Luu^

TELDIANA

Anthropology

SERIES, NO. 35

Villages Before Aggregation: The Merrigan Site (LA 110971), a Developmental Period Hamlet, El Rancho, New Mexico

Winifred Creamer Jonathan Haas

June 30, 2003 PiibliiJifion 1523

PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY

Information for Contributors to Fieldiana

Fieldiana is a pccr-rcvicwed monographic series published by the Field Museum. The series publishes the research of staff members and our research associates. The page charge currently is $65 per page. This figure is subject to change. All authors are encouraged to provide funding to support the production of their works.

Submission procedures: A submission procedures document is available from the scientific ditor of the journal and on the Museum's web site (by 2004). Manuscripts submitted for review hould be prepared using standard electronic software and submitted to the scientific editor in liree photocopies. The scientific editor also requires the names, addresses, telephone numbers, and e-mail addresses of at least five suggested reviewers. All elements of the work, including the illustrations, must be completed before the manuscript is submitted. References should be styled according to a recent Fieldiana publication, which may also be consulted for order of elements (title page, table of contents, list of illustrations, abstract, and the like).

The decision of the scientific editor to accept or reject a work is final.

Length: As a monographic series, Fieldiana publishes mid-length works. We do not publish short, joumal-article-length works. As a general rule, we are unable to accommodate manuscripts of less than 75-100 manuscript pages. Authors seeking an exemption for a manuscript of fewer pages should consult the scientific editor. Fieldiana also is unable to publish extremely long works without substantial support from the author. The publication of long works should be arranged well in advance with the scientific editor.

Please bear in mind: Changes on page proofs are very expensive. Author-generated changes on page proofs can only be made if the author agrees in advance to pay for them.

UNIVERSITY OF WaJa

LIBRARY AT URBANA-CHAMP/- ^nOKSTACK?

FIELDIANA

Anthropology

NEW SERIES, NO. 35

Villages Before Aggregation: The Merrigan Site (LA 110971), a Developmental Period Hamlet, El Rancho, New Mexico

Winifred Creamer	Jonathan Haas
Department of Anthropology	Department of Anthropology
Northern Illinois University	Field Museum of Natural History
DeKalh, Illinois 60115	1400 South Lake Shore Drive
U.S.A.	Chicago, Illinois 60605-2496
	U.S.A.
Department of Anthropology
Field Museum of Natural History
1400 South Lake Shore Drive
Chicago, Illinois 60605-2496
U.S.A.
Accepted February 5, 2003
Published June 30, 2003
Publication 1523
PUBLISHED BY FIELD MUSEUM OF NATURAL HISTORY

