Wahluke was a village of
some 30 houses and a burial area. It is situated on the
west bank of the Columbia River about 60 air miles north­
west of the Wallula site.
The Wahluke site was apparently of the late pre­
historic period.
Although specimens of copper were re­
covered, Krieger identifies it as native copper.66
theless, the total artifact assemblage is clearly that of
the late period just prior to European influence.
As in
the case of the artifacts of The Dalles region, the tools
and ornaments oommon to all oif the Plateau during the late
prehistorio are not listed; otherwise Table V, below,
compares the artifacts of Wahluke and the McNary region.
The close similarity of the material oulture of Wahluke
to that of the MoNary region is apparent.
Krieger*s report
does not give the frequency of artifaot types, but the
illustrations clearly show the similarities.
While he may
not have recognized chipped stone fetishes as diagnostic
artifacts, the polished slate fetishes would have been
No one else in the Plateau seems to have
Krieger, 1928
Ibid., p. 13
Four notched net weights •
Grooved net weights
Discoid choppers
Polished celts
Side notched points
Lozenge blades
End scrapers
Sandstone shaft smoothers
Bone shistles
Bone flakers
Bone beads
Digging stick handle
Bone dice
All kinds of shell
Lap stones -------Stone bowls
Chipped fetish
Polished slate fetish
Bone needles
Elbow pipes ------Shouldered points
Native copper
Stone olubs
Both Regiona
recognized them as artifacts, even when they were numerous
in the region.
The appearance of elbow pipes in a prehis­
toric site is puzzling.
Since Krleger did not always make
it clear which artifacts were recovered in situ and which
artifacts were merely typical of the region, it would he
reasonable to suspect that elbow pipes were not in direct
association with the Wahluke occupation.
Elbow pipes are
not a typical Plateau trait until the Plains influences with
horse and gun come in.
Native copper is not reported from
any other archaeological excavations in the area.
differences can be seen in types of projectile points and
the artifacts mentioned above, but the bulk of the material
culture shows close affiliations with the MclTary region.
This is especially true of the heavy stone torils used for
net weights, hammers and choppers.
The Yakima Region:
The Yakima River enters the
Columbia from the west, about ten miles north of the mouth
of the Snake River.
It rises in the Cascade Mountains about
120 miles away, and flows through an open valley most of the
The field work and report by Harlan I. Smith were
accomplished betv/een 1908 and 1910.
His method was a
combination of surface survey, excavation and examination
of private collections.
The region that he investigated
extended nearly 100 miles from Ellensburg, Washington, to
Smith, H. I., 1910
Bone needles
A.11 kinds of shell
Chipped fetish ----------------X
Polished slate fetish
Sandstone shaft smoothers
Bone whistles
Stone olubs ---------------------X
Elbow pipes
Sculptured pipes
Elaborately oarved bone
Elaborately oarved pestles
the Columbia River.
Part of the Yakima region lies just
north and west of the McNary region and south of the Wahluke
Since the Wahluke material culture was so similar to
that of the MoNary region, it might "be expected that the
Yakima materials would also be similar.
The similarities
were found, and can be seen in Table VI.
The Yakima region appears to differ- only slightly from
the McNary region in material culture.
The differences
in almost the same.traits that distinguished the McNary from
The Dalles.
It is remarkable that Smith did not find bonei whistles
and sandstone shaft smoothers in the Yakima Valley.
artifacts have been recovered both north and south of the
Yakima region, and Smith was unable to explain their absence.
It is possible that the elaborate carving in stone and bone
belonged to a late period, perhaps as late as 1875, to 1900.
Smith opened several graves which were quite recent at the
Most of the complex sculpture was in the hands of
private collectors, and thus of unknown provenience.
these exceptions, however, the Yakima material culture is
nearly identical to that of the McNary region.
The Chief Joseph Reservoir:
About 100 air; miles north
of Wahluke, considerably farther via the Columbia River,
is the Chief Joseph Reservoir. It is downstream from the
Grand Coulee Dam.
During the summer of 1950,
Osborne, Crabtree and Bryan, 1952
River Basin
Surveys parties excavated several sites along the river.
Although some information on architecture was recovered,
artifacts were few in number.
All of the sites appeared to
be late, and even the graves were poorly furnished.
few artifacts were found to permit a systematic comparison
with those of the McNary region, but the ones that were.
found did not differ significantly.
The Upper Columbia Regions
Archaeological reconnais-
sance in the Grand Coulee Reservoir began in 1939, after the
dam was under construction, and because of the rising water,
was not carried to completion.
The salvage archaeology
of Collier, Hudson and Ford recovered representative col­
lections of artifacts, but little information on prehistoric
Their region included both banks of the
Columbia River from the Grand Coulee Dam almost to the
Canadian border, and is referred to as the Upper Columbia
Although some of the sites were without European trade
goods, none of them appeared to be earlier than the late
prehistoric period.
This makes all of the material culture
suitable for comparison with that of the MoNary Region.
Table VII compares the material culture of the two regions.
The close similarity between the two regions is clearly
indicated, but a few reservations should be made.
Collier, Hudson, and Ford, 1942
Table VII
Upper Columbia
Lozenge blades --------X
Small side notohed points
End scrapers
Spall flakes
Tubular pipes
Polished Celts
Grooved net weights
Discoid choppers
Sandstone shaft smoothers
Digging stick handles
Bone points
Bone needles
Bone dice
Bone beads
Bone whistles
Chipped fetish
Polished slate fetish
Stone bowls
Pour notohed net weights
Large corner notched points ------------X
Barbed harpoon points
Stone clubs
Grooved mauls
Elbow pipes
and grooved net weights are present in the Upper Columbia
region, but are not at all numerous.
The same is true of
nearly all of the hammering and chopping tools made from
river worn cobbles.
In the.Upper Columbia most chopping
tools were chipped all over, and were made of cryptocrystalline stone.
As was the case in nearly every region
outside of the McNary, carving in stone and bone was im­
portant in the Upper Columbia.
In spite of the differences
in numbers of oertain artifacts, it is clear that both the
McNary and Upper Columbia regions shared many material
culture traits.
The traits included all types of artifacts,
utilitarian as well as decorative, and in most cases were
nearly identical in every small detail.
British Columbia«
Harlan Smith's archaeological
investigations in British Columbia were part of the Jesup
North Pacific Expedition, and were carried out in 1897•
Smith explored burial places and villages along the Praser
and Thompson Rivers, as well as making examinations of
private collections.
The bulk of the material culture
illustrated in his report came from a burial ground near
Lytton, British Columbia.
The graves contained no articles
of European manufacture, but appeared, according to Smith,
to be of the late prehistoric period.
The close resemblance
of the artifacts to those of the late prehistoric period in
the McNary region tends to confirm his conclusion.
