Women's Work and the Aztec Economy

Aztec women worked in a variety of contexts and occupations. They cooked and produced cloth. They worked in markets as vendors and administrators. They were healers, priestesses, prostitutes, and judges. They did agricultural work and tended animals. The list of occupations of Aztec women is long and varied. Ethnohistoric sources show that some of the most common occupations were weaving and cooking (Brumfiel Reference Brumfiel, Gero and Conkey1991, Reference Brumfiel, Brumfiel and Feinman2008b; Sahagún Reference Sahagún, Anderson and Dibble1950–Reference Sahagún, Anderson and Dibble1982:bk. 10, ch. 3). Both activities were partly done for household consumption, and sometimes also to sell items in the market. Cloth was an important product, used to fulfill the wide variety of functions of clothing for the family, from providing warmth and protection from the elements, to marking class, ethnicity, and gender (Anawalt Reference Anawalt1981; Brumfiel Reference Brumfiel and Klein2001b). Cloth was also a main form of currency among the Aztecs (Seeger Reference Seeger1978; Villanueva Reference Villanueva1985:20–21). It was rendered in large quantities as tribute to the Aztec king, and later to the Spanish (Brumfiel Reference Brumfiel1996a, Reference Brumfiel and Klein2001b; Villanueva Reference Villanueva1985:21). Women also sold cloth in the market (Villanueva Reference Villanueva1985), underscoring how difficult it can be to separate the domestic economy from the supra-household economy.

More germane to the present article is cooking. Aztec women were expected to “be a good cook, a preparer of good food” (Sahagún Reference Sahagún, Anderson and Dibble1950–Reference Sahagún, Anderson and Dibble1982:bk.10, ch. 3). They spent hours every day grinding maize and cooking a variety of foods for household consumption, for family members to take with them to agricultural fields and to work far from home, and for sale in marketplaces. The dishes prepared by Aztec women varied according to the needs of their household, their proximity to urban markets, and whether they cooked meals for the house, for religious festivities, for sale on the market, or to support men involved in tribute and agricultural labor far from their homes (Brumfiel Reference Brumfiel, Gero and Conkey1991). Women also participated in markets as vendors and administrators, along with men (Brumfiel Reference Brumfiel, Brumfiel and Feinman2008b).

Ceramic vessels were not objects of tribute. Their distribution took place mostly through market exchange, as indicated in ethnohistoric sources (Hodge et al. Reference Hodge, Neff, Blackman and Minc1993:132). The tools used by women involved them in market exchange when they purchased pottery, produce, and other ingredients needed for cooking. The products of women's work may have re-entered the market if they sold food whether for profit or to purchase items they needed to fulfill their tribute obligations (Brumfiel Reference Brumfiel, Gero and Conkey1991). Therefore, women's work was involved in the household economy, the market, and the tribute economy.

Cooking vessels

Aztec women worked with a variety of tools, and in this analysis we focus on ceramic containers for cooking and storage. The wares analyzed include plain and lead-glazed cooking and storage vessels. Plain Ware appears in the literature as “Plain Ware” (Tolstoy Reference Tolstoy1958), “Plainware” (Peters Reference Peters2002), “Plain Orange” (Parsons Reference Parsons1966:125, 133–137), as a plain variant of the “Orange Ware tradition” (Charlton et al. Reference Charlton, Otis Charlton, Fournier García, Kepecs and Alexander2005:58) and “loza anaranjada no decorada” (Fournier Reference Fournier1998:25). In spite of the varied terms used to refer to this ware, and the fact that most of the time authors do not present a description of the ware or types, scholars are rarely confused by these terms, and generally agree that they refer to earthenware with a fine to medium-coarse paste with 15–20 percent inclusions, and a surface color that ranges from orange to buff and brown, depending on the firing atmosphere. Orange is the modal surface color this ware, although other colors, including brown, and cream, are common. Parsons (Reference Parsons1966:133–137) provides a long list of the varied Munsell colors observed in this ware.

