« Previous Region | Index | Next Region »

SE Region

Table of Contents:

Lake Abert
Chewaucan Drainage
Dog Lake 35LK1941
Buchanan Springs 35HA64
Drews Summit 35LK2104
Warner Valley
Dunn 35HA1261
McCoy Creek 35HA1263
Big M 35LK2737
Ratz Nest 35LK2463
Paulina Marsh 35LK2676
Fort Rock
Penensula 35LK2579
Birch Creek 35ML181
Paulina Lake Site 35DS34 House Summary


"Winter villages consisted of one or two families up to about five families. Winter house structures varied from area to area, but were usually constructed with circular floors, a central pole, and a conical covering of bark, brush, grass or earth. Most groups did not excavate these floors, but some semisubterranean houses have been noted (Fowler 1982: 122, 133) ... Summer structures were more temporary, often nothing more than a roofless brush enclosure or windbreak" (Oetting & Pettigrew 1985: 7).


[back to top]

Lake Abert Sites

During a survey for a highway project along Lake Abert in 1976, Cole and Pettigrew described a series of sites with houses. 35LK478 contained thirty-four house depressions as well as cairns and a petroglyph. 35LK481 was on two terraces and fifteen housepits were counted with "a circular arrangement of basalt boulders that looks much like the remains of a stone-walled house, complete with entranceway" (Cole and Pettigrew 1976:9). Site 482 had one and perhaps two house depressions. A single house pit was called 35LK483. 35LK487 had ten houses with two rock circles, three rock semi-circles, a curved rock wall and a linear rock wall. Site 488 contained at least eleven housepits; 489 had three; 490 had three rock rings; 492 and 493 had two parallel curving rock walls with a cleared area between then about 15 by 4 meters in size; 495 had eleven house depressions and a rock circle; 496 had five houses in a single row on a terrace; 497 had thirteen houses in a row as well.

35Lk41 stone walled house
Cole & Pettigrew

"Of the twenty-five sites recorded, twenty appear to be aboriginal. Of these twenty, four are petroglyph sites, twelve have housepits, one has no housepits but does have three rock circles that seem to be house structures, and three have no housepits but are terrace sites that have domestic tools characteristic of village sites. Of the twenty aboriginal sites, then sixteen are certainly or possibly village sites....A density of thirteen or sixteen village sites in a distance of eight miles of lakeshore is a greater concentration than has been reported elsewhere in the Great Basin. The 119 possible house depressions that were counted in the survey are a very large sample, considering the relative paucity of such features in the region" (Cole and Pettigrew 1976:23).

Village plans
Lake Abert

In 1985, Pettigrew, Baxter and Connolly published on the Lake Abert east shore in the University of Oregon Anthropology series (#32). The work recorded 441 circular features that were probably houses, storage pits and communal structures. They were found mostly on the primary terraces of Lake Abert. The features ranged from 1 to 12 meters but most were in the 4-6 meter range and those from 3-7 meters were interpreted as house remains. "These structures tend to have entryways on the east or northeast side, have walls of piled boulders that were originally up to 1.5 m high, and often include in their walls a very large, naturally deposited boulder, almost always located at the south side and very frequently incised with petroglyphs. Very often there is a well-used grinding slab or mortar ground onto one of the house-wall boulders at or very near the doorway, and occasionally the stones in the doorway are smoothed, apparently by foot traffic. In at least three cases, several stone rings are joined together, possibly indicating either additional rooms to a single house or several houses with adjoining walls" (Pettigrew, Baxter & Connolly, 1985: 85).

Stone circle
Pettigrew, Baxter & Connolly 1985:24
Map of features
Pettigrew, Baxter & Connolly 1985:32

They found some differences in patterns of feature size by terrace elevation. Below and elevation of 9.5 meters above the current lake, 75% of the features were from 4-6 meters in diameter and overall size is more homogeneous. The larger and smaller features tended to be found above the 9.5 meter elevation. If the lake gradually dried up, the changes may indicate that the proportion of very large and very small features decreased over time. "It may be assumed that different CF (cultural feature) sizes represent structures that were functionally or socially distinct (e.g., houses, storage facilities, sweat lodges or menstrual lodges, communal structures, chief's houses, etc.) If this assumption is made, the settlement at Lake Abert may indicate a larger proportion of such specialized functions at an earlier period, and that this proportion decreased through time" (Pettigrew, Baxter & Connolly 1985: 89).