Smith, 1899
Both Reglons
Lozenge blades ---------- X
Side notched points
End scrapers
Polished celts
Stone bowls
Sandstone shaft smoothers
Lap stones
Digging stick handle
Bone needles
Bone dice
Bone beads
All kinds of shell
Chipped fetishes - - - - - - - - - - - - - - - x
Polished slate fetishes
Diso choppers
Pour notched net weights
Grooved mauls -------------------Eccentric points
Fancy carved bone
Harpoon sockets
VIII compares the Lytton material culture with that of the
McNary region#
Unless Smith failed to illustrate most of the Lytton
projectile points, there are few resemblances to be seen in
the points of the two regions.
Since Lytton is situated in
the extreme northwest corner of the Plateau, some 350 miles
from the MoNary region, it is significant that the simi­
larities in material culture are so great.
There is no way
of determining the relative frequency of the different types
of artifacts from Lytton.
However, the presence of so many
oomplex artifacts that are identical in detail to those from
all over the Plateau indicates close cultural connections.
The Snake-Clearwater Region:
Spinden's investigations
among the Nez Perce' Indians were made between 1897 and 1900,
with additional work done in 1907.
In the region ex­
tending from the forks of the Clearwater River to, and along
the Snake River, he examined excavations made by railroad
He also made surface surveys and studied
collections that belonged to private collectors.
like H. I» Smith, was more interested in presenting an aresl
view of the material culture, and somewhat neglected the
time factor.
However, from the stratigraphic studies made
in the McNary Reservoir, it is possible to give time place­
ment to the most of Spinden^ materials.
Spinden, 1908
In Table IX,
Both Regions
Snake River
Pour notched net weights - - - - X
Grooved net weights
Discoid choppers
Polished celts
Side notched points
End scrapers
Lozenge blades
Sandstone shaft smoothers
Stone bowls
Lap stones
Bone whistles
Bone flakers
Bone beads
Digging stick handles
Bone dice
Bone needles
All kinds of shell
Chipped fetish - - - - - - - - - - - - - X
Polished slate fetish
Elbow pipes -------------------- X
Gorge hook
Spear tines
similarities and differences between the material cultures
of the McNary and Snake-Clearwater regions are listed.
, The correspondence of the material culture of the
Snake-Clearwater region to that of the MoNary is probably
closer than that of any other region discussed to this point
The similarity may actually be more in the nature of identi­
Chipped stone fetishes were not listed by Spinden, but
may have been present in the Snake-Clearwater region.
elbow pipes, which were reported for that region, very
possibly are typical of the historical period and not of the
late prehistoric.
Gorge hooks and fish-spear tines have
not been recovered in the McNary region, but early travelers
journals give the impression that both were typical arti­
facts of that region (see section on historical period).
The gorge hook is a short piece of bone, pointed at each
end, and secured at the center to a fishing line.
The bone
turns sideways in the throat of a fish, and thus acts as
a hook.
If these differences can be eliminated, the SnakeClearwater and the McNary regions would have been nearly
identical in material culture.
Historical sources have
verified the resemblance of the Walla Walla, Umatilla and
Nez Perce'Indians for the period from 1805 to 1835, so a
correspondence in prehistoric times might be expected.
Comparable materials were available from all sections
of the Plateau except the northeast.
Archaeological surveys
and test excavations were made there by River Basin Surveys,
but insufficient artifacts were recovered for comparison.^2
Prom the limited collections made, there is good reason to
believe that the region from Coeur d*Alene Lake north to
Lake Pend (^rielle differed significantly in material culture
from the rest of the Plateau.
The region, although within
the Columbia Basin, is atypical.
It is heavily wooded and
is well provided with lakes and running streams.
Ray»s analysis of Plateau political and social organization
indicates that the region had strong Plateau affinities,
its cultural position in prehistoric times cannot be defined
until further excavations are made there.
Early Plateau Sites: Here and there in the Plateau
there are traces of cultures that are probably much more
ancient than the materials from the Eat Greek site.
man has recently recovered evidence of "early mann near
rt a
The Dalles Oregon.
Details are lacking at the present
time, so comparisons cannot be made.
At Lind Coulee, near
Moses Lake, Washington, Daugherty has recently recovered
evidence of "early manl,o,',5
Again, details are not available,
but Daugherty states that his materials bear no resemblance
to the artifacts from Hat Creek.
Radio-Carbon dates for
the Lind Coulee site averaged in the neighborhood of 8,700
72. Shiner, 1952
73. Ray, 1939, p. 145
74. Cressman, 1953, Personal communication
75. Daugherty, 1953, AAA Annual Meeting
The theory that a widespread Paleo-Indian culture
existed in the northwest was first advanced by Cressman.
Although the majority of his field work has been carried
on in- the northern Great Basin, Cressman until recently
has been the only active archaeologist in the Pacific North­
He has, therefore, been an authority on the Plateau,
the northern Great Basin and much of the Northwest Coast.
Cressman^ interest has centered in early man and in the
many problems connected with such a study.
Availability of
comparative material for a time focused his attention to the
He made comparative studies of Oregon cave materials
with early cave materials of the southern Great Basin and
with certain artifact types from the southwest.
He did not,
however, lose interest in the areas to the north and north­
east of the northern Great Basin.
Cressman had formulated a theory that early man in the
western part of North America was divisible into two cul­
ture types.'76
East of the Rocky Mountains was a hunting
type of culture typified by Folsom man.
On the west of the
mountains the, culture represented was a mixture of hunting
and gathering, characterized by an abundance of grinding
tools. The presence of these tools in the west spelled a
fundamental difference in culture to Cressman.
Cressman, 1950, p. 369
He felt that
out of the early hunting and gathering culture of the west,
there developed a group of localized cultures, and that the
development of these had little if any relationship to the
Folsom and later developments east of the Rooky Mountains.
Looking to the future, he wrote,
"We shall eventually, by systematic extension of
this work north and south and east and west, discover
whether there was a province of culture west of the
Rookies extending from far toward the Southwest well
toward the Plateau region of the north, and, if there
was such a province, what the cultural substratum was
from which the later localized types of culture
Daugherty's discoveries seem to be bearing out Cressman's predictions except for the fact that grinding tools
were not an integral part of the Lind Coulee culture.
fact, grinding tools of stone may not even exist north of
the Columbia River,
They did not appear in any of the
Columbia and Snake River sites investigated by River Basin
This does not prove Cressman to be wrong.
He was
searching for the goegraphical limits of the northern Great
Basin Culture, and it now appears that the northern limit
may be the Columbia River.
In fact, it has been reported"''®
that Cressman found grinding tools in an early site on
the south bank of the Columbia River near The Dalles, Oregon.
Where the materials from sites like Hat Creek fit in
the picture, it is not possible to say.
The culture of. that
Cressman, 1940, P. 15
Cressman, 1953 Verbal Communication
period could not be called Plateau, for it was not adjusted
to Plateau ecology.
It supports Osborne1s theory about the
early prehistoric period.
"My present impression is that the Plateau was
originally occupied by groups similar in culture to
the Great Basin (the Plateau is, in many ways, little
more than an extension of the Great Basin)."79
Until . Cressman,s and Daugherty»s new materials are
studied, and other "early man" sites are excavated in the
Plateau, the nature of Paleo-Indian culture in the area
cannot be properly determined.