Plain ware was mostly made into jars and griddles. Jars are globular in shape with or without a neck (Figure 2). Ethnohistoric sources show that they were used for storing food, for cooking wet stews and tamales, to cook maize drinks, and for fermenting pulque, an alcoholic beverage made from maguey sap. Griddles (Figure 3), commonly known as comals, were flat round cooking vessels, sometimes with an upcurved rim, with a highly burnished upper surface. They were used to cook dry foods, including a variety of tortillas, pinolli (toasted maize), and toasted nuts and vegetables (Brumfiel Reference Brumfiel, Gero and Conkey1991; Fournier Reference Fournier1998). Jars and comals were often used together to cook different dishes in a meal, or even used in different steps of cooking one dish. For example, pots were used in the nixtamalization process, in which maize is soaked in warm water and mineral lime, in preparation for grinding. After soaking, the nixtamal, or soaked maize, could later be used to make tortillas and other dishes made with ground corn and cooked on comals (Biskowski Reference Biskowski2000; Brumfiel Reference Brumfiel, Gero and Conkey1991; Fournier Reference Fournier1998; Katz et al. Reference Katz, Hediger and Valleroy1974; Rodríguez-Alegría Reference Rodríguez-Alegría, Graff and Rodríguez-Alegría2012). Aztec sites typically yield an abundance of comal and jar sherds. We include jars and comals in our sample, giving continuity to Peters’ study of Aztec jars and comals. Colonial contexts in Xaltocan have yielded a total of 4,597 Plain Ware rims, making up 51.90 percent of the total rims from colonial units.

Figure 2.

Plain jar excavated at Xaltocan.

Figure 3.

Plain comals excavated at Xaltocan.

A second ware analyzed as part of this study is lead-glazed earthenware (Figure 4). It was made into a variety of cooking, storage, serving, and decorative forms, including jars, cazuelas or wide basins, bowls, dishes, cups, miniature jars, and a seemingly endless variety of forms and decorative treatments (Sodi Miranda Reference Sodi Miranda1994). The forms of lead-glazed earthenware do not typically include comals. One may also find glazed molcajetes or dishes with striated bottoms used mostly to grind chiles and other vegetables for making the hot sauces that have made Mexican food famous. We include jars, basins, and molcajetes in our sample. Colonial contexts in Xaltocan have yielded 653 lead-glazed earthenware rims, making up 7.37 percent of all rim sherds in colonial units.

Figure 4.

Lead-glazed earthenware sherds excavated at Xaltocan.

Lead glazing was introduced by the Spanish, and it is found in post-Conquest contexts, whether Colonial (1521–1820) or Republican (1821–present) (Charlton Reference Charlton1976; Charlton et al. Reference Charlton, Otis Charlton, Fournier García, Kepecs and Alexander2005; Hernández Sánchez Reference Hernández Sánchez2012). The earliest contexts containing lead-glazed earthenware in Xaltocan have not been dated by radiocarbon, but they include Aztec III and Aztec IV pottery (associated with the sixteenth century) as well as sixteenth and seventeenth century tin enameled ceramics (Rodríguez-Alegría Reference Rodríguez-Alegría2009). We feel confident that the sample represents the material from the Colonial period, including the sixteenth and seventeenth centuries, with little or no modern material.

Sampling and Analysis

The current analysis focuses on 111 cooking and storage vessels excavated in Xaltocan during 2003 and 2007. It is part of a broader study of colonial ceramics in Xaltocan that included 251 pottery samples and three clay samples (Rodríguez-Alegría et al. Reference Rodríguez-Alegría, Millhauser and Stoner2013; Stoner and Glascock Reference Stoner and Glascock2011). The 140 samples not included in this article are either serving vessels of different types or lead-glazed earthenware body sherds that we could not classify as cooking or storage vessels, either because the fragment did not allow a classification by form, or because it was part of a serving vessel. Although all 251 samples form part of the statistics used in analysis, we only discuss the 111 cooking and storage vessel samples.