© 2003 Field Museum of Natural History

ISSN 0071-4739

PRINTED IN THE UNITED STATES OF AMERICA

Table of Contents

I. The Merrigan Site 1

Physical Environment 3

Research Objectives 3

Prehistory of the Region 4

Research Approach and Methodology .... 6

II. Excavations 6

Field Methods 7

Architecture 8

Room 1 8

Room 2 10

Room 3 11

Room 4 11

Room 5 12

Room 6 12

Room 7 13

External Features 13

Construction Sequence at LA 1 1 097 1 ... 16

III. Artifacts 18

Lithic Materials 18

Chipped Stone 19

Ground Stone 22

Other Stone Materials 23

Ceramics 24

Unpainted Ceramics 24

Painted Ceramics 26

Other Ceramics 30

Faunal and Macrobotanical Remains 31

IV. Chronology, Seasonality, and

Aggregation 32

Site Chronology 32

Chronometric Dating 32

Relative Dating 33

Seasonality 34

LA 1 1 097 1 and the Aggregation

Process 35

V. Literature Cited 36

Appendix 1 . Participants in the Field

School 38

List of Figures

1. Topographic map of LA 110971, the Merrigan site 2

2. Plan map of excavated structures at

LA 110971 7

3. Section view of Room 1 9

4. Plan and section views of collared

hearth in Room 1 10

5. Plan and section views of exterior

hearth, feature 7 14

6. Plan and profile views of small exteri- or hearth, feature 20 14

7. Trench profile for LA 110971 15

8. Scenarios for occupation of LA

110971 17

9. Chert distribution 20

10. Basalt distribution 21

1 1 . Obsidian distribution 22

12. Obsidian and chert points 23

13. Grooved maul of vesicular basalt 24

14. Ground stone distribution 25

15. Turquoise bead fragment from Room 1 ... 25

16. Culinary ceramic distribution 27

17. Miniature vessel of culinary ware 27

18. Corrugated ceramic distribution 28

19. Corrugated ceramics 28

20. Sherds of Black-on-white types 29

21. Other decorated ceramics 30

22. Corrugated sherds 31

23. Clay pipes 31

24. Archaeomagnetic date 32

List of Tables

1. Features in Room 1 9

2. Features in Room 4 12

3. Features in Room 6 13

4. Exterior features around Rooms 1-7 13

5. Summary by raw material of chipped stone 19

6. Chipped stone by room 20

7. Ground stone 24

8. Hammerstones 25

9. Concretions 26

10. Ceramics 26

11. Sherds identified as "other" 30

12. Fauna 31

13. Macrobotanical remains 32

111

Villages Before Aggregation: The Merrigan Site (LA 110971), a Developmental Period Hamlet, El Rancho, New Mexico

Winifred Creamer and Jonathan Haas

I. The Merrigan Site

The Merrigan site is one of many thousands of small sites scattered across northern New Mexico. These hamlets, occupied by one or a very few families, were the building block of ancient so- ciety in this region. The earliest houses found in northern New Mexico, around a.d. 500, were cir- cular pit houses about 1 meter deep. Around a.d. 1000 rectangular, aboveground rooms were added to these, first as storage rooms and then as living rooms. The Merrigan site includes both circular and rectangular rooms, which suggests that the transition from one type of room to the other was still under way when the site was occupied.

The site was first identified during an archae- ological survey of property, which was about to be subdivided (Legare 1995, 1996). The buyer of the property, Mr. Tom Merrigan, planned to build a house in a location that would disturb the ar- chaeological materials. Construction could not be- gin until the site was excavated, and Mn Merrigan invited the Northern Illinois University archaeo- logical field school to conduct the excavations, as a cost-effective way to carry out the work.

It was the surface ceramics on the Merrigan site that made it of interest as a research project, part of a series of studies carried out by the Northern Rio Grande Research Project (NRGRP). (This project, a collaboration between the Field Muse- um and Northern Illinois University, has focused on the development of large protohistoric pueblo villages and the impact of European exploration on Pueblo life.) Ceramic types recorded by the first survey included St. Johns Black-on-red, Wiyo Black-on-white, Galisteo Black-on-white, glaze-on-red (type uncertain), and Biscuit A. These materials suggested the site's occupation was transitional between the Late Coalition and

Early Classic periods of the regional chronology, approximately a.d. 1300-1360. During this period of time, the first very large pueblo villages were being built, and examination of a small site was expected to yield information about the change from family-sized hamlets to villages with hun- dreds of residents. Although the ceramics on the surface suggested a fourteenth century date, our excavations determined that the ceramics record- ed on the surface were not typical of the site as a whole, which appears to have been in use during the late eleventh century and early twelfth cen- tury. The earlier date is supported by the size of the site and the types of rooms. Rather than focus on the aggregation of small groups into larger ones, we excavated the site to look at the transi- tion from pit houses to mostly aboveground rect- angular rooms, the traditional "pueblo" of the American Southwest. Hamlets like the Merrigan site have been excavated, some as early as the 1950s (Vytlacil and Brody 1958) and some quite recently. Publications focusing on such excava- tions are relatively few in number, however, and the present report adds to what we know of the population of northern New Mexico during this period, when society was developing a distinctive regional character

The bulk of the Merrigan site, or LA 1 10971, was excavated in 1997 by students from a North- em Illinois University archaeological field school (Fig. 1 ). Funding for this research came from Mn Merrigan, the landowner at the time, the Field Museum, the Northern Illinois University anthro- pology department, the College of Liberal Arts and Sciences External Programs division of Northern Illinois University, and private dona- tions. Project staff is listed in Appendix 1. Al- though the intent of the research was to lift a non- disturbance easement and allow a house to be

FIELDIANA: ANTHROPOLOGY, N.S., NO. 35, JUNE 30, 2003, PP. 1-38

The Merrigan site. LA 1 10971

1997 excavations

Northern Illinois University. Field Museum

O -•n=«s

H - excavated area

— - contour interval 50 cm

Fig. 1. Topographic map of LA 110971, the Merrigan site.

built on the property by Mr. Merrigan, the parcel was not developed after the easement was re- moved. The land was sold to the state of New Mexico in 2000 and is now managed as undevel- oped state land.

Architecturally, the Merrigan site consists of three circular living rooms and four rectangular storage rooms located on a rise overlooking the Pojoaque River. The rooms were aligned from northeast to southwest, with circular rooms locat- ed on the south and the rectangular rooms wrapped in an L-shape around them to the north. There was a hearth in each of the circular rooms as well as outdoor hearths of two sizes. Various pit features and post holes were present both with- in and outside the structures, some of which sup- ported a ramada, or roofed work area (Fig. 2).