Other than the comparison between the early
levels of Hobo Cave and the Cold Springs site, it is not
possible to discuss cultural developments elsewhere in the
The late prehistoric period, however, is well
represented by sites all over the area.
The tables, which
were given for sites and regions in nearly every part of
the Plateau, clearly indicate that the late prehistoric
period was one of widespread uniformity of material culture.
Some of the artifaots listed were crudely made and more or
less generalized.
Many of them, however, were highly-
specialized and of complex form.
It is the wide distribution of the specialized arti­
facts that is significant, for they are far less likely to
be independently invented.
A large series of complex tools,
which are similar in small detail, spread all over the
Osborne, 1951, p. 302
Plateau, is substantial evidence of the homogeneity of
Plateau culture in the late prehistoric period.
In addition to the list of artifacts common to the
entire Plateau, others can be given which are almost as wide­
Lozenge blades vary slightly in shape from region
to region, but are present in each.
Sandstone shaft
smoothers were found in every region except the Yakima
Bone beads and polished celts were recovered in
every region except The Dalles.
Hobo Cave did not give a
rounded sample of material culture, but did have a large
collection of projectile points.
Side notched points did
not appear in the cave but were present elsewhere in the
A survey of historical references to Plateau Indian
culture is essential for several reasons.
First, it es­
tablishes that certain items of material culture were used
in certain ways, and second, it gives a trait list for the
historical period which can be compared with that of the
prehistoric period.
In view of ohanges that take place in
material culture through time, it is vital that the culture
IF I. (
described for the historical period be limited in its time
When the Plateau is spoken of as a culture area, re­
ference is made not to present day culture, but to culture
of the period prior to European acculturation.
of the Plateau Area was first available when Lewis and Clark
returned from their expedition in 1806.
Fortunately, they
were careful observers, and were not primarily interested
in trade or religious conversion.
Other travellers followed
in rapid succession, and more than a few of them wrote of
their experiences.
It is difficult to determine just when
the early contact period ended in the Plateau.
Ray, in
reference to the middle Columbia basin, wrote,"A few villages in the area had been abandoned
as the result of white encroachment as early as 1880,
"but major displacements did not occur until after the
turn of the century."80
This, however, has little bearing on the problem of accultu­
The disturbance of aboriginal Plateau culture began
even before Lev/is and Clark, at the time that European trade
goods began to appear in quantity.
The earliest trade goods of European origin were glass
beads and bits of copper.
They were probably traded into
the Plateau by other Indians who obtained them from eastern
fur traders.
The earliest post nearby was established at
Nootka on Vancouver Island by John Meares in 1787.
would have given Plateau Indians a source of trade goods
only 350 miles away.
In 1791, Robert Gray entered the
Columbia River in a ship named the Columbia.
He named the
river after his ship, and traded extensively with the
Among the things he carried were copper sheets,
iron bars and buttons.
It was about 1811 to 1812, when
trade goods in large quantities began to reach the central
part of the Plateau.
At this time, parties from the post
at Astoria (at the mouth of the Columbia River) traveled
up and down the River, trading as they went.®®
In 1818,
Ft. TSez Perce'was established near the mouth of the Walla
Walla River, and other posts were soon scattered throughout
the area.
Ray, 1936, p. 99
Winther, 1947, p. 25
Ibid*, p. 25
TKWTites, 1904
By 1835, when the Reverend Samuel Parker toured the
Plateau, it waa rather clear that acoulturation had taken
His account is full of details of Indian life,
although colored by religious philosophy.®^
Parker*s ob­
servations, several of which will be discussed later, lead
to a conclusion that about 1835 to 1840 should mark the
close of the early historic period.
Parker, and soon after­
ward, Marous Whitman, began to convert Plateau Indians away
from their religion.
Settlers, traders and trappers, by
this time, had begun to push the Indians around, and European
tools and weapons had become common over most of the area.
By the 1840_T;s, around the Hudson Bay posts,
"Schools for the native children are attached to
all the principal trading posts, and particular care
is extended to the education of the Half-breed children,
the joint offspring of the traders and the Indian
women, who are retained and bred, as far as possible,
among the whites, and subsequently employed, when
found capable, in the service of the company. The
policy of course is obvious. The savage is gradually
cured of his distrust and coaxed into new connections.
He abandons the use of his bows, his arrows and all his
former arms, and the result is that he soon becomes an
absolute dependant upon those who furnish him his guns,
ammunition, fish-hooks, blankets, etc."85
Therefore, as far as the present discussion is concerned,
between 1835 and 1840 will be considered the end of the
early historic period and the "ethnographic present".
following quotations and comments will apply to that period,
Parker, 1844
Wilkes, 1845, p. 86
beginning in 1805 and ending about 1835 or 1840»
Seasonal Migration
None of the writers on Plateau archaeology have pro­
perly evaluated the effect of seasonal migration on archaeo­
logical sites.
There is evidence to show that practically
all of the Plateau Indians made seasonal migrations in order
to avail themselves of different kind3 of food.
In 1805,
Lewis and Clark wrote of the Uez Perce',
"During the summer and autumn they are busily
occupied in fishing for salmon and collecting their
winter store of roots. In the winter they hunt the
deer over the plains, and towards spring corss the
mountains to the Missouri for the purpose of traf­
ficking for buffaloe robes."86
Seasonal migration was noted in the same area by another
"The spot where we landed (on the Snake River)
was an old fishing establishment, of which there re­
mained the timbers of a house carefully raised on
scaffolds - - - the property of the Indians who still
remained in the plains hunting the Antelope."8^
On the Columbia River among the Tenino Indians the following
spring, they noted,
"Since we left them in the autumn they have re­
moved their village a few hundred yards lower down
the river, and have exchanged their cellars in which
we then found them, for more pleasant dwellings on
the surface of the ground. They are formed by stioks
and covered by mats and straw - - -."88
Hosmer, 1905
Biddle, 1904, Vol II, p. 185
Ibid. Vol III, p. 63
By the time they had arrived in Nez Perce' territory again,
"The salmon not having yet called them to the f
rivers, the greater part of the Chopunnish (Wez Perce)
are now dispersed in villages through this plain, for
the purpose of collecting quamash (camas) and oows
There are numbers of references by other observers in
the area, all pointing to seasonal shifts in habitation
which meant the movement of the entire village.
Even as
late as 1885, the pattern was still being followed by some
Nez Perce' on the lower Snake River, According to one of
the writer's informants, an early settler, the Nez Perce^
used to spend the winter in the river bottom and move out
in the spring.
Ray summed it up as follows,
the whole, Plateau life involves wintertime
occupancy of river villages and summertime camping
at fishing, berrying and root digging spots."90
What would be the effect of seasonal migration on the
archaeological record?
One result would be found in archi­
tecture, since only the semi-subterranean pit houses would
be preserved.
The summer mat house would leave little or
no trace after a few years.