Pettigrew, Baxter & Connolly 1885:36

"The overwhelming impression obtained from inspection and excavation of the sites and tabulation and analysis of the data is that sedentary or semi-sedentary society maintained permanent and fixed settlements on the lakeshore for a considerable period of prehistoric time" (Pettigrew, Baxter & Connolly 1985:183).

"... the upper terraces represent an early period of abundance, when occupational intensity (in terns of the length of the annual period of occupancy) was relatively great and social differentiation was at a peak. At such a time the functional and status diversity among CFs would have been high, resulting in a greater range of CF sizes. The CFs on the lower terraces, then, were created at a later time when environmental abundance was not as great, occupational intensity was lower, and societal structure was more atomized and egalitarian" (Pettigrew, Baxter & Connolly 1985:186).

I suggest that at the earlier time, when populations were lower and the lake fresher, that resource abundance could have allowed extended family structures and storage pits. At later times, with growing populations and degrading lake resources, that nuclear families and fewer storage features were the norm.

In Vol 2 of the east shore of Lake Abert study, Oetting concluded that "Of the 13 depressions tested, only three contained relatively well defined floors or occupational surfaces and only one pit rim was exposed... All are considered to be the remains of prehistoric pithouses, but at present no evidence for the superstructures over these pits is available.... No functional conclusions could be reached for six, or nearly half, of the depressions tested... Four of the depressions did not function as housepits, based on the archaeological evidence" (Oetting 1988:245-246).

"Internal features were not very common. No postholes were identified, nor were any cache or storage pits. There was no evidence of timbers or other roofing and no entryways were discerned. There was no indication that the superstructure of the houses had burned. Identified floors, fire pits, or rims were, for the most part, rather vague or indistinct" (Oetting 1988:250).

[back to top]

Chewaucan Drainage

During a 1984 survey funded with a SHPO grant, Oetting & Pettigrew recorded sites near Lake Abert in the Chewaucan drainage. They recorded 31 sites with depressions and 2 sites with stone walls. Within the 31 depression sites, they counted over 140 features ranging from a single one to two clusters of 21 and 23 possible housepits. Diameters ranged from 3.5 to 12 meters but most were between 5-7 meters. Depth was generally 50-70 centimeters, but as great as a meter (Oetting & Pettigrew 1985:42-43).

[back to top]

Dog Lake 35LK1941

During a timber sale survey near Dog Lake in 1985, two sites with housepits were recorded. Site 35LK1941 contained two depressions, one 6 meters acroos and the other 3 meters. Site 35LK1944 had two bedrock mortars and tow rock-lined depressions 6 meters in diameter and 50 cm deep (Hoard & Hemphill 1985).

[back to top]

Buchanan Springs 35HA64

Chip Oetting (1987:66-67) reported on a possible housepit at site 35HA64 (Buchanan Springs) during testing for a highway project in Stinkingwater Pass. A test pit hit the edge of a large buried feature that was tentaively identified as the edge of a housepit. The feature was about 30 cm deep and nearly a meter of the feature was exposed in the profile.

[back to top]

Drews Summit 35LK2104

During a survey and evaluation of sites for the Drews Summit section of OR-140, Musil noted site 35LK2104 had three large rock rings with dense surface scatters of lithic debris inside and outside the "house structures". Projectile points 1000-6000 years old were observed associated with the features (Musil 1990:6).

[back to top]

Warner Valley Land Exchange

Bill Cannon recorded sites with rock rings during the Warner Valley Land Exchange project in 1989. Site 35LK2584 had 7 stone rings with associated lithic scatter. 35LK2585 consisted of a ring built against a large boulder with flakes and a bedrock metate. 35LK2588 had three depressions ranging from 2 to 3 meters in size. 35LK2599 had three possible housepits but no measurements given (Cannon 1989).