First identified on September 23, 1995, during a survey by DSS Consulting (Legare 1995), the Merrigan site was tested in 1996 to ascertain the size of the site and the extent of cultural deposits (Legare 1996). During that work, all visible sur-

face materials were recorded and limited subsur- face testing was conducted. The site was estimat- ed to be approximately 1,000 square meters in size, with cultural materials that included ceram- ics and lithics, ground stone, burned adobe, and fire-cracked rock.

During the initial phase of fieldwork, a con- certed effort was made to establish the age of the site using surface ceramics. It could be the case that the site dated to the later Coalition-Classic transition period as identified by Legare if the sur- face assemblage had been depleted by repeated scavenging of decorated sherds. In that case, we would have expected to find decorated sherds in the subsurface deposits. We might also have found tiny decorated sherds on the surface. Dec- orated sherds are recognizable at a much smaller size than undecorated sherds, yet the smallest pieces are not particularly attractive to collectors. Neither subsurface nor surface deposits yielded glaze-paint-decorated or Biscuit ware sherds in a consistent pattern. A single glaze ware sherd was

FIELDIANA: ANTHROPOLOGY

found by a project participant on the surface, just east of the driveway into the property. This find is consistent with the interpretation of the Biscuit ware sherds recorded by Legare as later additions to the site.

A disturbed area of soil that could have been a heavily eroded adobe structure, located east of the driveway, was closely investigated to see if it could be a structure. No artifacts of any kind were found around or on the deposits. The shape was amorphous, and it could not be determined wheth- er this was a small structure, a disturbed area, or an area where loads of earth from another location had been dumped, which might have introduced artifacts to the property. We concluded that this area did not appear to originate as a Pueblo struc- ture and was not the locus of fourteenth century or later occupation of the site. It was decided that excavation would be necessary to establish the age of the site, and that surface collection was not an adequate alternative.

Physical Environment

The Merrigan site is situated on a knoll overlook- ing the Pojoaque River, the principal tributary of the Rio Grande, in the area today known as El Rancho, New Mexico. Both the ancient and mod- ern hamlets are located south of Espanola and just west of Pojoaque, New Mexico (Espanola Quad- rangle, Township 19 N, Range 8E, Sections 11 and 14).

The location of LA 1 1 097 1 takes advantage of intersecting environmental zones and their re- sources, among them the riparian resources along the course of the Pojoaque, the grasslands in the immediate vicinity, and patches of pinon-juniper woodland that are present nearby. The river pro- vided a source of water some 300 meters distant from the site. The area beyond the river and its associated swath of riparian vegetation is domi- nated by grassland with piiion and juniper pines. Other grasses, four-wing saltbush, sagebrush, rab- bitbrush, prickly pear, and choUa cactus are pre- sent in the ara. Although it covers a relatively small area, the riparian habitat includes a greater range of plants, those mentioned along 'with a number of species having economic value. These species include Rocky Mountain beeweed, Indian ricegrass, three-awn grass, side-oats grama, and flax. The Pojoaque is one of the few perennial streams in the region and also supports cotton-

wood, willow, tamarisk, rushes, and sedges (Moore 2000:10-11).

The choice of this location indicates the resi- dents' interest in taking advantage of resources of the ecotone. The raised knoll location rather than a low-lying site afforded a way to maximize vis- ibility over the surroundings, take advantage of occasional breezes, and build structures in a high- er and therefore dryer place. Building on a dryer site could have been a means to escape overly moist housing sites or to preserve wetter sites for cultivation and farmland.

LA 110971 is situated on the western margin of the Espanola Basin, a portion of the Rio Grande Rift, a series of basins formed by faulting. The faulted rocks in this area are Tertiary sedi- mentary deposits (Hudspeth 1997:10). Soils in this area are predominantly loam, including fine, sandy, and gravelly loams with varying potential for dry farming and irrigation (Post 1996:7). Ellis has suggested that occupations like the Merrigan site might have been seasonal habitations by groups that spent the rest of the year living at LA 835, a cluster of habitations farther east, near the modern town of Pojoaque (Ellis 1975), although other alternatives will be discussed. Information on the physical environment reflects people's choices of where to settle, when, and for how long. These choices in turn bear on where and when aggregation might begin, and what a per- manent autonomous extended-family settlement might look like, as opposed to a seasonal en- campment.