Our sampling strategy can be best described as a stratified random sample. We selected samples randomly within the types of interest. Sample selection began by identifying excavation contexts that date to the colonial period, based on the absence of modern ceramics and artifacts, as well as the absence of Aztec pottery. Selection of contexts was based on the relative chronology of the material, and not on whether the context was a feature associated with cooking activities. Given that the pottery chosen for the study is known to have been used as cooking pottery and associated with women's activities (e.g., Brumfiel Reference Brumfiel, Gero and Conkey1991; Fournier Reference Fournier1998), the choice of archaeological context would do little to alter our designation of this pottery sample as cookware. Once a context was identified as dating to the colonial period, a random sample of plain and lead-glazed earthenware was chosen for this study within the bag of pottery associated with the context.

The sample was not structured to reflect the proportion of different types in the assemblage. For example, Plain Ware rims make up 51.90 percent of all colonial rims excavated in Xaltocan, but it makes up 73.87 percent of the fragments included in the sample discussed in this article (83 out of 111 samples). Lead-glazed earthenware rims make up 7.37 percent of all rims in colonial contexts excavated in Xaltocan, but 26.12 percent (29 out of 111) of the fragments included in this article. It is not unusual for an NAA sample not to reflect the proportions of different types in an assemblage. Given that the sample was not entirely random the results of this analysis may not be representative of all colonial ceramics at Xaltocan. Nevertheless, the results can inform us of the relative proportions of samples within each ceramic type that were derived from different loci of production.

This sample includes a variety of contexts from excavations directed by Rodríguez-Alegría that targeted colonial occupations in Xaltocan, including PC1 (“PC” stands for Pozo Colonial), PC2, and PC4. These three units consisted mostly of post–1521 layers (Rodríguez-Alegría Reference Rodríguez-Alegría2009, Reference Rodríguez-Alegría2010, Reference Rodríguez-Alegría, Graff and Rodríguez-Alegría2012). The sample also comes partially from excavations co-directed by Brumfiel and Rodríguez-Alegría, including the units designated as Zoc A and Zoc B, where a large colonial trash pit intruded into domestic architecture (Brumfiel and Rodríguez-Alegría Reference Brumfiel and Rodríguez-Alegría2010). A smaller portion of the sample comes from several excavation units excavated by Brumfiel, including Operations G, H, I, T, and Z1 (Brumfiel Reference Brumfiel2005). All of these excavations were designed to sample Aztec households, and the colonial material was found in a variety of contexts. The contexts sampled are either trash pits or fill, but of course, it would be unusual to find large fragments of domestic pottery in primary use contexts. Given that the purpose of the study is to understand trade in Xaltocan in the colonial period, and that the archaeological context would do little to further our interpretation of these pots as cooking pots, the contexts sampled are appropriate for this study.

Production and Trade of Plain Ware before the Spanish Conquest

Peters (Reference Peters2002) did a chemical characterization study of Plain Ware in Xaltocan with the goal of understanding how political changes affected commoners and their work. The ceramic sample included four distinct phases of occupation at the site, each associated with different conditions of political domination (Brumfiel Reference Brumfiel2005). For each phase, she selected 20 comal and 20 jar fragments. The results reveal changes in cooking pot production and trade through time.

Phase 1 (a.d. 900–1100) is the first occupational phase at Xaltocan (Brumfiel Reference Brumfiel2005:349) and it corresponds to the period when Xaltocan was independent and self-sufficient. Population grew slowly during this period and reached approximately 4,200 inhabitants. Local production of maize and other crops thrived, in part due to the construction of 1,000–1,500 ha of chinampas in Lake Xaltocan (Morehart and Eisenberg Reference Morehart and Eisenberg2010). The local economy was diverse and included hunting, salt production, chipped-stone tool production, and ceramic production and exchange (Brumfiel Reference Brumfiel2005; Millhauser Reference Millhauser and Brumfiel2005). During this time period, most of Xaltocan's Plain Ware was made locally: 70 percent of the sample analyzed by Peters was made in Xaltocan, including 14 jar sherds and 14 comal sherds (Peters Reference Peters2002:152) (Table 1). The people of Xaltocan also used Plain Ware from Cuauhtitlan, west of Xaltocan, which accounts for 12.5 percent of the sample, including two comals and three jars. The loci of production of the rest of the sample from Phase 1 could not be identified.

