Dr. Geoffrey McCafferty

After 25 years of archaeological investigations in Pacific Nicaragua, one of the nagging issues has remained what happened in the final centuries before the Spanish conquest, beginning in 1522. Ethnohistorical sources report a large multi-ethnic population in the region between the Pacific Ocean and Lakes Cocibolca and Xaltocan, yet from an archaeological perspective this period is virtually invisible. The chronicles describe Indigenous cultures relating to the Nahuat-speaking Nicarao and the Oto-mangue-speaking Chorotega, as well as autochthonous Chibcha-speakers, though the information about the Nicarao is most explicit. Furthermore, the past 25 years have seen several large-scale excavation programs working in the region, including the Leiden project on the east side of Lake Cocibolca, the Japanese projects along the shore of Lake Xolotlan, and especially the Calgary program along the western shore of Lake Cocibolca. Despite this relatively intensive effort, almost no La Paz phase occupation has been encountered (apart from some evidence at the La Paz site, itself). Consequently, the archaeological record provides little support for the ethnohistorical interpretations found in sources such as Oviedo y Vazquez. 

To address this gap I returned to the site of Tepetate, located just north of modern-day Granada and where we had excavated in 2008. Tepetate is supposedly the site of the Contact-period community of Xalteva, which according to sources was a center for the Chorotega when Spanish conquistador Gil Gonzalez first entered the region and met with Chief Diriangen in 1522. That account indicated that Xalteva was a large and wealthy community. After initially welcoming Gonzalez, the Chorotega attacked and drove the Spanish away, only to succumb open their return the following year. The Spanish established their 'new' city on the lakeshore, where it became an important trading center and eventually the capital of the Spanish colony. Early 20th century accounts indicate that mounded architecture of Indigenous Xalteva remained, though expansion of Granada in later years covered over the ancient settlement. 

In 2008, the Calgary program conducted excavations in Tepetate, investigating one of the last remaining mounds and several burial clusters. Diagnostic artifacts and two radiocarbon dates indicated that these contexts dated to the Santa Isabel phase (1000-1250 CE). Because of the poor preservation due to high soil acidity, urban development, and extensive looting it was decided to suspend further investigation and move on to the El Rayo site on the Asese peninsula south of Granada. Thus, although the ethnohistorical sources predicted a Contact-period settlement, no such occupation was encountered.

In 2025, however, we decided to return to Tepetate to continue the search for the La Paz phase town of Xalteva. Several loci were identified within the modern urban setting where green-spaces preserved what was hoped to be undisturbed cultural deposits. These were investigated using a combination of shovel-testing followed by horizontal excavation. 

Locus 1

This was a large field next to a municipal basketball court and playground, located about 400 m southwest of the 2008 excavation area. Shovel-tests were laid out on a 10m grid. Artifact density was very low, and there was a mixture of modern and pre-Columbian material. Three 1 x 1m excavation units were dug around STPs with the highest artifact concentrations, but no features were encountered. The few diagnostic sherds found included Papagayo Polychromes, but these could not be identified as to the specific phase. It was decided to move on to a more productive area.

Locus 2

Two city-blocks south of Locus 1 was another open green space with three low mounds elevated about 1.5 m above street level. The central mound (Mound 2) was designated Locus 2. It measured 37 x 14 m. A series of shovel tests were dug at 10 m intervals across the top, and five trenches were dug into the sides in search of evidence of mound construction. A 1 x 1 m unit was also placed on top of the mound to expose construction stratigraphy. 

The two trenches on the south side of the mound encountered a brick-lined drainage channel that probably dated to the 20th century (Fig.  ). The drain ran the entire length of the Mound 2 and ended at a municipal water line at the southeast corner. 

The trenches into the sides of the mound revealed a complex stratigraphy of different colored and consistency soils, probably indicating sequential layers of construction fill (Fig.  ). The fill layers were almost sterile, suggesting that the soils were transported from outside of the urban area. We speculated that the different soil types may have come from different areas and could be an indication of tribute contributed for the mound construction. 

Unit 6 began as a 1 x 1 m unit dug into the top of Mound 2. Differential levels of compaction was interpreted as evidence of an intrusive pit, but upon excavation to a depth of 2 m no cultural material was encountered to explain the purpose of the feature. At bottom of unit the fill layer ended at a layer of reddish talpuja (natural soil). On the north side of the unit a more hard-packed surface was encountered at 20 cm below the surface. This was interpreted as a walking surface and several more units were opened to expose the surface until ultimately 4 m2 of the surface was exposed. An alignment of three postholes were found along the north side of the walking surface. These measured about 30 cm in diameter, and are interpreted as structural posts to support a roof  (probably of thatch). 

