AS consumers choose among cities, they may trade off higher earnings against differences in the consumption of environmental goods. Commuter travel time, crime and air quality, the quality of educational and health facilities, each may involve unpurchased environmental goods. Decisions about -the consumption of such goods are made simultaneously with the choice of city of residence. By examining the compensating earnings differentials relative to differences in environmental goods across cities, one can estimate hedonic prices for each environmental attribute. The hedonic prices may be useful in valuing the benefit of environmental improvement, and as weights in constructing an index of the quality of life, as Tobin and Nordhaus (1972) propose. By relating the index of the quality of life to city size, something may be learned about the effect of urban growth on the quality of life. This method is a substantial improvement over the index of Liu (1975). Previous efforts to estimate these hedonic prices by Izraeli (1974) and by Hoch and Drake (1974) have been unsatisfying for several reasons. First, they model only the consumer side, while ignoring the possibility that differences in productivity may influence wage determination. Thus, they do not indicate the conditions necessary for their equations to be identified. Second, they do not include variables representative of a wide range of environmental attributes. Exclusion of important categories of environmental attributes may unduly bias the estimates. Hoch and Drake (1974), for example, focus on climate while ignoring many other environmental attributes. Third, if one includes a wide array of environmental attributes, one is confronted with a serious multicollinearity problem. Kelley (1977) takes explicit account of the demand side of the labor market in estimating hedonic prices for amenities. His analysis, however, does not account for differences in the cost of living in different cities, apparently assuming that all goods are traded in national markets. In addition, Kelley does not account for differences in labor force quality in different cities. This essay considers the production activity of cities in identifying the supply of labor. A rich set of environmental attributes is then used to develop new estimates of the hedonic prices for urban amenities.