Research Objectives

This work is part of continuing research on change in population in the northern Rio Grande region by the Northern Rio Grande Research Pro- ject, sponsored by Northern Illinois University and the Field Museum. This fieldwork developed out of an interest in the emergence of centraliza- tion following aggregation of population in ever- larger villages in the American Southwest. Field- work at LA 1 10971 investigated the role of small sites in permitting and fostering population aggre- gation, which reached a peak in the large northern Rio Grande villages of the fifteenth century. One of our research interests in exploring the Merrigan site was to document the initial stages in the pro- cess of aggregation, which began in earnest dur- ing the fourteenth century with the construction

CREAMER & HAAS: MERRIGAN SITE, EL RANCHO, NEW MEXICO

of Arroyo Hondo (LA 12), outside Santa Fe. Ex- cavations elsewhere have been aimed at identi- fying the factors that fostered aggregation (threat, large-scale immigration, competition for hunting and cultivating spaces and for water, cooperation, common religious beliefs) and at recognizing countervailing forces (desire for privacy, family- level production, lack of leadership, and absence of centralized economic or political institutions). These factors are examined at LA 1 10971 as well.

Prehistory of the Region

Since archaeological work began in the northern Rio Grande region, not long after the turn of the twentieth century, a number of chronological frameworks have been proposed for the region. The first scheme was suggested by H. P. Mera, based on archaeological survey in northern New Mexico (Mera 1934, 1935). The chronology he presented was revised by Stubbs and Stallings (1953). A chronological sequence that relied less on contacts with Chaco and Mesa Verde and fo- cused on cultural developments within the north- em Rio Grande region was developed by Wendorf and Reed (1955), and it is their sequence that has been widely adopted for use across the region. It is ironic that although the Pecos classification was developed for the northern Rio Grande, it works poorly within the region, and the outline suggest- ed by Wendorf and Reed has become the pre- ferred chronology. Divisions include the Precer- amic (pre-A.D. 600), Developmental (a.d. 600- 1200), Coalition (a.d. 1200-1325), and Classic (A.D. 1325-1610) periods.

The earliest known occupants of northern New Mexico were Paleoindian bands of big-game hunters who relied on mammoth, bison, and other now extinct animals as an integral part of their subsistence (Judge 1973). Sites of the Paleoindian period (9500-6000 B.C.) are nearly absent from the northern Rio Grande region, however. A few finds of isolated late Paleoindian artifacts have been reported from the Galisteo Basin (Honea 1971; Lang 1977). Elsewhere in the state, how- ever, Paleoindian sites have been recorded, in- cluding the type sites for Clovis and Folsom, along with examples of the Midland and Cody complexes (Stuart and Gauthier 1988).

The end of the Paleoindian period saw the emergence of the more generalized hunter-gath- erer subsistence strategies of the Archaic period.

from 7500 b.p. to about 1500 b.p. Until recently, Archaic sites were best known from the Albu- querque area, including sites in the Sangre de Cristo range (Wendorf and Miller 1959), in the foothills of the Sandia Mountains, and in the Rio Rancho area (Reinhart 1967). Sites consisted of semipermanent base camps and brief-use, special- purpose sites, like those described by Fawcett (1974). Most sites were located well away from the Rio Grande (Skinner et al. 1980:13). More recent research in the Santa Fe area (Post 1996) has revealed more extensive Archaic occupation in the northern Rio Grande. The transition to hunting smaller game and greater utilization of plant food may have made Archaic sites among the least visible to archaeologists (Post 1996:11). During this lengthy period, sites are generally identified from distinctive projectile point styles. In his study of the Archaic period in the Santa Fe area. Post (1996) notes, "Early and middle Ar- chaic period materials are similar to the Jay (5500 to 4800 B.C.), Bajada (4800 to 3200 B.C.), and San Jose (3200 to 1800 B.C.) phases of the Os- hara tradition (Irwin-Williams 1973)" (1996:11). As identified by Irwin-Williams (1973), the Os- hara tradition demonstrates the first steps toward regionalization, as over time, differences in settle- ment pattern and material culture can be distin- guished as hunting and gathering groups adapted to local conditions such as availability of water, range of plant foods, and movement patterns of hunted animal species. The later phases of the Ar- chaic/Early Basketmaker period are similar to lat- er phases of both the Oshara and Cochise tradi- tions (Post 1996:11).

