Prompted by the lack of attention by sociologists and the challenge of materialist explanations of warfare in “precivilized” societies posed by Keeley (1996), this paper tests and finds support for two materialist hypotheses concerning the likelihood of warfare in preindustrial societies: specifically, that, as argued by ecological-evolutionary theory, dominant mode of subsistence is systematically related to rates of warfare; and that, within some levels of technological development, higher levels of “population pressure” are associated with a greater likelihood of warfare. Using warfare measures developed by Ember and Ember (1995), measures of subsistence technology originally developed by Lenski (1966, 1970), and the standard sample of societies developed by Murdock and White (1969), this study finds evidence that warfare is more likely in advanced horticultural and agrarian societies than it is in hunting-and-gathering and simple horticultural societies, and that it is also more likely in hunting-and-gathering and agrarian societies that have above-average population densities. These findings offer substantial support for ecological-evolutionary theory and qualified but intriguing support for “population pressure” as explanations of cross-cultural variation in the likelihood of warfare.