Table 1.

Provenance of Plain Ware samples excavated in Xaltocan, divided by occupational phase. All percentages are row percentages. Data from Phases 1–4 from Peters Reference Peters2002

The pattern from Phase 1 reflects a greater reliance on local production of Plain Ware when compared to decorated serving vessels. In general, the people of Xaltocan made less than 50 percent of their decorated serving vessels during Phase 1 (Nichols et al. Reference Nichols, Brumfiel, Neff, Hodge, Charlton and Glascock2002:65), but made 70 percent of their plain cooking pots, indicating more trade in decorated serving vessels than in plain cookwares.

Phase 2 (ca. a.d. 1100–1300) shows much continuity in terms of population and the economy in Xaltocan. The economy remained as diverse during Phase 2 as it had been during Phase 1 (Brumfiel Reference Brumfiel2005:353). Beginning in Phase 2 there is a decrease in the percentage of locally produced Plain Ware in the sample studied by Peters, and also a decrease in reliance on Plain Ware from Cuauhtitlan: 60 percent of the sample is of local production (10 comals and 16 jars), and only 5 percent of the sample (two jar sherds) was produced in Cuauhtitlan. The rest of the sample consists of ten comals and four jars of unknown provenience (Peters Reference Peters2002). That comals were traded more than jars may relate to the easier transport of comals stacked on one another in canoes. This may also indicate that there was less interest in trading jars as containers for food in comparison with trade of comals as cooking vessels.

Chemical characterization of decorated serving vessels reveals that in Phase 2, six out of 17 Black-on-Orange sherds (including Aztec I and Aztec II pottery) were made in Xaltocan, a decrease in comparison with Phase 1. Most of the Red Ware analyzed by Nichols and colleagues (2002) came from the southern Basin of Mexico. Thus, the data indicate a trend of decreased reliance on local production of both, plain cooking pottery and decorated serving vessels between Phases 1 and 2 in Xaltocan.

In Phase 3 (a.d.1300–1430), Xaltocan's population grew and the site reached its peak in terms of population, economic prosperity, and political influence. Chinampa agriculture thrived (Morehart Reference Morehart2012a; Morehart and Eisenberg Reference Morehart and Eisenberg2010) and salt production expanded (Brumfiel Reference Brumfiel2005). The people of Xaltocan continued producing chipped-stone tools (Millhauser Reference Millhauser and Brumfiel2005) and other goods (Brumfiel Reference Brumfiel2005:359). During this time period, Brumfiel (Reference Brumfiel, Gero and Conkey1991) noted an important shift in the use of cooking vessels in Xaltocan. Comals became more common than in previous phases, suggesting that the women of Xaltocan cooked more dry portable foods (including tortillas and pinolli). The increase in production of dry portable food was likely to support agricultural labor in chinampas that were located far from the town's center, work in markets, and military campaigns (Brumfiel Reference Brumfiel, Gero and Conkey1991, Reference Brumfiel2005:359). During Phase 3 Xaltocan entered a long war with neighboring Cuauhtitlan. The people of Xaltocan lost the war and they were later incorporated into the Aztec empire in 1428 (Brumfiel Reference Brumfiel2005; Morehart Reference Morehart2012b).