In the area just south of LA 1 10971, Post iden- tified Armijo-phase (1800-800 B.C.) sites which are suggested to indicate increasing use of pied- mont and woodland environments, a shift sup- ported by excavations elsewhere in the Santa Fe region (1996:440). Even more numerous were projectile points characteristic of the subsequent En Medio/Basketmaker II period (800 b.c.-a.d. 400). Sites identified from this period included a range of functions, from habitations to special- purpose localities (1996:441). Sites of this period, ranging from pit houses a meter deep to structures built in shallow pits, can be found in riverine, piedmont, foothill, and montane settings. The in- creased diversity in settlement types and locations suggests an overall population increase (Boyer and Lakatos 2000:70).

During the subsequent Developmental period (a.d. 600-1200), there was an increase in the

FIELDIANA: ANTHROPOLOGY

number of permanent settlements, and the pro- duction of ceramic vessels began. In the earliest phase (a.d. 600-900), sites consisted of small groups of one to four pit houses, which sometimes had associated extramural storage pits or activity areas. Sites were located on low river terraces, such as the earliest jacal structures at Pindi (Stubbs and Stallings 1953:24-25) and the pit house in Area B at the Tesuque By-Pass site. The overall trend toward decreased mobility and in- creased storage and the introduction of artifacts such as trough metates suggest an increased reli- ance on cultivation. Ceramics associated with Ear- ly Developmental sites include plain gray and brown wares, and San Marcial Black-on-white (Cordell 1979:43; Boyer and Lakatos 2000:72).

The Late Developmental period (a.d. 900- 1 200) was characterized by greater variety in both construction and material culture. Settlements typ- ically consisted of one or more pit houses and sometimes a number of aboveground structures. Unlike earlier pit houses, which tended to be a meter deep, these later structures ranged from 30 cm to over 1 meter in depth (Boyer and Lakatos 2000:73). Aboveground rooms were generally made of adobe and appear to have been first con- structed for storage, as they lacked interior fea- tures. Features such as hearths or ash pits were apparently added when the aboveground rooms began to be used as living quarters. Characteristic ceramics of this period include Red Mesa, Kwa- he'e, Gallup, and Escavada Black-on-white dec- orated types. Types found infrequently at Late De- velopmental sites include Chupadero, Socorro, Chaco, and Chuska Black-on-whites. Neckbanded and plain ceramics continued from the Early De- velopmental period, with the addition of corru- gated and incised types. Some intrusive examples of red wares are also found, part of the high var- iability in ceramics and presumably reflecting ex- ternal contacts among the occupants of the region during this period.

Late Developmental period sites include pit structures (LA 9140), sometimes combined with jacal (LA 191) or masonry (LA 10614) above- ground structures (Dickson 1979) . Perhaps the best known Late Developmental period site is LA 835, where ten units of seven to 15 contiguous adobe rooms were identified. These were distin- guished as both living rooms and storage rooms (McNutt 1969). Additional examples of the Late Developmental period include sites in the Las Campanas project area (Post 1996:442-443) and the Santa Fe to Pojoaque Corridor, including ad-

ditional work at LA 835 (Boyer and Lakatos 2000; Boyer, Moore, and Lakatos 2001). The shift from pit house to aboveground architecture was first delineated by Dickson (1979). In his survey transect from the Rio Grande to Arroyo Hondo Pueblo (LA 12), he noted a mix of pit house sites and sites with aboveground pueblos (1979:30-31) that illustrated the period of transition. The most intensive occupation during this phase was along the Santa Fe and Tesuque Rivers and their tribu- taries (Wiseman 1989).

In Bandelier National Monument, there was great variability among Late Developmental struc- tures, which ranged "from pithouses and small surface structures to 100-room pueblos" (Powers and Van Zandt 1999:23). Late Developmental sites, those bearing Kwahe'e Black-on- white ce- ramics, first appear around a.d. 1 150 (Powers and Orcutt 1999:553). Unlike the pit structures de- scribed for the valley areas north of Santa Fe, the early structures at Bandelier were rectangular, aboveground living and storage rooms, that is, rooms both with and without a hearth. Sites in Bandelier National Monument appear to have consisted largely of rectangular surface rooms, and no pit structures were recorded during the sur- vey. The earliest structures built in the Bandelier survey area are suggested to date to around a.d. 1150, and the ceramics associated with them in- clude corrugated ware and Kwahe'e Black-on- white (Powers and Orcutt 1999:553). The ceram- ics are similar to those found at LA 1 10971, but the architecture is different.

The time from about a.d. 1200 to around 1325 is known as the Coalition period. During this time aggregation into large villages began in earnest in the northern Rio Grande region. Maximum site size increased from approximately 100 to 1,000 rooms. Grid gardens and other forms of gravel mulch were introduced during this time as reli- ance on horticulture became more common. The earlier, Pindi phase (a.d. 1200-1300), was char- acterized by small villages and farmsteads. The later Galisteo phase was defined to recognize sites having the Galisteo variant of Santa Fe Black-on- white as part of the ceramic assemblage. Galisteo Black-on-white appears to have been introduced around a.d. 1300, and is associated with large pueblos, which became typical during the Classic period.