Artifact density was very low. A few diagnostic sherds of Papagayo Polychrome were recovered, including one identified as Casares variety. 

Locus 3

Just east of Mound 2 was a larger mound, Mound 1, that was designated Locus 3. Mound 1 was comparable in height (about 1.5 m) and about 80 x 80 m in surface area. Investigation of Locus 3 consisted of a transect of shovel tests at 10 m intervals along the western edge, and a 1 x 1 m unit at the southwest corner. Again, the artifact density was very low, and consisted of modern debris in the upper levels with non-diagnostic pre-Columbian artifacts below. 

Locus 4

About 150 m north of Mound 1 is a deep ravine with a stream that flows into Lake Cocibolca. Inspection of the side wall of the ravine revealed a lens of polychrome ceramics about 2 m below the surface. Three  1 x 1 m units were excavated along the edge of the ravine to arrive at the level of the sherd lens. Artifact densities were higher in these units, with diagnostic ceramics from the late periods. 

Material culture

Interpretations

The objective of the 2025 field season at Tepetate was to sample deposits related to the terminal pre-Columbian and Contact periods. The most notable result was the discovery of Mounds 1, 2, and 3, which based on the stratigraphy (especially from Mound 2) were artificial mounds constructed during the late pre-Columbian period. No diagnostic artifacts nor useful carbon samples were found to narrow that chronological assessment. However, the walking surface and alignment of postholes on Mound 2 resembles the layout of an elite structure depicted by Oviedo from the Nicarao site of Tecoatega, dating to the early Contact-period (Fig. ). While highly speculative, this association may indicate that this mound complex may have been an elite compound. 

The archaeology of Nicaragua remains poorly developed despite advances since the 1990s, yet the pre-Columbian past is an important foundation for the modern culture and political identity of the country and its people. The Tepetate Archaeological Project (TAP) produced important results to better understand the late prehistory of the indigenous population at the site of Tepetate, identified as the ancient town of Xalteva, located on the outskirts of the Colonial capital of Granada on the northwest coast of Lake Cocibolca (Fig. 1). Xalteva was a centre of the Chorotega culture at the time of the Spanish conquest, and had previously been surveyed and described by Silvia Salgado (1996). Since the Chorotega allegedly originated in central and southern Mexico, the project considered questions of migration and ethnic identity as revealed in the archaeological record, building on insights compiled in the previous project in Santa Isabel, Nicaragua.

Nicaraguan culture history highlights ethnohistorical accounts of migrations from central Mexico in the final centuries of the prehispanic era, including the Oto-Manguean speaking Chorotega and the Nahua Nicarao. Yet despite the well-documented historical accounts relatively little archaeological attention has been paid to the question of cultural origins. Recent anthropologically oriented interpretations tend to reject the significance of these migrations and prefer an autochthonous development of prehispanic culture as part of a Greater Chibchan culture area relating to lower Central America and northern South America. Excavations at Santa Isabel, in the Rivas district south of Granada, recovered evidence to support both of these competing interpretations, resulting in a more complex scenario than either position had previously proposed, and one that requires further exploration.

The TAP research continued to investigate these issues by moving to a site that was believed to have been occupied by the Chorotega at the moment of first European contact. Shovel testing and horizontal excavations exposed remains of domestic and public spaces. Household archaeology is widely regarded as one of the best means for inferring social identity and thus proved useful for interpreting the development of Chorotega ethnic identity. Since this is the same methodology employed at Santa Isabel it provided a valuable means for comparison between these contemporary and probably culturally related sites. However, because Santa Isabel was abandoned before the arrival of the Spanish, Tepetate has the potential to fill in a missing piece of the chronological puzzle of Pacific Nicaragua.

Archaeological investigations during the summer of 2008 included intensive shovel testing and excavation at three site loci on the northern extent of the Villa Tepetate neighborhood of Granada.

Shovel testing consisted of a 10 m grid in the 50 x 150 m field that was Locus 1 (Fig. 2).  Artifacts clustered around Mound 1, but several other concentrations were detected in non-mound areas, suggesting other activity areas, or at least midden deposits.  Additional shovel testing was done in transects to determine site boundaries and to discover other loci.  These were begun at a datum located about 1 km north of Locus 2 where a landowner gave permission to investigate an area of high artifact density.  The transect then extended south to connect with Locus 1.  In the process of shovel testing these transects two other areas of interest were identified, Loci 2 and 3. 

