A poverty line helps focus the attention of policymakers on the living conditions of the poor and may inform policy decisions about targeting development or poverty programs. Poverty lines serve two roles. One is to determine the minimum living standard before a person is not considered "poor." The other is to make interpersonal comparisons, such as for families of different sizes and compositions, for those living in rural or urban places, or for different dates. Economists give much attention to the second role, but surprisingly little to the first one. The first section of this paper examines how a poverty line can be interpreted within mainstream approaches to welfare measurement in economics. Subsequent sections look at the main methods found in practice. While these methods generally try to implement the objective ideas from economics, they also attempt to surmount some theoretical problems of welfare measurement by drawing on ideas from outside mainstream economics. That information includes data on "capabilities" (activities of a healthy life that a person is able to perform), which fall between the more familiar economic notions of "utility" and "commodities." In addition, measurement practice has sometimes turned to subjective perceptions of what poverty means in a given county. Some methods found in practice make more sense than others; some work in one setting but not others. This paper examines, and attempts to resolve, ongoing debates about poverty measurement, emphasizing those issues with the greatest bearing on policy discussions. (Contains 57 references, 5 figures, and 25 statistical formulas.) (TD)