During Phase 3, Plain Ware from Cuauhtitlan makes up 45 percent of the sample, whereas Plain Ware made in Xaltocan makes up only 35 percent of the sample. This is the first time period in which imported sherds from Cuauhtitlan outnumber sherds from Xaltocan. Of the Plain Ware samples produced in Xaltocan, nine are comals and five are jars. Of the samples produced in Cuauhtitlan, eight are comals and 10 are jars. Provenance could not be determined for the rest of the sample. The results of the study of Plain Ware are similar to the results of an analysis of decorated serving vessels. During Phase 3, 14 out of 17 Black-on-Orange sherds analyzed were from Cuauhtitlan, two from Tenochtitlan/Culhuacan, and one from Chalco (Nichols et al. Reference Nichols, Brumfiel, Neff, Hodge, Charlton and Glascock2002:67). Patterns for Red Ware were quite different. Half of the samples analyzed by Nichols et al. (Reference Nichols, Brumfiel, Neff, Hodge, Charlton and Glascock2002) came from Xaltocan, and the rest came from the southern Basin and Tenochtitlan, but not from Cuauhtitlan. The increase in Plain Ware and Black-on-Orange pottery from Cuauhtitlan is most likely an indication that trade and perhaps economic competition between Xaltocan and Cuauhtitlan increased through time. The two towns may have competed with each other until friction was enough to precipitate warfare between the two towns (Brumfiel Reference Brumfiel2005:361).

During Phase 4 (1430–1521), when Xaltocan was under Aztec domination, the economy of Xaltocan declined, the site was smaller than during Phase 3, and the population was impoverished. Economic activity and standards of living suffered during Phase 4 (Brumfiel Reference Brumfiel2005:361). Under Aztec domination, there was a significant increase in the use of Plain Ware from Cuauhtitlan. During this phase, locally produced Plain Ware makes up only 15 percent of the sample analyzed by Peters (3 out of 20 comals, and 3 out of 20 jars), and Plain Ware from Cuauhtitlan makes up 70 percent of the sample (14 out of 20 comals and 14 out of 20 jars; Peters Reference Peters2002). The provenience for the rest of the sample could not be determined. There were changes in the provenience of decorated serving vessels too: the source of Black-on-Orange shifted to Tenochtitlan/Culhuacan (eight out of 14 samples from Phase 4), and only three samples came from Cuauhtitlan. Red Ware, however, was mostly made locally. Eleven out of 16 samples analyzed were from Xaltocan, and only one was from the southern basin, whereas the other four remain unassigned (Nichols et al. Reference Nichols, Brumfiel, Neff, Hodge, Charlton and Glascock2002:68).

The increase in Plain Ware from Culhuacan is perhaps a reflection of more time spent in markets, especially the market at Culhuacan, selling food and obtaining the tools needed to cook more. Perhaps participation in the market at Culhuacan was forced directly or indirectly by the Aztecs, who depended on market participation to ensure movement of goods across the Basin and into their city. Or perhaps participation in the market at Culhuacan increased because the conquest of Xaltocan by the Aztecs dealt the town a severe economic blow and it may have obliterated the market in Xaltocan, and strongly affected local production of ceramics. It is unlikely that an increase in importation of ceramics is related to increases in agricultural production during Phase 4 for two reasons. First, Morehart (Reference Morehart2012a) has argued that the chinampas of Xaltocan were mostly abandoned after the conquest by Cuauhtitlan. We believe it is unlikely that agricultural production increased after that time. And second, people can schedule pottery production around agricultural work. There is no reason to believe that one would have competed with the other in Xaltocan. Regardless of the specific causes, the trend of an increase in use of Plain Ware from Cuauhtitlan through time, and a decrease in use of Plain Ware from Xaltocan is clear in the data.