Locus 1

Locus 1 was located about 300 m from the shore of Lake Nicaragua, behind the Brisas del Mar condominium development and at the northern edge of the urban zone of Granada.  A low mound was preserved in an open field near a large ceiba tree.  Since this was one of the only mounds that survive from the Postclassic site of Xalteva, we felt it important to target it for intensive investigation in order to preserve any information that remained as a sample of what had already been lost to urban development.  

Mound 1 was approximately 1.5 m in height and covered an area of about 30 m in diameter.  Large flat stones littered the surface, similar to the lajas that were described on the surface and sides of pre-Columbian mounds from the site.  Excavations commenced as an east/west trench across the top, and another trench north from the E/W axis (Fig. 3).  1 x 1 m units were excavated along these axes, with more opened to further explore particular features. 

The upper levels of Mound 1 were badly disturbed by modern activities, probably looting.  As a consequence there was an abundance of random stones found within the upper 30 cm, probably architectural fragments discarded from deeper looter's pits (Fig . 4).  Other evidence of modern disturbance was the partial skeleton of a cow, which had apparently fallen into one of these looter's pits, and impressions from digging tools found at a depth of more than a meter below the surface.  

Deeper levels encountered sections of tepetate floor (Fig. 5); tepetate is the local term for the conglomerate bedrock that hardens when exposed to the air.  The west trench exposed an alignment of vertical lajas (sheets of a non-local gray metamorphic stone.  Associated with this alignment was another alignment of lajas that outlined an area appropriate for a human burial (Fig. 6), though only a single human tooth was found. Horizontal excavations in the northwest quadrant exposed additional architectural features, including paving stones and a wall foundation (Figs. 7 & 8).

Shovel testing at Locus 1 identified several other subsurface artifact concentrations away from the mound.  An extensive concentration was located near the large ceiba tree to the northeast of Mound 1.  Excavations recovered high concentrations of lithic and ceramic fragments.  Beneath a massive root from the ceiba tree was a complete vessel and several large potsherds (Fig. 9).  Although the surface finish was poorly preserved, traces of paint and the vessel form suggest that this was probably an example of Papagayo Alfredo variety.

Other artifact concentrations likely indicate refuse disposal or other activity areas.  At the far north end of the field one shovel test yielded several hundred artifacts, and an excavation unit found that these were localized in a discrete layer about 30 cm below the surface (Fig. 10).  While shovel testing has not been widely used in Central American archaeology, we are finding them to be very useful for identifying subsurface concentrations where surface visibility is poor due to lush vegetation.

Locus 2

In the course of shovel testing at the northern end of the project area, roughly 1 km north of Locus 1, two low mounds were discovered that included artifact concentrations.  Locus 2 was located about 300 m from the lake shore.  Excavations focused on a low mound of about 1 m in height.  Possible packed-earth walking surfaces were encountered at a depth of about 80 cm below the surface, adjacent to a well-defined stone wall foundation (Fig. 11).  Diagnostic ceramics indicate that this mound also dates to the early Postclassic Santa Isabel phase (900-1250 CE).

Locus 3

Shovel testing also identified an artifact concentration about 500 m west of Locus 1, where the landowners indicated there may have been a pre-Columbian cemetery.  Excavation units identified stone rubble and a section of preserved floor paving (Fig. 12), and a broken urn with human remains (Fig. 13).

About 100 m to the north of these architectural remains was a shallow drainage ditch, where crew chief Jorge Zambrana detected additional urn fragments eroding out of the bank.  Excavation units were laid out over the urn fragments, and several clusters of urns were uncovered (Figure 14 & 15).  Urns appear to have been placed one on top of another, resulting in a complex stratigraphy of urn fragments intermixed with human skeletal remains (Fig. 16).  Again, high soil acidity resulted in badly deteriorated bones, to the extent that they could not be removed from the ground without complete disintegration (Fig. 17 -19).  Several grave offerings were encountered, including miniature vessels, a tripod bowl, a greenstone bead, and a ceramic earspool (Figs. 20-23).

Two separate clusters of urn burials were excavated, the first with roughly 20 urns and the second with about eight.  Due to the compressed nature of the deposits it was impossible to excavate the urns as discrete features, and in fact large portions of the clusters were removed in blocks for excavation in the lab at Mi Museo in Granada.  The urns were typically of the Sacasa Striated type, in both 'shoe-pot' forms and large jars.  These are diagnostic of the Postclassic El Rayo and Santa Isabel periods (800-1250 CE), as were other associated potsherds and the complete tripod bowl.  Burials in shoe-pot urns are a characteristic mortuary pattern of Pacific Nicaragua.  At Santa Isabel, urn burials were exclusively  associated with infants, but at El Rayo and Tepetate the burials were of adults (this was also the pattern found in urns from Managua). 