It is possible that the multi­
ple floor levels found in the house pits at Cold Springs,
Teohumtas Island and elsewhere are records of year to year
After the pit became too shallow, it may have
been cleaned out, and some of the floor levels may record
only one winter's occupation.
Doubtless, the seasonal
Biddle, 1904, Vol II, p. 83
Ray, 1939, p. 14
migration pattern is an old one in the Plateau, for the
Indians held to it long after other oustoras had disappeared#
There is no way of determining how far back in time the
custom was followed.
One of the best descriptions of Indian dwellings in
the Plateau was also the earliest.
In the spring of 1805,
near the mouth of the Snake River, Lewis and Clark wrote,
"The houses of the Sokulks are made of large mats
of rushes, and are generally of a square or oblong
form, varying in length from fifteen to sixty feet,
and supported on the inside by poles or forks about
six feet high: the top is covered with mats leaving
a spaoe of twelve or fifteen inches the whole length
of the house, for the purpose of admitting light and
suffering the smoke to pass through: the roof is
nearly flat - - -, and the house is not divided into
apartments, the fire being in the middle of the large
room and immediately under the hole in the roof; - - '*91
The explorers also mentioned a Nez Perce' sweat lodge, racks
for drying fish, deep pit houses near The Dalles, Oregon,
and a large burial vault on the Columbia River.
There is no early record of winter houses for the
northern Plateau but summer mat lodges were described.
In 1811, at the mouth of the Sanpoil River and near the
present site of Grand Coulee Dam, David Thompson observed,
"Their huts are of slight poles tied together,
covered with mats of slight rushes «w92
Biddle, 1904, Vol II, p. 189
Sperlin, 1913, p. 8
Farther down stream he mentioned a Wenatchee hut which was
209 yards long.
Material Culture
Actual descriptions of articles of material culture
are rare in the early reports.
Some of the closest ob­
servations were made of clothing which, being perishable,
was not recovered in the archaeological excavations.
and Clark, in 1805, noted that the Indians near the mouth
of the Snake River wore buffalo or elk-skin robes, beads and
feathers, leggings, and moccasins.
"The dress of the women is more simple, consisting
of a long skirt of argalia or ibex skin, reaching down
to the ankles without a girdle: to which are tied
small pieces of brass and shells and other small
articles. M 93
Stone mallets were apparently observed by Clark just
above the mouth of the Snake River in 1805.
He wrote,
"He began by bringing in a piece of pine wood
that had drifted down the river, which he split into
small pieces with a wedge made of the elkfs horn, by
means of a mallet of stone curiously carved."94
About the same time, Lewis mentioned a group of articles,
"The rooms are ornamented with their nets, gigs
and other fishing tackle as well as the bow for each
individual, and a large quiver of arrows, which are
headed with flint stones."^®
Bows and arrows were described by Henry, as of the year
Biddle, 1904, Vol II, p. 174
Ibid, p. 192
TEH, p. 189
His remarks apply to the Plateau Indians in general.
"The bows used by the natives W. of the mountains
are nearly made, and of three kinds - the horn, the
red cedar, and the plain wooden bow. The horn bow is
made of a slip of ram's horn. - - - overlaid with
several successive layers of sinew - - -. The red
cedar bow is made pf a slip of that wood overlaid
with sinew and glue - - -. The plain wooden bow is of
cedar, willow, or ash - - - it is well smoothed but not
so much esteemed by the natives - - -."96
"The arrows are muoh longer than those of our
Indiana E. of the Mountains, being nearly three feet,
very neajbly made, slim pointed, and well feathered;
They are usually tipped with flint- but of late iron
has been seoured for that purpose."97
While there is no description of dice, John Work, in
1825, near the present city of Lewiston, Idaho, noted that,
"There are about our camp near 250 or 300 Indians.
- - - they pass the greater part of their time gambling,
horseracing & footracing."98
In a burial vault on ar^ island in the Columbia River,
Lewis and Clark saw an array of grave offerings,
"Prom the different boards and canoes which formed
the vault, were suspended on the inside fishing nets,
baskets, wooden bowls, robes, skins, trenchers, and
trinkets of various kinds, - - -."99
Mentions, in passing of artifacts are rather common.
The early journals frequently speak of baskets, pipes, bows
and arrows, canoes, stone hatchets and stone knives; but
no details are given.
Henry, 1897, p. 713
Ibid, p.714
Work, 1825, p. 94
Biddle, 1904, Vol II, p. 204
Lewis and Clark made several references to Indian
In the spring of 1805, they noted,
"The dead are wrapped up in robes of skins, and
deposited in graves, which are covered over with earth
and marked or secured with little pickets or pieces
of wood stuck promiscuously around It."100
Thirty years later, in the same vicinity, Parker wrote,
"The grave was two feet deep. A mat was laid
in the grave, and the body was wrapped in a blanket.
A horn cup and a spoon was placed in it. and a mat or
rushes above and then it was filled in."101
Another kind of disposal was that of the burial shed or
On Blaloclc Island, some 30 or 40 miles west of the
McNary reservoir, Lewis and Clark visited one of the vaults,
"This place in which the dead are deposited is
a building about sixty feet long and twelve feet wide,
- - - so as to form a shed." "We observed a number
of bodies wrapped carefully in leather robes, and
arranged in rows on boards - - - and in the center of
the building was a large pile of them heaped pro­
miscuously on each other."102
These vaults were seen in later years, and so thoroughly
robbed by collectors that no trace of them remains.
Various journals mention the gathering of roots and
Biddle, 1904, Vol II, p. 178
Parker, 1844, p. 285
Biddle, 1904, Vol II, p. 203
berries by Indians all over the Plateau.
Camas, Camaasla
sp. and kouse, Lomatium Kaus were staples.
At one time or
another, all of the early travelers subsisted on these
tuberous rooted plants whioh grew in marshy lowlands.
There are no details available on how these roots were
gathered in the early hlstorioal period.
Later on, it was
noted that digging sticks and baskets were employed.
Preparation of food in the early historic period has
been described (see page 56 on earth ovens), but little
mention wa3 made of artifacts used for that purpose.
of them must have been of a perishable nature for no con­
tainers are found.
been recovered.
Pestles are not rare but no mortars have
If baking and roasting had been extensively
used, containers would not have had to be plentiful.
of the stone bowls that have been found have been far to
small to have been used for cooking but may possibly have
been used for serving.
Almost any of the stone tools that
have been classified as hammers and choppers oould have
been used in pounding roots, berries and meat.
many of the artifacts used for preparing food were either
made of perishable materials or were of a generalized tool
Details are rare on hunting techniques although fre­
quent mention is made of Indians engaged in that activity.
Apparently the Plateau Indians employed techniques that were
common over most of North Amerioa.
They used disguises
suoh as skins of animals,^03 and such tactics as surround­
ing, driving and running down game on horseback.
Pishing is well documented all over the Plateau in the
early historical period.
