The concepts underlying numerical taxonomy and factor analysis are examined in relation to the goals and requirements of archaeological classification. Numerical taxonomy, represented in the archaeological literature most commonly by cluster analysis on raw data, is found to be inappropriate for those goals. In contrast, the technique of r-mode factor analysis fits very well with the nature of archaeological data and the general aims of archaeological classification. The results of three experiments are presented which illustrate the drastically different interpretations that can be engendered by the two procedures from the same data.

During the past five years rock painting researchers concluded that a means of cataloging and classifying rock paintings was necessary. Research at the University of Winnipeg has led to the development of a punch card code and card design for Canadian Shield rock paintings that may also be applicable to other forms of pictography within the region. This design went through many stages of development and many problems were encountered by the developers. Despite the problems a set of basic computer-oriented categories for the comparison and correlation of rock paintings has been developed.

In 1974, a sweathouse was discovered in the Postclassic site of Los Cimientos-Chustum, department of El Quiché, Guatemala. It is rectangular, provided with an interior hearth, a drainage canal, and a large water reservoir. It is, thus, quite different from the El Paraiso sweathouse, the only Precolumbian structure of this type known today in the Highlands of Guatemala. The Los Cimientos arrangement is similar, although on a more modest scale, to the Lowlands sweathouses, especially those of Piedras Negras.

Obsidian artifacts from Edzna, Dzibilnocac, and Santa Rosa Xtampak in northern Campeche, Mexico, have been analyzed by means of X-ray fluorescence in order to determine their trace element composition. The results are compared to the trace element composition of obsidian sources in Guatemala and Mexico. It is postulated that the obsidian from the three sites in northern Campeche came from highland Guatemala, mostly from El Chayal. Our data are compared to the published data of other archaeological sites in Mesoamerica from which obsidian samples have been analyzed. Finally, obsidian transport routes are postulated between northern Campeche, Mexico, and highland Guatemala.

Archaeologists dealing with large faunal samples have all faced the problem of recording and retrieving masses of data regarding animal bones. This paper presents a comprehensive computer coding system for all vertebrate skeletal parts. The system utilizes alphabetic abbreviations that can be automatically translated into a fully hierarchic numerical code. The format is readily learned and can be used in conjunction with standard canned computer programs available at many institutions.

Plants metabolize carbon dioxide photosynthetically either through a 3-carbon (Calvin) or 4-carbon pathway. Most plants are of the C-3 type; C-4 plants are primarily grasses adapted to hot, arid environments. Since C-4 plants have a higher 13C/12C ratio than C-3 plants, animals and humans with a significant C-4 plant food-intake will have higher 13C/12C ratios as well. Maize is a C-4 plant, hence maize cultivators living in predominantly C-3 plant environments should show significant isotopic differences from local hunter-gatherers in their skeletal remains; the importance of maize in their diet should also be measurable. The practicability of this method is demonstrated for New York State archaeological materials and wider implications are mentioned.

Knapping experiments and examination of flaked stone objects distributed throughout portions of southwestern Imperial Valley and northern Baja California indicate that criteria other than presence or absence of percussion bulbs can be used to support the artifactual nature of certain fractured stones.

Developments in environmental quality testing have revealed the need for a reappraisal of the methods of phosphate determination employed by archaeologists. The results of such a reappraisal are presented with recommendations for the implementation of a new technique of quantitative phosphate determination called sequential fractionation. With this technique three discrete fractions are determined by differential solubility criteria. These fractions closely approximate in amount the major types of inorganic phosphate known to be retained by soils. Sample analyses are presented which indicate that the method can be employed to distinguish between natural and human deposited phosphate and to identify features.

Weakly bound composite materials such as low-fired pottery, bricks, mortar, and indurated soils can be disaggregated by ultrasound with no significant chemical or physical alteration of individual component grains. The components may then be separated by size, shape, density, magnetic properties, etc. This allows mineralogical, bulk chemical, trace-element, thermoluminescence, or other analyses of individual separates as well as determination of their relative proportions in the composite.

A reexamination of all the published reports on split-twig figurine sites in light of the excavation of Cowboy Cave where 20 whole and fragmentary specimens were recovered under good stratigraphic control provides a foundation for discussing the possible cultural affiliation of the Grand Canyon figurine complex. A hypothesis is presented which may account for the known distribution of figurine finds and their lack of cultural associations at some sites.

The Early Formative period divisions proposed by Meggers, Evans, and Estrada for the coast of Ecuador were incorrectly shown in a recent paper by Myers; the correct version is supplied.

Dissection of certain animal specimens indicated that meat poundage figures based on modern processing practices could be unreliably high.