Production and Trade of Cooking Vessels in Post–1521 Xaltocan

Xaltocan was conquered by the Spanish and burned to the ground in 1521 (Cortés Reference Cortés1970:118; Hassig Reference Hassig2006:142–143). At the time of the Conquest and during most of the sixteenth century, it was an impoverished town due to severe tribute demands from the Aztecs and to the attack by the Spanish (Brumfiel Reference Brumfiel2005:365; Montúfar Reference Montúfar1897:91–96; Rodríguez-Alegría Reference Rodríguez-Alegría2010, Reference Rodríguez-Alegría2016). Xaltocan became an encomienda, and provided tribute to the Spanish for a long time after the conquest (Bejínez Juárez Reference Bejines Juárez1999:97; Hicks Reference Hicks and Brumfiel2005). Based on historical documents, Gibson (Reference Gibson1964:366) has argued that Xaltocan's economy recovered during the Colonial period. The people of Xaltocan took advantage of new opportunities found in the colonial market and they exploited lake resources for commercial purposes. Participation in colonial markets may have led to a period of prosperity in the town, although archaeological evidence does not support the idea that Xaltocan was prosperous for a long time. A better interpretation of all the available data is that Xaltocan went through cycles of economic well-being and impoverishment during the colonial period (Rodríguez-Alegría Reference Rodríguez-Alegría2016; Rodríguez-Alegría et al. Reference Rodríguez-Alegría, Millhauser and Stoner2013).

After the Spanish conquest, there were changes once again in the production and trade of cooking vessels at Xaltocan. The 111 cooking vessel samples submitted for analysis were classified into seven compositional groups, belonging to four sites and one general geographical area: 32 samples were made at Xaltocan (3 belong in the Xaltocan 1a group, and 29 in the Xaltocan 1b group), 19 were made at Cuauhtitlan, six were made at Otumba, three at Tenochtitlan, and two in the southern Basin of Mexico (1 in the Southern Basin-1 group, and 1 in the Southern Basin-3 group) (Figures 5, 6, 7). We note here that all samples in the Basin of Mexico assigned to the Xaltocan 1b chemical group were sampled from Xaltocan. It has not yet been identified outside the site. A total of 47 samples remain unassigned due to a relatively equal probability of membership in more than one of the compositional groups, and our conservative approach of group assignment. Unassigned specimens are most likely of local (Xaltocan) origin, as discussed below. In general, the colonial pattern at Xaltocan is of increased proportion of locally made Plain Ware compared to the proportion of Plain Ware from Cuauhtitlan, reversing the pattern seen during Phase 4. Most of the ceramics analyzed were produced in the northern Basin of Mexico. The patterns become more meaningful when examined by ware and form. Plain comals from the colonial period were made mostly at Xaltocan. Out of 43 comal samples analyzed, 19 were made at Xaltocan and 20 other samples remain unassigned but we believe they are of local origin because they show a high and relatively equal probability of membership in both Xaltocan chemical groups. If our assignment of these 20 comals is correct, that would make it 39 out of 43 comals made in Xaltocan. Only four of the comals analyzed were made at Cuauhtitlan. We analyzed 31 jar samples, and of those, eight belong compositionally in the Xaltocan groups, four were made at Cuauhtitlan, and two were brought from the southern Basin. Seventeen of the plain jar samples remain unassigned, and just as was the case with comal samples, they show a high but relatively equal probability of membership in both Xaltocan compositional groups. They are most likely of local production. Finally, we also sampled eight plain cazuelas, or wide basins with open necks. The plain cazuelas were made locally (n = 3), at Tenochtitlan (n = 1), and at Cuauhtitlan (n = 1), and three samples remain unassigned. The higher proportion of the sample that was made locally in comparison with imported Plain Ware during the colonial period is clear in the data for the three vessel shapes sampled, whether comals, jars, or cazuelas.

Figure 5.

Canonical discriminant (CD) functions 1 and 2 of colonial ceramics from Xaltocan analyzed in this study. Ellipses represent 90 percent probability of membership in the named chemical groups.

Figure 6.

Scatterplot of Co (ppm) and Na (ppm) of colonial ceramics from Xaltocan analyzed in this study. Ellipses represent 90 percent probability of group membership.

Figure 7.