Material culture

Due to extremely high soil acidity few organic materials survived, and even the painted surfaces of polychrome pottery were poorly preserved (Fig. 24).  Identifiable ceramic diagnostics were predominantly of the Papagayo type, most characteristic of the Santa Isabel period (1000-1250 CE).  The most obvious diagnostics due to the poor preservation of painted decoration were Papagayo Mombacho Incised variety bowls, characterized by often complex incised decoration on the interior surface (Fig. 25). A notable characteristic of the Tepetate ceramic assemblage was molds used in the manufacture of figurines and vessel supports (Figure 26). Several miniature vessels were found associated with the Locus 3 Feature 7 burials (Figure 27), as well as objects of adornment such as a ceramic ear spool (Figure 28) and a greenstone bead. Lithics were found in relative abundance, particularly mixed in around the floor paving stones.  These consisted of chert debitage and occasional obsidian blade fragments; it was noted that obsidian made up about 8% of the lithic assemblage, significantly more that has been found at other contemporary sites such as Santa Isabel and El Rayo.. 

 Preliminary Conclusions

The three loci excavated were on the northern edge of the ancient settlement, and along the shore of Lake Cocibolca  All diagnostics indicated a Santa Isabel phase occupation (1000-1250 CE), which is consistent with two radiocarbon dates from the site.  This is in contrast to the expectation of Tepetate being the Contact-period town of Xalteva.  Perhaps there was later occupation in other parts of the site. 

It is unfortunate that a site as important as Tepetate has been so severely impacted by recent development.  Additional urban archaeology, working in open spaces of the residential neighborhoods, could provide greater insights onto this important site.  Looting is another factor that has impacted the archaeological resources, and Mound 1 is an unfortunate example of that.  Finally, the high soil acidity has destroyed important artifact classes.  We were fortunate to be able to recover what information that was available before these loci were also destroyed by development.
 

While awaiting our excavation permit in 2003, the University of Calgary archaeology team was invited to assist the National Museum in the excavation of prehistoric 'mammoth' bones in the region of Veracruz de Zapotal, in the Department of Rivas. Farmers excavating a deep well for their banana fields had encountered fossil bones of a large mammal, and reported them to museum officials. Thus began an unlikely adventure.

 The site was remote, and had to be accessed via slippery farm paths.  The well was about 3 m deep, and an area of roughly 3 m x 4 m in area (Fig. 1).  Since we were working in the rainy season the water table was quite high, and in order to work on the exposed bones it was necessary to run a pump full time or else the pit would fill with water within about half an hour.

Because of the short amount of time allocated to this project the goal was to remove the exposed bones without additional excavation. This was precarious due to the wet conditions, though fortunately the walls of the pit were quite solid so there was no danger of collapse.  Over the course of several days the bones were exposed and carefully removed (Figs. 2-4). We also checked the backfill in search of possible cultural artifacts, without success.

The in situ deposits indicated that the skeleton was at least partially articulated, as segments of the vertebral column was aligned, and in correct anatomical position relative to the pelvis, scapula, long bones, and ribs.  No evidence of the cranium was found, unfortunately, and if it was also articulated it may still be buried adjacent to the pit.

The skeletal remains were removed to Rivas, where they were cleaned and conserved (Figs. 5-7). A display of the bones was exhibited at the Rivas Museum of Anthropology, where it is tentatively identified as a "mammoth."

In 2002, a team of archaeologists from the University of Calgary assisted Dr. Marc Winter of the Oaxaca Regional Center of INAH on a salvage archaeology project at the site of Macuilxochitl. Macuilxochitl is located in the Tlacolula arm of the Valley of Oaxaca, midway between Oaxaca City and Mitla (Fig. 1). Improvements to that section of the Pan-American highway necessitated mitigation of mounds located on either side of the road, especially at the turn-off to the town of Macuilxochitl.

Macuilxochitl was one of the largest pre-Columbian sites in the Valley of Oaxaca. It sprawls over a large area, with distinctive concentrations at different time periods; for example, Dainzu is a Formative period center withing the area of 'metropolitan' Macuilxochitl, while the modern town is located at the far northern extent of that same area. The project area included mounds relating to the Late Classic Xoo and Early Postclassic Liobaa periods. Several mounds were bisected when the Pan-American highway was initially constructed in the 1930s, and plaster floors and mound fill have long been visible along the side of the road. It was gratifying to be able to finally answer questions about those mysterious mounds.

