During the 1960s in Sweden, socioeconomic differentials decreased sharply and both the labor force participation of married women and aggregate divorce rates increased more rapidly than during any other period of the twentieth century. The aim of this paper is to investigate how the socioeconomic composition of the couple influenced the probability of divorce during this period of rapid restructuring. The study uses a large data set covering the entire married population of Sweden in 1960 and applies a binary model whereby the couples are analyzed as units rather than separate individuals to model divorce during the period from 1960 to 1965. The main results show that the equalization process between genders and social classes during this period contributed to the decrease in marital stability. Dual-provider families exhibit substantially higher probabilities of divorce as compared to traditional single-provider families. We also find that the socioeconomic gradient of divorce had become negative by the early 1960s and that couples with low socioeconomic status contributed more to the increase in divorce than did couples in the higher strata. A difference between the results reached in this study and those from divorce research covering later decades is that children do not reduce the probability of divorce when the wife’s labor force participation is controlled for. The results indicate that the determinants of divorce have varied across different phases of the divorce transition during the twentieth century and that a historical perspective is necessary if we are to understand the long-term process that has produced current marital behavior.