One of the most detailed de­
scriptions is from Lewis and Clark in 1806, near the mouth
of the Walla Walla.
"Near our camp is a fish-weir, formed of two
curtains of small willow switches, matted together
with withes of the same plant and extending across
the river in two parallel lines, six feet asunder,
These are supported by several parcels of poles in the
manner already described, as in use among the Shoshonies, and are either rolled up or let down at
pleasure for a few feet, so as either to suffer the
fish to pass or detain them. A seine of fifteen or
eighteen feet in length is then dragged down the river
by two persons, and the bottom drawn up against the
willows. They also employ a smaller seine like a
scooping net, one side of which is confined to a
semicircular bow - - -."105
While the bulk of the fish taken were probably captured
with seines or dip nets, some were speared and others were
taken by angling.
Spear tines have been recovered in all
regions of the Plateau, but no archaeological evidence of
hook and line has been recovered in the McNary region.
Angling was reported at least two times in the region that
is now within the MoTTary Reservoir.
In 1806, Lewis and
Clark described the use of a gorge-hook,
wSoon after we halted, an Indian boy took a piece
of bone, which he substituted for a fish-hook, and.
caught several chub, nine inches long."106
Thwaites, 1849, p. 297
Ibid, p. 316
Biaffle, 1904, Vol III, p. 80
Ibid, p. 74
Again, in 1811, just above the mouth of the Snake River,
Ross described an Indian fisherman,
MPor this purpose, the fisherman cut off a bit
of his leather shirt, about the size of a small bean;
then pulling out two or three hairs from his horse's
tail for a line, tied the bit of leather on one end
of it, in place of a hook or fly. Thus prepared, he
entered the river a little way, sat down on a stone
and began throwing the small fish, three or four inches
long, on shore, - - -."107
There, is no doubt that by the beginning of the early
historical period, the Plateau Indians were well versed in
the art of taking fish.
Nets, traps, spears and angling
were all used with skill.
It is not clear just where horses fitted into Plateau
There is little question but that the first
Plateau horses came from the Shoshones.
Haines1 research
on the subject showed clearly that the Shoshones were re­
sponsible for the transfer of horses both to the north and
to the northeast of their territory.
Haines quotes
the narrative of an early explorer to demonstrate that the
first horses reached the Plateau about 1735.
This date is
acceptable for the first acquisition of horses, but there
is good reason to believe that they did not become numerous
or important to the economy until 50 or more years later.
Horses were seen among the Walla Walla Indians at the mouth
of the Snake River in 1805.
Lewis and Clark noted that
Thwaites, 1904, Vol VII, p. 142
Haines, 1938, p. 435
nThe sokulks possess but few horses, the greater
part of their labours being performed in oanoes."109
Yet only a few years later, in 1811, Ross visited the
same vicinity and observed,
plains were literally covered with horses,
of which there could not have been less than four
thousand in sight of the camp."HO
What seems to have taken place is a rapid build-up in
numbers between 1805 and 1811.
The increase would probably
have to be attributed to raids rather than to natural multi­
At any rate there is no direct evidence of many
horses in the Plateau until 1811.
The archaeological picture is much the same.
bones have been found in a number of Plateau sites, particu­
larly in the McNary Reservoir.
Osborne, in a recent paper,
pointed out that in no case were horse bones found in pre­
historic sites, but they were present in two late sites.
In the two cases mentioned by Osborne, sites 45-BN-3 and
45-BN-6 contained large numbers of glass beads, copper and
Sites 45-BN-53, and 35-UM-17, which appear to be
slightly earlier, contained no horse bones.
This small
amount of evidence tends to show the archaeological oc­
currence of the horse to be little if any earlier than
1800 A.D.
Returning to Haines1 statement regarding evidence of
109. Biddl-e,. 1904, Vol XI, p. 191
110. Thwaitos, 1904, Vol VII, p. 138
111. Osborne, 1953, p. 262
the horse in the Plateau by 1835, it should be noted that
this observation was partly based on a statement made to
David Thompson in 1787 by an Indian who. he estimated to be
75 or 80 years old.
If the Indian, who recalled seeing
horses in his youth had been a little younger than Thompson's
estimate, and if his memory had not been of the best, the
first occurrence of horses may have been nearer 1750 than
This later date tends to fit better the archaeological
and ethnological evidenoe.
Travel on horseback did not, however, replace oanoe
travel for Parker noted,
"My three Indians were well acquainted with the
river and the art of managing the canoe."113
Trade in the Plateau was most important at The Dalles,
a situation recognized to date from prehistoric times.
Americans and the British, however, came out to meet the
They established posts in various parts of the
Plateau, and traveled up and down the rivers, trading
as they went.
The fur trade which began on the coast in
1790 and in the Plateau by 1810, did not last much beyond
Several ceremonies were witnessed by the early travelers,
Haines, 1938, p. 435
Parker, 1844, p. 133
but only a few of them were described in any detail.
use of burial goods has been oited, and it conforms to
archaeological patterns.
Other ceremonies however are less
easy to recover archaeologically.
The "first fruits" cere­
mony was observed by Lewis and Clark in 1806.
They wrote,
"The whole village was filled with rejoicing
today at having caught a single salmon, which was
considered as the harbinger of vast quantities in
four or five days."114
Puberty rites were also recorded for the early his­
torical period.
In 1806, Lewis and Clark wrote,
"The daughter of the man is now about the age of
puberty, and being incommoded by the disorder incident
to that age, she is not permitted to associate with
the household or kitchen furniture, or to engage in
any occupation."!15
Lewis and Clark also mention several dances, but failed
to describe them fully or determine their function.
Biddle, 1904, Vol III, p. 64
Ibid, p. 89
The argument over whether the Plateau should "be con­
sidered a separate culture area or a peripheral area is not
merely academic.
It is not dependent entirely on the defi­
nition of what a culture area should entail, but on an
understanding of Plateau Culture and Plateau Culture History#
Certain misconceptions in the past have oaused a few anthro­
pologists to overemphasize extra-areal influence on the
One of the first to disparage the individuality
of the Plateau was Spinden, who characterized the Plateau
from his observations among the Nez Perce'.
He wrote,
"The cult-are of the Basin area, as shown by one
of its representative tribes, was purely a transitional
culture. Its elements were drawn in nearly equal
proportion from the Plains and from the Pacific Coast.
Only a small residuum autochthonous ideas are found
when the borrowed ones are excluded."116
Such an extreme view was not shared by Spier, who refuted
the importance of the overlay of Plains traits,
"There can be little doubt that much, if not all
the overwhelming part of the Plains Traits among thera
(Nez Feres'), dates only from the introduction of the
horse into the Snake-Columbia basin sometime between
1750 and 1800."117
In addition to the fact that the list of traits from
the Plains is of recent acquisition, there is another con­
It seems to be common practice in the space-
time consideration of traits, to look for sources.
If a
trait appears in a relatively unknown area, someone always
seeks to trace it to an adjacent area which happens to be
better known archaeologioally and ethnologically.