A small group of University of Calgary graduate students accompanied me for a two month field season: Danny Zborover, Melissa Knight, Bobbi Parry, and Hollie Brooks. We stayed in the town of Macuilxochitl with the Zapotec family of one of the laborer chiefs. We greatly enjoyed the opportunity to learn about the local culture and appreciated the wonderful hospitality.

The salvage project concentrated on three large mounds: 35, 36, and 55. Mound 35 was located on the south side of the highway, while Mound 55 was on the east side of the road leading into the town of Macuilxochitl. The U of Calgary team focused on Mound 36 (and part of Mound 37), located on the north side of the highway and west of the road to town.

Mound 36 was the mound most severely impacted by previous road construction, with less than half of the original structure intact. Several plaster floors were exposed in the road cut (Fig. 2).  Excavations were directed at clearing the exposed surfaces and the upper floor level.  Additional clearing exposed a staircase on the north side of the mound facing Mound 37 (Fig. 3).  Located at the midpoint of the staircase was a human skeleton in a flexed, seated position, placed over the final construction layer and with a metate fragment over the skull (Fig. 4).

Although Mound 36 was badly damaged by highway construction and subsequent erosion and looting, nevertheless there were multiple wall and floor levels encountered.  One of the most complete features was a wall facing to the north, made of small pebbles covered with a layer of stucco, and in the characteristic talud/tablero form best known from Teotihuacan (and not common in Oaxaca) (Fig. 5).  This feature was found on both the east and west side of the central staircase.

 The outermost facade featured a decorative pattern of horizontal masonry interspersed with larger stones placed vertically (Fig. 6).  This style is known locally as a "Mixtec" trait, supposedly linked with the Postclassic incursion of Mixtec speakers into the Valley of Oaxaca.  Macuilxochitl is specifically named as one of the sites with Mixtec rulers  in the Late Postclassic period.  If this (admittedly tenuous) ethnic identification is true, then the Macuilxochitl archaeological evidence would suggest a very early date for this influence.  A more complete example of the "Mixtec" facade was found at Mound 35 (Fig. 7).

North of Mound 36 was Mound 37, a very large mound that was likely residential.  Separating the two mounds was a modern drainage canal that also impacted the prehispanic architectural remains.  Excavations in the canal exposed several plaza floor layers (Fig. 8), as well as ancient water management systems (Fig. 9).

A small area of Mound 37 was excavated to explore a floor surface exposed by the modern canal.  A patio was cleared that included a sunken area filled with wall rubble (Fig. 10).  To one side of this patio was a drain that presumably fed run-off from upper levels of the structure (Fig. 11).  A similar arrangement of small patios is known from Monte Alban.

The material culture associated with Mound 36 was consistently Late Classic (Xoo phase), but with Early Postclassic Liobaa material associated with post-abandonment levels.  This was best demonstrated in a trench dug west from the mound, which encountered a rich midden deposit including well-preserved corn cobs. For the Late Classic, by far the most abundant ceramic type was the ubiquitous G-35 conical bowl, making up roughly 90% of all pottery fragments.  Liobaa material culture included numerous examples of what have been identified as ritual offerings (Winter et al. 2007), including miniature vessels and handled incense burners.  Also present were many fragments of anthropomorphic urns (Fig. 12).  Interestingly, there were virtually no stone tools or evidence of lithic debitage.  One notable object that was found (at Mound 35) was a carved human mandible similar to those depicted in Mixtec codices (Fig. 13).

La Arenera is located at the base of a volcano on the northwest side of Managua, off of the highway to Leon. In 2000 we excavated at a La Arenera phase (c. 500 BCE-1 CE) site that had been covered over by volcanic sandy ash. La Arenera (literally "the sand quarry") lies beneath a two meter thick deposit of coarse sand that is being quarried for construction material (Fig. 1). Work crews remove the profitable sand, leaving the lower levels intact. The lower level consists of the occupation layer that was apparently buried by the volcanic eruption.

The archaeological deposits are easily exposed beneath the sand. The soil is hard-fired, probably as a result of the heat of the volcanic deposit. House floors and on-floor artifacts are identifiable, again suggesting that the occupation was terminated as a result of the eruption (Figs. 2 and 3). During our brief visit to the site we surveyed surfaces exposed by the quarrying, to conclude that the pre-Columbian occupation covered a large area, at least 40 hectares.