The Rio Grande Classic period lasted from about A.D. 1325 until 1610. During this period most of the region's population lived in large vil- lages. Late in the Rio Grande Classic period, in

CREAMER & HAAS: MERRIGAN SITE, EL RANCHO, NEW MEXICO

A.D. 1540, the first European explorers arrived in the region. The results of exploration included eventual population decline and cultural change (Creamer and Haas 1998; Creamer 2000).

Research Approach and Methodology

Research at LA 1 10971 has examined the process of centralization that occurred in tandem with population increase and aggregation in the north- ern Rio Grande valley. Beginning in the late 1200s with the abandonment of the San Juan Ba- sin and Four Corners region, an increase in the population of the northern Rio Grande region is apparent from an increase in the number and size of sites. A tremendous aggregation of population followed, continuing into the mid- 1400s.

There would be no reason to expect a transition from dispersed hamlets to large aggregated villag- es unless common concerns fostered or forced such a change in settlement. Our excavation was aimed at identifying factors that might have fos- tered aggregation, such as threat, large-scale im- migration, and competition for water, hunting ter- ritory, and cultivable land. Cooperative activities may have included pooling labor to intensify pro- duction, the development of common religious beliefs, and the construction and maintenance of alliances. Countervailing forces that would have tended to drive people apart could have included the desire for personal privacy, family-level pro- duction, lack of leadership, the absence of cen- tralized economic or political institutions, and re- ligious secrecy.

Although excavations at LA 1 10971 were orig- inally intended to examine the role of small sites during the fourteenth century, the early dates for the site forced a shift in research focus. The re- vised goals were to look at the range of functions present in small sites prior to the onset of aggre- gation and to assess the degree of local self-suf- ficiency of a small, independent site. Such sites have been suggested to have been part of a larger but decentralized socioeconomic and political community. For example, Ellis has suggested that LA 835 was the principal or permanent center uniting the smaller sites in the area of LA 1 10971, the upper Tesuque Valley (1975:41). Thus, we ex- amined the evidence for permanence of LA 1 1097 1 . If the site appears to have been a seasonal farmstead, it will support Ellis's theory. If the site was a year-round settlement, we need to consider

what its role may have been during subsequent regional integration.

II. Excavations

To identify excavation units across the site, a grid of stakes laid out at 2-meter intervals was placed on a N/S axis aligned on magnetic north. An ef- fort was made to place the central datum where it was located during Legare's (1995) work at the site, but the former datum stake could not be iden- tified. The excavation datum was placed at the top of a low rise where the structures are situated. Test units outside the site center were selected to iden- tify the maximum extent of site debris. Units were designated by the distance from the 0,0 point to the corner farthest from 0,0. Excavation of a 162- square-meter area revealed rooms and outdoor work space. Testing began with grid units exca- vated with shovels until room walls were identi- fied.

Excavations began in unit Nl El with the goal of relocating and defining the possible work sur- face that had been tentatively identified during the testing operation (Legare 1996). Unit Nl El did not yield clear evidence of a work area but did yield three artifacts, a hammerstone, a large plain ware sherd, and a quartzite grinding stone, all at about the same level below datum in the south- west corner of the unit. Although there were no signs of a surface beneath these objects, at about this same level along the south margin of the unit was a cluster of slabs. Two of the slab fragments fit together, and all the pieces exhibited some ev- idence of grinding. As a result of these finds, ex- cavation was extended to the adjacent unit, Nl Wl. A curving wall segment was recognized in the excavation of the unit, enclosing a room that was numbered Room 1 . The other materials found were associated with Room 4.

After Room 1 was identified, excavation ex- panded outward, resulting in the discovery of Rooms 2 through 7 to the north and east of Room 1 . A number of units were excavated to the west and south. While most of these units contained some artifacts, the frequencies were low and the soil was very shallow. We concluded that the con- centration of rooms was the entirety of the site architecture, forming a cluster that trended south- west to northeast across the crest of the knoll at LA 110971. Based on the frequency of artifacts, we also concluded that site midden was concen-

FIELDIANA: ANTHROPOLOGY

The Merrigan site, iJV 110971

1997 excavations

Northern Illinois University, Field Museum

Post Holes ^ Datum

Roclts -- Dark Soil

Groundstone O Outdoor Hearth

O Indoor Hearth

Fig. 2. Plan map of excavated structures at LA 1 10971, the Merrigan site.

trated around the structures and to the east. It ap- peared that trash was discarded to the east outside the structures. The subrectangular structures may have opened to the east, which would make dis- posal to the east the path of least resistance. Fur- ther, the hill sloped most steeply in that direction, and it was the side of the site away from the river, which passes to the south and west of the site.

