Patterns of relationships among the variables of pregnancy status at marriage, marital dissolution probabilities, residence and migration status, and race were ascertained. The data source was the 1967 Survey of Economic Opportunity, a large national probability sample expanded to U.S. population parameters. Relationships among these variables were rather different for rural to urban migrants than for either the rural population of origin or the urban host population, generally exceeding these 2 populations for marital dissolution probabilities. Tentative explanations were offered for: (1) the lower probability of marital instability among rural women; (2) the finding that, although among women who marry very young (under 18), those premaritally pregnant have more stable marriages than their conventional counterparts, the probabilities of marital dissolution tend to decrease as age at marriage increases; and (3) the striking differences in marital formation and dissolution patterns between Negro and white women (i.e., urban white women had slightly higher proportions with conventional family patterns while rural white women had the smallest percentage postmaritally pregnant; exactly the opposite held for Negro women). (Author/KM)