House profile
Musil 1990b:40
[back to top]

Dunn Site 35HA1261

Robert Musil's (1990) testing project revealed a house floor in the Dunn site (35HA1261). Component II contained the remains of a pithouse. The house had been bisected by a backhoe trench and then excavated with block units. It was 4.1 by 4..3 meters across and 36 cm deep with a central hearth. In addition a storage pit, four post holes, a block metate and a dense concentration of artifacts was associated with the wall slope which was relatively steep. The house was more bowl shaped than saucer shaped. The flat living floor was about 3 meters across. A C14 date of 3255±65 was recovered form the house floor and was consistent with the Elko series points recovered.

House detail
Musil 1990b:44

The hearth was 22 by 15 cm and 13 cm deep. The four post holes were concentrated near a large unaltered rock set in the wall slope. He felt this feature may have been a step with associated side entrance defined by the posts. The floor appeared to have been regularly cleaned and debris tossed onto the slope.

[back to top]

McCoy Creek 35HA1263

The McCoy Creek site (35HA1263) was dug in conjunction with the Dunn site, and reported separately. It was part of a proposed land exchange into private hands. The site had been tested in 1984 and backhoe trenches were combined with block excavations to locate and excavate subsurface features. The trench hit two house floors and a wickiup (feature 11).

east wall profile
Musil 1991:53

In Trench 2, an unexcavated house floor was noted in the profile. The house was 2.3 meters and 24 cm deep, shallow and basin shaped.

House floor #1 was a 5 meter profile consisting of a shallow bowl shaped darker charcoal stained soil. "The combination of evidence provided by the wall profile as Features 1,2,3 and 4. The combination of evidence provided by the wall profile and the position of the various features at basically the same level resulted in the designation of these buried features, and in situ artifacts as House Floor #1 (Feature 9)" (Musil 1991:75). House 2 was a post-field reconstruction of a house floor derived from study of the profiles and association of features 6,7 and 8 and in situ artifacts recovered during excavation. It was clearly associated with feature 6, a hearth.

West wall of house
Musil 1991:60

The most interesting feature was 11, the wickiup exposed in Trench 2 in profile, and then block excavated. "At the edges of the wickiup depression a number of small burnt poles were uncovered in situ where they had apparently burnt off in the postholes and fallen towards the center of the floor. A total of 10 small burnt posts were exposed at the edge of the circular wickiup depression. The scattered remains of a number of burnt poles, undoubtedly portions of the posts and superstructure of the roof, were found lying on the wickiup floor" (Musil 1991: 78). The posts were identified as willow and mock-orange and were about 5 cm in diameter. "Near the northern edge of the wickiup in Units 3 and 4, a small concentration of burnt grass stems and charcoal flecks was encountered lying on the sloping floor. This material was identified as burnt stems and tissue of wheatgrass. The thatch-like nature of this grass concentration suggests that it was the remains of a twined or bundled mat of some kind" (Musil 1991: 79).

The feature was 3.4 by 4 meters with a basin-shape that reached a maximum depth of only 19 cm. Two shallow pits had been dug in the floor. A large central hearth was found containing charred sage and bunchgrass. The points recovered in the house were small projectile points from the Rosegate and Desert Side-notched forms. A C14 date of 480±70 was derived from charcoal on the floor. The house was a typical winter wickiup known historically.

Wickiup feature 11
Musil 1991:81

House floor #1 was associated with a C14 date of 1270±70 and house #2 with dates of 990±90, 1340±?, and 1140±60. The hearths contained seeds such as goosefoot, knotweed, dogbane, fleabane, and wheatgrass (1991: 202).

Musil notes that the differences in the house forms and the sequence corresponds to the theorized Numic expansion and a change from a more sedentary and intense economic strategy to a more mobile and extensive economy relating to drier conditions (1991:227). The author points out the need to locate and excavate more houses from this critical period of time to establish a sequence of changes at regional levels.