The remainder of the fieldwork carried out in 1997 focused on excavating all the structures that we identified, on excavating a test trench along the north side of the site, and on testing the re- mainder of the hilltop for other possible struc- tures. Descriptions of the individual room exca- vations provide greater detail about those exca- vations.

Field Methods

The entire site surface was covered with loose sand and thick grass. For this reason, the first few

centimeters of each unit were designated as a sur- face (S) level. Each surface level is likely to have been disturbed by historic and modern activities, including trampling by horses and cattle as well as human activities associated with nearby struc- tures. There was broken glass in a number of plac- es across the site surface, and several casings of spent ammunition were found in surface levels as well.

Stratigraphy was shallow throughout most of the site, with wall fragments and floor surfaces located about 30 cm below the present ground surface. The site was dug in natural layers when visible and in 20-cm arbitrary levels when natural stratigraphy could not be discerned. As structures were identified during excavation, each was num- bered sequentially, with structures numbered 1 through 7 (Fig. 2). Room I, the best preserved structure, was excavated in four units. First, the room was divided in half across its N/S and EAV axes. The eastern and western quadrants were ex- cavated first, and a section was drawn across the room. The remaining quadrants were then exca-

CREAMER & HAAS: MERRIGAN SITE, EL RANCHO, NEW MEXICO

vated. Each of the other Rooms 2 through 7 were excavated in halves in natural levels. In areas where no rooms were identified, grid squares were excavated down to sterile soil.

All features were mapped and excavated sepa- rately. Feature excavation was carried out in two stages. One-half of each feature was excavated, a section drawing was made of the bisecting profile, and the remainder of the feature was then exca- vated. During excavation, a 2-liter soil sample for flotation and a pollen sample were collected from each 20-cm level. Additional samples were col- lected from each feature, hearth, and any unusual concentration of ash or dark-colored materials. Additional pollen samples were taken from the corners and center of each room floor.

Extramural features were numbered sequential- ly from 1 to 24. Features within the boundaries of a room were numbered with three digits: first the number of the structure, then two digits iden- tifying the order in which the features were un- covered within the structure. Thus, feature 101 is the first feature recorded in Room 1 , for example. Test units, including the materials over the struc- tures, were numbered with a north or south co- ordinate and an east or west coordinate, e.g., Nl E 1 . Any features revealed during excavation of a test unit were assigned a one- or three-digit num- ber, depending whether or not it was within the boundaries of a structure.

Architecture

The primary occupation area at LA 1 10971 con- sisted of seven rooms in close proximity to each other. There were three living rooms overlapping or connected to each other, and four storage rooms, also connected to each other. Each living room had a hearth feature located near the center. The walls of the rooms were made of adobe, al- though only the bases of the walls were found during excavations. The walls had mostly eroded away, and there was no evidence of the roof struc- ture in any of the rooms. All of the rooms ap- peared to have been built in the same way, with puddled adobe. The fill of all the rooms included soil deposited by wind and erosion. The adobe from the walls was eroded by rainfall, which re- sulted in the puddling of adobe on room floors and outside room walls.

Room 1

Length: 3.6 m Width: 3.9 m Area: 14.04 sq m

Room 1 was a square structure with rounded corners (see Fig. 2) located at the south end of the occupation area next to Room 4. Patches of floor were detected in Layer IIL An adobe wall bisected the room on a diagonal NAV to S/E axis. The wall appears to have been made of puddled adobe, although the pattern of erosion made it ap- pear superficially to have been assembled of ado- be blocks. There was some slumped adobe on the floor of the room and outside the wall of Room I, but the quantity was relatively small and sug- gests the room had a jacal, or wattle-and-daub, superstructure over a low adobe wall. However, the preserved portion of the wall did not contain post holes that might have supported such a su- perstructure, so this reconstruction is somewhat speculative.

A hard-packed level that included a portion of the room floor may indicate a former use surface, one that was associated with the occupation of Room I . If that is the case, then the room was a shallow pit structure, with a slightly excavated or dish-shaped floor. This is the only excavated room where a large section of floor surface was pre- served, which we interpret as indicating the final occupation of the site.

