volve the use ofkeys, comparative collections, and sorting of materials into lots based on color, lus ter, texture, and inclusions (such as fossils, oo lites, and vugs). While Morrow does not apply his system to materials derived from till, Bakken's system is more or less explicitly directed at the examination of till-derived raw materials. Lithic collections may be characterized using two different methods, here termed the macro scopic and microscopic. The macroscopic method is dependent on the use of comparative lithic col lections, on comparison of lithic features visible to the unaided eye, or with the aid of a 1 Ox hand lens. The microscopic method is essentially pet rographic, involving the use of polarized light microscopes and the identification of mineralogi cal components in lithic thin sections. Over the past few decades various petrographic and geochemical descriptions of northern Plains lithic materials have appeared in the literature. Some relevant examples are Porter (1962) for Tongue River silica and Bijou Hills quartzite, Loendorf et al. (1984) for Rainy Buttes silicified wood, Church (1994) for Ogalalla orthoquartzite (formerly Bijou Hills quartzite), Campling (1980) for Swan River chert, and Clayton et al. (1970) for Knife River flint. Others have described or discussed northern Plains lithics in more conventional archaeologi cal terms. Examples are Ahler (1977), Anderson (1978), D. Fredlund (1976), and Low (1996). Bakken (1997) has assembled a review of lithics used in prehistoric Minnesota, and Morrow (1994) provides a key for identification of lithics com monly found in Iowa. While archaeologists are becoming more so phisticated in description and analysis of lithic material, many archaeological studies are done without the benefit of petrographic analysis of thin sections, X-ray Diffraction (XRD), Scanning Elec tron Microscopy (SEM), or other laboratory tech niques. The most useful of these techniques for archaeological identification of raw materials is petrographic description of thin sections. In many cases, optical microscopy using polarized light allows an investigator to identify the minerals that comprise a sample. Details of the textural rela tionships between mineral grains are clearly seen when they are magnified between l(M500x. The abundance of each mineral species present can be estimated, and in many cases, the composition and growth conditions of the rock can be interpreted. For extremely fine-grained rocks, optical micros copy does not provide the necessary magnifica tion, and SEM analysis must be undertaken. A review of the applications of petrography to ar chaeology is provided by Kempe and Harvey (1983). Given these two methods, questions sometime arise regarding the reliability of the macroscopic method. Specifically, are lithic raw material names assigned by archaeologists on the basis of macro scopic examination corroborated by thin section descriptions? For this reason a simple experiment was set up during the analysis of the Rustad lithic materials. The lithic collection was divided into raw material classes macroscopically, following a procedure similar to that described by Bakken (1997) and Morrow (1994). The procedure devel oped out of the recommendations of these research ers is to inspect lithic collections in aggregate, that is, to review large quantities of lithic material al together, noting the range of variation across the collection. As this is done, the rough outlines of a lithic classification are developed. Next, a com parative collection of lithic material is used, along with basic descriptive information about the rock types that appear to be represented. Items from each provenience unit are then placed in one or another of the categories, paying special attention
Read full abstract
Feb 9, 2017
Mariana Sacchi 
Open Access