[back to top]

Big M 35LK2737

Jenkins (ND) summarized the University of Oregon field schools of 1989-91 done in cooperation with the Lakeview BLM in the Fort Rock Basin. He hypothesized that : "Middle Archaic villages were located along marsh peripheries, particularly in ecotonal settings at the base of ridges. relatively small populations were 'mapped on' stable, concentrated, storable lowland resources. Reduction in the reliability of these resources and increases in human populations led to a gradual shift in residence to upland root grounds beginning around 3,000 BP. Late Archaic populations were 'mapped on' upland root crops until about 900 BP when drought greatly reduced lowland resources causing local populations to disperse. A final surge in population may have occurred about 700 to 500 BP. Upland settlements during this time period may have served as spring/summer camps in a seasonal village pattern similar to that of the ethnohistoric Klamath/Modoc. Sometime after 500 BP the area was no longer occupied by the Klamath and the Northern Paiute apparently moved into the vacuum left by their departure" (Jenkins ND: abstract).

Excavations at the Big M site (35LK2737) encountered a structure identified as a wickiup. It was oval, 3.5 by 4.5 meters and defined by a hard packed soil. There was suggestion of a doorway in the NE and a nearby hearth inside the structure. A C14 date of 4880±110 and Northern Side-notch, Elko and Gatecliff points indicate an early date for a "wickiup" structure. The absence of superstructure remains makes its interpretation as a wickiup questionable.

Later excavations encountered a shallow house pit 6 by 5 meters and 35-40 cm deep. A ramp-like feature in the NE suggested a doorway. A bulk sample produced a date of 3530±100. Artifacts included a Humbolt and Elko points. A second house floor gave a date of 4550±120 associated with a Northern Side-notched point.

Structure 3 Big M site
Jenkins ND

"In short, the Big M site appears to have been a small village site occupied by 5,000 RCYBP and periodically reoccupied until at least 3,500 RCYBP. Small to medium sized houses, up to 6 m in diameter and 25 to 40 cm deep, were excavated into the culturally sterile substratum of the site. One of these houses was reoccupied at least twice. The cultural materials sampled by our excavations apparently relate to 2 periods of relative sedentism when the site was likely to have been occupied for a substantial portion of the year (Jenkins ND: 22).

Ratz Nest house
Jenkins 1989:9
[back to top]

Ratz Nest 35LK2463

The 1989 University of Oregon field school excavated a small semi-rectangular house ring 2.5 meters in diameter in the Ratz Nest site (35LK2463). The walls were dry-laid basalt boulders reaching a maximum height of 70 cm. Two C14 dates were obtained from the house floor: 500±70 and 110±60. Small narrow-necked points were recovered. "The presence of charred juniper berries in the ashy lens roughly 10 cm below the surface and 5-10 cm above the house floor suggests the super-structure was at least partially comprised of juniper branches.. Soil flotation samples, processed by Nancy Stenholm at Botana Labs for identification of botanical remains, produced evidence for juniper, ponderosa pine, Douglas fir, aspen and/or willow, elderberry, sagebrush, and maple (Stenholm 1989). The juniper and pine probably provided the main structural support for the house. They may have also served as fuel for cooking fires, as the sagebrush, aspen, and elderberry probably did. The sagebrush, aspen, and elderberry, being lightweight, and structurally too weak for use as supports, were probably also used as wall and roofing materials. The maple is not believed to be indigenous to the area and probably was brought to the site in the form of various wooden tools or stone and bone tool handles" (Jenkins 1989: 8).

The floor of the house had been subjected to a great deal of rat turbation, hence the name for the site. Artifacts included drills, scrapers, a bone bead, utilized flakes, and two grinding slabs. Nearly 1500 bone fragments were recovered, but not subject to analysis for this report.

[back to top]

Paulina Marsh 35LK2678

During the survey of Paulina Marsh, Aikens and Jenkins reported a site with a single depression (35LK2678) 6 m in diameter and 30-40 cm deep associated with Rose Springs points, mortar and metate fragments (Jenkins & Aikens 1991: 14).

[back to top]

Fort Rock Valley

Jenkins, Prouty, McDowell & Singer (1994) placed test excavations in nine sites in the Fort Rock Valley for a private corporation mining claim on BLM lands. The Bowling Dune Site was trenched, revealing semi-subterranean houses. "Features 1 and 5 represent interior features of a semisubterranean house (Feature 2) that was approximately 4.5 meters in diameter and had a shallowly excavated floor 45 to 50 cm deep. Feature 1 was a small hearth or post hole roughly 25 cm in diameter and 8 cm thick. It was located at the base of the west wall of a small structure. Feature 1 was composed of a pocket of charcoal situated below the house floor at a depth of roughly 75 to 83 cm. A sample of charcoal recovered from Feature 1 produced a date of 2830±70 BP (Beta-67520)" (Jenkins, Prouty, McDowell & Singer 1994:5).