Scatterplot of Fe (ppm) and Co (ppm) of colonial ceramics from Xaltocan. Ellipses represent 90 percent probability of group membership.

The lead-glazed earthenware sample does not contain any comals, given that comals were not typically glazed, but it contains a small sample (n = 6) of molcajetes. Of the 29 samples analyzed, 16 were jars. They were made mostly at Cuauhtitlan (n = 7), and to a lesser extent at Xaltocan (n = 1), Tenochtitlan (n = 1), and Otumba (n = 1) (Table 2). Six of the glazed jars remain unassigned (discussed below). The sample of glazed cazuelas is too small to present any definite patterns, and we assigned one sample each to Xaltocan and Tenochtitlan, two to Otumba, and two to the chemically variable and distinctive Xaltocan D group, and one is unassigned. Finally, the molcajetes were made at Otumba (n = 3) and Cuauhtitlan (n = 3). None of the molcajete samples is of local production, which could indicate that it was not a form that was produced locally, or that the sample is too small to find them. Larger samples of molcajetes are needed to clarify whether they were made in Xaltocan at all.

Table 2.

Provenance of lead-glazed earthenware samples excavated in Xaltocan, by form. All percentages are row percentages

A major finding from this study is that there was production of lead-glazed earthenware at Xaltocan. If we take into account all forms, and not just cooking and storage vessels, the total number of samples of lead-glazed earthenware made at Xaltocan is four out of a total of 76 samples analyzed (Rodríguez-Alegría et al. Reference Rodríguez-Alegría, Millhauser and Stoner2013). We did not have any evidence of local production of lead-glazed earthenware prior to this study. Future excavations may enable us to locate the places where lead-glazed earthenware was produced, whether by identifying production facilities, raw materials, wasters, or other kinds of evidence. Production of lead-glazed earthenware required the usual raw materials used to make Plain Ware (clay, tempering materials, tools), as well as the ingredients necessary for glazing, primarily lead ore. The people of Xaltocan must have obtained these additional materials in markets, although we currently do not know where. Still, local production implied trade in additional raw materials, and also knowledge of the glazing process.

Lead-glazed earthenware came mostly from other towns, including Cuauhtitlan and Otumba. During Phase 4, as well as in the colonial period, Xaltocan did not import Plain Ware from Otumba, but it imported glazed pottery in the Colonial period. This pattern shows that Plain Ware and lead-glazed earthenware were not necessarily traded in tandem.

The two samples of lead-glazed earthenware that belong in the Xaltocan-D group are difficult to interpret. The group is named provisionally as Xaltocan-D until further research can help identify whether it belongs to a specific provenance or whether it is the result of contamination, differential leaching or weathering of these sherds, or some other taphonomic or behavioral process. The group has very high concentrations of REEs, but at the same time, the specimens in the group have a rather heterogeneous REE composition. Whereas some samples have REE concentrations similar to other samples in the study, other samples in the Xaltocan-D group have values that are four or even five times higher than those of samples that belong in other chemical groups. Furthermore, Xaltocan-D samples have the lowest concentrations of calcium and strontium in the study, and they have low concentrations of transition metals as well (Rodríguez-Alegría et al. Reference Rodríguez-Alegría, Millhauser and Stoner2013).

Lead glazing was one of the most widely adopted ceramic technologies in Mexico after the Spanish conquest (Fournier and Blackman Reference Fournier and Blackman2007; Hernández Sánchez Reference Hernández Sánchez2012:143). Given that an unknown number of towns adopted this technology, it is no surprise to find that seven of the lead-glazed samples remain unassigned, which is not an unusually high percentage of unassigned samples in chemical characterization studies anyway. The number of unassigned samples could mean that we have samples from a variety of production loci that remain unidentified, or that a limited number of production centers made these samples with clays that were obtained from different places and show a range of compositional differences as a result (e.g., Arnold Reference Arnold2008:226). Future studies may help clarify this issue.