This paper examines pottery technology and change through the eyes of the experimental archaeologist. A new vision is presented of experimental archaeology and the role its findings can play in archaeological explanation. It is argued that the most useful results of experimental archaeology are best obtained with long-term research programs. This perspective is illustrated by a case study of the relations between surface treatments (interior and exterior) and thermal performance in cooking pots. The experiments indicate that surface treatments like texturing, organic coatings, and smudging have marked impacts on thermal shock cracking and on thermal spalling in simulated cooking. It is emphasized that the findings of experimental archaeology, expressed as correlates, can be employed in explanations of prehistoric technological change, but only when embedded in more inclusive correlate theories and coupled with the requisite contextual information.

Structural changes in Archaic landscape use are examined using a large-scale survey data set from the dissected forested uplands of southwest Indiana. Spatial patterning of hafted bifaces, as indicated by proximity to streams of different order and distance from major rivers, was found to vary between the Early and Middle Archaic periods. Early Archaic settlement emphasized the upper reaches and divides of basins, but also included near-river locations, while Middle and later Archaic groups focused on locations near basin outlets and major rivers. These changes are argued to be a function of reduced residential mobility and a shift from a largely foraging to a collecting strategy. Increased travel and processing costs associated with a logistical strategy resulted in a deemphasis of more distant extravalley resource patches. Several explanations for this shift in strategies are explored, including possible increased spatiotemporal heterogeneity of the environment associated with postglacial climate and vegetation change.

Ground-stone tools and hunter-gatherer subsistence in late Pleistocene southwest Asia are examined in light of ethnographic and experimental data on processing methods essential for consumption of various plant foods. In general, grinding and pounding appear to be labor-intensive processing methods. In particular, the labor required to make wild cereals edible has been widely underestimated, and wild cereals were unlikely to have been “attractive” to foragers except under stress conditions. Levantine ground-stone tools were probably used for processing diverse plants. The earliest occurrence of deep mortars coincides with the glacial maximum, camp reoccupations, the onset of increasingly territorial foraging, and the earliest presently known significant samples of wild cereals. Two major episodes of intensification in plant-food processing can be identified in the Levant, coinciding respectively with the earliest evidence for sedentism (12,800-11,500 B.P.) and the transition to farming (11,500-9600 B.P.). The latter episode was characterized by rising frequencies of grinding tools relative to pounding tools, and suggests attempts to maximize nutritional returns of plants harvested from the limited territories characteristic of sedentary foraging and early farming. This episode was probably encouraged by the Younger Dryas, when density and storability of foods may have outweighed considerations of processing costs.

This study examines the relation between raw-material procurement and subsistence behavior among foraging societies. “Embedded procurement” of raw materials may characterize many or most modern foraging societies (Binford 1979). Past societies, however, present economic configurations different than those of any contemporary society. The Early Paleoindian societies of North America present extreme examples in this regard, and were characterized by high mobility, low population density, and high weapon reliability. A lithic-debitage analysis of a portion of the Nobles Pond site (33ST357) supports the argument that the acquisition of lithic raw materials was not embedded in subsistence behavior, but rather, was a specialized activity required by the particular demands of band aggregation in a location far removed from sources of acceptable lithic materials.

A paleodietary analysis of the mid-Holocene mortuary site, Windover (8BR246), based on carbon and nitrogen bone-collagen values and archaeobotanical information is consistent with a subsistence strategy that utilized river-dwelling fauna and a range of terrestrial flora, such as grapes and prickly pear. The isotopic analysis does not support the extensive human dietary use of either marine mammals or classic terrestrial fauna such as deer or rabbit. Seasonal (late summer/early fall) use of the site is indicated by the range of flora found in association with the burials.

Artifact collection has become increasingly expensive, especially with respect to curation costs. Consequently, many government agencies as well as researchers have turned to an alternative, “on-site” analysis approach in which many analytical procedures usually reserved for the laboratory are conducted in the field. Using data gathered at an andesite quarry in eastern Nevada, we evaluate such an approach in terms of feasibility, reliability of results, and cost, as well as its impact on the surface archaeological record. Although we still believe that artifact collection is central to archaeological research, we argue that in circumstances where collection is not possible, a well-designed and systematically implemented on-site approach can yield reliable results, although not without some visible impact on the surface archaeological record.

A series of experiments making and using bone and antler tools show that functional identifications of these tools can be made with confidence in some circumstances. Using principles from the field of tribology, the experiments demonstrate that different uses leave different microscopic traces on bone and antler. They also show that when the materials used are similar, the wear produced will be similar. In particular, wet materials, including snow, ice, wet hide, and wet antler all produce nearly identical microscopic patterns. Other groups of similar materials, such as bone, antler, and wood, or fish scales and hair, present the same problem. Although differences can be detected, these may not be preserved on archaeological tool specimens. Application of the experimental results to bone and antler tools from the Mackenzie Delta illustrates that functional identifications of tools can be made with confidence, despite the problem of similar microscopic patterns, when other lines of evidence (ethnographic and historical accounts, distribution of wear) are taken into account. When such information is lacking, functional identifications are more difficult and must be made with more caution.

Red-slipped pottery and a multiterrace platform mound at Trempealeau, Wisconsin, indicate the presence of an early Mississippian outpost in the upper Mississippi Valley ca. A.D. 1000. Trempealeau apparently represents a Mississippian elite site-unit intrusion from the American Bottom, and it probably served as a nodal point of early contact between Cahokia and peoples of the upper Mississippi Valley. By establishing a mound center at Trempealeau, its founders not only secured access to material goods but also facilitated the flow of information from the northern Mississippi Valley to the newly emerged elites in the American Bottom.

Pollen deposited on the ground surface is carried down into the soil by percolating groundwater. Such postdepositional pollen transport preserves the record of historical vegetation and land use in slowly or nonaggrading sediment profiles by separating the pollen spectra of successive ground covers. This is demonstrated at Great Meadows, Pennsylvania, where pollen spectra in hillside cores indicating a preagricultural-era forest are succeeded during the clearance and agricultural period by weed and cereal pollen during the nineteenth and early twentieth centuries. These are, in turn, replaced by grass during the park period of the last 65 years. The preagricultural pollen spectra indicate that differences in historical ground cover across relatively short horizontal distances and elevations can be reconstructed with pollen analysis.