This paper describes the results of research examining the racial and socioeconomic segregation of public schools in two very different cities, Philadelphia (Pennsylvania) and Houston (Texas). Factors that explain the degree of racial and economic segregation and the consequences of this segregation for student academic achievement were examined. Data for the analyses were derived from census figures, the annual reports of the city school districts, and achievement test scores for the cities. In spite of the many differences between the two cities, there were striking similarities in the composition and character of their school districts. Both provide educational services to higher proportions of minorities and the poor than are observed in their respective metropolitan area populations. Results observed for the impact of neighborhood segregation, parental-student choice, and the development of magnet school programs were very similar. Analysis indicated that overall the most important determinant of academic success for city schools was the proportion of students from low income families. The failure of such schools is the result of a series of characteristics such as withdrawal of resources, diminished teacher commitment, and disinterested families and communities. (Contains 9 tables and 17 references.) (SLD)