Preliminary analysis of the material remains indicated a high number of Usulutan-like resist painted pottery, similar to the ceramics found at the type-site in El Salvador (Fig. 7). There were also examples of Rosales Incised Black on Red pottery, diagnostic of the La Arenera phase (Fig. 8). The most abundant lithic material was obsidian, especially large flakes. Both the Usulutan-like pottery and the obsidian indicate trade with the southern periphery of Mesoamerica. 

Historic Newport is a shining, even gilded, example of conspicuous consumption during the mid- to late -19th century. Mansion-like "cottages" along Bellevue and Ochre Point Avenues attest to an amazing display of wealth through their architecture, landscaping, and exotic objets d’art.  Excavations by students of Salve Regina University’s Archaeological Field Methods class in 1998 investigated a map-documented historical structure that  provides evidence for the early development of Newport’s Gilded Age.

A mid-19th century article in Newport’s Mercury described plans for the construction of a "large and expensive mansion house" by William Beech Lawrence (Fig. 1). It went on to discuss his "charming" residence, and "among other objects of convenience an orchard house, 300 feet in length, fully covered in glass" where he raised such exotic plants as strawberries, pineapples, peaches and pears, even during the winter.

Lawrence was a prominent international lawyer, state congressman and Lieutenant Governor (Fig. 2). He was a renowned author of legal studies and had one of the largest law libraries in the United States. In 1850, Lawrence purchased a 69 acre parcel of undeveloped Ochre Point for the construction of his estate, a beautiful frame house. Note that this was one of the earliest mansions on the point, and that Lawrence lived there year-round as opposed to the later use of the "cottages" as summer homes.

The earliest map of the area dates to 1876, and shows the intact Lawrence estate including a large structure on the drive off of Lawrence Ave., several small out buildings across from the structure, and another large frame structure presumed to be the primary residence near the cliff overlooking the ocean (Fig. 3). In 1882, one year after the death of Lawrence, the bulk of the estate was sold to the Van Allen’s and the Lorillard’s. The subsequent results of this sale, found on the 1883 map, show that the large, unidentified rectangular structure is still present, and that the smaller, seaside, frame-style house has been removed in favor of Lorillard’s Vinland. By 1893, the large, rectangular structure had also been removed, this time for the construction of Van Allen’s Wakehurst (Fig. 4).

The large wooden structure first came to our attention as the result of a student paper dealing with the occupation history of Wakehurst Mansion, now part of the Salve Regina University campus. Determining the function of the map-documented structure posed a question that became one of the research objectives of the Field School program: was it a residential structure, perhaps the original Lawrence house? Or was it the "orchard house" described in the Mercury article? Historical archaeology would hold the key to evaluating this problem.

Historically, greenhouses first appeared in America in 1764. Subsequently, throughout the 19th century, the construction of these houses steadily rose, with a particular surge in popularity after 1850. The historical dating, in accord with the Mercury article, would place Lawrence’s "orchard house" at the forefront of this luxury craze.

The goal of the excavation was to obtain a greater understanding of the function of the site. The collection of a substantial artifact assemblage was thought to be the best way to understand the questionable role of the structure, and in order to accomplish this goal effectively and efficiently, a series of 23 shovel test pits were dug on a 50 ft grid in front of the SRU Library (Fig. 5).

Based upon the preliminary analysis of test pit data, especially the presence/absence of natural subsoil, the outline of the building was established and artifact concentrations and possinle features were identified. Four excavation units were laid out to further expose potential features.  One of these units exposed a thick concrete pad, possibly the foundation for a furnace used to heat the orchard house (Fig. 6) 

Upon comparing the types of ceramics found at the site, it became apparent that we were not dealing with a typical domestic artifact assemblage.  It was found that the bulk of the assemblage (70%) was made up of unglazed redware (Fig. 7). Redware was the typical material used for planting, and the forms were consistent with 'flower pot' forms (Fig. 8).

The relative frequencies of glass fragments also provided a distinctive pattern.  The great majority was of flat window glass, consistent with the description of the orchard house (Fig. 9).  But there was also an unusually high proportion of colored bottle glass associated with wine bottles (Fig. 10).  Also present were fragments of stemmed wine goblets.

Excavations on the campus of Salve Regina University identified the architectural footprint of an anomalous building that appeared on historic maps dating to the mid/late 19th century, and was associated with the estate of William Beech Lawrence.  Artifacts recovered from these excavations included a high proportion of unglazed redware interepreted as planting pots, and a high frequency of window glass.  These material classes are consistent with the historical description of the Lawrence Orchard House.  An unexpected insight onto the social activities that may have been associated with the orchard house came via the wine bottle and goblet fragments.  Did dinner parties include a visit to the orchard house to show off the exotic plants and their all-season fruit crops?  This possibility adds a new dimension to the concept of 'conspicuous consumption.'