Room I was excavated in quadrants established once the walls were identified. Transects divided the room into quadrants: north, east, south and west. The north and east quadrants were excavat- ed in artificial levels and the south and west quad- rants were dug in natural layers identified in the cross-section profile (Fig. 3).

North Quadrant — This quadrant contained a large amount of melted and crumbled adobe which was probably once part of the walls. The first level of the room produced lithics, sherds, adobe, and a carbon sample. A charcoal layer was found in level 3. Below the ash was more melted adobe. There was a post hole feature (feature 101) in this quadrant, along with portions of the wall that bisected the room, the hearth, and the asso- ciated ash pit.

East Quadrant — The first level produced charcoal flecks, ash, ceramic material, and lithic material. The ashy deposits were located in the

FIELDIANA: ANTHROPOLOGY

NW-SEProaieofRooal

I Charcoal I Dark Layer ! Adobe I Light Ash I Rock Kxcavated Prior to Drawing

Fig. 3. Section view of Room 1 at LA 1 10971.

south and west corners. Levels 2 and 3 included a number of lumps of amorphous melted adobe. Level 4 produced an ash deposit in the west cor- ner that may have been part of the hearth. The entrance to the structure may have been along the east quadrant of the room. Although the remains of an adobe wall were found around the circum- ference of Room 1, the presence of such a wall was difficult to detect in the eastern quadrant. Fur- ther, the soil along the eastern margin of the struc- ture was more pebbly than elsewhere and more mottled in color. These characteristics are consis- tent with that zone having been an entrance. In addition, feature 102 is a possible sipapu. It was approximately 10 cm in diameter and just under 10 cm deep. No artifacts were found in this fea- ture, although the soil in it was slightly darker than the soil outside the feature. Feature 103 was a pair of post holes, perhaps a post hole that had been reworked at some point. Portions of the cross-wall and the hearth were in this quadrant.

South Quadrant — The fill in Layer I consisted of charcoal-flecked soil and adobe chunks. The layer ended at an ashy, charcoal-flecked lens. A hammerstone, ground stone, and small sherds were removed from Layer I. Layer II produced a concentration of artifacts, charcoal, and also burned adobe in the north corner of the quadrant. A small turquoise bead was found in Layer II at 139 cm below datum in the east comer of the quad. Layer III produced a smaller number of sherds and lithics. The floor was difficult to dis- cern, as it was thin and eroded. Bedrock was reached at the bottom of the layer along the south side of this quadrant. The south quadrant also had a pair of post holes (feature 107) that appeared to indicate that a post had been repositioned at some point. As in the other quadrants, feature 106, the

hearth, was part of the south quadrant excava- tions.

West Quadrant — The top layer excavated produced a large amount of ash in the east comer, along with a hammerstone and a large sherd. Some lithics and sherds were found in Layer II. A patch of floor was found at the bottom of Layer III. A collared hearth was located in the east cor- ner, along with various hearthstones. A burned corncob was recovered from the west portion of the quadrant. In addition to a comer of the hearth and rock-filled pit, a post hole (feature 104) was excavated in this quadrant of Room 1 .

Features — Features 101 to 107 were located in Room 1 (Table 1). These features include post holes, a hearth, and an associated rock-filled pit. A collared adobe hearth, feature 106, was found in the southwestern half of the room (Fig. 4). This feature was set into the floor of Room 1, against the adobe wall in the center of the room. The hearth had been remodeled, and an earlier collar was covered by ash and the more recent collar A depression in the floor, filled with smoke-black- ened rocks, was uncovered adjacent to the hearth, and no collar of adobe separated the hearth from

Table I. Features in Room 1. LA 1 10971.

Feature Dimensions Depth

no. (cm) (cm) Function

Post hole-roof support Post hole

Post hole-roof support Post hole-roof support Rock-filied pit Adobe-collared hearth Post hole-roof support

F 101	17 X 14	28
F 102	11X9	10
F 103	36 X 24	28
F 104	10 X 10	10
F 105	61 X 57	11
F 106	52 X 48	15
F 107	46 X 30	27

CREAMER & HAAS: MERRIGAN SITE, EL RANCHO, NEW MEXICO

D g [TQ

LA 110971 Room 1 Feature 106 Collared Hearth

Adobe Adobe Adobe Ash Wall Collar Collar

Remodeling

Fig. 4. Plan and section views of collared hearth in Room 1 .

the pit of rocks. This feature could have been a cooking area, where materials could be heated with rocks from the fire.

The cross-wall was a later addition to the room, based on its abutment to the subrectangular walls of the room. A quantity of ash was found on the side of the center wall opposite the hearth. This and the position of the hearth against the