This study evaluated the relative association of socioeconomic, minority group and housing characteristics of census tracts with the racial composition of residential areas within one southwestern city between 1950 and 1980. The unit of analysis was the census tract; the data were taken from the U.S. Census of Population and Housing 1950-1980 for the Fort Worth, Texas SMSAs. The Index of Dissimilarity compared racial segregation in the Fort Worth urbanized area for blacks with all others (1950-1980) and for Spanish and non-black minorities with all others (1960-1980). The data show little change in the extent of residential segregation over 30 years. The multiple regression showed that the degree of segregation in census tracts became increasingly predictable based on past minority concentration in the same neighborhood. Lagged social status and minority group variables significantly predicted the percent of the population that was black or Spanish in census tracts ten years later. Beta weights for percent black or percent Spanish were always the strongest in each tract regression and largely determined the level of segregation that existed in tracts ten years later. This paper asserts that social status characteristics must approach more equal levels between minority and majority groups before integrated neighborhoods can reasonably be expected. Yet many of these variables are still highly associated with black and Spanish areas. Rising income and improved housing in black census tracts give some basis for believing that in time these variables will narrow sufficiently to give more choice in residential housing. Although Spanish tracts are only about 65% as segregated as black census tracts and although the association of the variables with Spanish residential areas are never as strong as with black census tracts, still, with increased Spanish immigration in recent years and the downward trend found in social status factors within areas of Spanish concentration, Spanish residential areas may face increasing risks of greater segregation. It is recommended that emphasis on educational attainment and occupational training be continued, possibly augmented with civic programs designed to facilitate movement away from dynamic clustering.