In collaboration with Dr. Sergio Suarez Cruz of the INAH's Centro Regional de Puebla, mapping and limited excavation on the northeast platform of the Great Pyramid at an elite residential structure first excavated by Eduardo Noguera  in the 1930s. A sequence of six construction stages were identified around a miniature pyramid altar that contained two adult skeletons plus grave goods.The facade of the altar was decorated with stucco skulls. The successive construction stages included earlier staircases and a previous miniature altar that was partially dismantled and covered over for subsequent construction. Ceramics from the different construction stages were seriated to indicate changes from the Late Classic period into the Epiclassic (ca. 600-800 CE). Later stages included types including Cocoyotla Black on Natural and stamp-bottom bowls. These data indicate that elite context was occupied during a time period when traditional culture history of Cholula interpreted abandonment. The platform where the structure is located is depicted in the Historia Tolteca-Chichimeca as the palace of the Aquiach (high priest) Amapane of the Olmeca-Xicallanca when the Tolteca-Chichimeca migrated into Cholula, ca. 1200 CE.

In collaboration with Dr. Sergio Suarez Cruz of the INAH's Centro Regional de Puebla, an excavation of a Classic period residence and associated features. Excavated features included a plaster floor, adobe walls, and a sub-floor burial crypt. Four C14  dates placed the house occupation between 400-600 CE. Ceramics and figurines were consistent with that date, and indicated cultural similarities with Teotihuacan. Associated features included a midden deposit with a high concentration of green obsidian debitage, and another midden with early Colonial ceramics and fauna. A high concentration of comal and small olla fragments suggested a non-domestic function for the midden, possible from a tavern because of the specialized consumption pattern. A large fragment of a San Pedro Polished Black on Red Incised olla featured an incised feathered serpent pattern, which I interpret as a form of native resistance to Spanish/Catholic colonization.

When I was a graduate student at SUNY Binghamton I worked for the Public Archaeology Facility, a Cultural Resource Management firm affiliated with the university. I directed several large projects, including a highway department survey in Alpine, New York. Under CRM legislation, archaeological survey must be done prior to federally funded construction. As it happened, an old bridge over Swan Hill Creek had to be replaced, so a cultural resource survey was required. This resulted in two seasons of archaeological research, including excavation at several site loci.

The Swan Hill Creek bridge spans the seasonal stream of the same name on a small road leading out of the town and into mountainous state forest to the east. The town of Alpine in the late 1980s was a quiet community of about one hundred inhabitants, living largely in house trailers. Historical research, however, indicated that Alpine had been a booming mill town in the mid- to late-19th century. This project, then, was aimed at documenting the rise and decline of a rural industrial center.

Alpine was first settled in the early 1800s, and first appears on the Tompkins County map of 1835. By 1857, the Beers' map shows that Alpine was home for about two dozen homes and several mills (Fig. 1). The town continued to boom, as indicated on the 1874 Schuyler County map (Fig. 2). In addition to the mills, a hotel and several businesses were also indicated, including blacksmith shops. The end of the boom was precipitated by the exhaustion of local forests, and then when the railroad by-passed Alpine in the late 19th century; this is seen on the USGS map from the turn of the century (Fig. 3).

Archaeological Investigations

In order to identify and evaluate possible archaeological resources, an initial program of shovel testing was done around the bridge over Swan Hill Creek. The terrain on the east side of the creek was heavily overgrown, with minimal surface visibility, whereas the west side of the creek included maintained yard lots. Shovel testing identified multiple artifact concentrations, labelled 'loci".

Locus 1 was located in the northeast quadrant of the project area, and consisted of late-19th century artifacts, butchered animal bones, and architectural rubble. Excavations identified intact, dry masonry wall foundations (Fig 4). No structures are identifiable for this locale on any of the historical maps, but the Wixsom photographic archive depicts a wooden structure on this property. Based on the material culture, Locus 1 is interpreted as a short-lived residence.

Locus 2 was located in the northwest quadrant of the project area, along the main street of the town. Based on the 1857 map, it probably corresponds to the S & G Mill. There were large, well built foundation walls (Fig. 5), and a thick concentration of industrial refuse deposit (Fig. 6). A well was discovered that had also been filled with refuse, including parts of a Model A automobile, milk cans, and a baby carriage. The refuse consistently dated to the early 20th century, suggesting that the mill may have served as a garbage dump in the final years of the community.

