Archaeologists have studied buildings for decades, but a very narrow range of analytical approaches has been applied to prehistoric architecture. This paper presents a basic theory of architectural meaning, which has as its principal focus the role of access patterns within buildings as a medium of social control. On the north coast of Peru, Andeanists have held an implicit notion about the relation between access patterns and social control, but have lacked the analytical tools to explore this question further. A set of basic tools borrowed from graph theory and locational geography is presented, and discussed, and then applied to a specific set of Prehispanic constructions—the monumental compounds of Chan Chan, the capital of the Chimu state (A. D. 900-1470). The methods illuminate differences and similarities in the organization of access within the compounds, and they are used to test a hypothesis about the role of a specific architectural element—the U-shaped room—in controlling access within royal compounds at Chan Chan. The results have implications for understanding the nature of social control in the Chimu state and suggest the potential that new analytical methods may have for the study of prehistoric constructions.

Clothing styles, design motifs, and techniques of cloth production found in codex illustrations and on pottery and extant textile fragments suggest diffusion of culture traits from the northern coast of South America to West Mexico and on into the American Southwest. The non-mesoamerican garments depicted in a West Mexican sixteenth-century manuscript and on mortuary figurines buried more than 1,000 years earlier in an adjacent area find analogs only in styles that were present in Ecuador from 1500 B. C. up to the time of Spanish contact. Clothing and textile design motifs represented on figures found in the West Mexican shaft tombs of Ixtlán del Río, Nayarit, indicate that these parallels existed as early as 400 B. C. A variety of other data suggest that intermittent maritime contact persisted between Ecuador and West Mexico through the intervening period and into the sixteenth century.

The technique of obsidian-hydration dating contains great potentials for error, from both laboratory determinations of rate constants and measurements of effective hydration temperatures (EHTs) in the field. The rate constants used to determine these dates are of questionable validity and need to be independently verified. Significantly, no measurements of EHTs were taken at the site of Copán, Honduras, until the majority of the obsidianhydration dates were calculated. An error of but a few Kelvins in estimated EHT can lead to dates that are in error by several centuries. In view of the likelihood of large errors in the Copán obsidian dates, the assertion that the Late Classic Coner phase should be extended beyond A. D. 900 (Webster and Freter 1990a) is premature.

Some prominent Mesoamericanists long considered lowland central Veracruz to have been agriculturally unproductive prior to the Totonacs' construction of a canal-irrigation system at Zempoala during the Middle Postclassic period (A. D. 1200-1400). This evaluation reflects a long-standing negative predisposition toward tropical lowlands and a preoccupation with the significance of canal irrigation in the emergence of urban societies. However, an appreciation of mesoenvironments and their ecological interrelations has led to a reevaluation of agriculture in central Veracruz. In wetlands to the south of Zempoala, evidence of canals and planting platforms supporting maize cultivation by A. D. 500 demonstrates that people were intensively cultivating that mesoenvironment by the Classic period (A. D. 1-850). Moreover, vestiges of sloping-field terraces occur throughout some 1,000 km2 of piedmont west of those wetlands. Direct evidence for crops and a chronology are still lacking for these latter fields, but ethnohistorical data and plant ecology suggest cotton, maize, agave, and a Prehispanic origin.

Due to the poor conservation of domestic structures in tropical and subtropical environments, the study of households has received little attention from Caribbean archaeologists. However, recent studies have produced good quality household data in the form of post-mold distributions that can be used to address this topic. A method for the definition of houses using the post-mold data is introduced, and three cases from Puerto Rico are used to investigate changes in house shape and size through prehistoric times. These changes are discussed in the context of sociocultural and political trends in Precolumbian social formations.