This is
done in spite of a complete lack of any information on how
old the trait may be in either area.
Some of the Plains
traits in the Plateau can be shown to be recent, and to have
been derived from that area.
The horse, gun, tribal organ­
ization, honors by war achievement and the elbow pipe have
been proven to be of Plainj origin.^®
Kroeber apparently was not willing to oonoede that the
Plateau was an authentic culture area.
He considered it
to be a hinterland of the Northwest Coast and influenced
by the Plains Area since,
- - the relatively poor subsistence conditions
and consequent low level of culture along the Columbia
and Snake would have strained out many of the more
specialized traits, and most or all of the luxury
developments, of both eastern and western culture."119
There is no doubt, now, that Kroeber made two mistakes.
Since there were few data available on the culture of the
Plateau, he assumed that it did not exist, except on a very
low plane.
Secondly, he was apparently misled by Boaz
glowing descriptions of northwest Coast culture.
the spectacular
ceremony, the large plank houses
Ray7 1939
Kroeber, 1931, p. 57
with elaborate wood carving and the seagoing canoes, the
Northwest Coast had only a few high cultural attainments, and
these were restricted to a minority of coastal peoples, the
Haida, Tlingit and a few others.
Most of the Plateau cere­
monies were never recorded for comparison#
Lacking wood
over most of the Plateau, the Indians did not develop carving
or elaborate houses.
Whether they did or did not, does not affeot the
question at hand.
Relative attainment of cultures is not
involved, only classification.
If Wissler's concepts were
followed no one would question the fact that the Plateau
was a true culture area during the "ethnological present",
the era referred to here as the early historical period.
The problem has been to determine what the cultural situation
was in prehistoric times.
Can this be done through the
historical approach of archaeology?
Prom the data presented in the body of this paper,
certain significant points may be made concerning the Plateau,
both prior to and subsequent to European contact.
in the late prehistoric period:
Cultural development leading up to the late pre­
historic period was measured and progressive.
Material culture was uniform
Economy was similar over the entire area.
Influences from outside the area were much less
important than in the early historic period.
Similarly, in the early historic period:
Plains traits were encroaching on the southeastern
part of the Plateau.
Northwest Coast traits were influencing the south­
western part of the Plateau.
Influences from both areas were recent and super­
Political organization was homogeneous.
Social organization was uniform.
Material culture had widespread similarities.
Economy was the same over the whole area.
In order to present the data with v/hich the nature of
Plateau culture can be discussed, it is neoessary to review
the sources from which they were drawn.
The new materials,
from which much of the discussion has been drawn, are the
results of exoavations and surveys performed by River Basin
Surveys between 1947 and 1952.
Prom these materials, it was
possible to construct a local sequence for the McNary region,
and present a reasonably complete picture of Indian life
during the late prehistoric period.
By "reasonably complete"
is goes without saying that no archaeological reconstruction
can really be complete.
Comparative archaeological data were drawn from nearly
every section of the Plateau, north, east and west of the
MclTary Reservoir.
The most useful data were those obtained
from a series of excavations in a single region; as for
example, the work at The Dalles, by Strong, Schenclc and
S t e w a r d , a n d that of Collier, Hudson and Ford on the
Upper Columbia River
The local sequence developed for the McNary Reservoir
does not begin with "early manM»
Paleo-Indian sites are
present in the Plateau, but have1 not been found in the McNary
The earliest culture in that region seems to bear
a resemblance to the culture of the Great Basin.
It was
surely non-sedentary and more adjusted to hunting than to
fishing or gathering.
The fact that fish and shellfish were
"eaten indicated that some adjustment to Plateau ecology had
been made.
Following a heavy fall of volcanic ash, members
or descendants of the same culture took up residence at Cold
At this time specialized fishing tools began to be
made, and a suitable house type was constructed.
From that
time, up to about 1800, and the appearance of Europeans, the
material culture became more diversified and complicated.
Highly specialized tools, houses, fishing apparatus and burial
procedures were adopted.
European tools and ornaments did
not immediately replace those of the Indians,.but after 1805,
no new aboriginal innovations appeared.
Elaborate stone
carving was probably a late development elsewhere in the
Plateau, but it never became popular in the McNary region.
The local sequence in the McNary region, even though
dates are not available, shows a gradual development from
Strong, Schenck and Steward, 1930
Collier, Hudson and Ford, 1942
simple to complex.
Although there seemed to be two horizons
in which a number of new artifacts appeared, at Cold Springs
and again, at the beginning of the late prehistoric period,
there i3 no reason to believe that cultural changes of a
revolutionary nature took place.
There is no evidence of a
migration or of a new culture appearing on the scene*
persistence of old artifact types and the gradual acqui­
sition of new ones tends to confirm a local development of
Turning to the Plateau in general, it was found that
only for the late prehistoric period, was there comparable
material culture.
Architecture could not be treated, for
house remains had not been excavated in other regions.
Material culture trait lists, comparing other regions of the
Plateau with the McNary region, found much in common in
every case.
Not only were the artifacts of nearby sites
similar, but also those of British Columbia, 350 miles away.
Close approximation in simple generalized artifacts might
be expected, whatever the cultural situation.
A wide distri­
bution of complex and specialized artifacts, however, re­
quires some explanation.
ized and unspecialized
The following table lists special­
artifacts that are more or less typi­
cal of the Plateau during the late prehistoric period.
every type is typical of every region, however.
Generalized Artifacts
Specialized Artifacts
Cobble hammers
Cobble choppers
Lap stones
Stone bowls
Side notched points
Leaf shaped blades
Shouldered points
Corner notched points
Polished pestles
Spall knives
End sorapers
Splinter awls
Polished awls
Bone beads
Marine shells
Grooved net weights
Notched net weights
Discoid choppers
Shaft smoothers
Lozenge blades
Polished celts
Tubular pipes
Pish spear tines
Digging stick handles
Horn wedges
Bone dice
Bone whisties
If information on architecture and burial customs were
available, the table would probably be much larger.
distribution of chipped stone fetishes may have been Plateauwide, but they were probably not recognized by the earlier
workers in the area.
The least than can be said of the
table is that is shows close internal consistency in the
Plateau, as far as material culture is concerned.
It was possible to make inferences on aboriginal
economy from two sides, the artifacts recovered and the
bones of fish and animals that were used for food.
region in the Plateau showed, where- data were available,
that hunting, fishing and gathering were of nearly equal
No region showed any significant departure
from the pattern.
The early historical period in the Plateau was one of
European fur traders were active over the
entire area and some influence, especially in art forms,
was being felt from the Northwest Coast.
Plains influence,
which probably began about 1750, has been much discussed.
The traits which accompanied this influence were those as­
sociated with the horse and gun.
They included tribal
organization, status and chieftanship achieved by success
in war, and Plains-type warfare.