The houses were associated with a number of large cache pits, suggesting substantial storage capacity. If the tested pits were used at the same time, they would have stored 7.3 m3 of food and materials (Jenkins 1994:11).

[back to top]

DJ Ranch Site

At the DJ Ranch site, "A portion of what appears to be a sloping house floor was located at a depth of 140 cm near the west wall of Unit M. It was exceptionally light colored and unusually compact. A Gatecliff Split Stem projectile point and an Olivella shell bead were recovered from its surface. Two, or possible three, thin lenses of charcoal were distinguishable in the wall above this feature suggesting multiple occupations of this apparent house floor" (Jenkins, Prouty, McDowell & Singer 194:18). Nearby C14 dates of 4900±80 BP (Beta-67527) and 4050±100 (Beta-71832) were taken.

[back to top]

Penensula Site 35LK2579

In her University of Nevada, Reno, thesis, Sunday Eiselt (1997) summarized the work at the Peninsula Site (35LK2579). It is described as "an intensively occupied, possible winter aggregation site, based on the nature of the site structure, diversity of artifact types, and zooarchaeological and microbotonical remains. Depressions lie at the bases of hillocks where sediment deposition is high. Twenty-one such depressions were noted during 1990 and 1996 mapping activities, and three were sub-surface tested during the 1990. 1995 and 1996 field seasons" (Eiselt 1997: 124).

Depression 1 contained in-situ burned logs, mats and had artifacts on the house floor. The depression was 3 by 5 meters. The north wall was steep-sided, but the south was a shallow disc-shaped wall. No post holes were recorded, but "large, subangular rocks (measuring 15-20cm in width) were found in clusters along the periphery just outside of the depression in what may be formations related to structure pole support" (Eiselt 1997: 129). A 40-50 cm wide bench was found along the north and east walls containing a metate and a cache of tools.

Depression 1
Eiselt 1997:130

"Burned logs and matting capped the floor.... The structure was made by first creating a skeletal frame of western juniper (Juniperus occidentalis). Two layers of twined matting were then placed over the frame and a layer of grass was loosely piled to cover this. The layers of matting in relation to the frame are such that the first layer was laid vertically, the next, horizontally, and the last layer of piled grass was laid vertically to the ground, probably to shed rain water" (Eiselt 1997: 130-132). This hade been covered with a daub earthen cover that baked hard when the structure burned. No hearth was observed. Radio-carbon dates of 625±90, 450±80, 440±80 and 240±65 were obtained.

House reconstruction
Eiselt 1997:137

The reconstruction is a triangular tent-like structure in sloping ground with a variable depth pit.

Depression two was sampled with six 1 by 1 meter units. It was disk-shaped and similar in shape and size to depression one. Depression three was tested with five 1 by 1 units. The depression was 35 cm deep, and "a trampled, depressed area was identified in the field as a doorway to the depression situated on the east side facing Hart Mountain" (Eiselt 1997: 142).

A single rock ring was also noted and tested with a single unit. The results were unclear if the feature was modern or ancient.

"Depression one yielded a fairly intact structure that, when compared to ethnographic accounts, shows marked similarities with both Klamath and Northern Paiute houses. Klamath influences are seen in the oval planview shape, square-sided walls on the north and east sides, the presence of benches, techniques of matting construction and layering, and the application of an earthen covering. Paiute influences include the presence of a disc-shaped wall on the south side, and the application of woven matting. Some variability in depression one characteristics could also be related to seasonal practices since the structure displays characteristics of both winter and summer houses. It is apparent, however, that the depression one feature is a house structure used for protection from the elements and domestic household activities" (Eiselt 1997:150).