Locus 3 was also located in the northwest quadrant of the project area, on the west side of the road leading across the Swan Hill Creek bridge. Historic maps and images from the Wixsom photographic archive showed this to have been a commercial district, including a blacksmith shop. Excavations discovered the foundations of several adjacent buildings, and especially deposits associated with the blacksmith (Figs. 7 & 8). An area behind the shop featured a thick deposit of coal ash and clinker, mixed with hundreds of snipped horseshoe nails. There was also a round metal object in an area of fire reddened soil that was interpreted as part of the forge. Interestingly, the scrap heap included many metal parts from horse buggies. The timing of the blacksmith shop in the late 19th and early 20th centuries means that there may have been adjustments in action as the owners/smiths shifted from horses to motorized vehicles.

Across the street from Locus 3 was Locus 5, where domestic remains indicated a yard midden probably associated with the residence on L.B. Herrick (1874 map).  Excavations also encountered remains of a small structure that was likely the outbuilding shown on the same map.  The floor was made of split but unfinished logs, where the bark was visible in the side wall of the excavation.  A high abundance of small nails suggest that the small structure may have been a specialized shop, probably of a cobbler.

One final locus received intensive excavation.   Locus 12 was located at the far northeast part of the project area.  A concentration of domestic artifacts indicated another residence, probably that of H. Wood from the 1874 map.  Excavations encountered part of a foundation wall (Fig. 9), and also a stone-lined well (Figs. 10 & 11).

The excavations at historic Alpine provided an interesting insight into a rural community that prospered and then faded over a span of just over 50 years.  Several mills were established along Swan Hill Creek, exploiting the lumber of the nearby forests.  The town built up around that prosperity, including a hotel for visitors, and several commercial ventures.  Of particular interest was the blacksmith shop as it reflected technological and economic factors with the advent of the automotive age.  The final nail in the economic coffin seems to have been a politically motivated decision to by-pass the town with the rail line.  A clue to why this may have occurred was found in the historic photos of the Wixsom Archive, which featured a prominent 'Greenback" pole indicative of the Greenback movement

I spent 10 months in 1977 assisting Bruce Byland, a PhD candidate at Penn State University, in his dissertation research to survey the Tamazulapan valley in the Mixteca Alta region of Oaxaca. The Tamazulapan valley is a quaint, rural area that specialized in raising maize. Settlement pattern survey is done by walking transects over the countryside, using an air photo to document the location and characteristics of archaeological features encountered on the surface. We averaged about 15 km/day, in transects separated by about 30 m. In the process we discovered nearly 300 archaeological sites -- some with monumental architecture and others down to the size of a household. Artifacts were collected and brought back to the lab house in Tamazulapan town. I organized the ceramic analysis, using a well-worn copy of Ronald Spores' book on the archaeology of the nearby Nochixtlan valley. Besides getting me in excellent physical shape from all the walking, the experience helped develop my interest in ceramics and also (again thanks to Spores) my interest in the Mixtec pictorial manuscripts. One highlight of this experience was working with Spores and Byland in the analysis of a 16th century Relacion Geografica map of Tejupan, another town in the valley, that was depicted with European-style town grid, but also architectural glyphs using the Mixtec codex style (Byland published this). 

A bridge replacement project in the Finger Lakes region of New York required a cultural resource survey of the Halseyville historic district.  The existing bridge spans Taughannock Creek  (Fig. 1). This bridge had replaced a covered bridge built in the 19th century (Fig. 2).

Historical research indicated that the area had first been settled in the early 19th century by Nicoll Halsey, a prominent statesman and member of Congress (Fig. 3).  Nicoll Halsey was one of the early advocates of what became the Greek Revival tradition, believing that the democratic foundations of the new republic should be based on the ideals of the Golden Age of Greece, and that artistic and architectural styles based on Greek models would enhance that ideology.   The home that he built  characterized this concept (Figs. 4 & 5), as did other buildings of his self-styled community, including the Halseyville grist mill (Fig. 6). Remains of the grist mill foundation are stil visible from the river (Fig. 7).

Historical maps indicate the growth of the community in the mid-1800s around the Halseyville mill (Fig. 8). Excavations in the lawn in front of the Halsey mansion encountered evidence of an earlier log structure consistent with historical accounts of the initial settlement in 1807 (Figs. 9 & 10). North of the Halsey house, overburden from bridge construction covered other remains of Halseyville. A trench through the overburden (Fig. 11) exposed foundation remains of some of these buildings, including the Halseyville store (Fig. 12). Artifacts associated with the store included both objects associated with the historical settlement as well as some objects more typically associated with Indigenous cultural practices, suggesting an integrated social network in this frontier community.