Actually, by the time that Plains traits reached the
Plateau in strength, European acculturation had also takew
Ray, in his thorough analysis of Plateau cultural
relations, found strong Plateau individuality in the northern
and central portions of the area.-1-22
Plains type political
organization was found among the southern tribes, but weakly
Coastal influences, in the form of rank and
caste attitudes, had been accepted only among the Indians
of the southwest corner of the Plateau*
The central Plateau
Indians were completely opposed to warfare, and those of the
far north had no knowledge of Plains-type warfare.
According to Ray, political organization of the Plateau
was based on village autonomy.3*23
Chieftanship was deter­
mined by heredity, and social equality was the Ideal.
Ray, 1939
Pacifism was typioal of the entire Plateau, until a few of
the eastern tribes "began to adopt some of the traits of typi­
cal Plains warfare.
Ray's analysis of political and social organization in
the Plateau demonstrates, beyond any doubt, the individuality
of Plateau culture.
He concludes,
"The plateau is seen to possess distinctive char­
acter in its own right. Many aspects of culture which
are integral to its organization are not to be found in
adjacent areas. This indicates both the individuality
of the area and the relatively slight influence which
the region has exerted upon neighboring areas. This
is not to deny the existence or importance of cultural
elements of foreign origin. The Plateau has borrowed
much from the Coast and it has taken much from the
Plains but the importance of coastal influence has
probably been overrated, and diffusion from the Plains
is in large part recent and superficial."124
The refutation of Spinden^ speculation on the peri­
pheral nature of Nez Perce* and the other Plateau cultures
was a matter of the application of new data to the problem.
Kroeber's concepts regarding the Plateau stemmed partially
from the lack of data and partially from an undue emphasis
on cultural attainments.
If the impressions of early European visitors to the
area can be accepted, the people in the Plateau had physical
and cultural traits which distinguished them as a group;
traits whioh set them apart from the Northwest Coast, the
Northern Great Basin and the Western Plains.
Lewis and
Clark and Parker noticed these traits, and commented on
Ray, 1939, p. 145
them loqg before anyone developed the concept of culture
The most outstanding of these traits, according
to Parker were : 1.
Non-sedentary life among the Plateau
people as opposed to the coast,
high standard of living
as opposed to the poverty of the Basin Shoshoni and,
general oleanliness and health as opposed to both Basin
and Coast.-1-^
The economic pursuits of fishing and gather­
ing were sufficient for Parker to differentiate between the
Plateau and the Plains Indians.
These and other Europeans
also remarked on the uniformity of dress, tools, economic
pursuits and religion.
With the information now available it should be clear
that the Plateau in early historio times, was a distinct
culture area.
It.had uniformity of both climatic and
cultural traits and it contrasted with adjacent areas.
The homogeneous state of Plateau culture in the early
historic period must also have existed in the late pre­
historic period.
As widespread and identical as were the
patterns of politioal and social organization, it is im­
possible to conceive of a situation any different only a
few years earlier.
It is true, that in 100 years, the
Plains Area changed into a relatively homogeneous culture
area, but that area was unified by the horse, the gun and
Plains-type warfare.
In the Plateau, pacifism was the idea}.,
Parker, 1845
and no such unifying force could have existed.
Without the
influence of Plains traits, the Plateau in the late pre­
historic period was certainly more clearly a culture area
than it was in the early historic period.
If, by definition, a culture area should have homo­
geneity of culture and olimate, as well as a distinct culture
history, the Plateau certainly qualifies.
The excavations
made by River Basin Surveys demonstrated how the Indians of
one region progressed from a relatively simple hunting and
gathering economy to one far more complex.
During this
period, the inventory of material culture grew by addition
and diversification, and a superior adjustment to the en­
vironment was made.
Throughout the span of time that was
recorded by the McNary sites, there was demonstrated a conservativeness on the part of the local inhabitants.
artifacts were constantly added, but the older tools that
seemingly should have been replaced, were not readily
This retioence to drop the apparently obsoles­
cent tools does not indicate a people who were dependent
upon outsiders for their culture.
reoeivers of ideas.
They were not entirely
Indeed, it oan be shown that the most
characteristic artifacts of the Plateau were probably in­
digenous, and that the adjacent areas may well have drawn
on the former for a significant amount of their material
Plateau material culture in the late prehistoric
period, in spite of small local variations, was homogeneous
over the entire area and the influences from the Northwest
Coast and the Plains had not yet made significant pene­
trations of the area.
Berreman, J.
"Tribal Distributions in Oregon", American Anthro­
pological Association Memoir, Menasha
Biddle, Nicholas
History of the Expedition Under the Command of
Captalns'Tiewis and Clark, Vol• II, New York
Carithers, W.
"Pumice and Pumicite Occurrences of Washington",
Report of Investigations No# 15, Division of Mines
ana Geology, State of WasETngioh
Chauberlain, A.P.
"The Kootenay Indians", Annua1 Archaeological
Report 1905, Toronto
Cline, Walter
"The Sinkaietk or Southern Okanogan of Washington",
General Series in Anthropology, No. 6, Menasha
Collier, D., A. Hudson, and A* Ford
"Archaeology of the Upper Columbia Region",
University of Washington Publication in Anthro­
pology, Vol• IX No. I, Seattle, Washington
Cressman, L. S.
"Early Man in Oregon", University of Oregon Mono­
graphs No. 3, Eugene, Oregon
"Archaeological Researches in the Northern Great
Basin", Carnegie Institution of Washington. Pub.
538, Washington
"Archaeological Research in the John Day Region
of North Central Oregon", Proceedings of the
American Philosophical Society, XCIV," ITo. "4,
Daugherty, R. D.
"Arohaeologioal Investigations in O'Sullivan
Reservoir, Grant County, Washington", Amerioan
Antiquity, Vol. XXVII, No. 4, Menasha, Wise.
Daugherty, R. D. (CorjH)
"The Archaeology of the Lind Coulee Site, Wash­
ington", Paper read to the Annual meeting of the
American Anthropological Association, Tucson,
Drucker, P.
"Appraisal of the Archeologioal Resources of the
McNary Reservoir, Oregon and Washington", Columbia
Basin Project, River Basin Surveys, Smithsonian
Garth, T. R.
"The Middle Columbia Cremation Complex", American
Antiquity,, Vol. XVIII, No. 1, Salt Lake City, Utah
"Archaeological Excavations at Port Walla Walla",
Pacific Northwest Quarterly, Vol. XLIII, No. 1
Seattle", Wasnington
Haines, Francis
"The Northward Spread of Horses Among the Plains
Indians", Amerioan Anthropologist, Vol. 40, No. 3
Henry, A. and D. Thompson
New Light on the Early History of the Greater
Northwest,"TlE&ited by Elliot Coues), "New York
•Hosmer, J. K.
History of the Expedition of Captains Lewis and
Clark, cKToago
Jaoobs, M.
"A Sketch of Northern Sahaptin Grammar", University
of Washington Publications in Anthropology, Seattle
Jenness, Diamond
"The Indians of Canada", Bulletin, National Museum
of Canada, No. 65, Ottawa
Krieger, H. W.