[back to top]

Birch Creek 35ML181

The work at Birch Creek (35ML181) by the Center for NW Anthropology (WSU) under Andrefsky encountered two and perhaps three house pits in a long stratigraphic trench dug by backhoe. One of the houses was partially excavated within a subsequent block. The house contained at least four living surfaces representing several use and abandonment (evidence for the house being burned twice) sequences. The deepest house occupation level "D" was dated 4480±70, while an upper level "B" was dated to 2420±70 and 2410±60 (see illustration of levels below).

"Single and double house pit sites have been found to occur as early as 5,500 year (SIC) ago on the Columbia Plateau and even earlier in parts of northern California. However, house pit villages, or sites that have dozens of house pits used contemporaneously are first dated to about 2,300 years ago on the Plateau" (Andrefsky & Presler 2000: 89).

"Again, if there is a house pit village at 35ML181, and preliminary analysis indicates the two and possible many more house pits are located on the site, it is at least 2,500 years old. All other house pit villages on the Plateau are dated after 2,500 year (SIC) ago. House pit villages on the Plateau are always associated with bow and arrow technology, which begins at approximately 2,500 years ago. Initial laboratory sorting of artifacts from the 1998/1999 season indicates no evidence of bow and arrow technology at the site"(Andrefsky & Presler 2000: 89).

House profile
Andrefsky & Presler 2000:69

The site is located on a riverine shoreline terrace and is about 60 by 200 (12,000 m2) in size so only a very small sample has been taken. Preliminary results included 101 points, 67 point tips, 19 drills, 254 bifaces, 33 scrapers, 277 cores, 37 pebble cores, 5 pestles, 13 mortars, 27 hammer stones, 100 other ground stone objects, 1 shell ornament, 795 other pieces of shell, 11 bone awls, and 11 bone beads.

Photo of house
Andrefsjy 7amp; Presler 2000:67
[back to top]

Newbery Crater: Paulina Lake Site (35DS34)

Thomas Connolly and the Oregon State Museum of Anthropology under an ODOT contract, excavated the earliest dated surface house in Oregon at Newberry Crater. The Paulina Lake Site (35DS34) encountered a wickiup-style house or windbreak with a central firepit and five charred support posts in a semi-circular pattern. Three of the posts (20 cm in diameter) were dated: 8540±90, 8460±110, and 8670±110 RCYBP. The central hearth was about 30 cm deep and 70 cm in diameter and was dated to: 9060±80 and 8080±110 RCYBP.

Paulina lake site house
Connolly 1999:104

The projectile points from the pre-Mazama assemblage included 113 projectile points: Western Stemmed (Windust) and unstemmed foliate/lanceolate (Cougar Mtn, Haskett) points.

[back to top]

Houses and Pole & Thatch Houses

Eiselt did a comparison table on Klamath and Northern Paiute houses (1997: 151):

FormKlamathKlamathN. PaiuteN. PaiuteDepression One
Above FormConicalConicalDomePorchUnknown
PlanviewOvalOvalCircularNone visibleOval
Dimensions4-10m3-10m3-10mUnknown 3x5m
Depth50-100cmNo PitShallow PitNone20-30 cm
PostholeCentral/PeripheryCentralPeripheryPeriphery None
MattingTwined/SewnWoven MatsTwined/SewnTwined/PiledWoven Grass Piled
HearthCentral No pitNoneCentral PitNoneNone
Storage PitsYesUnknownUnknownUnknownNone
BenchesPeripheryUnknownNoneNone One Side
CoveringEarthBanked SidesOccasionally NoneBanked Sides

Willig (1981) summarized Great Basin pole & thatch houses as a master's paper at the University of Oregon. She notes the unroofed windscreen, the domed pole shelter, the conical pole shelter and subconical (open top) pole structure. The house excavated at Dirty Shame Rockshelter was used as the primary example. The house was over a shallow saucer shaped depression 5 meters in diameter. Paired willow poles (2-2.5 cm across) had been placed into the ground about every 30-40 cm and lashed horizontally with branches tied with twigs or withes of willow. Bundles of Great Basin rye grass were then lashed to the framework with willow pins bent in a U-shape. There was a central fire hearth, and it had a side entrance. The thatch was dated at 1140±95, 1175±70 and twigs or poles at 1480±75, 1715±70, 2005±75 and 2545±80.

Thatch house
Willig 1881:12b

« Previous Region | Index | Next Region »