The chipped stone industry of Cihuatan and Santa Maria, Postclassic sites of north-central El Salvador, is discussed briefly in terms of behavioral/technological typology. The industry was based almost exclusively on obsidian and emphasized specialized production of prismatic blades. Source determinations by X-rayfluorescence and neutron activation analysis of 20 obsidian specimens from Cihuatan indicate that at least three highland Guatemalan sources supplied obsidian to the site. Twelve specimens are attributed to the Ixtepeque source, seven to El Chayal, and one to San Martin Jilotepeque. The multiple-source procurement pattern is interpreted as a hedge against the fragility of Postclassic sociopolitical alliances and shifting exchange networks. One of the most important advances in Mesoamerican lithic studies of recent years has been the development of behavioral and technological typologies based on linear reduction models of manufacturing trajectories (Clark 1979, 1981 la; Clark and Lee 1979; Moholy-Nagy et al. 1984; Sheets 1972, 1975a, 1978a). These typologies focus on the behavior involved in the manufacture, use, and recycling of lithic artifacts, thus attempting to model the real technological structure of lithic industries. Behavioral/technological typologies are now available for about 20 sites in the Maya area and the southeastern periphery of Mesoamerica (Fowler 1985). Another significant advance has been the development of techniques for accurate, high-precision chemical characterization of obsidian as a means of source determination (Stross et al. 1968; Stross, 1976). The results of precise geochemical analyses to determine the sources of archaeological obsidian in Mesoamerica have led to better understanding of the nature and degree of long-distance exchange in the area. Three factors make obsidian an excellent indicator of long-distance exchange contacts: (1) good-quality, workable obsidian occurs in a relatively limited number of geological deposits; (2) obsidian artifacts deposited in archaeological contexts are virtually imperishable; and (3) it is possible to determine, within limits of statistical certainty, the geological sources of obsidian artifacts through precise geochemical characterization. It is a fair assumption that obsidian exchange is an indicator of the exchange of other items that left no trace in the archaeological record. At the very least, the buyers had to offer something to the sellers in exchange for the obsidian. Whether or not perishable goods moved in appreciable quantities along the same routes of exchange as obsidian (see Drennan 1984), the reconstruction of Precolumbian obsidian exchange in Mesoamerica sheds light on prehistoric interregional contacts and exchange networks in general and provides a foundation for hypotheses concerning ancient sociopolitical ties that operated over long distances. When behavioral/technological typologies and obsidian source determination are used conjunctively, the analyst leams more about the cultural context within which a given lithic industry developed and functioned (Moholy-Nagy et al. 1984). The resulting reconstruction is more detailed and is a more fruitful source of hypotheses than either of the approaches used alone. The chipped stone industry of the ancient settlements of Cihuatan (Bruhns 1980; Fowler 1981; Kelley 1985) and Santa Maria (Fowler 1981), in the Paraiso (or Cerr6n Grande) Basin of north