"Smithsonian Institution Exploration and Field
Work, 1927" Smithsonian Institution Miscellaneous
Collections No. 78
"A Prehistoric Pit House Village Site on the
Columbia River at Wahluke, Grant County,
Washington", United States National Museum Publi­
cation No. 73, Washington
Kroeber, A. L.
"The Culture-Area and Age-Area Concepts of Clark
Wissler", Methods in Social Scienoe, (S. Rice
editor), University of Chicago Press
"Cultural and Natural Areas in Narive America",
University of California Publications in Amerioan
Archaeology and Ethnography» Berkeley,T?alifomla
Lewis, A. B«
"Tribes of the Columbia Valley and the Coast of
Washington", Amerioan Anthropological Association
Memoir No. 1, Lancaster
Martin, G. C.
"The Recent Eruption of Katmai Volcano in Alaska",
The National Geographic, Feb., 1913, Washington
Martin, P.S., G.I.Quimby, and D. Collier
Indians Before Columbus, University of Chicago
Mills, J. and C. Osborne
"Material Culture of an Upper Coulee Rockshelter"
Amerioan Antiquity, Vol. XXVII, No. 4, Salt Lake
City, Utah
Osborne, H. D.
"An Appraisal of the Aroheological Resources of
the Ice Harbor, Lower Monumental, Little Goose and
Lower Granite Reservoirs, Snake River, Washington',1
Columbia Basin Frojeot, River Basin Surveys,
Smithsonian Institution
"The Archeologioal Investigations of Two Sites in
the McNary Reservoir, Washington", Columbia Basin
Projeot, River Basin Surveys. Smithsonian
"Excavations Near Umatilla, Oregon: The Archeo­
logy of the Columbia Intermontane Province".
Unpublished Ph» D. Dissertation. University of
California, Berkeley, California
"Archaeological Occurrences of Pronghorn Antelope,
Bison, and Horse in the Columbia Plateau." The
Scientific Monthly, Vol. LXXVII, N.Y., N.Y.
Osborne, H. D., R. Crabtree, and A* Bryan
"Aroheological Investigations in the Chief Joseph
Reservoir", American Antiquity, Vol. XXVII, No. 4
Salt Lake City, Utah
Osborne, H. D. and J. L» Shiner
"River Basin Surveys-State College of Washington
Aroheological Exoavations in the Lower MoNary
Reservoir, Oregon", 1949, Columbia Basin Project,
River Basin Surveys, Smithsonian Institution
"The 1950 Excavations in Two McNary Sites, Wash­
ington and Oregon", Columbia Basin Project, River
Basin Surveys, Smithsonian institution
Parker, S.
Parker's Exploring Tour Beyond the Rocky Mountains,
Ithaca, New York
Ray, V.P.
"The Sanpoil and Nespelem: Salishan Peoples of
Washington", University of Washington Publications
in Anthropology, vol. v.""Seattie. Washington
"Native Villages and Groupings of the Columbia",
Pacific Northwest Quarterly, Vol, XXVII, Seattle,
"Cultural Relations in the Plateau of Northwestern
America", Publications of the Frederick Webb Hodge
Anniversary Publication"TuncT7 Vol, III. Los Angeles,
Rollins, P. A.
Editor, The Disoovery of the Oregon Trail, by
Robert SEuart, flew YorE
Ross, Alexander
Adventures of the First Settlers on the Oregon
or Columbia""Kiver, London
The Fur Hunters of the Far West, London
Sapir, E.
"Wishram Texts", American Ethnological Society
Publication, No#"~2
Shiner, J. L»
"An Appraisal of the Archeological Resources of
the John Day Reservoir on the Columbia River,
Oregon and Washington", Columbia Basin Project,
River Basin Surveys, Smithsonian,' Institution
"Hobo Cave; A Stratified Site on the Columbia
River", TJnpublished Manuscript, University of
"The Excavations at Site 35-UM-5 in the McNary
Reservoir, Oregon", Columbia Basin Project, River
Basin Surveys, Smithsonian Institution
"A Preliminary Report on the Archeology of Site
45-WW-6 on the Columbia River, Washington",
Columbia Basin Project, River Basin Surveys,
Smithsonian Institution
"The 1950 Excavations at Site 45-BN-6, McNary
Reservoir, Washington", American Antiquity, Vol#
XVII, No. 4
"Excavations at Site 35-WS-5 on the Columbia
River, Oregon", Columbia Basin Project, River
Basin Surveys. Smithsonian Institution
Smith, H. I.
"Arohaeology of Lytton, British Columbia",
Amerioan Museum of Natural History Memoir. Vol 2.
part 3
. 1900
"Archaeology of the Thompson River Region, British
Columbia", American Museum of Natural History
Memoir, Vol. 2, part 6
"Arohaeology of the Yakima Valley", American
Museum of Natural History, Anthropological Papers.
Vol. 6, part 1
Sperlin, 0. B.
"Exploration of the Upper Columbia River", Wash­
ington Historical Quarterly. Vol. IV. No. 1.
Seattle, Washington
Spier, Leslie
"Klamath Ethnography", University of California
Publications in Arohaeology and kt'hho'gyaphy1;
Vol. XXX, Berkeley, California
Spier, Leslie (Conft)
"Tribal Distribution in Washington", General Series
in Anthropology, No. 3, Menasha
Spier, L. and E. Sapir
"Wishram Ethnography", University of Washington
Publication in Anthropology, Vol. TTl, Seattle,
Spinden, H. J.
"The Nez Perce*.;indians", Amerioan Anthropologi pal
Aasoolation Memoir. Vol. II, part 3, Menasha, Wise.
Strong, W. D., W. E. Schenck, and J. H. Steward
"Arohaeology of The Dalies-Deschutes Region"
University of California Publications in Amerioan
Arohaeology and Ethnology, Berkeley. California
Swanton, J. R«
"The Indian Tribes of North America", Bureau of
American Ethnology, Bulletin 145, Washington
Teit, J. A*
"The Thompson Indians of British Columbia", American
Museum of Natural History Memoir, Vol. II, Part 4
"The Salishan Tribes of the Western Plateaus",
45th Annual Report of the Bureau of American
Ethnology, Washington, D. "Cl
Thwaites, R. G.
Original Journals of the Lewis and Clark Expedition
1861 - 1856, Vol." TTlTo 7oTT"VTT7 ftewTork
Turney-High,.JEI. H.
"The Flathead Indians of Montana", American Anthropologioal Aasoolation Memoir. No. 48, Menasha
Wilkes, G.
"The History of Oregon"> Reprinted in the Washingt on Historical Quarterly. Vol. II, 1/907, Seattle
Williams, H.
The Ancient Volcanoes of Oregon, Eugene, Oregon
Winther, 0. 0.
The Great Northwest. New York
Wissler, C.
The American Indian, 3rd Edition, New York
Work, J,
Indiana of the Uhited States; Four Centuries of
History and Culttire. Hew York
"Journal of John Work'*, Reprinted in Washington
Historical Quarterly, Vol. V, No. 2, Seattle,
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF