Prominent theories converge in suggesting that a wife's resources are positively related to marital dissolution (i.e., the wife's independence hypothesis), whereas a husband's resources ale inversely related to dissolution (i.e., the husband's income hypothesis). Using data from the Panel Study of Income Dynamics (1968-1985), a discrete-time event history analysis identifies modifications required of both hypotheses. First, wife's earnings have a nonlinear, U-shaped relationship to the risk of marital dissolution. Second, the impact of husband's earnings varies as a function of wife's earnings. In particular when the wife has no earnings, lower husband's earnings have a disruptive effect on the marriage. By contrast, when the wife has earnings, lower husband's earnings have a nonsignificant impact on marital dissolution. Finally, results fail to support the hypothesis that better economic prospects for a wife, measured by education and time worked increase the risk of marital dissolution separately from her actual economic standing measured by her earnings.